Main

April 25, 2008

AccuWeather sees "slightly" more active hurricane season

Yet another crowd of tropical weather forecasters has chimed in with their predictions for the upcoming Atlantic hurricane season. Two weeks ago it was the Colorado State University team of Phil Klotzbach and William Gray, calling for a "well-above average" season.

Now it's AccuWeather.com's hurricane trackers, led by Joe Bastardi, who's expecting only "slightly more storms than average," with increased risk of U.S. landfalls, during the six-month season that opens, officially, on June 1.

For the record, "average" in the Atlantic basin during the period from 1950 to 2000, means 9.6 named storms, with 5.9 of those growing to hurricane force, and 2.3 of those, on average, reaching Category 3, with sustained winds of 111 mph or more.

Bastardi and his crew say both the waning La Nina conditions in the tropical Pacific Ocean, and a continuing warm-water cycle in the Atlantic argue for slightly more activity than the average, and increase the chances for U.S. landfalls.

More specifically, they note that the warm conditions are not uniform across the basin. "In some areas where hurricanes normally form - the central and eastern tropical Atlantic," Bastardi said, "ocean water temperatures are near or below normal. This should limit the number of storms." So don't expect a blowout season like 2005.

But "the warmest waters relative to normal will be in the northern areas of the Atlantic, especially toward the North American continent. This could potentially increase the threat of major landfalls to the U.S. coast."

Where the spread of storm tracks last year shifted southwest, sending a batch of powerful storms across the northern Caribbean, "this year, early indications show that the spread will move north and east, with a target closer to the Southeast U.S."

In their April 9 forecast, Klotzbach and Gray said there would be a 45 percent chance that a Cat. 3 storm or bigger would make landfall somewhere along the east coast, including Florida. The long-term average for the last century is 31 percent per season.

Bastardi and company say the conditions this year most resemble those in 1955, 1996 and 1999. (Klotzbach and Gray agreed on 1999, but also found analogs in 1950, 1989 and 2000.) NOTE: An earlier version of this post, relying on a release from AccuWeather included inaccurate dates. AccuWeather has since corrected its release, and those fixes are reflected here.)

In 1955, Hurricanes Connie and Diane struck North Carolina. And 1999 saw both Hurricanes Floyd and Dennis strike North Carolina. Here's the storm track for Floyd.

NOAA

Hurricane season forecasting is, of course, a very young science. Weather and climate are vast, chaotic systems with more variables than even the most advanced computer models can capture with any certainty. These forecasters do their best with the knowledge they have (or think they have). And their efforts get lots of news play because hurricanes are big threats to life and property. There is always some benefit to alerting the public to the risks we face every year, in the hope we will pay attention, and prepare, when storms are on the move.

NOAA - AndrewWhen seasonal forecasts fall short of perfect, as they have in recent years, there's a risk that the public will scoff and pay less attention to the hazard in the future. That would be a mistake. Even a slow storm season can cough up one or two calamitous storms. See Hurricane Andrew in 1992. That's some of Andrew's aftermath at left.

Next up is the National Oceanic and Atmospheric Administration's forecast, due out next month.

September 25, 2007

Summery for now; Fall arrives Saturday

It was a beautiful morning for watching the International Space Station fly over Baltimore and right past a gleaming Venus. But summer resumes this afternoon, with even hotter, muggier weather tomorrow. This early-autumn reprise of summer will be short-lived however.

The big mound of high pressure that's brought us these gorgeous days (which I've enjoyed through the sealed, double-paned windows of the newsroom) will be packing up and heading out to sea shortly. As it departs, Maryland will find itself on the west side of the clockwise circulation around the high. That will bring our air up from the south. Daytime highs today and tomorrow will rise toward 90 degrees and humidities will climb. It's going to feel a whole lot like summer.

As the high moves away, of course, it will be trailed by an approaching cold front and falling barometer. That will bring us increasing clouds and rising chances for rainfall late Wednesday and Thursday. Forecasters out at Sterling have been boosting their estimates. We're looking at 50 to 60 percent rain chances Thursday - showers and even a possible thunderstorm as the front and daytime heating stirs the air. Obviously, we need the rain, so every drop will be welcome. Right now this September ranks as the fourth-driest on record for Baltimore, with just 0.35 inch of rain on the meter at BWI, and 0.42 inch here at Calvert & Centre.

Once the front goes by, pressure will start to rise again Friday and cold air will pour in from the northwest. Skies will clear, and daytime highs will rise only into the mid-70s for the weekend, with sleeping-weather lows in the 50s. That's much more fall-like,  just about exactly in line with the long-term averages for Baltimore at this time of year. 

September 10, 2007

Hurricane season peaks today

NOAA 

Today (Sept. 10) marks the statistical peak of the Atlantic hurricane season. From here, the likelihood that tropical storms will form begins to decline. The season runs through November.

That's not to say we can't still have a busy few months. On the chart above, there's an interesting spike in activity in Mid-October. In fact, forecasters expect the backside of this season will be very active.

So far, we've had 7 named storms, 2 of which became hurricanes - both of them (Dean and Felix) making landfall as Category 5 storms. That's never happened before.

National Weather Service forecasters expect a total of 13 to 16 named storms before the season ends, of which 7 to 9 will become hurricanes. If they're right, we have 6 or 7 more named storms (and 5 to 7 hurricanes) to go. Here's the full report.

Phil Klotzbach and William Gray, at Colorado State University, have predicted 15 names storms and 7 hurricanes this season. That would leave 8 named storms and 5 hurricanes to go. Here's their webpage. A link to their latest report is at the upper left hand corner of their main page.

For now, Gabrielle is the only active tropical system on the charts. There is a disturbance in the Atlantic that forecasters are watching. It could become a tropical storm in the next several days. If so, it will be named Humberto.

 

September 6, 2007

La Nina coming, with SW drought, SE hurricanes

Federal climatologists say that conditions in the tropical Pacific are shifting toward a full-fledged La Nina event this fall. That could mean a deepening drought in the southwestern United States, and ripening conditions for an active autumn hurricane season.

In a release this morning, the National Oceanic and Atmospheric Administration's Climate Prediction Center said surface waters in the eastern and central Pacific are cooling. The green color in the map, along the equator west of South America show the cooler-than-average sea-surface waters.

"While we can't officially call it a La Nina yet, we expect that this pattern will continue to develop during the next three months, meeting the NOAA definition for a La Nina event later this year," said Mike Halpert, acting deputy director for the Climate Prediction Center, in Camp Springs, Md.

La Nina events typically mean wetter-than-average weather in the Pacific Northwest, and drier-than-average weather in the Southwest. "These conditions also reinforce NOAA's August forecast for an above-normal Atlantic hurricane season," said Gerry Bell, NOAA's lead seasonal hurricane forecaster.

La Nina winters in Baltimore are typically poor snow-producers. But there have been exceptions. Since 1950 there have been 9 winters with moderate-to-strong La Ninas. None produced a storm with 8 inches of snow or more. There were 7 winters with weak La Ninas. They produced only two storms of 8 inches or more. But those were sizable. They included the 22-inch storm in January 1996, and a 15-inch snowfall in January 2000. El Nino winters - when the Pacific is unusually warm - tend to be bigger snow-makers in Baltimore.

La Nina events are declared when average sea surface temperatures in the eastern and central equatorial Pacific, measured over three months, move more than 0.9 degrees Fahrenheit below the average. For months, forecasters have been predicting weak La Nina ocean temperatures for this summer. Those conditions now appear to be developing, with some portions of the east-central Pacific now a degree or two below long-term averages.

September 4, 2007

Is global warming igniting more Cat. 5 storms?

Global warming theory and complex computer climate models predict that a warming ocean will increase the intensity of hurricanes. Both of this year's hurricane to date have made landfall at Category 5 on the Safir-Simpson Scale, the first time that's ever been recorded in the Atlantic basin.

Is there a connection here? Maybe. But it's not as clear as you might think, says Chris Landsea, science and operations officer at the National Hurricane Center.

All the numerical models say we should see increased hurricane intensity as the oceans warm. But Landsea notes that the "sensitivity" of that response is not very impressive - on the order of two percent. In other words, for every one-degree increase in global ocean temperatures, we would expect to see a two-percent increase in hurricane intensity.

Since the tropical oceans are about 1 degree warmer now due to global warming, he said, that means a 160-mph hurricane - the hurricane center's estimate of Felix's top sustained winds at landfall - will instead strike land at 162 mph.

On the other hand, it seems to me, that's an average. Some storms will be more dramatically intensified by warmer oceans, some less so. It's the really bad ones that we worry most about. It's like rising temperatures. The global average is one thing, but the impact is expected to be - and already is - more dramatic in the Arctic and Antarctic. And a seemingly modest rise of a few degrees in the average temperature for Baltimore means some summer days (and some winter days) will be dramatically warmer. And the difference between 95 degrees and 105 degrees on a July afternoon can kill people.

In any case, Landsea said the "real driver" behind the busy hurricane seasons of recent years, and the more numerous Category 5 storms, is a long-term natural cycling of ocean surface temperatures and atmospheric conditions that began in 1995 and is expected to continue for another 10 to 30 years. It's a cycle that looks much like another that began in the 1920s and continued into the 1960s.

"In fact," Landsea said, "there is one year we may have had two Category 5's making landfall."

Those were Hurricane Hilda and Hurricane Janet, in 1955. Both followed the same paths Dean and Felix have this season. "Janet we know was a Category 5," he said. Hilda struck in a sparsely populated region of Central America where measurements were scarce. Hurricane hunter aircraft in those days didn't fly through storms that strong, and there were no satellites to estimate central winds. "We don't have any idea how strong it was," Landsea said. "It could have been a 3, 4 or 5."

We do know it crossed the Yucatan and reintensified in the Gulf of Mexico, becoming a borderline Cat. 3 or 4.

We also know that seasons like this one, in the middle of a multi-decadal upswing in hurricane frequency, and La Nina conditions, the season is "not only busy, but long-lasting," Landsea said. "October and November are quite active as well."

August 21, 2007

Dean was 3rd most intense ever at landfall

Dean at landfall - NOAA

Hurricane Dean had a central barometric pressure of 26.75 inches just before making landfall near the Mexico-Belize border early today. That made it the third-most-intense Atlantic hurricane at landfall since record-keeping began in the 1850s, according to the National Hurricane Center.

The air pressure in the eye of a hurricane is an indicator of a storm's ferocity because it's the low central pressure that powers the winds swirling around the eye, and the storm surge. 

The only storms with lower barometric readings at landfall were the unnamed 1935 Labor Day storm that ravaged the Florida Keys, with a pressure of 26.35 inches; and Hurricane Gilbert, which struck Mexico's Yucatan peninsula in 1988 with a central pressure of 26.22 inches.

The lowest barometric pressure ever recorded in an Atlantic hurricane was 26.05 inches, for Hurricane Wilma while at sea in 2005.

If Dean had made landfall in the United States, it would have ranked as the second-most intense ever to come ashore in the U.S., after the Labor Day storm in 1935. The next-most intense at landfall in the U.S. was Hurricane Camille, a deadly Category 5 storm that tore up the Gulf Coast of Louisiana and Mississippi in 1969. Its pressure at landfall was 26.84 inches.  

In the meantime, a weakened but still deadly Dean continues its trek across the Yucatan. Hurricane warnings are going up along the Mexican Gulf Coast. Here's the latest advisory. Here's the predicted storm track, and here's the satellite view.

 

July 19, 2007

Is dust over Atlantic stifling hurricanes?

The 2007 Atlantic hurricane season continues to be very tranquil. After an early start that spawned two quick storms in late May and early June, the tropics have settled down quite nicely. (Of course, as soon as I note such trends, they reverse. Beware.) Here's all the National Hurricane Center is watching at the moment.

Saharan dust over E. Atlantic - NASA Terra satelliteOne explanation for the failure of the eastern Atlantic to generate tropical storms at this time of year has been the presence of large clouds of Saharan dust in the atmosphere. Some scientists believe such clouds stifle the formation of the kinds of tropical storms that can eventually reach the Caribbean and the U.S. coast. Here's another shot.

Satellite imagery in recent weeks has been documenting exactly such clouds off West Africa. At left is a shot of the eastern Atlantic, taken Monday by NASA's Terra Earth-observing satellite. Some African dust has been tracked as far west as the Caribbean.

Officially, this has been forecast to be an abnormally active hurricane season - not as busy as the record season of 2005, but busier than last year's unexpectedly quiet summer. Here is the latest National Weather Service estimate for the current season.

May 3, 2007

Katrina costliest, 3rd deadliest U.S. hurricane

With the 2007 hurricane season less than a month away, the National Hurricane Center has updated its listing of the costliest and deadliest hurricanes to strike the U.S.

Not surprisingly, perhaps, Katrina, which struck New Orleans and the Gulf Coast in late August 2005, is ranked as the costliest hurricane in U.S. history, with damages exceeding $81 billion. Even after adjustment for inflation, Hurricane Andrew, which trashed South Florida in 1992, still comes in a distant second, at $42 billion.

In terms of deaths, Katrina ranks third, with an estimated 1,500 dead. The death toll from the 1900 hurricane that ravaged Galveston, Tex. remains the highest at some 8,000. (The true number has never been determined.) In second place is a 1928 storm that killed an estimated 2,500 in South Florida and Lake Okeechobee. 

Katrina also ranks third in storm intensity at landfall, with a barometric reading of 27.17 inches.

For the full report, click here.

It's interesting to note that the 1900 and 1928 storms were both Category 4 storms, while Katrina was a weaker Category 3 at landfall. Andrew - the second-ranking storm in terms of storm damage, does not even rank among the top 50 for storm fatalities.

The standings reflect what hurricane scientists have been saying all along - that while modern early-warning and forecasting skills are saving lives, intense development along our vulnerable coastlines is accelerating the property damage totals from comparatively weaker storms.

April 29, 2007

Hurricane Hunters at Martin Airport

Got the afternoon off Wednesday?  Why not drive out to Martin State Aiport and see the National Oceanic and Atmospheric Administration's "Hurricane Hunter" aircraft? The Orion WP3 will be visiting as part of NOAA's effort to call attention to its role in forecasting the track and power of hurricanes as they approach the U.S. coastline.

The airplane will be open to visitors, and local and state emergency services personnel will have displays on site, too. Here's the lowdown from NOAA. And here's more on NOAA's Hurricane Hunter squadron. They're part of the nation's smallest uniformed service, the NOAA Commissioned Corps., based at MacDill Air Force Base, in Tampa. Here are some of their coolest weather photos.

The NOAA Hurricane Hunters are distinct from the Hurricane Hunters of the 53rd Weather Reconaissance Squadron, part of the Air Force Reserve, based at Keesler AFB, in Biloxi, Miss.

April 3, 2007

East Coast hurricane chances rise

Continuing warmth in Atlantic surface waters and the dissipation of El Nino conditions in the Pacific will combine to bring the U.S. a "very active" hurricane season in 2007, according to hurricane forecasters at Colorado State University. Katrina That is, busier than last year, when no hurricanes made U.S. landfall, but not as active as 2004 and 2005, when storms pummeled Florida and devastated New Orleans and the Gulf Coast. (That's 2005's Katrina at left.)

The CSU forecasting team of Phil Klotzbach and William Gray today boosted the numbers they forecast in December for the upcoming Atlantic hurricane season, citing the faster-than-expected disappearance of the El Nino conditions in the Pacific. They're now anticipating 17 named storms between June and November, besting the long-term average of 9.6. Of those 17 storms, the forecasters expect nine will grow to hurricane strength (The average is 5.9.) Five of those will become Category 3 storms or stronger. (The average is 2.3.)

The CSU forecast also predicts a 74 percent chance one of those major hurricanes will strike somewhere along the U.S. Coast, higher than the average of 52 percent for the past century. There is a 50 percent chance one will hit somewhere on the East Coast, including Florida, more than the 31 percent average for the last century. And, there is a 49 percent chance the Gulf Coast will see a storm that big, the forecast states, up from the long-term average of 30 percent.

Here is the full report. Looking back, Klotzbach and Gray fell far short of predicting 2005's record hurricane season, but so did everyone else. Their forecasts in December 2005, April and June 2006 also badly over-estimated the actual storm counts in last year's season. That season fizzled, meteorologists say, because of the unanticipated development of El Nino conditions in the Pacific, and dust storms from North Africa that blew out over the eastern tropical Atlantic.

This year, Klotzbach and Gray anticipate neutral, or weak-to-moderate La Nina conditions in the Pacific, which would support hurricane formation in the Atlantic. The Atlantic basin also continues to experience unusually warm surface waters, which is another factor supporting hurricane formation. The Atlantic Basin has seen unusually active hurricane seasons in most years since 1995 thanks to warm water conditions there, the CSU team says. That trend, part of a long-term Atlantic climate cycle, is likely to continue for decades to come.

Unlike many climate scientists, Klotzbach and Gray say there is insufficient scientific evidence that global warming is playing a role in the recently heightened hurricane activity in the Atlantic. Their reasoning can be found in their current forecast.

March 28, 2007

Dust or El Nino: What killed the 2006 hurricane season?

Even the nation's top hurricane forecasters were caught off-guard when last year's hurricane season - forecast to be another active one - fizzled. The early explanation was that the development of El Nino Mitch conditions in the central and eastern tropical Pacific Ocean stifled hurricane development in the Atlantic. But some scientists are arguing that dust storms blowing off North Africa were to blame.

As the theory goes, that dust wafts off the desert and out over the eastern Atlantic where many hurricanes are born. It absorbed solar heat and cooled the ocean waters beneath it, cutting off the energy needed to spawn new storms.

Not everyone agrees, of course. This is science, after all. Some say El Nino was a bigger factor. Some say it was mostly the dust. You can read about it for yourself. Just click here.

The 2007 Atlantic hurricane season starts June 1, and the El Nino conditions have ended, so forecasters expect a busier season this time around. The National Hurricane Center's official forecast comes out later this spring. Here's the December prediction from Colorado State University's hurricane forecasting team.

June 12, 2006

Video: 27 storms in 5 minutes

NASA's Goddard Space Flight Center has produced a fascinating video from 2005 Atlantic hurricane season data that recaps the record-smashing six-plus months of storms. It features a speeded-up satellite view of the entire season, tracking each storm as it forms over an overheated ocean, steams toward shore or wanders aimlessly at sea, and then dissipates.

It's very similar to the Science on a Sphere exhibits at the Maryland Science Center and the Goddard visitors center. Nice musical background, too. To watch it, click here.

June 8, 2006

Storms starting later than 2005

OK, so we're ahead of the game, at least so far. On this date last year, meteorologists were already tracking what would become Arlene, the season's first tropical storm in the Atlantic basin. For now, the tropics remain mercifully quiet. Here's a satellite view of the Gulf and the Caribbean and part of the Atlantic.

Arlene was born June 8, 2005 in the western Caribbean Sea, off the coast of Honduras. On June 9th it reached tropical storm strength and received its name. It crossed over western Cuba and headed north across the Gulf of Mexico. About halfway across the gulf it reached its maximum strength, with top winds of about 69 mph.

Arlene made landfall near Pensacola, Fla., with gusts as high as 59 mph. There was little property damage, but one fatality - a Russian exchange student caught in a rip tide at Miami Beach - was blamed on the storm.

What was left of Arlene continued north across Alabama, Tennessee and Illinois, finally merging with other weather systems and passing into hurricane history near Flint, Mich.

But the 2005 season was just getting wound up. Tropical Storm Brett was born on June 28 and became the second named storm. The first hurricane was Cindy, which arrived July 3. Before the season finally wheezed to a close - when Tropical Storm Zeta fell apart in the Atlantic on Jan. 6, 2006 !! - there would be 28 named Atlantic storms - an all-time record.

May 31, 2006

Hurricane forecast unchanged; first storm weakens

The latest pre-season forecast for the 2006 Atlantic hurricane season, which opens tomorrow, was released today. And it shows no change from previous versions. The forecast team led by Phil Klotzbach and William Gray, of  Colorado State University, continues to predict a very active season, with 17 named storms, including 9 hurricanes, of which five are expected to be "major" - meaning Category 3 or higher.

In addition, the Colorado State group is setting an 82 percent chance that at least one major hurricane will make landfall in the United States (the long-term average is 52 percent), and a 69 percent chance it will land on the U.S. East Coast, from Florida northward - more than twice the average. You can read the whole report here.

The good news is that sea surface temperatures across the hurricane-birthing area of the Atlantic, while warmer than average, are cooler than they were at the start of last year's hurricane season. These images from May 2005 and May 2006 tell the tale.

In the meantime, the 2006 Hurricane Season for the Eastern Pacific is already underway, with the first named storm - Aletta - weakening south of Mexico's Pacific coast. Here's the latest (and last) forecast discussion. Here's the satellite view. What's left of Aletta is the mass of clouds at lower right.

May 25, 2006

Tropics calm, for how long?

By now everyone knows we're just a week away from the opening of the 2006 Atlantic hurricane season. Fewer remember that we're just two weeks away from the date, last year, when the first tropical storm of that record season popped up in the western Caribbean.

Tropical Storm Arlene was born June 8, one of the earliest storms on record and the first of a record-shattering 28 named storms last year. It tracked northward across the western tip of Cuba. and across the Gulf of Mexico. It went ashore on the Florida panhandle, near Pensacola. Arlene caused little damage, but a rip current stirred up by the storm was blamed for the death of an exchange student swimming in Miami. Arlene was the first of two tropical storms in June last year. TS Bret formed on June 28, lasted three days and killed one person when it went ashore in Mexico.

Fewer still will remember Tropical Storm Ana, in 2003, which defied the hurricane calendar and emerged in the Atlantic on April 20 - yes, April. It was the first North Atlantic tropical storm ever recorded in April. Ana wandered in the open Atlantic for five days before dissipating.

Forecasters are expecting a very active season this year, with 13 to 16 named storms. Among other factors, the sea surface temperatures in the Atlantic are slightly lower than last year. It's also statistically unlikely that we would see two consecutive record seasons. On the other hand, their pre-season forecast is slightly more pessimistic - calling for slightly more storms - than last year's May forecast (which, at 12 to 15 named storms, turned out to be too conservative by half.) 

May 5, 2006

Northeast hurricane overdue

AccuWeather meteorologists have concluded that the Northeast is overdue for a major hurricane - one like the 1938 storm that ripped across Long Island and tore deep into inland New England. They seem to see a pattern of serious northeast storms following bad Gulf Coast storms by a year. The dense population and the value of development in the Northeast have increased so much since 1938 that a similar strike there today would have huge consequences, perhaps more costly than Hurricane Katrina, the forecasters argue.

One of the best accounts of the 1938 storm was "A Wind to Shake the World," written by the late Everett S. "Joe" Allen. He was a cub reporter on his first assignment for the New Bedford (MA) Standard-Times when the storm struck. He also wrote books about New England whaling and the history of Martha's Vineyard. He was still on the job and a colleague of mine in New Bedford during the 1970s, and one of the finest newspapermen and historians I have ever known. Great beach reading. But keep your NOAA Weather Radio handy.

May 4, 2006

La Nina ends; fewer hurricanes expected

Residents of hurricane country may catch a break in the upcoming 2006 Atlantic hurricane season. NASA says the La Nina phenomenon, a pattern of cool sea surface temperatures in the tropical Pacific that encouraged Atlantic hurricane formation during the 2005 season, won't be a factor this year. The weak La Nina that prevailed last year has ended, and sea surface temperatures returned to normal in April. Forecasters still expect the coming hurricane season will be more active than the long-term average for other reasons. But they say it's unlikely to be as busy as the record-breaking 2005 season. To read more, click here.

April 7, 2006

Infamous storm names retired

We won't have Katrina to kick us around anymore. Or Dennis. Or Rita. Or Stan. Or Wilma. All those tropical storm names from the 2005 storm season have been officially retired by the World Meteorological Organization's international hurricane names committee, which includes officials from the National Oceanic and Atmospheric Administration's National Hurricane Center. The decisions came out of their recent meeting in Puerto Rico.

Filling in for the retired names will be Don, Katia, Rina, Sean and Whitney.

The international committee can retire names when storms cause very serious death tolls and damage. Otherwise, the names get recycled and reappear every seventh year in the new list of names. In this case, Katrina, Rita and the others would have popped up and haunted us again in 2011.

Here are the name lists for the upcoming Atlantic Hurricane seasons. The NHC typically assigns 21 names for each season. They run in alphabetical order, skipping Q, U, X, Y and Z, and alternating genders. The names are a mix of names from the languages and cultures of the affected region. If they run through the list, as they did in 2005, the NHC turns to the Greek alphabet.

Last year the busy season ripped right through Alpha, Beta, Gamma, Delta, Epsilon and Zeta before quiet returned to the Atlantic Basin.

February 23, 2006

The Katrina Report

The White House today released the Bush Administration's critical assessment of its own performance in the aftermath of Hurricane Katrina last August, and the lessons learned (we hope).  Here's a link to the full report.

December 22, 2005

Katrina struck as Cat. 3, not 4