Flying an aircraft through a hurricane is risky business, even if the plane is specially equipped for the job. In the hurricane’s eye, skies are clear and calm prevails, but in the ring of intense storms surrounding the eye—the eyewall—rain falls in thick sheets and winds gust to 300 kilometers per hour.
In 2005, despite those perils, the pilots of three “hurricane-hunter” planes flew repeated missions into the cores of the monster storms Katrina and Rita as well as the much tamer Ophelia. During the missions—collectively dubbed the Rainband and Intensity Change Experiment, or RAINEX—scientists on board the instrument-laden aircraft collected unprecedented data on the structure, configuration, and interaction of clouds within the massive hurricanes. Probes dropped from the planes garnered additional information.
In one case, the aircraft were the first ever to encounter and directly observe a ring of intense thunderstorms just outside the storm’s eyewall. Such secondary eyewalls, which appear to have significant effects on hurricanes’ strengths, had often been detected by satellites and radar but had never been seen in the fine detail achieved during RAINEX.
Analyses of data from that encounter may enable researchers to identify the features within a hurricane that most affect the storm’s intensity. With that information in hand, meteorologists could do a far better job of forecasting wind speed and ocean surge as a storm approached land. Also, scientists say, the new techniques that RAINEX researchers employed on shore to coordinate hurricane hunters’ flights could transform how such missions are flown.
Whenever meteorologists announce a new tropical storm or hurricane, two questions immediately arise: Where’s the storm headed? and How strong will it be when it gets there?
“The first question is by far the easier of the two,” says Hugh E. Willoughby, an atmospheric scientist at Florida International University in Miami. The path that a hurricane takes depends largely on prevailing weather patterns throughout the surrounding region, including factors such as the strength, configuration, and movement of high- and low-pressure areas. Recent improvements in forecasting hurricane paths stem primarily from enhancements in the computer models used to predict weather in general, he says.
Meteorologists gauge the accuracy of path predictions by their “track error”—a measure of how far off its predicted line a hurricane’s eye wanders, explains James Franklin, a forecaster at the National Hurricane Center in Miami. In the 1970s, the average track error in the 3-day forecasts for hurricanes and tropical storms was 700 km. So far this decade, 3-day forecasts have missed the mark only by 300 km on average, he notes.
Predictions of hurricane intensity haven’t improved nearly as much. In the past 2 decades, errors in the National Hurricane Center’s 2- and 3-day forecasts for wind speeds within hurricanes and tropical storms have dropped only a couple of kilometers per hour. That’s because computer models that aim to represent hurricanes must pack data points close together to accurately simulate the small-scale, rapidly evolving features that swirl around the core of a storm. If a computer model has weather-data points spaced no closer than 5 km apart, for example, the theoretical storms it portrays turn out to be “larger, weaker cartoons of their counterparts in nature,” says Willoughby.
“It’s critical for forecasters to get a hurricane’s track right, but it’s an even bigger challenge to predict the strength of its winds,” says Bradley F. Smull, a research meteorologist at the University of Washington in Seattle.
Accurate wind forecasts are vital for several reasons. As well as directly affecting how much damage a storm inflicts on structures, wind speed dramatically influences the height of a hurricane’s storm surge, the mound of water its winds push ashore. However, hurricanes are notorious for their sudden, and sometimes severe, variations in intensity.
Some of the factors behind such changes are well understood, says Willoughby. For instance, three of the four hurricanes that struck the Gulf Coast in 2005—Dennis, Katrina, and Rita—intensified as they passed over the Gulf of Mexico’s Loop Current, whose warm waters provided a ready source of energy for the storms. Rita strengthened from category-1 status (wind speeds between 121 and 153 km/hr) to category-5 (sustained winds exceeding 250 km/hr) in less than a day.
Then there are murkier influences on storm intensity, such as the interactions between thunderstorms immediately surrounding a hurricane’s eye and those arranged in bands that, seen from space, lend hurricanes a pinwheel appearance. The dearth of information about such interactions led researchers to propose the 2005 RAINEX missions, which ended up differing from previous hurricane-hunter flights in several ways, says Robert A. Houze Jr., an atmospheric scientist at the University of Washington in Seattle.
First, one of the three aircraft deployed during each RAINEX mission was equipped with a type of Doppler radar that hadn’t been used before inside hurricanes. The system has two antennas that look in slightly different directions and take measurements at a faster rate than normal Doppler radar does. The result is a high-definition look at clouds.
Second, information from the three aircraft was transmitted during the mission to earthbound scientists, who combined it with data from satellite images and ground-based radar to create a composite map of the storm.
Third, the ground team ran high-resolution computer simulations of the hurricane, which the scientists then used to direct the pilots toward parts of the storm where interesting features were present or likely to appear. “Using this technique, the pilots aren’t flying blind,” says Houze. He, Smull, and their colleagues described the experiments and their findings in the March 2 Science.
The RAINEX missions targeted hurricanes Katrina, Ophelia, and Rita, a trio of storms that certainly offered the scientists variety.
Katrina, which reached peak strength while over the Gulf of Mexico on Aug. 28, 2005, was a category-5 storm and the fourth strongest on record for the North Atlantic basin. After Katrina’s eye moved off the Loop Current and over cooler waters, the hurricane weakened to category-4 status. Sustained winds at landfall measured about 200 km/hr.
Ophelia, which formed just east of Florida in early September, alternated several times between tropical storm and hurricane status as it wobbled its way slowly up the Atlantic Coast. With top winds of 140 km/hr, the hurricane never exceeded category-1 status.
Rita, one of the storms that strengthened to exceed Katrina, put on the best show for researchers in late September. Immediately after a single-day growth spurt to category-5 level, the hurricane underwent a strength-sapping process called eyewall replacement. That’s just the kind of sudden intensity variation that RAINEX scientists were hoping to observe at close range, says Houze.
Meteorologists know in general terms how hurricanes typically evolve. The thunderstorms that ring a hurricane’s eye are usually more intense than storms elsewhere in the system. The warm, humid winds that fuel these thunderstorms spiral toward the center of the hurricane at low altitude, says Willoughby. When that soggy air reaches the eyewall, it rises and some of its moisture condenses out as rain. The energy released during that process heats the air further and causes it to rise even faster. By the time the air has risen to the tops of the eyewall clouds, where air temperatures are normally 100°C cooler than they are at the ocean’s surface, condensation has wrung all of the moisture out of the air.
But sometimes this circulation is interrupted, as it was inside Rita while it was still far out over the Gulf of Mexico. In that hurricane, thick bands of thunderstorms sweeping toward the eye coalesced to form a ring of storms about 20 km outside the eyewall. “The processes that occur during the formation of such a secondary eyewall aren’t well understood, but their effects are clear,” says Houze.
First, the thunderstorms in a secondary eyewall grab much of the humid air that’s headed eyeward. That robs the inner eyewall of fuel, which causes the thunderstorms there to weaken overall. Instruments dropped into the region between Rita’s inner and outer eyewalls revealed that the air there was warmer and much less humid than expected, says Houze. That meant less condensation of moisture in the clouds of the inner eyewall, further stifling precipitation in the thunderstorms there.
Eventually, as Rita approached the western portions of the Gulf Coast, the inner eyewall collapsed completely. “In 12 to 24 hours, it was gone,” says Houze.
Before the secondary eyewall formed, Rita’s top winds raged at about 275 km/hr. After the thunderstorms in the inner eyewall had subsided, wind speeds in the secondary eyewall measured only 180 km/hr or so. Soon, however, the eye began to contract, and wind speeds picked up again. This process—in which an inner eyewall breaks up and an outer eyewall then draws inward to take its place—is what meteorologists call eyewall replacement.
Guided by ground-control personnel, the pilot of a RAINEX aircraft equipped with the high-definition Doppler radar flew through the moat between Rita’s inner and outer eyewalls. During that 3-hour mission, the aircraft made at least one complete circuit of the moat, says Houze. The researchers discovered that the doughnut-shaped region wasn’t simply a void where nothing much was happening. Instead, the air everywhere inside the moat was moving downward. “It’s not just a passive region that’s stuck between two eyewalls,” he notes.
The high-resolution data gathered during the RAINEX missions could improve scientists’ simulations of hurricanes, says Willoughby. For example, meteorologists might be able to better estimate how external influences, such as ocean temperatures, and internal influences, such as cloud interactions, affect a storm’s strength.
Timing is everything
The unpredictability of a hurricane’s hour-to-hour strength—especially as affected by still poorly understood phenomena such as development of a secondary eyewall—hobbles forecasters. Incorrect estimates of the timing and magnitude of such changes can result in inadequate storm warnings that can cost lives. The trick is in knowing what a hurricane’s strength will be at or near its landfall.
Many storms suddenly intensify just before they strike land. Notable examples include Hugo, which slammed into South Carolina in 1989, and Charley, which raked across Florida in 2004. In 1992, Andrew completed an eyewall replacement shortly before it struck Florida’s Miami–Dade County as a category-5 storm. In contrast, other storms have unexpectedly weakened just prior to landfall—1999’s Floyd, for example. Over the course of several days, a couple of eyewall replacements shrank the hurricane from near-category-5 status, transforming Floyd into a category-2 hurricane that caused more flooding than wind damage.
The RAINEX missions of 2005 may serve as a model for future hurricane-hunter flights, says Houze. “Focusing future aircraft observations in the same way should make it possible to identify small-scale areas in a storm where the processes that affect intensity are occurring,” he notes.
Willoughby agrees, noting that the data gathered in such targeted missions could yield new insights into hurricane behavior. “We haven’t solved all of the problems [of hurricane forecasting] in the past few years, but we’ve made a lot of progress,” he notes.