Seeing into the Heart of a Hurricane
by John Weier
October 12, 2000
  Page 2
Despite the forecasts that Hurricane Opal would hit their town in a little more than 24 hours, the residents of Pensacola, Florida, remained relatively calm on October 3, 1995. They pulled their boats out of the water, boarded up the beachfront businesses and went about their daily routines, fearing no more than perhaps a few fallen trees and a missed day of work. At that time, the National Hurricane Center predicted that Opal would remain a Category 1 storm, packing peak winds of around 90 miles per hour—a veritable creampuff as far as hurricanes go.

Then overnight, as the hurricane moved across the Gulf of Mexico towards the city, something happened that no one predicted. The hurricane gathered energy from some then unseen reserves, jumped up in intensity to a strong Category 4 storm with peak winds of 150 mph, and threatened to turn Pensacola into a deluge of seawater and rain. The whole region went into a frenzy. The residents gathered what they could, evacuated their homes, and lined up bumper to bumper on Interstate 110 in an effort to flee.

Though the storm dropped down a notch to Category 3 when it made landfall, Opal ended up being the fourth costliest hurricane in U.S. history, totaling $3 billion in damage and claiming four lives. Many feel that this loss of life and property damage could have been avoided if folks had been alerted of the storm's possible change in intensity the day before.
 

  Hurricane Opal
This GOES image shows Hurricane Opal as it sweeps over the Gulf Coast of the United States on October 10, 1995. Image produced by the NASA GSFC Laboratory for Atmospheres using data from GOES-8. click to enlarge (209kb)

Also see:
Hurricanes: The Greatest Storms on Earth

What happened to the forecast and the forecasters? The National Hurricane Center (NHC) was at full alert and the personnel were monitoring the storm as best they could. The problem was that there was very little the forecasters could do to predict such a jump in intensity. While over the years government agencies have gotten more proficient in tracking storms and predicting their paths, they have not really gotten a handle on how intense the storms are going to be as they travel. Forecasting intensity requires a better understanding of the ongoing physical interactions between the ocean and atmosphere within the storm. Researchers need the ability to see through the cloud cover and into the hurricane as well as the ocean below it. Satellites and research aircraft have only infrequently been able to peer through the heavy cloud cover, so scientists have gotten too few data from inside hurricanes to fully understand how they work. Over the next few years, however, this may change. With the launch of NASA's Tropical Rainfall Measuring Mission in 1997 and several other weather satellites that can actually measure what goes on inside of cloud formations, scientists now have the tools to determine what causes a hurricane to intensify or weaken and to predict how intense a given storm will be as much as a day in advance.

next  A Deadly Cycle

The data used in this study are available in one or more of NASA's Earth Science Data Centers.

  TRMM spacecraft The Tropical Rainfall Measuring Mission (TRMM) is the first mission dedicated to measuring tropical and subtropical rainfall through microwave and visible infrared sensors, and includes the first spaceborne rain radar. (see TRMM Fact Sheet)

  A Deadly Cycle Page 1Page 3
"The thing that makes hurricanes tick is the warming of the upper atmosphere in the eye of the storm. This heat comes from the condensation of water vapor into cloud droplets," says Jeff Halverson. He is a meteorologist at NASA's Goddard Space Flight Center who has spent most of his professional life studying hurricanes. He explains that hurricanes are essentially a ring of thunderstorms swirling around the eye of the storm. To keep going, these thunderstorms need a continuous supply of water to replace what they lose in the form of raindrops. They are replenished by evaporated water that condenses in the air above the hurricane to form cloud droplets, which are then subsumed by the thunderclouds.
 
 
Hurricane Bonnie

Though water vapor in the atmosphere may seem harmless, en mass it contains an enormous amount of energy. The energy comes from the sun, which initially excites the water molecules on the surface of the Earth and causes them to evaporate. Each time a little of this water vapor comes together in the air to form a cloud droplet, the process of evaporation is essentially put in reverse and a little energy is released into the atmosphere in the form of heat. While the release of heat to form one raindrop is miniscule, the amount of heat released to replenish the cloud droplets in a hurricane is massive.

It is this heat that drives a hurricane. "The rapid spinning motion of the hurricane keeps this heat in the center of the storm," says Halverson. The heat causes the air in the upper reaches of the hurricane eye to warm and expand vertically in the center of the hurricane, creating a reduction in pressure near the surface. To fill this void, air at the ocean surface streams inward toward the eye. Much like a giant vacuum cleaner, the hurricane's eyewall sucks up air from the ocean's surface, giving rise to the spinning motion of the hurricane and drawing in even more evaporated water from the surface of the ocean. The fresh evaporated water from the sea's surface then typically travels up through the hurricane's eyewall and splays out at the top where it cools and condenses. The formation of ice particles below freezing temperatures releases more heat energy and adds more water to the thunderclouds, often times causing them to grow even bigger.

"So hurricanes are like car engines. The fuel is evaporated water on the ocean's surface. The cylinders are the thunderclouds inside the eye wall. They take in the fuel and convert it into heat energy which keeps the hurricanes alive," says Halverson. As long as the hurricane encounters new sources of moist sea air as it moves above the ocean's surface, it will continue to thrive.
 

  This TRMM Precipitation Radar image of Hurricane Bonnie, acquired August 22, 1998, shows where the highest concentrations of precipitation are within the storm. The red areas indicate the highest levels of precipitation, yellows show intermediate values, and blues indicate lower values. The highest level of precipitation, indicated by the prominent red "finger" in the center, is from a convective burst towering 11 miles into the atmosphere. (Image produced by Shirah/Morales, NASA GSFC; data courtesy NASA/NASDA) click to enlarge (243kb)
Bonnie Temperature Anomaly graph Bonnie Temperature Anomaly

Without a steady supply of evaporated water, however, the big thunderstorms that make up a hurricane would eventually empty their contents into the ocean and the low-pressure system at the center of the hurricane would die out. Cold waters that haven't been heated by the sun and land surfaces in general do not produce much evaporated water. So a hurricane will fade away if the water is less than 26.5 degrees Celsius (79.7°F) or if the hurricane makes landfall (University of Illinois, 1998).

But if the hurricane wanders over an area of the ocean where the sea's surface has been heated and the air is humid and stagnant, the thunderclouds can grow extremely violent in a matter of hours. The worst possible situation, Halverson explains, is when hurricanes undergo what is known as a "convective burst." In this instance a sort of positive feedback loop is triggered wherein the hurricane passes over a large pool of warm water, grows more intense, draws in more moist air, gets even nastier and so on until it reaches Category 4 or Category 5 strength. "Once one of these convective bursts gets going it can last ten hours or more," says Halverson. This is what happened with Hurricane Opal in 1995. It also occurred in the South Pacific in 1997 when Tropical Storm Paka suddenly intensified and became Super-Typhoon Paka. The typhoon was so powerful that it even shut down the Joint Typhoon Warning Center in Guam.

next The Trouble With Intensity
back Seeing into the Heart of a Hurricane

  This graphic plots data from NOAA's Advanced Microwave Sounding Unit (AMSU) and shows a cross-section of Hurricane Bonnie to reveal its internal temperatures as a function of altitude and atmospheric pressure (both Y-axes), and as a function of west longitude (X-axis). The false colors represent temperature, with pink and red showing the highest values, greens and blues are intermediate values, and purples are the lowest values. Notice the high temperatures toward the upper central part of the hurricane, which are due to the latent heat being released from the water vapor being drawn upward from surface. The colder areas toward the lower central part of the storm are caused by the relatively cold rain drops. (Image courtesy Mitch Goldberg, NOAA [published in Kidder et al. 2000])

  The Trouble With Intensity Page 2Page 4
Though forecasters had trouble predicting the intensity of both Paka and Opal, they were more confident in terms of predicting where the hurricanes would hit land. Over the last thirty years, forecasting the movement of hurricanes has steadily improved. The number of deaths per year from hurricanes in the United States has dropped by half over this time period despite the fact that coastal populations in the eastern U.S. have doubled.

Improvements in storm tracking have had everything to do with technology. Since the late 1960s, a number of satellite networks such as the NOAA's GOES (Geostationary Operational Environmental Satellite) and AVHRR (Advanced Very High Resolution Radiometer) have been launched into orbit for the specific purpose of monitoring weather. The primary sensors on most of these craft measure the visible and infrared light (light just beyond visible red on the color spectrum) emitted and reflected off clouds and the surface of the Earth. By analyzing these satellite data, scientists can observe such variables as sea surface temperature, atmospheric temperature, and wind speed all around a hurricane as it makes its way to shore (NESDIS Public Affairs, 1995).

The other big advance has been with the computer models forecasters use to predict a hurricane's path. These models are essentially complex computer programs. Researchers feed known variables such as wind current direction collected by the satellites and reconnaissance aircraft into the models. The models then run the measurements through a series of algorithms and various other calculations and come up with the most likely predictions of hurricane trajectories. Due to ever more powerful computers and improved programming techniques, the models have become more accurate each year (National Hurricane Center Public Affairs, 1996).

Despite all these new tools and resources, researchers still have difficulty predicting intensity. The problem is that to forecast intensity accurately, researchers have to understand what is going on inside and directly underneath the hurricane. Visible and infrared imaging satellites such as AVHRR cannot see past the upper cloud layer of a hurricane. The clouds block nearly all visible and infrared radiation coming from the surface of the Earth and from within the clouds, thus all the scientists receive from those sensors are images of reflective white and gray clouds. While aircraft and balloons can retrieve internal readings, they cannot do so quickly across the entire hurricane and not for a sustained period of time.

Most current models that forecast intensity do so by relying on those conditions that lie sometimes days in front of the storm or by comparing these readings to a jumble of statistics on past hurricanes in the area (Henson, 1998). To get a truly accurate prediction of intensity, however, scientists need to know how warm the ocean is below the hurricane, how deep the warm water extends and, most importantly, if the heat released inside the eye wall of a hurricane is increasing.

next TRMMing Off the Tops of Clouds
back A Deadly Cycle

  hurricane engine image
The driving engine of a hurricane converts heat from tropical ocean waters to raging winds and heavy rain. In this animation, notice how heat is drawn up by central, cyclonic winds and distributed to the swirling clouds surrounding the eye. The warmer the ocean waters beneath, the more powerful the storm is likely to grow. view animation (3.3MB)

water evaporation in hurricanes
As water evaporates from the warm ocean surface, it's forced upward in towering convective clouds that surround the eyewall and rainband regions of the storm. As the water vapor cools and condenses from a gas back to a liquid state it releases latent heat. view animation (1.3MB)

hurricane cloud growth
The release of latent heat warms the surrounding air, making it lighter and thus promoting more vigorous cloud development. It's believed that rapid bursts of cloud growth, particularly in the eyewall region of hurricanes, may relate to the intensification phase of a storm. It is also interesting to note that due to the intense winds surrounding the eye, clouds are often absent from the center of a hurricane; they're simply thrown out from the center.view animation (1.3MB)

Animations courtesy NASA Goddard Space Flight Center Scientific Visualization Studio


  Unwrapping the Clouds

back The Trouble With Intensity

  The driving engine of a hurricane converts heat from tropical ocean waters to raging winds and heavy rain. In this animation, notice how heat is drawn up by central, cyclonic winds and distributed to the swirling clouds surrounding the eye. The warmer the ocean waters beneath, the more powerful the storm is likely to grow.

Animation courtesy NASA Goddard Space Flight Center Scientific Visualization Studio


  Releasing Energy from Water Molecules

back The Trouble With Intensity

  As water evaporates from the warm ocean surface, it's forced upward in towering convective clouds that surround the eyewall and rainband regions of the storm. As the water vapor cools and condenses from a gas back to a liquid state it releases latent heat.

Animation courtesy NASA Goddard Space Flight Center Scientific Visualization Studio


  Towering Clouds

back The Trouble With Intensity

  The release of latent heat warms the surrounding air, making it lighter and thus promoting more vigorous cloud development. It's believed that rapid bursts of cloud growth, particularly in the eyewall region of hurricanes, may relate to the intensification phase of a storm. It is also interesting to note that due to the intense winds surrounding the eye, clouds are often absent from the center of a hurricane; they're simply thrown out from the center.

Animation courtesy NASA Goddard Space Flight Center Scientific Visualization Studio


  TRMMing Off the Tops of Clouds Page 3Page 5
With the help of NASA's Tropical Rainfall Measuring Mission (TRMM) satellite, the trouble with intensity may soon end. "The TRMM satellite is essentially a weather radar put into orbit," says Halverson. Launched in 1997, the satellite orbits the Earth traveling from west to east, back and forth across the equator, taking measurements each day from 40 degrees latitude north of the equator to 40 degrees south. In the Western Hemisphere, this zone extends roughly from the northern border of Virginia to the southern tip of Uruguay.
 
 
While TRMM does have an instrument called a Visible and Infrared Sensor (VIRS) that makes measurements similar to those made by the GOES satellites, it also contains two additional instruments that can see into and through clouds. The first, known as the Precipitation Radar, does exactly what its name implies—it bombards the clouds with radio waves and then receives the signals that bounce back. The radar is rigged so that the only signals that bounce back are from droplets that are at least the size of raindrops. By retrieving data from this instrument, scientists can tell whether a cloud is precipitating or not. The other instrument is known as the TRMM Microwave Imager (TMI). This instrument detects a number of microwave frequencies that the ice crystals and water droplets in clouds absorb and reflect. By precisely measuring these microwave signals, scientists can determine how much ice exists in clouds as well as how warm the ocean water is beneath the clouds (NASA TRMM Web Site, 1996).

Both of these instruments are particularly well suited for measuring the increase or decrease in activity in the hurricane eye wall. The Precipitation Radar can estimate the number of raindrops in the eye wall. A rise in the amount of raindrops is an indication that a hurricane is growing more intense. "In general the more rainwater there is in a hurricane, the more heat is being released into the center of the hurricane, and the more likely the hurricane will grow in intensity," says Halverson. Microwave Imager readings, on the other hand, can give forecasters an idea as to how much ice has formed in the upper regions of the storm clouds. A large increase in ice in the upper levels of a hurricane indicates that the clouds are growing so large and powerful that they have extended high into the atmosphere where the temperatures are well below freezing.

Although scientists already have access to TRMM data of hurricanes as they occur, simply glancing at computer images of precipitation and ice formation can only lead to a rough approximation of whether the hurricane is intensifying. "Eventually, we'd like to have a [computer] model that will take the TRMM data and crank out numbers on the amount of heat forming in the eye wall," says Halverson. He and a team of scientists from NOAA and NASA are developing a theory on hurricane intensification using both TRMM satellite data and data from NOAA's orbiting Advanced Microwave Sounding Unit (AMSU), which can measure the temperature of the atmosphere through cloud cover.

The hypothesis they are testing will take in TRMM precipitation and ice data from a hurricane, put these data through a number of complex calculations, and return readings on the rate at which heat is being produced in the eye wall. The heat readings would then be related to the existing temperature readings inside the eye from the AMSU data. "If we know how much heat is being released and we can say how warm it is now, we should be able to tell if the eye is heating up or cooling down," says Halverson. If the heat is increasing rapidly, for instance, a convective burst could be the culprit.

Already, Halverson's team is testing their hypothesis using TRMM data from Opal, Bonnie, and Paka. They have also been taking airplanes into hurricanes to understand better the finer points of hurricane convection. Halverson says they may be able to track and even predict the heat levels in a hurricane's eye in a few years.

next Testing the Waters
back The Trouble With Intensity

  precipitation in Hurricane Mitch
This view of Hurricane Mitch was produced using data from TRMM's precipitation radar. High rates of rainfall appear in red, with lesser amounts appearing in blue. By mapping the structure of storms, experts can "take them apart" in the laboratory as they try to understand how they work. view animation (2.9MB)

Animation courtesy NASA Goddard Space Flight Center Scientific Visualization Studio


  Precipitation in Hurricane Mitch

back TRMMing Off the Tops of Clouds

  This view of Hurricane Mitch was produced using data from TRMM's precipitation radar. High rates of rainfall appear in red, with lesser amounts appearing in blue. By mapping the structure of storms, experts can "take them apart" in the laboratory as they try to understand how they work.

Animation courtesy NASA Goddard Space Flight Center Scientific Visualization Studio


  Testing the Waters Page 4
Another way in which the TRMM satellite may aid in future intensity forecasts is in its ability to measure the temperature of the sea through cloud cover. Before a jump in intensity can occur in a hurricane, the hurricane must first pass over an area where the water is especially warm. Traditionally, scientists have used satellite sensors' measurements of thermal infrared radiation—aboard satellites such as GOES and AVHRR—to retrieve sea surface temperatures. In general, the warmer the water the more infrared radiation it emits. But again, the problem with testing the waters under a hurricane is that infrared radiation cannot penetrate the thick cloud cover. Most forecasting models now use sea surface temperature data gathered up to a week before a hurricane moves into an area.
 
 
sea surface temperature following Hurricane Bonnie

Frank Wentz, a physicist and president of Remote Sensing Systems in California, explains that knowing sea surface temperatures directly underneath a hurricane is crucial to predicting its intensity. All types of unexpected changes can occur. In Opal's case, some researchers believe an undetected warm eddy floating through the Gulf wandered into the hurricane's path and thus provided it with an unexpected supply of warm water. In other instances, warm waters in the ocean will run deeper than scientists expect. The hurricane will intensify because it has a much greater supply of undetected fuel.

For the past three years, Wentz's company has been using data from the TRMM TMI instrument to create daily global data sets of sea surface temperatures below both clear skies and cloud cover. (In particular, they use TMI's 10.7 GHz channel.) Wentz explains that when the sun heats up the water, the water emits microwave radiation as well as infrared radiation. "Unlike infrared radiation, nearly 97 percent of the microwave radiation that comes from the Earth's surface goes through the clouds. Using the other channels on TMI, we can penetrate the clouds to see the water vapor down a column of atmosphere, liquid water within a cloud, and rain rates." Wentz uses these values to account for the remaining 3 percent of the 10.7 GHz signal that is lost as it travels through the atmosphere. In short, Wentz and his associates observe the lower frequencies of the microwave spectrum that the instrument detects-the more radiation it measures, the warmer the water is. Already they have uncovered some interesting findings by analyzing TRMM data of hurricanes Bonnie and Danielle. In August of 1998, hurricane Bonnie struck the coast of North Carolina with Category 3 winds, causing some moderate flood damage. Danielle followed on Bonnie's heels and was expected to brush up against the North Carolina coast before heading north out into the Atlantic. Yet, when Danielle hit Bonnie's wake, it dropped steeply in intensity from a high Category 2 storm to a low Category 1. While the forecasters told North Carolina residents to expect another torrential rain, all they got were some good surfing conditions and late summer showers.

At the time, scientists suspected that Bonnie's wake somehow altered the sea surface temperatures, but no one knew exactly how. Upon reviewing the TRMM data of sea surface temperatures under both hurricanes, Wentz and his team discovered that Bonnie stole all of Danielle's fuel! "Bonnie gathered much of the heat energy from the ocean's surface. The friction from Bonnie's winds also churned up the warmer surface water and brought up cool water from deep in the ocean," says Wentz. When Danielle hit this patch of cold water, it immediately dropped in intensity. The TRMM reading told the whole story and provided evidence that one hurricane can slow down another that follows in its wake.

At this point, Wentz says he and his team are carefully reviewing the data to see what they can find on a case-to-case basis. By analyzing previous model forecasts and comparing them to actual data throughout a given hurricane's lifetime, his team can correlate errors in the input data with errors in the model predictions. This process, over time, will help them refine their model.

"Eventually what we'd like to see are numerical models that predict both the track and intensity of a hurricane from sea surface temperatures. One way this could happen is if we start using microwave measurements of sea surface temperature," says Wentz. It is not simply enough to plug these new data into existing hurricane models that use sea surface temperatures taken by infrared remote sensors. Instead, agencies like NOAA need to develop new forecasting models that are tuned to these new data and incorporate them operationally.

Halverson explains that his work and Wentz's could be used in conjunction to help the forecasting community predict a storm's intensity. With a model that uses continuous sea surface temperature data, forecasters could see how a hurricane is developing at all times. If warm water turns up, then the model could be employed to monitor the hurricane as it passes over and look for sudden flare-ups in intensity. "The idea is that all this information will converge and we will get a complete picture of the mechanisms that drive a hurricane," says Halverson. "Then we may be able to better predict disasters such as Hurricane Opal."

References

Kidder, S. Q., M. D. Goldberg, R. M. Zehr, M. DeMaria, J. F. W. Purdom, C. S. Velden, N. C. Grody, and S. J. Kusselson, 2000: Satellite analysis of tropical cyclones using the Advanced Microwave Sounding Unit (AMSU). Bulletin of the American Meteorological Society, Volume 81, pp. 1241-1259.

Henson, R., 1998: The Intensity Problem, Weatherwise, September/October, pp. 1-7.

NASA TRMM Website, 1996: Nasa Facts: TRMM Instruments, Greenbelt, MD. (http://trmm.gsfc.nasa.gov/)

National Hurricane Center Public Affairs, 1996: Hurricane Tracking Models: Helping to Forecast Severe Storms, Miami, FL.

NESDIS Public Affairs, 1995: NOAA's Geostationary and Polar-Orbiting Environmental Satellites, Suitland, MD.

University of Illinois, 1998: WW2010, Urbana-Champaign, IL. (http://ww2010.atmos.uiuc.edu/)

Wentz, Frank J., Chelle Gentemann, Deborah Smith, Dudley Chelton, 2000:"Satellite Measurements of Sea-Surface Temperature Through Clouds." Science, Volume 288, Number 5467, pp. 847 - 850.

back TRMMing Off the Tops of Clouds

  One of the major stumbling blocks for forecasters has been the precise measurement of sea surface temperatures under a storm as it forms and evolves over time. In this scene, clouds (acquired by GOES) have been made translucent to allow an unobstructed view of the surface. Notice Hurricane Bonnie approaching the Carolina Coast (upper left) and Hurricane Danielle following roughly in its path (lower right). The ocean surface has been falsely colored to show a map of water temperature, measured by the TRMM Microwave Imager (TMI). Dark blues are around 75°F, light blues are about 80°F, greens are about 85°F, and yellows are roughly 90°F. In the animation, notice that as Hurricane Danielle followed in Bonnie's path, the wind speed of the second storm dropped markedly, as available energy to fuel the storm dropped off because the surface waters in Bonnie's wake were cooler. But when Danielle left Bonnie's wake, wind speeds increased due to temperature increases in surface water around the storm. view animation (6MB)

Image courtesy TRMM Project, Remote Sensing Systems, and Scientific Visualization Studio, NASA Goddard Space Flight Center


  Sea Surface Temperature and Hurricane Bonnie

back Testing the Waters

  One of the major stumbling blocks for forecasters has been the precise measurement of sea surface temperatures under a storm as it forms and evolves over time. In this scene, clouds (acquired by GOES) have been made translucent to allow an unobstructed view of the surface. Notice Hurricane Bonnie approaching the Carolina Coast (upper left) and Hurricane Danielle following roughly in its path (lower right). The ocean surface has been falsely colored to show a map of water temperature, measured by the TRMM Microwave Imager (TMI). Dark blues are around 75°F, light blues are about 80°F, greens are about 85°F, and yellows are roughly 90°F. In the animation, notice that as Hurricane Danielle followed in Bonnie's path, the wind speed of the second storm dropped markedly, as available energy to fuel the storm dropped off because the surface waters in Bonnie's wake were cooler. But when Danielle left Bonnie's wake, wind speeds increased due to temperature increases in surface water around the storm. view animation (6MB)

Image courtesy TRMM Project, Remote Sensing Systems, and NASA Goddard Space Flight Center Scientific Visualization Studio