Why do hurricanes happen?
What's the difference between a hurricane and a typhoon or tropical cyclone?
When is hurricane season?
How big and how strong can hurricanes get?
Can we control hurricanes?
Hasn't the number of hurricanes been going down lately?
Are hurricanes striking in new places?
Is global warming affecting hurricanes?
Advances in hurricane research
Most hurricanes start life as areas of rough weather and thunderstorms in the tropics. Many of these disturbances, or tropical waves, produce little more than heavy rain and gusty winds. But if a tropical wave succeeds in spinning into a complete circle of winds rotating around an area of low air pressure at its center, it's given the name tropical depression. When a depression's peak sustained winds reach 39 miles per hour (34 knots), it's called a tropical storm.
As a tropical system strengthens, its winds spiral inward, concentrating moisture near the center. This spiraling, a result of Earth's rotation, can't happen near the equator. To benefit from the curving winds produced by the Coriolis effect, a storm needs to be at least 300 miles (500 kilometers) north or south of the equator.
When a tropical storm maintains wind speeds of at least 74 miles per hour (65 knots), it's known as a hurricane, at least in North and Central America (see What's the difference between a hurricane and a typhoon or tropical cyclone?).
Each year, hundreds of tropical depressions spin up over warm waters worldwide. But more is needed for a depression to grow into one of Earth's most powerful storms. Besides warm ocean temperatures, the depression needs lots of warm, moist air to feed on.
Winds also matter. Strong differences in wind direction and strength at different heights in the atmosphere, known as wind shear, can pull a tropical system apart, dissipating its energy. So low wind shear is also essential to tropical storm formation. (But note that high wind shear plays a role in tornado formation, including tornadoes that form over land as a hurricane dissipates.)
For more about the life cycle of tropical storms, from their birth as tropical depressions, to full-blown hurricanes or cyclones, to their ultimate demise, see Hurricanes & Tropical Cyclone Life Cycles (from UCAR COMET's Hurricane Strike!).
When a tropical disturbance organizes to the point where its sustained winds top 34 knots (39 miles per hour), it's known as a tropical cyclone. But various parts of the world use a variety of terms once a tropical cyclone packs winds of at least 65 knots (74 mph) in a particular "basin" (an oceanic region where these storms occur).
Around North and Central America, they're called hurricanes. The god of evil for the Carib people was named Hurican, according to the authors of Hurricane Strike! That's the source, with a slight twist in spelling, of the name used in the Atlantic Ocean, Caribbean Sea, Gulf of Mexico, and Northeast Pacific Ocean.
In the Northwest Pacific, the same powerful storms are called typhoons. In the Southeastern Indian and Southwest Pacific Oceans they're called severe tropical cyclones. In the North Indian Ocean, they're called severe cyclonic storms, while in the Southwest Indian Ocean, they simply keep the name tropical cyclone.
Hurricanes that make Category 3 status on the Saffir-Simpson Hurricane Wind Scale (winds of at least 96 knots or 111 mph) are labeled intense hurricanes. If a typhoon hits 132 knots (150 mph), it becomes a supertyphoon.
When a system weakens below hurricane strength, it is typically reclassified as a post-tropical cyclone. This was the case with Hurricane Sandy in 2012 just before landfall. In order to reduce confusion, the National Weather Service introduced a new policy in 2013 in which hurricane warnings may be continued even after a storm has technically become post-tropical.
Hurricane season in the Atlantic Basin (including the Caribbean and the Gulf of Mexico) runs from June 1 through November 30, with most activity clustered from August to October.
The typhoon and cyclone seasons follow their own patterns. In the Northeast Pacific, the official season runs from May 15 to November 30. In the Northwest Pacific, typhoons are most common from late June through December. The North Indian Ocean sees cyclones from April to December, with peaks in May and November.
The Southwest Pacific and South Indian oceans (including the waters bordering Australia) get most of their activity from November to May.
If the conditions are right, tropical cyclones can develop outside their official seasons, especially in the Northwest Pacific, where they occur year round.
NOAA's Hurricane Research Division has more about seasonal timescales.
Whatever you call them, these monster storms are the most powerful atmospheric phenomena on Earth. Hurricanes gather energy from water vapor in the atmosphere stretched for hundreds of thousands of square miles across the warm ocean water of the tropics. The storms themselves can bring heavy rains and gale-force winds blowing 38 to 73 miles per hour (62–117 kilometers per hour) across an area the size of Louisiana. Hurricane-force winds can extend 50 miles (80 km) or more from the storm center.
According to NOAA's National Hurricane Center, the average hurricane eye—the still center where pressure is lowest and air temperature aloft is highest—stretches 30 miles (48 km) across, with some growing as large as 120 miles (200 km) wide.
The eye typically shrinks as a hurricane intensifies, sometimes narrowing to less than 10 miles (16 km) in width. Eventually, a new eye may form around the old one; hurricanes often weaken during this transition but can intensify again as the new eye contracts.
Hurricane intensity is categorized on different scales around the world. The strongest storms, equivalent to Category 5 on the Saffir-Simpson scale, have sustained winds that exceed 155 miles per hour (135 knots).
In order to grow, hurricanes need plenty of warm, moist air to form showers and thunderstorms, along with winds that don't change direction much with height, which allows the central circulation to develop undisturbed. If these atmospheric conditions are right, then hurricane strength is dictated largely by the presence or absence of deep, warm ocean water (ideally 79°F or warmer [26°C or warmer]). As winds strengthen, more water evaporates, releasing energy stored in the warm seas.
Using ocean data, scientists can estimate the maximum potential intensity (MPI), or the peak hurricane strength one might expect at a given location. Scientists are now exploring how best to calculate and predict MPI in computer models.
Size does not determine intensity. Some of the most destructive hurricanes to hit the U.S. coastline, including Hurricane Andrew in 1992, extended over a relatively small area.
Whatever their size or strength, long-lived hurricanes normally "spin down" after they come ashore. The land surface creates drag and the oceanic heat source disappears. The high winds gradually weaken and spread over a larger area. Even as they decay, hurricanes can still bring damaging winds more than 100 miles inland. Also, the frictional effects on surface winds can help create the wind shear needed to spawn tornadoes. One hurricane can produce dozens of tornadoes as it moves ashore.
But the worst hurricane damage is often the result of a storm surge that causes coastal flooding. Along parts of the Mississippi coast, the surge from Hurricane Katrina was 28 feet or more above mean sea level, putting it among the highest surges ever recorded in the United States. Large waves on top of a storm surge cause even more damage.
Surge risk varies depending on the strength and structure of a given hurricane as well as geography and tidal cycles along the coast where it strikes. Although Superstorm Sandy fell just below hurricane strength several hours before its center reached southern New Jersey in October 2012, Sandy was an extremely large system, and it struck near high tide. As a result, the storm pushed huge amounts of water into the New York and New Jersey coastlines, producing a catastrophic surge that set records in several places.
Originally, the Saffir-Simpson scale included typical storm surges expected for each category. In the case of Sandy and several other recent hurricanes, storm surges have been higher than the levels once assigned to each Saffir-Simpson category. The categories have been revised so that they now refer only to wind speed, and the National Hurricane Center is moving toward new ways of depicting the threat from storm surge, including storm surge watches and warnings that will be distinct from hurricane watches and warnings, as described in this 2012 article.
Storms that move inland often bring much-needed rain that farmers and water managers count on. Recent studies indicate that across parts of the U.S. Deep South, up to 15% of all warm-season rainfall typically comes from hurricanes and tropical cyclones. But if too much falls at once, the rain can quickly overwhelm stream and river beds, producing serious river flooding.
Even smaller hurricanes pack a mind-boggling amount of power. The heat energy released by a hurricane equals 50 to 200 trillion watts—or about the same amount of energy released by exploding a 10-megaton nuclear bomb every 20 minutes.
We are unlikely to come up with methods to control such overwhelming natural power for the foreseeable future—though that has not kept people from speculating about what it would take. Researchers instead focus on understanding every aspect of hurricane structure and behavior, with the hope their work will lead to better predictions of storm tracks and intensity so warnings can be issued to protect life and property.
The societal side of the equation includes communication between forecasters and emergency managers who make decisions about when and where to call for evacuation from threatened areas and then get the word out with help from local public safety offices and the mass media. Researchers at NCAR are collaborating with colleagues elsewhere to address the human side of hurricane forecasts and warnings through the Collaborative Program on the Societal Impacts and Economic Benefits of Weather Information (SIP).
Studies show 63% of hurricane-related deaths occur inland. NCAR scientists have been using tropical storm winds, census data, and flood maps to help emergency managers identify the most vulnerable populations and plan their response. The resulting 2012 map reveals the surprising extent of risk.
An overview in the Bulletin of the American Meteorological Society (August 2012) summarizes recent work by NCAR scientists and collaborators on the communication of hurricane risk.
On average there are about 70 to 110 named tropical cyclones per year across the world, including about 40 to 60 that reach hurricane strength. This range has held remarkably steady within the last 40 years. Within each basin, the numbers often vary more dramatically than the global average.
In part this is due to ocean-atmosphere cycles such as El Niño and La Niña. Because they affect where showers and thunderstorms develop, these cycles can suppress hurricane activity in one basin while enhancing it in another.
During El Niño years, the Atlantic tends to be less active than usual, while parts of the central and northeast Pacific are typically busier than usual. While the long-distance effects are not exactly opposite during La Niña events, there is a tendency for more activity in the Atlantic than the Pacific.
Even though the total number of tropical cyclones around the world holds fairly steady, some years are more active than others. Scientists often use an index called Accumulated Cyclone Energy, or ACE, to measure the overall intensity of a given year’s cyclone activity. ACE takes into account a storm’s peak winds at each six hours of its lifetime. After dipping to 30-year lows in the early 2010s, global ACE values have begun rising again (see graph).
Of all the hurricanes that build over the North Atlantic and Gulf of Mexico each year, only a small fraction make it to the U.S. coastline at hurricane strength. As of early 2013, the nation had not seen a major landfall (Category 3 or stronger) since 1995. However, the total number of hurricanes swirling across the North Atlantic remains unusually high.
In recent years, we’ve seen several hurricanes and tropical storms strike in unfamiliar places. This could be a result of improved monitoring, regional changes in ocean temperatures and upper-air circulation (perhaps linked to global warming in some cases), natural variability, or all of the above.
The South Atlantic had been considered free of tropical cyclones—that is, until March 2004, when a mysterious storm later dubbed Hurricane Catarina made landfall in Brazil. In October 2005, Vince became the first tropical storm ever recorded in Spain. And in the Arabian Sea, Gonu became a Category 5 in June 2007—that region’s strongest tropical cyclone on record. After weakening, Gonu brought unprecedented damage from rain, wind, and flooding to parts of Oman and Iran.
Although the Bay of Bengal has seen the world's deadliest tropical cyclones on record, most of these strike India and Bangladesh, while the coast of Myanmar tends to experience only weaker cyclones. The catastrophic Cyclone Nargis, which struck Myanmar in May 2008, appears to be the first major cyclone (Category 3 strength or higher) ever recorded in the populous Irrawaddy Delta. More than 90,000 people died in the storm and its aftermath.
The mid-Atlantic and New England can see intense hurricanes, but quite rarely. These are almost always moving north to northeast, often sideswiping the coastline. However, unlike any other hurricane in modern records, Hurricane Sandy moved west-northwest toward New Jersey, making landfall just after becoming post-tropical yet still delivering an unprecedented storm surge to the region.
Are hurricanes getting worse as the global average temperature rises? What can we expect in the future?
A growing amount of research is looking into these questions, and not all of the answers are in just yet. Here’s a summary of what research tells us so far.
The bottom line: As summarized in 2007 in the Fourth Assessment Report of the Intergovernmental Panel on Climate Change:
There is observational evidence for an increase of intense tropical cyclone activity in the North Atlantic since about 1970, correlated with increases in tropical SSTs [sea-surface temperatures]. There are also suggestions of increased intense tropical cyclone activity in some other regions where concerns over data quality are greater. —IPCC Working Group I, Summary for Policymakers, 2007
Sea-surface temperatures across the tropics have risen along with global temperature over the last 100 years and are expected to warm further in the next century. All else being equal, warmer oceans can support stronger hurricanes.
The extra water vapor evaporated by oceans in a warmer climate can be expected to boost rainfall from hurricanes by as much as 8% for every 1°C (1.8°F) of warming, according to Kevin Trenberth of NCAR (view a PDF of this paper).
Several studies in the last decade have reported enhanced hurricane activity in some regions. For example:
Changes in observing techniques pose a major challenge when studying past hurricanes. Tropical cyclones have been routinely monitored by aircraft for 60 years or less, and by satellite for only 20 to 40 years, depending on the ocean. In earlier years, many storms over the open ocean may have gone unobserved, as noted in a 2007 study by Christopher Landsea (NOAA, view a PDF of this paper). Thus, scientists must take special care in analyzing global hurricane records prior to the 1970s.
The bottom line: According to the 2012 IPCC report Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation (PDF):
Average tropical cyclone maximum wind speed is likely to increase, although increases may not occur in all ocean basins. It is likely that the global frequency of tropical cyclones will either decrease or remain essentially unchanged. (Section 3.4.4)
Because the inner core of a hurricane only spans a few miles, and the space between data points in a typical global climate model is wider than that, global models cannot yet produce realistic hurricanes.
Scientists are using several techniques to get around this roadblock. For example:
Several groups are attempting to produce global models that can directly depict hurricanes without an intervening small-scale model. The first such study, produced on Japan’s Earth Simulator computer and released in 2005, indicated more and stronger hurricanes in the Atlantic by later this century. Globally, it showed a 30% drop in the number of tropical cyclones, but a rise in the number and strength of the most intense hurricanes (view a PDF of this paper).
Meanwhile, research to boost the resolution of computer models continues. At NCAR, researchers are experimenting with the global Community Climate System Model (see Global model goes local), the Weather Research and Forecasting model (see Weather forecast goes global) and idealized models of hurricane structure that reveal details never depicted before, such as fine-scale turbulence in a hurricane eyewall.
Hurricane researcher William Gray (Colorado State University) and others have emphasized the role of a natural 20- to 40-year cycle in ocean temperature, dubbed the Atlantic Multidecadal Oscillation (see NOAA FAQ), in shaping hurricane activity across the North Atlantic. If only the AMO were considered, Atlantic hurricane counts would be expected to drop for several decades beginning in the 2010s or 2020s. However, analyses by other researchers, including Kevin Trenberth (see NCAR news release), have examined the AMO influences in light of global warming changes in ocean temperatures to conclude that the AMO is only a minor factor. Instead, they argue, the recent uptick in Atlantic hurricane activity is more closely related to overall global warming and thus may continue for decades to come.
Some scientists are exploring how interactions among different parts of the tropics will unfold in a warmer climate. For example, Gabriel Vecchi (NOAA) and Brian Soden (University of Miami) have found that wind shear over the North Atlantic may increase in the coming century, with the western tropical Pacific warming even more than the tropical Atlantic (see American Geophysical Union news release). This setup would help to inhibit Atlantic hurricanes. However, not all of the IPCC models examined by Vecchi and Soden show this pattern—a sign of the continued challenge in portraying tropical ocean circulation in global models.
One of the best ways to figure out how hurricanes operate is to fly right into them. Each year, skilled pilots steer research aircraft as close as safety allows to storms forming in the North Atlantic and Pacific. The planes are part of NOAA's hurricane hunter fleet and other research operations supported by the U.S. Navy, Air Force, universities, and research labs.
In 2008 NOAA adopted a technique called VORTRAC to provide detailed 3-D views of an approaching hurricane every six minutes. The three-dimensional views help to determine whether the storm is gathering strength as it nears land (see NCAR news release).
VORTRAC (Vortex Objective Radar Tracking and Circulation) was developed by researchers at NCAR and the Naval Research Laboratory. The technique requires no major new hardware, but instead relies on the existing NOAA network of Doppler radars along the Southeast coast to closely monitor hurricane winds. About 20 of these radars are scattered along the Gulf and Atlantic coastlines from Texas to Maine. Each radar can measure winds blowing toward or away from it, but no single radar could provide a 3-D picture of hurricane winds before now.
NCAR scientist Wen-Chau Lee and his collaborators developed a series of mathematical formulas that combine data from a single radar with general knowledge of Atlantic hurricane structure in order to map the approaching system's winds in three dimensions. The technique also infers the barometric pressure in the eye of the hurricane, a very reliable index of its strength.
Forecasters using VORTRAC can update information about a hurricane each time a NOAA Doppler radar scans the storm, which can be as often as about every six minutes. That could enable forecasters to monitor it for the critical 10-15 hours before landfall.
At UCAR’s COSMIC program, researchers have shown that computer models do a better job of predicting hurricanes when they’re fed extra data on how much moisture is in the air surrounding a developing storm. The data to help predict which potential hurricanes are most likely to develop come from SuomiNet, a UCAR-managed network of ground-based sensors that assess moisture by measuring tiny changes in GPS signals.
Meteorologists have hypothesized that ventilation, the injection of cooler and drier air into the core of a tropical cyclone, can significantly lessen the cyclone's intensity. To further understand this phenomenon, in 2012, NCAR scientist Brian Tang and colleague Kerry Emanuel (Massachusetts Institute of Technology) published work using a hurricane model to determine how much ventilation is needed to weaken a tropical cyclone and where this is most likely to happen within it.
Forecasters at NOAA’s National Hurricane Center make use of a wide array of computer models, including one called Hurricane WRF. Adopted by NOAA in 2007, Hurricane WRF was derived from the Weather Research and Forecasting model (WRF) and configured to reveal tropical storm activity in great detail. A multiagency effort, WRF is a next-generation computer model for weather prediction that can be used by both researchers and operational forecasters. See the roster of NOAA/NCEP model output to find Hurricane WRF forecasts of current storms.
NCAR has also refined research-oriented versions of WRF over the last few years, using them to study hurricanes and other forms of severe weather. To accurately depict the small but intense features within hurricanes, a special version of the Advanced Research WRF (ARW) sharpens the detail over targeted regions to 7.5 miles (12 kilometers) for forecasts out to 120 hours, with a resolution as fine as 0.8 mi (1.33 km) near hurricanes. Since 2005, NCAR's experimental forecasts of hurricane track and intensity have ranked among the most accurate of the computer models used by researchers and forecasters to predict the season's hurricanes. See the ARW modeling page for forecasts of current storms.
During August–October 2013, NCAR will be conducting 3- to 7-day hurricane forecasts using the experimental Model for Prediction Across Scales. When run on high-performance supercomputers, MPAS can simulate weather processes around the globe in very fine detail, capturing the development and evolution of individual cloud systems—a long-time goal of weather researchers. MPAS will help pave the way toward more detailed simulations by linking tropical storm and hurricane activity with global atmospheric conditions. The new simulations will provide better intensity forecasts than regional models at roughly the same cost for computing.
This side-by-side animation of Hurricane Katrina's path across the Gulf of Mexico compares the actual radar observations (left) with NCAR's Advanced Research WRF experimental forecast, issued 62 hours before landfall. In both frames, narrow rainbands can be seen pinwheeling counterclockwise into the storm's core. The model resolution for this animation was 12 kilometers (7.5 miles). The radar vantage point is stationary, on the Gulf Coast, while the model's viewpoint follows the hurricane itself. Click here or on the image to launch the animation in a new window.
Eyewall replacement during Katrina
About UTC: The Universal Time code is based on a 24-hour clock, with 0000 UTC equal to 7:00 p.m. Central Daylight Time. Convert other UT codes to U.S. time zones with this conversion chart from the U.S. Naval Observatory.
Track and Intensity Forecast Comparisons (Colorado State University)
NOAA/NWS Hurricane Storm Surge Probabilities (experimental forecast)
NOAA/NWS Glossary of Hurricane Terms (National Hurricane Center)
NOAA/AOML Hurricane Research Division FAQ (in English, en español, en français, auf Deutsch)
Hurricane categories: Category 1 through Category 5 differences
Hurricanes & Typhoons (NCAR Research)
Hurricane Forecast & Warning System: Social Science Research (NCAR Societal Impacts of Weather Program)
Hurricane & Typhoon Training Modules (UCAR MetEd, free registration required)
This extensive list of online training includes:
Hurricane Strike! interactive science & safety module (UCAR MetEd)
Look Out for Dangerous Weather: Hurricanes (UCAR Education & Outreach)
Hurricanes (Windows to the Universe)
Hurricane Features and Remote Sensing (UCAR COMET Program)
Hurricanes - Online Meteorology Guide (University of Illinois)
Updated May 2013
Backgrounders provide supplementary information and should not be considered comprehensive sources.
The University Corporation for Atmospheric Research manages the National Center for Atmospheric Research under sponsorship by the National Science Foundation. Any opinions, findings and conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.