Examining Rapidly Intensifying Hurricanes in the North Atlantic Basin
Christopher Friend
Greens Farms Academy
Hurricane forecasting, along with weather forecasting in general, has only begun to develop successful methods for the prediction of future weather events in the past 30 years through the introduction of GOES satellites and numerical weather models. The forecasting of hurricanes was notoriously difficult prior to the launching of the GOES satellites in the mid to late 1970s; before the launching of satellites, hurricane intensity could only be estimated by boat or plane reports, and most hurricane predications were based more so on climatology and less so on actual weather patterns (University of Illinois ) . These forecasting deficiencies were readily apparent during the 1938 Hurricane that made landfall in southern New England. In accord with climatology, the hurricane, moving westward just north of Puerto Rico, was predicted to hit Florida, and after the storm began to turn northward away from Florida, the hurricane was then predicted to turn due north, then northeastward and out to sea, a track favored by climatology. Instead, the storm continued traveling northward, and made a devastating landfall on Long Island; the lack of warning killed hundreds and caused millions in damage (Scotti, 2003) . Today, hurricane forecasting is significantly better, but challenges still remain in both track and intensity forecasting due not only to the large number of factors that influence both the track and intensity of hurricanes, but also because small changes in a hurricane’s intensity and track can have great effects on hurricane damage and casualties upon landfall. A one hundred mile forecasting error, which is close to the average 24 hour hurricane track forecasting error, equates to the difference between a storm’s making landfall at Jupiter, FL or Miami, FL. Rapid intensification of hurricanes can have grave effects on the impacts of a storm’s landfall, as well; 30kts of strengthening can be the difference between a few trees falling to hundreds of trees collapsing to the earth. The factors that affect the ability for tropical systems to rapidly intensify can be pared down into three main categories: atmospheric conditions, oceanic thermodynamics, and inner core characteristics of the tropical system. Within each of these main categories, there are individual characteristics such as wind shear, sea surface temperatures, and eyewall structure, most of which have been extensively analyzed (however, they are not always well understood in real-world applications). One of the less examined factors, storm size, is analyzed in this paper and found to have a notable effect on the potential for tropical systems to rapidly intensify. The goal in analyzing storm size is to try and improve upon a Rapid Intensification Index (RII) developed by leading hurricane researchers, by discussing the plausibility of adding storm size as a possible factor in the potential for rapid intensification.
Section A1: How Hurricanes Strengthen and Function
A hurricane is a very intense low pressure system that requires enormous amounts of heat, supportive atmospheric conditions, and organization within the storm in order for it to develop and strengthen[1]. The one characteristic of hurricanes that makes them unique amongst all cyclones is that hurricanes are warm-cored low pressure systems, which enable them to become some of the most powerful storms on the planet. A warm-core low is defined as a low that is “warmer at its center than at its periphery”; such lows generally form in the tropics and rely on the heat generated from the condensation of the tropical ocean water for their energy needs (American Meteorological Society ) . Cold-core lows, by contrast, are lows that are colder at their center than at their periphery; they generally form in the mid latitudes (nor’easters are a type of cold-core cyclone) and get their energy from pre-existing temperature differences within the atmosphere (Australian Government Bureau of Meteorology ) . Hurricanes, as warm-core lows, act like massive heat engines; to emphasize just how powerful hurricanes are, it is estimated that the average hurricane releases 6 x 10^14 watts of energy daily, approximately 200 times more energy than the daily energy production capacity of the entire world (Landsea) . These heat engines work when first, the sun evaporates the tropical ocean waters, which become water vapor and rise into the atmosphere. This water vapor condenses and forms thunderstorm complexes. These thunderstorm complexes lower the minimum sea level pressure (MSLP) and develop a low pressure system, which in turn strengthens the winds that travel inward towards the center of the storm as the developing atmospheric pressure differences act to ‘fill’ the low pressure system (Studio, 2007) . These winds carry energy in the form of water vapor towards the core of the storm, at which point these winds rise rapidly towards the stratosphere, carrying the energy upwards and creating intense convection in the form of hot towers. The stronger convection transports more energy to the tropical cyclone, creating a stronger storm; the MSLP is lowered, and wind speeds are increased, creating a positive feedback cycle, so long as environmental conditions remain favorable for strengthening. The warmer the ocean water, the more efficiently energy from the water can be transferred to the atmosphere and the intensifying tropical cyclone.
This is not to say that favorable atmospheric conditions and storm characteristics are not important factors. Oceans are just one part of the complex equation that constitutes hurricane intensity; however, oceanic heat an important factor in hurricane formation and intensity change, as warm ocean waters are the fuel for a hurricane’s heat engine. In a case study of the rapid intensification of Hurricane Opal, Bosart, et. al (2000) commented that
“Hurricane intensification can be broadly related to three physical processes: 1) large-scale environmental influences, 2) storm-scale internal dynamics, and 3) ocean–atmosphere interactions. These processes may act individually or collectively, and all three may be important at different times in a storm’s life cycle. Interactions between hurricanes and external larger-scale circulations have been shown to be important to hurricane intensity changes.” (Bosart, 2000)
In order for the heat engine that enables a tropical cyclone to function and intensify to work, ‘large scale environmental influences’ and ‘storm-scale internal dynamics’ cannot be overly inhibitive (high wind shear and a disrupted eyewall would both be overly inhibitive) towards strengthening. Further after a hurricane has formed, all three factors can play different roles in governing the intensity of a storm. Writing about large-scale characteristics of rapidly intensifying (RI) tropical cyclones, and the development of a methodology for predicting the probability of rapid intensification in hurricanes, Kaplan and DeMaria (2003) commented that “The fundamental impact of the ocean on [tropical cyclone] intensity has been stressed for many years…Previous studies have found that vertical shear plays a significant role in modulating [tropical cyclone] intensity” (Kaplan, 2003). How much of an influence these three factors have on hurricane intensity at different times is vital to understanding the potential for hurricanes to rapidly intensify.
A2 – Difficulties in Hurricane Intensity Forecasting
When compared with the great advancements made in hurricane track forecasting, hurricane intensity forecasting has greatly lagged in improvement in the past twenty years, mostly due to the complexity of the interaction of the above features that control hurricane intensity. Additionally, the intensity forecast is highly dependent on the track forecast. If the track forecast contains large errors, the storm could be located in a completely different environment than was predicted and the intensity forecast could be quite poor. Of the three factors that affected intensity, the only one that can alter the track of a tropical system is atmospheric influences, which change the direction of a storm. Neither heat of the ocean, nor characteristics of the storm itself will change the track that the storm takes (however, the intensity of a storm can have an effect on which layers of the atmosphere influence the tropical cyclone). These factors are generally well understood, and we owe an improvement in track forecasting to “better observations and much improved numerical models” (Emanuel, 1999). The factors that control the track of hurricanes also play a role in many other types of weather phenomenon, such as the track of mid-latitude cyclones and large scale weather patterns. Much effort has gone into producing computer models that can effectively track all cyclones; because tropical cyclone intensity is affected by more unique factors, less research has gone into modeling tropical cyclone intensity. This is largely due to the fact that large scale numerical models do not handle mesoscale events (such as hurricane intensity) very well, and small scale models for analyzing hurricane intensity have been only recently developed. As well, track forecasts can influence intensity forecasting. Oceanic conditions and atmospheric conditions are different everywhere; it would be very hard to predict the intensity of a hurricane if you had no idea know where it was going (Emanuel, 1999). The ability to make intensity forecasts lags far behind our skill with track forecasting. Emanuel (1999) states that the best hurricane intensity forecasts up to 1999 were “statistically based.” (Emanuel, 1999) In a paper evaluating long term trends in intensity forecasting, DeMaria et. al (2007) noted that
“To put these results in perspective, the intensity forecasts were compared to the track forecasts for the same data sample. The skill was comparable at 12 h, but the track forecasts were 2 to 5 times more skillful by 72 h, with the largest ratio in the west Pacific. The track and intensity forecast error trends for the two-decade period were also compared. Results showed that the percentage track forecast improvements were almost an order of magnitude larger than those for intensity, indicating that intensity forecasting still has a very long way to go.” (DeMaria, 2007).
The same paper concludes that major improvements to tropical cyclone modeling systems have yet to make a significant improvement in actual intensity forecasting, which influences the claims of other papers that we have little to no skill at making intensity forecasts.
Graphs from DeMaria (2007) and the National
Hurricane Center comparing improvements in track
and intensity forecasting over the past decades.
The need for further study intensity forecasting is readily apparent, as our best forecasting models rarely outperform simple statistical models in forecasting intensity. RI forecasting is a specialized part of intensity forecasting and is both especially important and difficult to predict. The importance of RI forecasting is that large swings in intensity, especially when they occur close to landfall, can have huge effects on the damage caused by a hurricane. Being able to predict where a hurricane will make landfall is a fantastic skill that society now has, but it is nearly worthless if it is unknown whether that landfalling hurricane will have winds of 70kts or 120kts. RI is difficult to forecast for the same reasons that normal hurricane intensity is difficult to forecast; in the words of Kaplan and DeMaria (2003)
“While the forecasting of TC intensity change in general has been quite difficult, the forecasting of rapid intensification (RI) has been particularly challenging…The inability to forecast RI is consistent with our limited understanding of TC intensity change in general. In the past, researchers have typically examined the role that the ocean, inner-core processes, and environmental interactions play in tropical cyclone intensity change. Although some intensity change studies have examined the importance of all three of these effects, most have focused on only one of these three areas” (Kaplan, 2003)
At the time the paper was published Kaplan and DeMaria stated that to date, very few intensity change studies had tried to put together all the puzzle pieces of intensity forecasting. Hurricane intensity forecasting today is where hurricane track forecasting was nearly two to three decades ago (that is, best predicted by statistics rather than with complex forecast models). It is necessary to continue researching hurricane intensity to improve intensity forecasting ability; increased knowledge of why hurricanes strengthen and weaken will make clearer why certain storms undergo rapid intensification at certain times.
B - The Atmospheric Factors that Influence Hurricane Intensity
Multiple researchers have already stated the three conditions that control hurricane intensity: oceanic heat, inner-storm conditions, and the state of the atmosphere. The basics of hurricane formation and factors that control intensity have been reviewed in advance of this paper’s analysis of how each of these three factors can change hurricane intensity and how their effect on hurricane intensity can further our knowledge about the propensity for storms to rapidly intensify. Although the atmosphere does not supply any of the heat to fuel tropical cyclones, certain atmospheric conditions, such as relative humidity and wind shear, can greatly influence the intensity of hurricanes and tropical storms. One of the most important factors that changes hurricane intensity is wind shear. Wind shear is defined as the amount of change in wind direction or speed with increasing altitude. (University of Illinois ) Wind shear alters the exterior of a hurricane by ‘shearing’ away the tallest thunderstorms away from the surface low pressure system. Convection can be sheared away from the actual low pressure center, leaving behind a low pressure system without much in the way of rain bands or convective clouds; the result is a skeleton of a tropical cyclone, abandoned by its heat engine. Strong wind shear rarely, if ever, promotes hurricane intensification; most studies have proven that there is less wind shear in the presence of intensifying storm systems. Merrill (1988) noticed that “vertical shear is less for intensifying hurricanes”, a point echoed in a later paper by Kaplan and DeMaria (2003). Wind shear can be especially damaging close to the center of the storm if the shear manages to disrupt the inner core of the hurricane.
A visual presentation of the data described below, taken from Frank (2001).
Frank and Ritchie (2001) in their report on wind shear and its effect on hurricanes say that “Recent modeling studies of tropical cyclone–like vortices in idealized environments…indicate that even modest amounts of vertical wind shear in the core region of a tropical cyclone can have strong effects on the asymmetric structure of the eyewall region, where the strongest winds and heaviest rains occur”, indicating the very destructive effects of shear on a hurricane (Frank, 2001). In that same paper, Frank and Ritchie modeled the effect that different magnitudes of shear had on an intensifying hurricane. Wind shear of less than 5m/s had little effect on the intensification process of the hurricane, while wind shear of 5m/s only had an effect on the intensification process of the hurricane 39 hours after the shear was induced. As the shear increased, so did the effects; 10m/s shear began affecting intensification about 12 to 15 hours after the shear was induced, while 15m/s shear began affecting the storm within three hours after the shear was induced. Their findings indicate that shear at low levels (5m/s or less) tends to have little effect on the intensification of a hurricane, anything around 5m/s of shear has a minimal effect on hurricane intensification, and when shear begins to approach 10m/s limit a hurricane’s intensification.
In fact, when ranking the eight most important factors to determine the potential of a hurricane to rapidly intensify in the North Atlantic, Kaplan and DeMaria (2008) ranked vertical wind shear, from 200mb to 850mb, as the second most important factor, nearly seven times as important as relative humidity (Kaplan, 2008). The only factor ranked higher on their scale than vertical wind shear was 200mb divergence. 200mb divergence, first linked to its ability to strengthen cyclones by Gibbs (1955), increases precipitation and convection, which in turn increases the potential of a hurricane to rapidly intensifying by utilizing the expanding the power of the heat engine (Gibbs, 1955). Increased divergence is one of the most important atmospheric factors for intensification in hurricanes for the opposite reason that wind shear is an important factor; wind shear inhibits intensification, while 200mb divergence promotes intensification. Low shear and increased divergence at 200mb, according to Kaplan and DeMaria (2008), provide for a ripe breeding ground for rapid intensification in North Atlantic hurricanes (the factors that affect hurricanes in the Pacific are different from those in North America, and this paper focuses on hurricanes in the Atlantic because Pacific hurricanes are unlikely to make landfall and have an impact). Another atmospheric factor, albeit with less impact, is the relative humidity in the atmosphere surrounding the hurricane. Higher relative humidity (which indicates a moister atmosphere) means there is more water vapor in the air, leading to greater convection and condensation. A higher relative humidity would increase the power of the heat engine. A lower relative humidity, indicating drier conditions, would inhibit the heat engine of the hurricane by limiting the available water vapor to the hurricane and causing downdrafts around the storm. These three factors, wind shear, 200mb divergence, and relative humidity, are all powerful atmospheric conditions that affect not only hurricane intensity, but also the potential for hurricanes to rapidly intensify. According to Kaplan and DeMaria (2008), wind shear and 200mb divergence are the two most important factors when considering the probability that a hurricane in the North Atlantic could rapidly intensify. The atmosphere, however, remains only part of the puzzle; the total effect that all atmospheric factors have can vary along with the effect of the ocean and storm characteristics.
C - Oceanic Factors that Influence Hurricane Intensity
Hurricanes survive on the heat that they are able to obtain from warm ocean waters, and the amount of heat available for intensification is determined by two main factors: sea surface temperature (SST) and oceanic heat content. The sea surface temperature is merely the temperature of the ocean; an accepted temperature typically needed for the support of tropical cyclones is 26C, although this is not a definite value (as storms have formed with sea surface temperatures of less than 26C) (AOML) . No matter the exact temperature, very warm waters are needed to support a warm-core system, and these temperatures are generally only found in the tropics (hence, tropical cyclones). Tropical Cyclone Heat Potential (TCHP) is a measure of oceanic heat content and is defined as a “measure of the integrated vertical temperature from the sea surface to the depth of the 26°C isotherm.” (AOML) Sea surface temperatures have long been tied to the maximum potential intensity of hurricanes; papers such as DeMaria and Kaplan (1994) have shown that the SST provides a sort of limit on hurricane intensity, based on the warmth of the ocean water. After this upper limit, however, the applications of SST for a possible intensity indicator are murky. Some storms can dissipate over 29C waters while other storms can peak in intensity while traversing over ocean temperatures of 26C, and so SSTs are not completely reliable for intensity prediction. Furthermore, until recently, most research has not taken into effect the possibility the storms themselves can cause changes in ocean temperatures.
Schade and Emmanuel (1999) state that “Nevertheless, the effect of this cooling on the intensity of hurricanes has received surprisingly little attention. In numerical hurricane models, the ocean was typically treated as a constant sea surface temperature (SST) boundary condition and the effect of the chosen SST field on hurricane intensity was investigated.” (Schade, 1999) This finding is a significant problem and flaw within both the models and their creators. It is well known that hurricanes affect their environment, especially by absorbing the heat of the ocean, transferring from the ocean to the storm. Strong winds create high waves that can alter the upper layers of the ocean, and hurricanes that travel too slowly can cause upwelling, bringing colder water to the surface that will invariably weaken the storm. All tropical cyclones will affect the ocean conditions around them because hurricanes are not a point on a map and tend to be hundreds of miles across. Rain bands and winds on the edges of hurricanes will affect the temperature and heat content of the ocean before the center of the storm reaches those points; the question is not so much a ‘will it affect the ocean’, but rather a ‘how much’. Later in the same paper, Schade and Emmanuel (1999) outline certain conditions or aspects of a storm that would increase “SST feedback” , or a lowering of SST before the arrival of the core of a hurricane. The paper goes on to give a list of SST feedback mechanisms, mentioning that “Deviations from the reference values change the SST feedback factor in a physically intuitive way: a stronger negative feedback effect occurs when: the oceanic mixed layer is thinner, the storm moves slower, the intensity potential is larger, the storm is of greater horizontal size, the storm occurs at lower latitudes, the thermal stratification below the oceanic mixed layer is stronger, and the relative humidity in the atmospheric boundary layer is higher.” (Schade, 1999) All of these factors lead to a greater negative SST feedback, casuing colder SSTs and therefore less heat for the hurricane to obtain, meaning less chance of rapid intensification. Emanuel (1999) goes on to highlight this in a subsequent paper he wrote, saying that
“Most of the research literature on hurricane intensity focuses on the pre-storm sea surface temperature and certain properties of the atmospheric environment, such as the vertical shear of the horizontal wind and dynamical features such as disturbances in the upper troposphere. This remains so, even though it is well known that hurricanes alter the surface temperature of the ocean over which they pass and that a mere 2.5 K decrease in ocean surface temperature near the core of the storm would suffice to shut down energy production entirely.” (Emanuel, 1999)
Emanuel’s comments reveal the fragile nature of the heat engine of a tropical cyclone since a mere decrease of 2.5C in surface temperature can change the state of a hurricane from rapidly intensifying to rapidly decaying. At this time in the paper appears our need to understand TCHP. The deeper and warmer the ocean layer, the more difficult it is for a hurricane to upwell colder waters from below to the surface. A high TCHP value indicates that the ocean has enough heat to support a rapid intensification, and that it is unlikely that the hurricane will be able to lower the water temperatures enough to cause a self-induced weakening. Two very famous cases of where a high TCHP helped a storm rapidly intensify and then keep its intensity were Hurricanes Camille and Katrina (Emanuel, 1999). Both storms rapidly intensified over a warm Gulf of Mexico eddy, also known as the Gulf Loop, which has a very deep layer of warm water (Elb, 2005). In any other place, the rapid intensification of these two storms may not have even happened, and if rapid intensification had occurred, the storms would have quickly weakened after their RI cycles and not have caused the serious damage that they did. TCHP is an important factor for rapid intensification, as it indicates where SST feedback is less likely to be significant.
Emanuel makes the point that it is preposterous that most of the influence of a hurricane’s intensity is derived from the upper atmosphere when very small changes in ocean temperature can have massive impacts on how hurricanes function. Emanuel is, for the most part, correct in his belief that the ocean plays a larger role than the atmosphere with regards to hurricane intensity. In support of Emanuel’s claim of the superiority of oceans over atmosphere with regards to intensity forecasting is that hurricanes undergoing rapid intensification tend to have a very sharp peak, with intensity rising rapidly (rapid intensification), followed by a rapid decline (as the hurricane quickly uses up available oceanic heat). Thus, rapid intensification in hurricanes is only available because of warm ocean temperatures. The atmosphere and the hurricane itself must be in conditions that are favorable for strengthening, but the rapid intensification would not be possible without the heat of the ocean. Oceanic heat, as the fuel for the hurricane’s heat engine, is necessary for formation of tropical cyclones, their function, and most important, rapid intensification.
D - Hurricane features that influence hurricane intensity
The organizational features within a hurricane, especially within the eyewall structure, have a great influence on the potential for hurricanes to intensify. The eyewall, probably the most important part of a hurricane, is also one of the least-understood and most dynamic parts of a hurricane. The eyewall of a hurricane has many vortices in it, which are created by the difference in wind speed between the eye and the eyewall (which can reach well over 100kts). Recently, NOAA has been experimenting with assimilating Doppler radar data from its hurricane hunter aircraft into one if its next generation forecast models to help that model better represent the inner core of the storm. However, this effort is fairly new and will not be fully operational for a few years. Until then, fake vortices are sometimes placed in the center of hurricanes in some numerical models because it is difficult to get an accurate measurement of these vortices within a hurricane. Placing fake vortices in a model in lieu of the actual vortices causes problems with predicting intensity as vortices tend to get underestimated in the models and the actual vortices in the hurricane will act differently than the fake vortices implanted in the model. In a case study on Hurricane Opal, Bosart et. al (2000) commented on eyewall structures, saying “Internal dynamical processes associated with eyewall convection have been the focus of several studies of hurricane intensity change. One unresolved theoretical issue is how axisymmetric and asymmetric processes in the hurricane contribute to intensity changes. Both axisymmetric and asymmetric convective structures are observed in hurricane eyewalls.” (Bosart, 2000) In simpler terms, Bosart et. al are commenting on how incredibly complex the eyewall of a hurricane is, and how the little information we have on eyewalls makes forecasting intensity changes due to subtle eyewall structures nearly impossible at this time. Eyewall structures remain an elusive part of hurricane intensity forecasting because the way that vortices in eyewalls behave are seemingly random and chaotic. Kossin and Schubert (2001) developed a model for examining eyewall “mesovortices” and rapid pressure falls because of these vortices. In one run of their model, a group of mesovortices would combine to form a single vortex, dropping the pressure of the hurricane by nine millibars. While this experiment shows that vortices can have a major effect on the intensity of hurricanes, the depth of that effect is still unknown. Inner-core disruptions, however, can be incredibly harmful to a hurricane attempting to intensify. An example of the effect of eyewall disruptions is the case of Hurricane Gustav in 2008. Hurricane Gustav hit western Cuba as a Category 4 Hurricane (nearly Category 5), but by the time it had crossed over the mountains and back into the Gulf of Mexico, the eyewall was nowhere near as organized as it had been prior to landfall. Although oceanic and atmospheric conditions were ripe for re-intensification, Gustav failed to strengthen because of an open eyewall. (Center, 2009) If a storm is not organized and lacks concentrated convection near its center, then rapid intensification is all but impossible.
II – The Influence of Storm Size on Rapid Intensification; A – Rapid Intensification
Recently, efforts have been made to move past the factors that affect just hurricane intensity and into the specific factors that affect rapid intensification. As mentioned previously, the need to further research into the causes of rapid intensification is caused by the great danger that rapidly intensifying storm systems can cause. The potential damage caused by a storm undergoing RI is shown in the landfall of Hurricane Charley (2004) near Fort Meyers, FL. Hurricane Charley strengthened from 90kts about six hours prior to landfall to 130kts at landfall, in addition to taking a slight jog east and impacting Port Charlotte, as opposed to the forecasted Tampa Bay. This strengthening and track change left most of the affected areas unprepared for a near-Category 5 storm as the maximum predicted intensity only six hours before landfall was 100kts (Center, 2005). Therefore, it is extremely important that we continue our research into what causes rapid intensification to occur as the effect that rapid intensification can have (especially close to shore) is devastating. In a 2006 conference paper updating their 2003 research, Kaplan and DeMaria identify seven different atmospheric, oceanic, and storm-based factors that they believe affect the ability of a storm to rapidly intensify. These seven factors are
“sea-surface temperature, the difference between the current and empirically derived maximum potential intensity, the 850-200 mb (~1,500-12,000m) vertical shear and the 850-700 mb (~1,500-3mrelative humidity both evaluated for the annulus between 200-800 km radius, and the previous 12-h intensity change. In addition, two inner-core predictors, the percentage of the area from 50-200 km radius with GOES infra-red brightness temperatures <-30º C and the standard deviation of the infra-red brightness temperatures from 100-300 km radius were also utilized.” (Kaplan, 2006)
and most of these factors have already been discussed in this paper. In his updated presentation, Kaplan showed how his Rapid Intensification Index (RII) was a significant improvement over both climatology and forecasts by the National Hurricane Center (NHC); however, his index still remains well (around 30%) below target goals of 90% for rapid intensification forecasting. The goal of the rest of this paper is to try and improve upon the RII developed by Kaplan and DeMaria by debating the merits of adding storm size as a possible factor in the potential for rapid intensification.
B – Storm Size and its Effect on Rapid Intensification
Storm size should be considered because there are a number of reasons that, hypothetically, a smaller storm should have a higher potential to rapidly intensify. One reason is that a smaller storm requires less energy to function and intensify. A hurricane needs heat (gathered from warm ocean temperatures) to function; naturally, it would follow that a smaller hurricane needs less energy to function and intensify than a larger hurricane. Energy is released through cloud formation and condensation, falling as rain. As previously mentioned, this amount of energy is massive, significantly larger than worldwide energy production. However, a smaller hurricane would require less energy, and so the amount of heat needed from the ocean would be significantly smaller for a small hurricane versus a larger hurricane. A smaller hurricane would mean that less heat would be necessary to fuel a possible rapid intensification. Another reason that a smaller storm has a higher chance to rapidly intensify is that a small hurricane is less likely to cause upwelling or large waves. Upwelling and large waves are likely to mix the warmer and colder layers of the ocean, pushing the warmer layer of the ocean below the surface of the ocean. A smaller hurricane, with a smaller wind field, creates smaller waves and less upwelling, which keeps the heat content of the ocean high ahead of a hurricane, so SST feedback is much less likely. Although hypothetically, there should be a relationship between a smaller storm and a larger potential for a storm to rapidly intensify, research must be done to find if this relationship holds up in reality, as a smaller storm can also be more vulnerable to environmental influences, such as shear or dry air.
Storm size can be defined in a number of different ways. One way to define the size of a storm is to measure the radius of the 64kt, 50kt, and 34kt wind fields from the center of the tropical system. A different method measures the radius of the outermost closed isobar, indicating the size of the hurricane’s pressure field, which is useful in determining the outer limits of the real effects of the hurricanes. The main method used in this study was the mean radius of 64kt, 50kt, and 34kt wind fields, because of the connection between winds and upwelling. The storms analyzed ranged from 2003 to 2008; this is because the size of wind radii was not accurately recorded, if recorded at all prior to the 2003 Hurricane Season. The record prior to 2003 is spotty at best – it is highly unlikely, for instance, that Hurricane Kyle (2002) had the same perfect radius (50 miles in all directions) of gale force (50kts) winds for 60 straight hours. Now, this lack of reliable data is not true for all storms and some storms have very detailed wind radii reports, but in an effort to reduce error, all storms prior to 2003 were not included in this study. The record from 2003 to 2008 is not perfect, but is the best available data. This hurricane record also has a significant amount of data; 100 tropical storms formed between 2003 and 2008, 50 tropical storms intensified to reach hurricane strength, and 25 hurricanes intensified into major hurricanes.[2] Of these 100 named storm systems, 32 underwent rapid intensification (defined as a strengthening of 30kts or more in a 24 hour period) at some period during their life cycle. This sample size is large enough to give us an idea of whether storm size affects the potential for a hurricane to rapidly intensify.
There are a few ways that storm size can be used to detect the propensity for cyclones to undergo future rapid intensification. One such method is the usage of storm size (the radius of 34kt winds) at the time a storm becomes a named storm[3]. Of the one hundred storms in the data set, the average initial tropical storm force wind radius was 40.2nm[4]. Tropical cyclones that underwent RI had an average initial 34kt wind radius was only 29.883 nautical miles (nm), while tropical cyclones that did not undergo RI had an average initial 34kt wind radius was 45.055nm. Put another way, cyclones that did not have a RI cycle had an initial 34kt radius that was, on average, 50% larger than cyclones that did have a RI cycle. This difference represents a large and real difference between wind radii of cyclones that undergo RI and storms that do not undergo RI, and indicates that small storm size is correlated with the potential for hurricanes to rapidly intensify. Taking the cases of storms that underwent RI, most of the cases seemed to be tightly clustered near the mean. Only 15.6% of storms that rapidly intensified had wind radii of less than 20nm, with the smallest being 7.5nm. Similarly, only 12.5% of cyclones that rapidly intensified had wind radii greater than 50nm, with the largest wind radii being 63.75nm. Only 14.7% of non-RI cyclones percentage of storms had a wind radius of less than 20nm, which means that the percentage of cyclones under 20nm was similar for non-RI and RI cases. By contrast, 26.5% of storms that did not rapidly intensify had wind radii greater than 50nm, with a maximum of 186.25nm (nearly three times as large as the maximum wind radii for rapidly intensifying system), and 8.8% of non-RI cases had wind radii of greater than 100nm. This data implies that a large initial wind radius is not conductive to rapid intensification, while a smaller initial wind radius does not necessarily have an effect on the potential for a cyclone to rapidly intensify.
It is easy to quantify and compare the initial wind radii between non-RI and RI systems because every tropical cyclones has an initial wind radius. The storm size at the beginning of rapid intensification in RI cases cannot be compared to storm size at any specific time in the lifecycle of non-RI tropical cyclones. Therefore, we are forced to research the effect of storm size on the potential for storms to rapidly intensify based on the given RI cases. Instead, the 32 available RI storms to see if any patterns emerge about the life cycle of rapidly intensifying hurricanes with regards to storm size. Some of the tropical cyclones analyzed rapidly intensified for more than 24 hours, such as Dean (2005), rapidly intensified more than once, such as Charley (2004). Of the 32 storms, there are a total of 39 cycles of rapid intensification, with some cases lasting more than 24 hours. The strength of the tropical cyclones at the onset of rapid intensification was varied; of the 39 cases examined, five were tropical depressions, fifteen were tropical storms, and nineteen were hurricanes. Tropical depressions have maximum winds of less than 34kts (they have no tropical storm force wind radii) and cannot be considered in our the examination of the initial wind speed radius of RI storms. Of the remaining 34 cases that were tropical storms or stronger, the average initial radius of the tropical storm force winds (>34kts) was equal to 79.96nm. While a value of nearly 80nm seems large in light of our previous data the number is actually small. The average maximum 34kt radius of each cyclone that underwent rapid intensification was 143.304nm, nearly double the radii of the tropical storm force winds at the time of rapid intensification. As with previous data, most of the values are closely clustered near the average; only six cases of the total 34 (or, 17.6%) that had a 34kt wind radius of greater than 100nm at the beginning of rapid intensification. This indicates that it is more difficult for a tropical cyclone to rapidly intensify with a tropical storm force wind radius of greater than 100nm.
At a wind radius of 50kt, the patterns shown at the 34kt wind radius only were strengthened. Of the 29 cases that involved storms with maximum wind speeds of greater than or equal to 50kts, the average wind speed radius was 34.612nm. For all storms that underwent RI, the maximum 50kt wind radius during the storm’s lifetime was 75.281nm, nearly double the 50kt wind radius at the time when the tropical cyclone underwent rapid intensification. Within the 29 RI cases, only four cases (13.8%) had a wind radius of greater than 50nm. It is interesting to note that all four of these cases that had wind radii of greater than 50nm occurred on a hurricane’s second or third rapid intensification; the largest 50kt wind radius of a storm that underwent RI for the first time was only 40nm. This indicates two things; first, that the 50kt wind radius must be very small for a hurricane to be able to rapidly intensify for the first time (this data set had a max of 40nm), and second, that hurricanes that rapidly intensify for a second or third time generally have a larger 50kt wind radius than storms that have not undergone RI. This disparity between wind radius in storms (and storm size) that have not rapidly intensified and storms that have rapidly intensified indicates that wind radius is not as important a factor to storms that have already had a cycle of rapid intensification. In fact, the average 50kt wind radius for the 6 second and third rapid intensifications in our study is 49.821nm, while the average 50kt wind radius for the 23 first rapid intensification is only 28.478nm. Given that the average 50kt wind speed radius and storm size for second and third rapid intensifications is nearly double the radius of first rapid intensifications, it can be postulated that wind speed radius has the largest effect on first RI cycles, and less of an effect on second and third RI cycles.
The wind radius for hurricane force winds (64kt or stronger) in rapidly intensifying cyclones, further reveals the connections between storm size and rapid intensification. There were 19 cases of rapid intensification that had an intensity of greater than or equal to 64kts when they began rapidly intensifying. These 19 cases had an average 64kt wind radius of 17.039nm at the time of rapid intensification. For all storms that underwent RI at one point, the maximum 64kt wind radius during a cyclone’s lifetime was 43.125nm, more than double the 64kt wind radius at the time the TC underwent RI. Of these 19 cases, 6 were second or third rapid intensifications, while 13 were first rapid intensifications. Like with the 50kt wind radius, the size of the 64kt wind radius seems to have the most effect on a storms ability to rapidly intensify for the first time. The 6 second or third rapid intensifications had an average 64kt wind radius of 27.708nm with a maximum of 37.5nm, while the 13 first rapid intensifications had an average 64kt wind radius of 10.469nm with a maximum of 25nm. The difference between first RI and second or third RI for the 64kt wind radius is larger than the 50kt wind radius; the second or third RI had an average 64kt wind radius nearly three times as large as the first RI 64kt wind radius! This data supports the findings of the 34kt and 50kt wind radii; that wind radii and storm size appears to have influence on the potential for storms to rapidly intensify, and this effect is seen greatest on storms undergoing their first RI, rather than their second or third RI.
The wind speed radius data above is worthless unless it can be used to help better predict rapid intensification in tropical systems. Given the data analyzed, wind speed radius can be best employed to reduce false positives in RI forecasting. The data indicates certain conditions that increase the potential for tropical systems to rapidly intensify. Named storms that initially have 34kt wind radii of greater than 60nm are very unlikely to undergo RI during their lifetimes, so predicting any storm that initially had a 34kt wind radii of greater than 60nm should be carefully considered. Similarly, predicting a first rapid intensification when the 34kt, 50kt, and 64kt wind radii are over 100nm, 40nm, and 20nm, respectively, is a dangerous forecast, as a very small percentage of tropical cyclones are able to rapidly intensify for the first time with wind radii greater than the above values. Second and third rapid intensifications for a tropical cyclone, according to the data set, are less tied to wind radii for their potential to rapidly intensify. Examining the wind radii of a tropical system is a helpful factor in determining when a tropical system has potential (or no potential) to undergo rapid intensification.
Hurricanes are heat engines, a quote oft-repeated in this paper. However, it is most likely the most poignant point of this entire paper, as everything about hurricane intensity relates back to the heat engine, which is both powered by the heat of the ocean and convection. If the heat engine of the hurricane is disrupted in any way, whether through cooler ocean temperatures, a shearing away of convection, or eyewall disruptions, the intensity of the hurricane is negatively affected. The better meteorology is able to understand how the heat engine of a hurricane functions and interacts within its environment, the greater skill with which we will be able to forecast hurricane intensity. With more data, more research, and more time, meteorology will one day be able to predict hurricane intensity with the same skill that hurricane tracks are predicted with now. Especially important in the future will be the prediction of rapid intensification, which, if can be more accurately predicted in the future, could save hundreds of lives through better forecasting. Hopefully, the end of this paper adds to a little bit to the research regarding rapid intensification, adding on a distinct set of numbers for storm size that would allow for rapid intensification. As the tiny bits of research add up, one day meteorology will be able to predict rapid intensification with accuracy and precision, giving society a semblance of power over an otherwise uncontrollable phenomenon.
Bibliography
American Meteorological Society . (n.d.). AMS Glossary . Retrieved January 1, 2010, from http://amsglossary.allenpress.com/glossary/search?id=warm-low1
Australian Government Bureau of Meteorology . (n.d.). BoM - Tropical Cyclone Frequently Asked Questions . Retrieved January 1, 2010, from http://www.bom.gov.au/weather/wa/cyclone/about/faq/faq_def_6.shtml
Barber, B. (2009). Sandbox for Mapping, Viewing, and Analyzing Atlantic Hurricane Data . Retrieved October 14, 2009, from http://db.hellohelp.net/hurdat_storms/storm_search.php
Booth, M. A., T. Lambert, J. Blackerby, and R. L. Elsberry, 2006: Accuracy of tropical cyclone intensity forecasts in the North Pacific and Atlantic. AMS 27th Conference on Hurricanes and Tropical Meteorology. 24-28 April, Monterey, CA. [available online at http://ams.confex.com/ams/pdfpapers/108805.pdf]
Bosart, L. F., Velden, C. S., Bracken, W. E., Molinari, J., & Black, P. G. (2000). Environmental Influences on the Rapid Intensification of Hurricane Opal (1995). Monthly Weather Review , 322-352.
Center, N.H. (2005, January 5). Tropical Cyclone Report – Hurricane Charley. Retrieved January 1, 2010, from http://www.nhc.noaa.gov/pdf/TCR-AL032004_Charley.pdf
Center, N. H. (2009, January 22). Tropical Cyclone Report - Hurricane Gustav. Retrieved January 1, 2010, from http://www.nhc.noaa.gov/pdf/TCR-AL072008_Gustav.pdf
DeMaria, M. (2004, January 29). Description of the Extended Best Track File. Retrieved January 1, 2010, from ftp://rammftp.cira.colostate.edu/demaria/ebtrk/readme16.txt
DeMaria, M., J.A Knaff, and C. Sampson, 2007: Evaluation of long-term trends in tropical cyclone intensity forecasts. Meteorol. Atmos. Phys., 97, 19-28.
DeMaria, M., & Kaplan, J. (1994). Sea Surface Temperature and the Maximum Intensity of Atlantic Tropical Cyclones . Journal of Climate , 1324-1334
Elsberry, R. T.D.B Lambert, and M. A. Boothe, 2007: Accuracy of Atlantic and Eastern North Pacific tropical cyclone intensity forecast guidance, Wea. Forecasting, 22, 747-762.
Emanuel, K. A. (1999). Thermodynamic Control of Hurricane Intensity. Nature , 665-669.
Erb, M.P. (2006). A Case Study of Hurricane Katrina: Rapid Intensification in the Gulf of Mexico. Preprints,20th National Conference on Undergraduate Research, Asheville NC.
Frank, W. M., and E. A. Ritchie, 2001: Effects of vertical wind shear on the intensity and structure of numerically simulated hurricanes. Mon. Wea. Rev., 129, 2249-2269.
Gibbs, W. (1955). 200mb Divergence Associated with Rapid and Intense Cyclogenesis. Central Meteorological Bureau, Melborne , 1-18.
Jarvinen, B.R., C.J. Neumann, and M.A.S. Davis, 1984: A tropical cyclone data tape for the North Atlantic basin, 1886-1983: Contents, limitations, and uses. NOAA Tech. Memo. NWS. NHC 22, Miami, FL, 21 pp. [Available from National Technical Information Service, U.S. Department of Commerce, 5285 Port Royal Road, Springfield, VA 22151.]
Kaplan, J. (Performer). (2008). A revised rapid intensification index for the Atlantic and E. Pacific basins. Orlando, FL, USA.
Kaplan, J., & DeMaria, M. (2003). Large-Scale Characteristics of Rapidly Intensifying Tropical Cyclones in the North Atlantic Basin. Weather and Forecasting , 1093–1108.
Kaplan, J., & DeMaria, M. (2006). Estimating the Likelihood of Rapid Intensification in the Atlantic and E. Pacific Basins using SHIPS Model Data. AMS Conference , 1-4.
Kossin, J. and W. H. Schubert, 2001: Mesovortices, polygonal flow patterns, and rapid pressure falls in hurricane-like vortices. J. Atmos. Sci., 58, 1079-1090.
Landsea, C. (n.d.). HRD: FAQs. Retrieved January 1, 2010, from http://www.aoml.noaa.gov/hrd/tcfaq/D7.html
Merrill, R. T. (1988). Enviromental Influences on Hurricane Intensification. Journal of the Atmospheric Sciences , 1678-1687.
Molinari, and, D. Vallaro, 1990: External influences on hurricane intensity. Part II:
Vertical structure and response of the hurricane vortex. J. Atmos. Sci., 47, 1902-1918.
National Hurricane Center . (2009 ). NHC Archives of Hurricane Seasons . Retrieved October 14, 2009, from http://www.nhc.noaa.gov/pastall.shtml
National Hurricane Center. (2009, January 1 ). Extended Best Track File . Retrieved January 1, 2010, from ftp://rammftp.cira.colostate.edu/demaria/ebtrk/ebtrk_atlc.txt
National Oceanic and Atmospheric Administration. (n.d.). 48 Hour Track Errors - "Early" Guidance . Retrieved January 1, 2010, from http://www.nhc.noaa.gov/verification/figs/Early_model_ATL_trk_error_trend.gif
National Oceanic and Atmospheric Administration. (2008 ). Easy to Read HURDAT 2008 . Retrieved October 15, 2009, from http://www.aoml.noaa.gov/hrd/hurdat/easyread-2008.html
National Oceanic and Atmospheric Administration . (2009). Daily Climate Composits. Retrieved October 12, 2009, from http://www.esrl.noaa.gov/psd/data/composites/day/
Schade, L. R., & Emanuel, K. A. (1999). The Ocean’s Effect on the Intensity of Tropical Cyclones. Journal of the Atmospheric Sciences , 642-651.
Scotti, R. A. (2003). Sudden Sea: The Great Hurricane of 1938. Little, Brown.
University of Illinois . (n.d.). History of GOES . Retrieved January 15, 2010, from GOES-1 through GOES-7: http://ww2010.atmos.uiuc.edu/%28Gl%29/guides/rs/sat/goes/oldg.rxml
[1] For the rest of this paper, the use of the term ‘hurricane’ is used loosely to refer to all tropical cyclones.
[2] As a side note, I’d like to point out how perfect the data set is from 2003 to 2008. 100 tropical storms, half became hurricanes, and half of the hurricanes became major hurricanes.
[3] The database the provides the data for the following pages of analysis, is known as the Extended Best Track File, and can be found at ftp://rammftp.cira.colostate.edu/demaria/ebtrk/ebtrk_atlc.txt
[4] For this study, wind radius was calculated by averaging the radius of each wind speed in each of the four quadrants of the storm.
