 |
 |
|
|
|
|||
|
|||||
 |
|||||||||||||||||
|
Printable version By Rob Mann and Jason Moreland As of May 25, 2008 1. Introduction This is the sixth Atlantic Basin hurricane season forecast issued by the Independent Weather Information Center. The primary parameters that are known to affect tropical cyclones activity in the Atlantic Basin have been closely analyzed throughout this off-season. Analogous pattern and parameter combinations derived from 60 years of historical climate data have also been examined to further justify the anticipated synoptic setup over the northern hemisphere. The aforementioned methodology has been utilized to forecast not only seasonal, but also monthly, regional, and landfalling tropical cyclone activity. While an overwhelming consensus of members within the meteorological community dismisses the idea of providing accurate landfall forecasts months in advance, it is the long-term goal of IWIC to prove that it is in fact possible. Countless hours of research and resulting statistical correlations combined with theory have led the authors to believe that several recent findings are significant. Verification reports for the successful 2006 and 2007 forecasts are available. Bare in mind this forecast is experimental and unofficial. Hence, IWIC is not liable for one's actions taken based on the information being presented. Please read the disclaimer. 2. Summer-Fall El Niño Southern Oscillation The El Niño Southern Oscillation (ENSO) is most notably characterized by significant variations in sea surface temperature anomalies (SSTAs) in the equatorial Pacific. These alternations impact atmospheric circulations and thus can have a drastic influence on the northern hemisphere summertime pattern. Current Status La Niña conditions became established in the fall of 2007. Although this onset was somewhat later than the climatological average, it strengthened at a quick pace. Weekly SSTAs in Region 3.4 (120-170ºW) dipped as low as -2.2ºC in January of this year. Such low equatorial Pacific SSTAs persisted the following month, resulting in one of the strongest February La Niña episodes on record. Since then, however, the event has been on a weakening trend. Region 3.4 SSTAs have recently risen to near the weak La Niña threshold, while SSTAs further east have actually been consistently above average over the past several weeks. Moreover, temperature anomalies below the surface of the equatorial Pacific have drastically warmed over the past several months. The expansive, intense cool anomaly pool that was present a few months ago has been replaced by patchy areas of cool anomalies near the surface with a moderate warm anomaly region further below. Yet another measure of ENSO magnitude is the SOI (Southern Oscillation Index), which is calculated using sea level pressure differences between Tahiti and Darwin, Australia. The SOI peaked in February with an astoundingly high monthly reading of 21.3. Like the surface and subsurface temperature anomalies in the equatorial Pacific, the SOI has been moderating, with the latest monthly value of 4.5 in April. The overall trend towards neutral values has continued this month. It is apparent that by all accounts the La Niña is fading. Climatology It is worth investigating whether the event is dissipating for good or will make a comeback later this summer. One vital component of long-range ENSO forecasting is climatology. Upon examining previous years, a few correlations pertaining to the current ENSO situation become evident. The most definitive relationship per climatology is between La Niña persistence and solar flux. For the purpose of this study, a La Niña event is classified when the -0.5ºC threshold is met for five consecutive three-month running seasons. Since 1948, there have been nine such La Niña events that have formed outside of winter, not counting the current one. Five of these events formed during a period of high solar flux (1949, 1970, 1973, 1988, and 1998). Four of these events persisted through the following year, the exception being the 1988 episode. On the other hand, out of the other four events that formed when solar flux was low (1954, 1964, 1984, and 1995), three of them dissipated the next year. 1954 was the exception of this group. While one might be tempted to end the research at this point, it is prudent to further explain the exceptions in order to justify the validity of the correlation. The 1954 event likely did not dissipate the next year due to the rapidly rising solar flux that took place in 1955, a unique situation among the sampled years. In attempt to understand the other exception, the 1988 event, it was noticed that that and the 1973 high solar flux La Niña both formed during a period of west Quasi-Biennial Oscillation (QBO) conditions (see section 4). It is interesting to mention that the 1973 event nearly dimished below the La Niña threshold at one point in 1974, after reaching record strength that previous winter. The rate of weakening in 1950, 1971, and 1999 was less extreme, and those are all events that formed during east QBO. Similar differences are seen in the La Niñas that formed during low solar flux. The events of 1954 and 1984 both developed during east QBO. The 1954 event was the one that persisted through the following year, while ENSO conditions remained strongly cool bias by the end of 1985. The 1995 episode on the other hand weakened quicker and 1965 saw the emergence of a strong El Niño. Thus, it can be postulated that solar flux and QBO at the time of a La Niña initiation play a crucial role in its sustainability the succeeding year. While there have been studies on possible ENSO-solar flux-QBO connections, the physical mechanism for such a relationship is poorly understand at this time. Nonetheless, the statistical analysis above is considered strong enough evidence for utilization in the current situation. The 2007 La Niña formed during a period of low solar flux and strongly east QBO. Solar activity is showing little signs of increasing over the next several months, and by this point in 1955 it was already rising. Thus, it is presumed that the exceptional 1954 event is not analogous to the 2007 event. This leaves 1964, 1984, and 1995 as the analog ENSO events based on solar flux. Considering QBO as well, which argues for somewhat more persistence, the sole analog becomes 1984. It should come to no surprise that this event so far is evolving rather similarly to what occurred in 1985, with the largest difference being the peak winter strength. This implies a continued steady moderation of the La Niña, with a transition into generally neutral conditions early in the summer. The possibility of an El Niño later this year was considered. However, there have only been two moderate-strong La Niña winters that were succeeded by an El Niño (1951 and 1976). 1951 was already under neutral ENSO conditions by April. The La Niña in place at the start of 1976 weakened at a more similar rate to the current one, but the subsurface temperature anomalies were significantly warmer during the spring with a strong warm anomaly pool already established over almost the entire length of the equatorial Pacific. Moreover, both of those La Niña events were well over a year in age. This brings up another point: it is unusual for a La Niña to form one year and be replaced by an El Niño a year later. The only such case of this is with the 1964 La Niña, and by this time the El Niño was within a month of establishing itself in the equatorial Pacific SSTAs. A strong warm pool centered below the surface eastern equatorial Pacific was also evident in the spring that year. Additionally, weak La Niña-esque conditions were present at the beginning of 1963, 1968, and 2006, years that later saw the initiation of an El Niño. However, not only were these much weaker than what was present this past winter, but they dissipated quicker and by this point the warm subsurface temperature anomalies were broader and focused further east. Thus, it is very unlikely that an El Niño will form at any point in the remainder of the year. 2008 ENSO Forecast With all the aforementioned data taken into account, the currently declining La Niña is expected to give way to neutral ENSO conditions in the next month or so. Equatorial Pacific SSTAs should still lean slightly cooler than average. All indications are that this setup will continue through the the end of the year. ENSO will therefore have little overall impact on Atlantic Basin hurricane activity in 2008. 3. Summer-Fall Atlantic Multidecadal Oscillation The Atlantic Multidecadal Oscillation (AMO) is a long-term pattern of SSTA variability in the North Atlantic Ocean. This factor can exert a very strong influence on seasonal tropical cyclone activity via modifying the SSTAs over the tropical Atlantic regions. Since 1995, the AMO has been in a predominantly warm cycle and will certainly remain in such during 2008. However, the AMO does occasionally fluctuate to the opposing phase of the long-term cycle, so a somewhat deeper analysis is necessary. The AMO weakened to just slightly warmer than average during the 2007 Atlantic Basin hurricane season, but has since rebounded. It has not returned to the much warmer level consistently observed from 2003 through 2006, however. This can be partially attributed to Pacific pattern evolution over the past year. Past studies indicate that the Pacific Decadal Oscillation (PDO) is positively correlated with the AMO, more so with the AMO lagging a few months. It is likely that the PDO switched to a long-term cool phase last year (see section 4). If this is in fact the case, then the current warm cycle of the AMO has probably already peaked. Another more direct link is noted between SSTAs in the equatorial Pacific, particularly the western section, and SSTAs in the tropical Atlantic. There is at least a 0.6 correlation between these two parameters with the tropical Atlantic SSTAs set to an approximately six month lag. Since the ENSO is a short-term variable oscillation related to the PDO, and tropical Atlantic SSTAs are often a more variable representation of the AMO, this correlation is not surprising when considering the aforementioned PDO-AMO connection. Indeed, tropical Atlantic SSTAs have likewise cooled over the past year and are currently running just slightly above average. An area of below average SSTAs has persisted just east of the Caribbean Sea, while above average SSTAs remain present off the west coast of Africa. This overall configuration is expected to continue over the next few months. The cool SSTAs may increase in coverage though as the region reacts to the cooling ENSO conditions that took place about six months ago. During the middle of the hurricane season, SSTAs in the area are expected to slightly rebound due to the La Niña weakening that began in March. Such evolution of tropical Atlantic SSTAs will take place in tandem with a more consistently moderate warm AMO. Although the AMO is less extreme than it was a few years ago, it will still have an enhancing influence on tropical cyclone activity. 4. Other Factors The combination of neutral ENSO conditions in the equatorial Pacific and a slightly warm AMO suggests that an active Atlantic Basin hurricane season is likely. However, there are several other parameters that have been considered in this forecast. Analysis of these factors combined with the knowledge of the ENSO and AMO have yielded a clear picture of the general pattern that is already setting up for the hurricane season. The differing influences of each individual factor will be thoroughly discussed in further sections. Spring Northern Hemisphere Pattern Analysis of the 60 year climatology database indicates that the springtime 500mb geopotential heights over North America and the Atlantic Basin have a lag affect on large-scale atmospheric conditions that evolve during the hurricane season. Geopotential heights in the mid-levels of the atmosphere help locate anomalous areas of high a low pressure and are thus crucial in examining both the formation and steering environment across the Atlantic Basin. Spring North Atlantic Oscillation The North Atlantic Oscillation (NAO) is an atmospheric pattern over the North Atlantic that is defined by differences in sea level pressure between the Icelandic Low and Azores High. The NAO is extremely variable on a monthly and even weekly basis, making it a difficult parameter to forecast in the short-term. However, an in-depth reanalysis of the last 60 years indicates that spring NAO values have a delayed influence over the synoptic conditions that progress through the hurricane season. The spring 2008 NAO remained near neutral, with pronounced ridging over Greenland and downstream troughing just east of Iceland. Winter El Niño Southern Oscillation Winter ENSO episodes are known to have a lag influence on large-scale atmospheric conditions well after they dissipate. Therefore, the winter ENSO signal is given weight when analog springtime patterns are selected. As aforementioned, strong La Niña conditions were present during the past winter. Summer-Fall Quasi-Biennial Oscillation The QBO is a periodic variation in the direction of stratospheric winds across the deep tropics. The two phases, easterly and westerly, generally last from 12 to 16 months, with the easterly phase often having a slightly longer duration. The QBO switched to its west phase at the end of last winter and is currently strengthening in magnitude. The QBO will continue to rise throughout this year, with the next transition not occurring until sometime in 2009. Summer-Fall Pacific Decadal Oscillation The PDO is a multi-decadal pattern of high and low pressure systems in the north Pacific Ocean, sometimes seen as a longer-term version of ENSO. Unlike ENSO, the PDO index is calculated by spatially averaged monthly SSTAs over the northern, as opposed to equatorial, portions of the Pacific Ocean. The PDO was in a long-term negative cycle from about 1944 to 1977, and remained predominantly positive until last year. Since the previous cycle lasted a little over 30 years, and it has been 30 years since the last shift, it is possible, if not probable, that the PDO has just switched back to its negative phase. 5. Activity By Region The Atlantic Basin has been divided into four regions to help portray the anticipated tropical cyclone formation and tracking tendencies of 2008. Tropical Atlantic (south of 20ºN, east of 60ºW) The key factors regarding tropical cyclogenesis in the tropical Atlantic are the NAO, ENSO, AMO, and QBO. It has been hypothesized by the authors that spring values of the NAO play an important role in tropical cyclone activity across this area of the basin, with positive NAO having the enhancing role. Spring 2008 featured generally neutral NAO conditions which will have little influence on activity in the tropical Atlantic. ENSO will likewise be neutral during most of the hurricane season and thus will similarly be of little consequence regarding development in this region. This leaves the AMO and QBO. It is logically and climatologically stated that warm AMO conditions increase the frequency and strength of tropical cyclones in the tropical Atlantic via increasing the regional SSTAs. The SSTAs in the area are expected to run slightly above average. The QBO became westerly this spring and will remain so through the 2008 hurricane season. Studies show that westerly QBO supports increased activity in the tropical Atlantic. With a warm AMO and westerly QBO both playing a favorable role, it is deduced that tropical cyclone activity will be above average in this region. To further back this idea, previous years were investigated. It was found that every neutral NAO year has featured average to well above average activity when combined with neutral ENSO conditions and a warm AMO. Analog years with similar conditions are 1953, 1961, 1980, 1996, and 2003. The aforementioned analog set still left a wide range of variability. A closer examination revealed that the 2008 spring pattern best favors 1961 and 1996, which happened to be the most active years out of the analog set. The other years observed above average spring heights over the tropical Atlantic itself, which likely reduced the amplitude of tropical waves traversing the area. Such a scenario is unlikely in 2008 given the southward-shifted weakness over the eastern Atlantic this past spring. With that considered, five to six named storms, three to four hurricanes, and one or two major hurricanes are forecast to originate from the the Tropical Atlantic. Subtropical Atlantic (north of 20ºN, east of 80ºW) All of the named storms that develop in the tropical Atlantic are expected to cross into the subtropical Atlantic at some point during their lives. The few that do not strengthen into hurricanes in the tropical Atlantic are expected to do so once north of 20ºN. This essentially guarantees five to six named storms, all five to six being hurricanes, and two to three major hurricanes in the subtropical Atlantic this season. However, there is not expected to be much tropical cyclogenesis in the region. Climatologically, this is a favored area during cool AMO or warm ENSO seasons, when extratropical cyclones frequently pass through and give way to incipient systems via cutoff lows. Neither situation is the case this year. The few generally neutral ENSO and warm AMO seasons that did support high activity in this area include 1969, 1980, 1981, 2000, 2001, and 2005. All of these years except 1980 and 2000 observed strongly negative NAO conditions during the spring. This greatly lowered heights across the subtropical Atlantic, which, in tandem with warm SSTAs, enabled nontropical lows to strengthen and develop warm cores. The other two years, 1980 and 2000, similarly saw a region of low heights across much of the central subtropical Atlantic. This past spring was generally dominated by ridging north of 20ºN, rather unlike the aforementioned years. Thus, only one to two named storms are forecast to originate in the subtropical Atlantic this season. Caribbean Sea With an above average season in the deep Atlantic in the cards, it is necessary to know whether any of these tropical cyclones will track into the Caribbean Sea. The height pattern immediately to the north of the Caribbean Sea would favor a storm to track westward across the sea, but there is the question of whether or not a storm will exist at such a low latitude in the first place. Indeed, with low deep Atlantic heights and a consequently northward shifted intertropical convergence zone, it is reasonable to argue against delayed Atlantic tropical cyclogenesis near the eastern Caribbean Sea. Such development typically occurs when a parameter results in southward-displaced east Atlantic ridging, such as an El Niño, cool AMO, negative NAO, or a combination of any of those indices. As mentioned above, none of this is present during 2008. Tropical waves will thus be permitted to develop further east in the deep Atlantic. Under favorable conditions and influence from an east Atlantic weakness, it is expected that they will all gain too much latitude early on to enter the Caribbean Sea. Thus, no tropical cyclones are expected to move westward through the Caribbean Sea in 2008. There is still the concern for in-situ Caribbean Sea development. Tropical cyclogenesis cannot be ruled out in the extreme northwestern Caribbean Sea during August and September. The real interest in this area is during October and November, when several major hurricanes have originated from this area in the past. Such late in-situ major hurricanes are most common during a La Niña, although some have developed during neutral ENSO conditions, as seen in 1948, 1952, 1996, 2001, and 2005. 1948 and 1952 observed ridges of high pressure over the southeast United States with a trough extending into the Caribbean Sea from the central Atlantic. 1996, 2001, and 2005 observed a deep trough concentrated near the east coast of United States. In contrast, there are no troughs that reach the Caribbean Sea in 2008. Given the fact that the height pattern is vastly different from the above years, and are instead more like years such as 1980 and 1989, no major hurricanes are forecast to develop in the Caribbean Sea. Having said that, one to two named storms are expected to develop in the western Caribbean Sea during the 2008 season. This is owing to the fact that in neutral NAO seasons, it is common for some healthy tropical waves to hold off on development until in the western half of the basin. Such tropical waves will be steered under the western Atlantic ridge and encounter favorable conditions upon reaching the western Caribbean Sea and Gulf of Mexico. Gulf of Mexico As mentioned in the Caribbean Sea sub-section, no tropical cyclones from the deep Atlantic are expected to move through the Caribbean Sea and later into the Gulf of Mexico. It is also unlikely that a deep Atlantic tropical cyclone enters the Gulf of Mexico after hitting Florida given the spring pattern. With no long-tracking tropical cyclone expected to reach the Gulf of Mexico, this leaves the late-season Caribbean Sea system mentioned earlier and in-situ development as the only ways to achieve Gulf of Mexico activity this season. In-situ tropical cyclone activity in the Gulf of Mexico is primarily influenced by the spring height pattern surrounding the region. During spring 2008, a weakness developed over the midwest United States with the western periphery of a ridge extending into the northeastern gulf. This pattern is typically very conducive for Gulf of Mexico or northwestern Caribbean Sea development. However, there are other points to consider that will prevent 2008 from featuring a violently above average gulf season. The warm AMO and neutral ENSO ensures that there will not be a plethora of baroclinically-induced tropical cyclones in the Gulf of Mexico. Moreover, the neutral to high Caribbean heights suggest that tropical waves will not significantly amplify as they pass through the area. This will put a limit on the number of tropical waves that are able to develop in the Gulf of Mexico. While not a major factor, it should be mentioned that seasons following a La Niña or neutral ENSO winter average slightly less gulf activity than seasons following an El Niño winter. With a somewhat favorable setup in the forecast, it is worth examining the prospects of in-situ major hurricane formation in the Gulf of Mexico. Since 1950, there have been 12 tropical cyclones develop and become major hurricanes in the northwest Caribbean Sea or Gulf of Mexico, notwithstanding late season systems. Four of these occurred under a pattern of low gulf heights and high Caribbean Sea heights, highly unlike what is the case for 2008. Two of them were in years with a trough dipping into the eastern Gulf of Mexico itself, also not analogous to the present situation. The remaining years (1950, 1953, 1961, 1964, 1977, and 1985) were then analyzed. 1953 and 1985 are similar in that they had a ridge centered over the Gulf coast with its eastern flank giving way to a trough over the Maritimes. No other years in the dataset had this setup. 1950 and 1961 are alike in that their ridges, while not completely extending into the Gulf of Mexico, are accompanied by low heights in the Caribbean Sea. A few other years had a similar overall theme, but the ridges were displaced too far northeast. 1964 and 1977 are the only two years in the above sample that did not have low Caribbean heights. Both of their ridges were centered in the vicinity of the Florida peninsula, with their western flanks rounding off in the Gulf of Mexico. This is important to consider because spring 2008 likewise does not have low Caribbean Sea heights. However, the spring 2008 ridge is centered further northeast than those seen in 1964 and 1977, with the southwestern flank only reaching one corner of the Gulf of Mexico. Two other years without low Caribbean Sea heights that are somewhat akin to this situation are 2002 and 2003, neither of which supported a major hurricane form in the Gulf of Mexico. 1980 had a similar ridge displaced further south, and it also did not observe an in-situ gulf major hurricane. The difference in outcome between 1964 and 1977 versus 1980, 2002, and 2003 indicates that the positioning of the ridge is crucial and argues against major hurricane activity in the Gulf of Mexico this year. Taking all of that into account, three to four named storms, one to two hurricanes, and no major hurricanes are forecast to develop or traverse the Gulf of Mexico. One to two of these are expected to enter the region from the western Caribbean Sea. 6. Activity By Bimonthly Periods The Atlantic Basin hurricane season officially spans six months, beginning on June 1 and ending on November 30. The season has been divided into three bimonthly periods to help portray the anticipated tropical cyclone formation and timing tendencies of 2008. June-July The pattern during the first two months of the hurricane season will be somewhat favorable for tropical cyclogenesis. Since winter ENSO has been shown to have a strong impact on June and July tropical cyclone activity, previous years that observed winter La Niña conditions were found. To narrow the list, years with generally neutral ENSO conditions by the summer combined with warm AMO were selected (1951, 1989, 1996, 2000, and 2001). 1951 and 2001 do not compare well to 2008 because they both had negative NAO conditions during spring. Even with NAO factored, there is still some discrepancy among the remaining years. This was solved by analyzing and comparing the 2008 spring pattern with the years above. First, troughing over the midwest United States is expected to persist through the summer. The base of this trough in tandem with the western flank of the subtropical ridge will help lower pressures in the Gulf of Mexico. A similar trough, albeit deeper and further east, occurred in 1996, while a less defined weakness further west occurred in 1989. Both of these years saw a tropical storm develop under the trough in June. 2000 on the other hand did not have troughing established in the spring anywhere near the climatologically favored western portion of the Atlantic basin. Unsurprisingly, no tropical cyclones developed in this area. There are also some differences in early activity in the tropical Atlantic. No named storms formed in the tropical Atlantic in 2000, a weak tropical storm occurred in 1989, and a full fledged hurricane plus an eastern Caribbean Sea hurricane formed in 1996. This variance can be attributed to SSTAs in the tropical Atlantic. 1996 had moderately warm SSTAs in this region, whereas SSTAs in 1989 and especially 2000 barely reached above normal before August. The tropical Atlantic SSTAs have stayed generally above normal over the past few months. The overall pattern that has been in place and will continue to remain in place favors warm SSTAs in the area. However, no significant increase in anomalies are expected because of the six month lagged correlation with ENSO. Taking into account the weakness extending in the Gulf of Mexico from the midwest United States and warm tropical Atlantic SSTAs, two to three named storms and up to one hurricane are forecast to develop before August. One of the named storms is expected to form in the Gulf of Mexico, with the other originating in the tropical Atlantic. This is somewhat above the long-term average. August-September Available data suggests that tropical cyclone activity in the peak two months of the season will not deviate far from the long-term average. As mentioned earlier, five to six named storms are forecast to develop in the tropical Atlantic. Most of these should occur during August and September when vertical wind shear is at its lowest. Additionally, a storm or two should originate in the Gulf of Mexico or northwestern Caribbean Sea. On the other hand, the subtropical Atlantic and the remainder of the Caribbean Sea are expected to remain tranquil during this period, as mentioned in their respective sub-sections. This idea is further supported by the neutral spring NAO and expected neutral ENSO conditions during the season. There were nine years in the dataset that had this combination (1953, 1959, 1960, 1961, 1966, 1980, 1985, 1996, and 2003), and the resulted amount of named storms in August and September was no more than eight in all of the cases. In fact, the only neutral NAO years that observed over eight named storms in August and September were 1955, 1984, and 1988. All of these seasons were under La Niña conditions and moreover had very different spring patterns that allowed high activity in certain regions of the basin. Given that 2008 is more like the other neutral NAO seasons and the already established expected amount of activity in each region, a total of six to seven named storms, five to six hurricanes, and two to three major hurricanes are forecast in August and September. Most of these tropical cyclones will originate in the tropical Atlantic. This is slightly above the long-term average. October-November The final two months of the 2008 season are expected to be somewhat underwhelming in tropical cyclone activity. One parameter to take into consideration is the spring pattern. Spring 2008 did not feature any troughs digging into the Caribbean Sea. This is expected to limit the number of named storms that develop in this region after September, as many are often initiated after a trough sweeps in and lowers the sea level pressures. Other neutral ENSO and warm AMO years that similarly lacked spring troughing into the Caribbean Sea include 1980, 1989, 1990, and 2003. All of these years saw one named storm form in or near the western Caribbean Sea during October or November. The possibility of late development in the tropical Atlantic was also considered. This occurred in the following neutral ENSO years: 1952, 1953, 1967, 1978, 1990, 2001, and 2003. Such late development in the cool AMO seasons of 1967 and 1978 was made possible by strong ridging that became established in the spring over the eastern Atlantic. This feature delayed and generally hindered the onset of tropical cyclogenesis through the peak months, but also kept shear lower than average as the season winded down. Such a feature is not present this year. The remaining years all observed very warm SSTAs over the Atlantic Basin during the last quarter of the calendar year, due to either a warm bias ENSO, persistent negative NAO, or a combination of both. While ENSO will gradually warm throughout the season, it is not expected to lean on the warm side. Furthermore, NAO averaged near neutral during the spring. All of this ensures that the tropical Atlantic will be finished with development after September. The height pattern, neutral NAO, and moderate AMO all strongly argue against a plethora of named cyclones in the subtropical Atlantic as the season closes. However, the weakness centered over the eastern Atlantic will be broad enough to possibly allow such a storm or two to form, as observed in 1990 and 2003. With that said, a total of two to three named storms, with one possibly becoming a hurricane are forecast to develop after September. This is slightly below the long-term average. 7. Landfall Activity By Coastal Areas Since the landfall forecasts issued in 2006 and 2007 were a great success, it has been decided that similar forecasts will be issued again this year. The coastal areas of the Atlantic Basin have been divided into different sections to help portray the tropical cyclone landfalling distribution. Central America and Yucatán Peninsula As mentioned in the Caribbean Sea sub-section, no tropical cyclones are forecast to traverse the Caribbean Sea from the deep Atlantic. This leaves any Central America or Yucatán Peninsula threat to western Caribbean Sea-originating systems. Fortunately for this region, none of these tropical cyclones are expected to grow into major hurricanes at any point, and probably will not become hurricanes while in the Caribbean Sea. Moreover, they are all expected to move into the Gulf of Mexico. On such a track, it would not be surprising to see a tropical storm landfall along the Yucatán Peninsula. However, no hurricanes are expected to hit Central America or the Yucatán Peninsula this season. Mexico and Texas No tropical cyclone landfalls in this area are supported by the research conducted during the offseason. The springtime pattern over North America featured a ridge over much of the western Atlantic, southeast United States, and eastern Gulf of Mexico. This is a good indication that strong southeasterly flow will be present over the entire Gulf of Mexico during the season. The presence of this feature alone does not dismiss the risk of tropical cyclone landfalls in Texas. However, low heights were observed near the Great Lakes area. Had this region of low heights been centered closer to the Midwest or Rockies, the southeast flow over the gulf would be directed more toward Texas. However, the dominant steering flow between the two height areas is expected to setup over the central Gulf coast, thus keeping all tropical activity east of Texas and Mexico. Louisiana, Mississippi, Alabama, and Florida Panhandle The central Gulf coast of the United States is one of the regions at highest risk for hurricane landfalls. As stated in prior sections, all tropical cyclones that traverse Gulf of Mexico waters this season are expected to either originate in the gulf itself or in the western Caribbean Sea. In other words, all threats posed toward the Gulf coast are anticipated to be homegrown. Moreover, two to three tropical cyclones are forecast to form in the Gulf of Mexico. An additional one to two tropical cyclones are forecast to enter the gulf from the Caribbean Sea. Basic arithmetic tells one that the central Gulf coast could be staring down the barrel of more than three tropical cyclones if the steering pattern permits it. Unfortunately, the springtime pattern indicates that tropical cyclones in the Gulf of Mexico will be aimed toward this region. An area of positive heights extends in the west Atlantic from 60ºW westward through the southeast United States. This feature alone would suggest that anything that does form in the western Caribbean Sea will not recurve over Cuba or the Florida peninsula. The final nail in the coffin is an area of low heights centered near the Great Lakes and up into the lower Hudson Bay. With no southwest United States ridge extending into the Gulf of Mexico, this weakness will prevent any tropical cyclones from continuing westward and striking Texas or Mexico. The path of least resistance is thus set up from Louisiana through the Florida Panhandle. While several tropical cyclones are forecast to impact this area, there is a bright side. As stated in the Gulf of Mexico sub-section, there is not enough evidence that would support a forecast of major hurricane formation in the Gulf of Mexico or northwest Caribbean Sea. Thus, three to four named storms and one to two hurricanes, perhaps as strong as moderate intensity, are expected to strike the central Gulf coast. West Florida Peninsula The western coast of the Florida peninsula is not expected to take a direct strike from any tropical cyclones. The springtime ridging noted over the western Atlantic and Florida itself suggests that no tropical cyclones are likely to recurve into the peninsula out of the western Caribbean Sea or Gulf of Mexico. Any storm that forms relatively close to the peninsula is much more likely to continue northward into the Florida panhandle or locations a bit further west. With tropical cyclone development forecast in the western Caribbean Sea after September, a more detailed analysis of the west Florida risk was warranted, given that it is a favored landfall location late in the season. Two years with similar west Atlantic subtropical ridges that observed a west Florida landfall in October are 1950 and 1964. However, the discrepancy lies in the western flank of the ridges. In 1950, the ridge just barely extended past the Florida peninsula, with most of the eastern Gulf of Mexico hence in southwesterly flow. The ridge in 1964 extended further west, but the flow in the Gulf of Mexico was similarly under strong southwesterly flow from the flank of the ridge. The position and alignment of the 2008 subtropical ridge is somewhat different. It is situated further north than what was observed in 1950 and 1964, and has a much wider western flank. These characteristics put the majority of the Gulf of Mexico under southeasterly flow. No year that had a west Florida landfall even after September observed such a height pattern. East Florida and Georgia One significant correlation regarding east Florida landfalls involves the spring NAO and winter ENSO. Years with winter El Niño conditions prior to the hurricane season average more east Florida landfalls, especially during positive or negative spring NAO. 2008 falls alongside 1950, 1955, 1956, 1963, 1968, 1974, 1984, 1985, 1996, and 1999 with a winter La Niña and neutral spring NAO. Of these years, only 1984 had an east Florida hit, and it was from an in-situ west Atlantic tropical storm. Such development, as mentioned in the subtropical Atlantic sub-section, is highly unlikely in 2008. Based on this correlation, one might be tempted to conclude that east Florida will escape a direct hit this season. However, the fact that this relationship is not perfect for any combination of ENSO and NAO suggests that other factors must be considered. Many of the above years featured a trough offshore the United States east coast, which would logically deflect any westward-moving tropical cyclones poleward and away from Florida. The only years above that did not have this pattern were 1955, 1963, and 1974. 1963 had a strong central Atlantic trough and El Niño during the season, which caused every deep Atlantic system to recurve early. 1974 observed a strongly cool AMO which severely limited the development east of the Caribbean. That combined with the fact that the ridge was focused west of Florida kept the deep Atlantic tropical cyclones south of Florida. 1955 is more difficult to rule out as a 2008 analog. Although it had a core of high heights west of Florida, there was also evidently a higher height corridor along and just offshore the United States east coast. It is no surprise, given the pattern, that three hurricanes hit North Carolina that season. With no real differences between the patterns in 1955 and 2008, it is a wonder how Florida could get hit this year even when it was not during 1955. The answer lies in the idea that deep Atlantic development will begin before August. As mentioned in the June-July sub-section, conditions will favor a tropical cyclone or two to form east of the Caribbean during or even before July. Under optimal conditions, such tropical cyclones will continue westward and generally follow where the spring height pattern dictates. One key difference with earlier systems compared to storms during the peak of the season is that since they occur in the middle of summer, the guiding subtropical ridge is usually stronger. Moreover, the mid-latitude troughs are weaker. During the heart of the season, the mid-latitude weakness over Canada will play a strong role in hurricanes approaching the western Atlantic, but this is not expected to be the case before then. While south Florida should remain safe, the above data suggests one named storm or perhaps hurricane landfall anywhere from central Florida to Georgia this season. The Carolinas The spring heights pattern favors tropical cyclones to hit the Carolinas this season. The dominant feature regarding hits in this region will be the west Atlantic ridge. The exact alignment and position of the ridge alone actually places east Florida through mid-South Carolina at the most risk, with east-southeasterly flow extending from the northeastern Caribbean islands all the way to the southeast United States coastline. This steering flow, without any mid-latitude influence, may indeed spell trouble for that region during July and early August, as mentioned above in the east Florida sub-section. However, at some point in August, this will cease to be the only real steering factor in the west Atlantic. One key feature during spring 2008 was a trough concentrated over the southern Hudson Bay area in Canada. The coverage of the weakness actually extends as far as the eastern flank of the subtropical ridge itself. This weakness is important to consider as it is known that tropical cyclones always follow the path of least resistance. In this case, the path of least resistance is equivalent to reaching the lowest, reasonably proximate heights. As summer fades into fall before the peak of the season, this feature will become increasingly dominant over the subtropical ridge by pulling any nearby hurricane northwestward, generally towards the Carolinas. With heights suggesting an increased risk for Carolina hits, curiosity drives one to determine just how many tropical cyclones will strike the region. Previous warm AMO years with a subtropical ridge in a position that favored a deep-Atlantic tropical cyclone to hit the United States include 1955, 1989, 1996, 1998, and 2003. During August and September, between 60 and 71 percent of the deep Atlantic-originating hurricanes bypassed the east Atlantic weakness to be steered by the west Atlantic ridge. In the cases of 1996 and 1998, the western flank of the ridge did not extend far enough to allow every storm to hit the United States. In 1989 and 2003, the mid-latitude low heights were concentrated directly north of the ridge, which allowed some hurricanes to similarly curve in the western Atlantic. Neither situation was the case in 1955, and to no surprise, every hurricane that made it under the subtropical ridge hit the North American continent. A similar scenario is expected to occur in 2008. Four to five hurricanes are forecast to form from deep Atlantic tropical cyclones during August and September. Considering the above range regarding the ratio of deep Atlantic hurricanes bypassing the east Atlantic trough, it is probable that up to two of these hurricanes will make it to the western Atlantic and eventually hit the Carolinas. One of these could be a major hurricane upon landfall. Mid-Atlantic and Northeast United States In spite of an increased risk to the southeast United States coastline, the states north of the Carolinas appear to be safe from a direct hit in 2008. In the 60-year dataset, the following years observed a deep Atlantic-originating named storm hit the northeast states: 1954, 1960, 1961, 1971, 1985, 1996, and 1999. Interestingly, cool bias to La Niña conditions were present in all of these hurricane seasons, with 1954 the only year that had even a warm bias ENSO the previous winter. Additionally, all of the years saw neutral or negative NAO conditions during the spring. However, since there were many other La Niña and neutral to negative NAO seasons that did not have a long-tracking northeast United States strike, further analysis of the spring pattern was warranted. With this, another correlation was discovered: all of the aforementioned years saw a deep trough extending from the Canadian Maritimes. In the case of 1961 and 1985, the trough was situated further east with its base over the western Atlantic, whereas the remaining years had the trough focused along the United States eastern seaboard. This is the type of pattern one would expect to yield an increased risk to the northeast states, as the troughs would act to curve hurricanes due northward once they near or hit the southeast states. Such a pattern is not at all the case this year, with spring 2008 featuring ridging from the eastern United States and into the western Atlantic. Thus, no deep Atlantic tropical cyclone is forecast to directly hit the northeast states. Every once in a while, an in-situ west Atlantic tropical cyclone hits the northeast states, the most famous examples occurring in 1976 and 1991. While the spring heights in these years are actually closer to 2008 than the years mentioned above, the synoptic pattern was very different. Cool AMO and El Niño conditions were observed and thus made such development possible. Other years with this caliber of tropical cyclones strikes were under similarly favorable conditions for formation to occur via non-tropical means off the southeast coast. With a warm AMO, neutral ENSO, and high heights across the western Atlantic this year, such development is unlikely to occur during the 2008 season. Atlantic Canada It was discussed in the Carolinas sub-section that deep Atlantic-originating hurricanes will either recurve in the east Atlantic or head under the subtropical ridge and hit the southeast United States. Any hurricane that curves in the east Atlantic will be too far east to directly hit the Canadian Maritimes. It is possible that a hurricane that hits the Carolinas will round the subtropical ridge and head northeastward back into the Atlantic. Such a storm could later hit the Canadian coastline. With that said, no more than one named storm is forecast to strike Atlantic Canada this season. Puerto Rico, Leeward Islands, and Windward Islands With an above average season in the deep Atlantic in the cards, there is legit concern for the eastern Caribbean islands. However, available data indicates that these islands will escape, perhaps just barely, a major hurricane hit. The warm AMO years with heights comparable to 2008 in the subtropical Atlantic are 1950, 1955, 1961, 1979, 1989, 1996, 1998, 2003, and 2004. A large portion of these years indeed observed major hurricane strikes in the eastern Caribbean islands. However, the subtropical ridges in years with a major hurricane hit extended as far east as 40ºW. On the other hand, in the above years when major hurricanes stayed north and east of the Caribbean, the eastern flank of the subtropical ridge was quite a bit further west. These differences, albeit somewhat subtle, are logical when one considers the influence they had upon the deep Atlantic hurricane tracks. A more eastward extending ridge took hold on tropical cyclones earlier and thus allowed them to remain on a westward course towards the Caribbean islands. 2008 is more like the other years in that the eastern Atlantic weakness is broader. This feature will not only quickly pull some tropical cyclones northward, but even the ones that do not completely recurve will still have a northward component to their motion. With this in mind, the one to two major hurricanes that develop in the deep Atlantic should stay north and east of the Leeward Islands and Puerto Rico, as was the case in 1955, 1961, and 2003. That said, a weaker impact cannot be ruled out. Any tropical cyclone that develops in the western half of the deep Atlantic will have enough northward component to avoid entering the Caribbean Sea, but it might be too far south to stay completely clear of all the islands. Therefore, up to one named storm, possibly a minimal hurricane, is forecast to hit the northeastern Caribbean islands. Bahamas and Turks and Caicos Islands The Bahamas and Turks and Caicos Islands run a risk of being struck by tropical cyclones that form in the deep tropical Atlantic. As noted in the tropical Atlantic section, five to six named storms are forecast to develop between Africa and Caribbean. A little over half of these named storms are expected to continue on a general west-northwest heading underneath a moderately strong western Atlantic ride. Any one of these tropical cyclones could impact the Bahamas and Turks and Caicos Islands, with the Bahamas at an especially elevated risk for a major hurricane strike. Cayman Islands, Cuba, Jamaica and Hispaniola The threat from deep Atlantic tropical cyclones to these islands will be largely diminished this season given that none are expected to traverse the Caribbean Sea. Any named storm that hits the northeast Caribbean islands will likely continue on a more west-northwestward path north of the Greater Antilles. Western Caribbean Sea-originating tropical cyclones are not expected to hit these islands either. These types systems have impacted this portion of the Caribbean beyond September in the past. However, in all the years when this occurred, a strong trough dipped into the Caribbean Sea the prior spring. This feature helped pull Caribbean Sea tropical cyclones on a north to northeastward path towards these islands. Such a pattern will not be the case for 2008. Moreover, it was aforementioned that any western Caribbean Sea-originating tropical cyclone will move into the Gulf of Mexico and avoid west Florida. On this track, it is possible that western Cuba will get sideswiped. However, any impact would be from a tropical storm at best. 8. Conclusion A moderately warm AMO combined with neutral ENSO conditions should allow for an above average Atlantic Basin hurricane season. The tropical Atlantic is expected to be the primary hotspot for hurricane development, with a few additional hurricanes also developing in the Gulf of Mexico. The weakness over the eastern Atlantic should act to keep several deep Atlantic tropical cyclones well at sea, but the ones that bypass this feature are forecast to hit the southeast United States. The Carolinas in particular are at risk for a major hurricane landfall. Moreover, the Great Lakes weakness will likely guide in-situ Gulf of Mexico and Caribbean Sea tropical cyclones towards the central Gulf coast, some likely at hurricane intensity. On the other hand, regions such as Central America, Mexico, Texas, the western Florida peninsula, most of the Caribbean islands, and the northeast United States are all expected to avoid hurricane landfalls this season. With that said, the forecast total number of tropical cyclones is presented below. If you have any questions or comments about this forecast, please email us at webmaster@independentwx.com. © 2008 Independent Weather Information Center. All rights reserved.
|
|
||||||||||||||||
 |