FACTORS GOVERNING TROPICAL CYCLONE FORMATION AND VARIABILITY IN THE NORTH ATLANTIC
It is not an easy matter for a hurricane to form, which is why the long-term average for the North Atlantic is only about 11 hurricanes annually (NHC, 2010b). For a hurricane to form, a weak disturbance must first form in a broad region of active thunderstorms and warm ocean conditions. The warmest ocean conditions are near the Equator, however tropical cyclones do not form close to the Equator due to the effects of the Earth’s rotation there. Finally, the winds circulating around the developing storm should not change very much with height or they will rip the system apart before it gains strength. In summary, then, the ideal environment for formation of a tropical cyclone arises from a convergence of (i) an initial weak cyclonic disturbance (i.e. counterclockwise air circulation in the Northern Hemisphere), (ii) a region of warm ocean conditions that is not too close to the Equator, (iii) weak winds that do not change much with height (weak “vertical wind shear”) and (iv) a broad-scale amount of thunderstorm activity (Gray, 1968 and 1979; Rappin et al., 2010). However, even if all four of these conditions are present, a tropical cyclone may not form: they are simply necessary, but not sufficient conditions. Thus, the complete set of sufficient conditions for hurricane formation has not yet been identified.
After its genesis, the storm system is known as a tropical depression until its intensity (measured by its strongest winds near the surface) exceeds 39 miles per hour (mph); once the system exceeds this wind threshold, it is labeled a tropical storm (Evans, 2010) and is assigned a name by the National Hurricane Center1 (NHC). As a storm intensifies further, it is classified as a hurricane and assigned a category2 based upon its maximum wind speed (NHC, 2010c). Once the tropical cyclone is formed, its motion is dominated by the large-scale winds in the storm environment “steering” it (the effects of the Earth rotation also have a small effect). Large-scale winds and effects of Earth rotation can transport a tropical cyclone from its genesis location to our region of interest, the Caribbean (Fig. 3). Along the way, variations in the same atmospheric conditions needed for tropical cyclone formation3 will cause the storm to strengthen or weaken and may also affect its size. These changes influence the spatial and temporal distribution of the severe weather associated with the storm.
The environment in which a tropical cyclone forms changes continuously, but we can also identify specific (longer) periods of time when systematic changes in the environment make it more or less favorable for storm formation. For example, the Madden-Julian Oscillation (MJO) is a roughly 4-6 week fluctuation in the tropical wind patterns over an area of thousands of kilometers (Madden and Julian, 1994). These changes in the winds influence the amount of thunderstorm activity and so render the conditions needed for storm formation more or less favorable. These studies of short-term variability help us to identify the types and magnitudes of changes in the environment that can effect changes on tropical cyclone formation likelihood, and inform our investigations of longer-term (years to centuries) tropical cyclone variability and trends. The El Niño Southern Oscillation (ENSO) phenomenon has also been shown to affect the incidence of tropical cyclones from year to year (Sabbatelli and Mann, 2007). While the links are complex, a relationship between North Atlantic hurricane activity and ENSO incidence has been established: there are less tropical cyclones during ENSO years (on average, of course). This basin-wide relationship is quite robust, especially for moderate to strong ENSO years. However, the frequency and strength of ENSO years varies from decade to decade (Table 1), due in part to the North Atlantic Oscillation (NAO) (Giannini et al., 2001). As we shall see this is reflected in the variability of tropical cyclone activity in the Caribbean basin.
CURRENT CLIMATE OF THE CARIBBEAN
The Caribbean has a wet/dry (rather than warm/cool) climate: dry in winter and wet in summer. During June to November each year, easterly waves form off the coast of West Africa and often cross the Atlantic Ocean into the Caribbean; indeed, they represent the primary rainfall source for the region at the wettest time of the year. These waves frequently mature into tropical cyclones in the latitude band ranging from 10°N to 20°N, especially if they are evolving in a favorable environment with high (> 25 o C) sea surface temperature (SST) and low vertical wind shear.
Inter-annual variability of the rainfall is influenced mainly by ENSO events through their effect on SST in the Atlantic and Caribbean Basins. The late rainfall season tends to be drier in El Niño years and wetter in La Niña years (Giannini et al., 2000) and tropical cyclone activity diminishes over the Caribbean during El Niño summers (Gray, 1984). However, the early rainfall season in the central and southern Caribbean tends to be wetter in the year after an El Niño and drier in a La Niña year (Chen and Taylor, 2002). On shorter timescales, hurricane incidence in the western Caribbean has been linked to the MJO (Maloney and Hartmann, 2000). On longer timescales, the inter-decadal variations in frequency and strength of ENSO years are reflected in the long-term variability of tropical cyclone activity in the Caribbean basin (Section 5.3).
Hurricanes impact the nations of the Caribbean regularly in the current climate. So the question arises: “How might hurricane activity and intensity change in a warmer mean climate?” Hurricane formation and motion depend on large-scale weather patterns and ocean temperatures across the Atlantic. We will show here that changes in the these atmospheric and oceanic conditions lead to variations of hurricane activity over years and decades, and these variations are evident in the Caribbean storm record. Understanding the underlying causes of these changes will give us the tools to explore potential changes in Caribbean hurricane activity in response to global climate change.
Our first step is to examine the historical records of tropical cyclone (tropical storm and hurricane) activity for the 1900-2009 period, both on the North Atlantic and local Caribbean basins (NHC, 2011). Then, we explore the spatial extent of the atmosphere that corresponds to hurricane activity in the Caribbean and the winds experienced in the region due to hurricane passage (NCEP, 2010). Finally, we consider the modulation of hurricane activity by the (Atlantic basin focused) Saharan Air Layer (SAL) and the (global) ENSO phenomena (NOAA, 2010a), before extending our understanding of the drivers of hurricane formation into the realm of climate change.