Introduction
Hurricanes are one of the deadliest natural
disasters on our planet, responsible for more deaths than any other
single phenomena (Abbott, 2002). Although the number of hurricane
deaths in the United States has been dropping, the costs associated
with hurricane damage have been increasing as more and more people move
to hurricane prone coastlines (Nicholls, 2001; Abbott, 2002).
Hurricanes follow a
generalized life cycle based on wind speed, beginning as a tropical
disturbance then progressing to a tropical depression (winds 37-62
km/h). This is followed by designation as a tropical storm (winds
64-117 km/h) and finally a hurricane (winds > 118 km/h) (Slattery
and Burt, 1997). The Saffir Simpson hurricane scale further divides
hurricanes into 5 categories, also based on wind speed, and to some
extent, pressure.
Typical hurricane structure includes a calm
eye 20-50 km in diameter characterized by low surface pressure
(typically ~965 mb), surrounded by the eye wall, and spiral wind and
rain bands (Slattery and Burt, 1997).
Several conditions are
required for a hurricane to form in the Atlantic Ocean. Sea surface
temperatures need to be greater than 27°C, surface and upper level
winds should be light. Easterly waves in the upper troposphere promote
hurricane formation in the eastern Atlantic, while upper level
westerlies restrict hurricane development. These conditions generally
occur between June-November, making these the peak months for Atlantic
hurricane activity, though this is subject to interannual
variability (Slattery and Burt, 1997).
Regional and global scale phenomena can have a
strong influence on hurricane number and intensity. La Nina events are
related to increased hurricane damage in the USA, while El Nino events
may help suppress hurricane formation, though exceptions certainly
occur (Nicholls, 2001). The transition to La Nina during 1995 resulted
in the second worst Atlantic hurricane season since late 1800’s, with
11 hurricanes, 8 tropical storms, 121 deaths, and $7.7 billion US in
damages. Slattery and Burt (1997) gave 4 reasons for this situation: 1)
the end of El Nino (and its suppression of tropospheric westerlies)
resulted in easterly winds at all levels, 2) high rainfall in West
Africa suppressed high level winds that otherwise shear off the tops of
developing hurricanes, 3) low pressure and high temperature in West
Africa caused southerly winds that intensify hurricanes, and 4)
stratospheric winds were in the westerly phase of the ~26 month
Quasi-Biennial Oscillation (QBO), which can cause twice as many
hurricanes as stratospheric easterlies.
In this project, I ask the
how the probability distribution of hurricane intensity and track paths
in Caribbean and the South-eastern seaboard of North America changes in
response to El Nino and La Nina events in the Pacific Ocean, and how
this changes the patterns of risk to life and property in the eastern
United States.