Tropical Cyclones and Climate Change

Tropical cyclones (TCs) are of the most extreme and dangerous weather phenomena on Earth. In the United States, landfalling TCs account for an average of $10 billion damages annually (Pielke et al. 2008). Hurricane Katrina (2005) alone caused $81 billion of damages and took more than 1,800 lives away. In developing countries, TC landfalls can be extremely detrimental. For example, Cyclone Nargis (2008) took more than 130,000 lives in Myanmer (Burma) . Due to the catastrophic nature of TCs, substantial efforts have been devoted to short-term predictions of TC track and intensity in an effort to minimize the damages and casualties.

In recent years, the relationship between TC activity and climate change has attracted strong attentions in the atmospheric research community. There are a number of evidences suggesting that regional TC activity is going under a substantial shift, which may be a response to changing climate. As discussed later in the chapter, however, detection and association of shifts in TC activity resulting from climate change have been controversial due to uncertainties in the observed data and the difficulty in separating natural variability and anthropogenic forcing. Projections of the future TC activity are also not unanimous; the sources of uncertainties are still being identified and tested for their relative effects in relation to the unanimous projections of TC activity. However, the need for reliable projections for the near-term future TC activities is strong among the governmental organizations and industries located near the coastlines that are often affected by TC landfalls. Therefore, identifying the source of uncertainties in the future projections of TC activity and developing a mechanism to reduc...

... middle of paper ...

...ccount when evaluating the simulated TC climatology. Further it is preferred to provide explanations outside the TC detection methodology (such as the simulated large-scale conditions) to support the simulated TC climatology.

Therefore, the specific goals for this work are to;

1. Quantify the range of sensitivity to various TC tracking criteria (Chapter II);

2. Identify model biases independent of TC tracking sensitivity (Chapter III), as well as new sources of uncertainties and inherent limitations that are undocumented thus far (Chapters III and IV);

3. Increase robustness of the future TC activity projection against the TC tracking sensitivity (Chapter IV); and

4. Present an alternative approach to render the limitation of TC intensity projection using a statistical model (Chapter V).

A summary and further implications can be found in Chapter VI.