1 Changes in Hurricane Climatology in Recent Decades Anais Orsi Dian Putrasahan Ha Joon Song James Means Scripps Institution of Oceanography Abstract Hurricanes can inflict catastrophic property damage and human life loss. Thus, it is important to determine how the character of these powerful storms could change with global warming. In this paper we examine different ways of measuring hurricanes, and look at how time series of various hurricane metrics have been analyzed in recent scientific literature. We conclude that the expected connections between hurricane activity and global warming are still equivocal. Introduction and Background A hurricane is an intense tropical cyclone that has sustained winds of greater than 33 meters sec-1. Hurricanes form in the world’s tropical oceans at latitudes from about 8° from the equator to about 20° from the equator—their structure is dependent on the balance between the Coriolis and pressure gradient forces and hence cannot occur too near to the equator, where the Coriolis force vanishes. They are known to occur in the North Atlantic, western North Pacific, eastern North Pacific, Southwest Pacific, and Indian oceans, as well as their associated marginal seas. Hurricanes are classified according to the Saffir-Simpson intensity scale, which grades hurricanes according to their maximum sustained surface wind speed. The Saffir-Simpson Scale is outlined in the table below. An important thing to understand about hurricane intensity and the corresponding Saffir-Simpson classification is that there is often no direct measurement of storm intensity. If information is available from “Hurricane Hunter” flights it is the preferred method of assigning intensity, although2 even in this case the wind speed at the surface is inferred from the flight level winds so it is also an indirect measurement. Saffir-Simpson Category Sustained Wind Speed (ms-1) 1 33–42 2 43–49 3 50–58 4 59–69 5 70≥ If flight information is unavailable then hurricane intensity is usually assigned by the Dvorak classification. In this scheme, the presentation of the hurricane in satellite imagery is compared to standard images of tropical cyclones and maximum wind speed is inferred from comparison with the standards. Allowance is made for storm “spin-up” and “spin-down” so that the intensity classification is not allowed to change too quickly. The original classification flowchart as described by Dvorak [1] is shown in Figure 1, and although the procedure has been updated [2], the present technique remains very similar. Before the Dvorak classification surface winds were estimated from central pressure and sea state estimated from visual observation from aircraft. Figure 1.Template and chart taken from Dvorak's paper showing how to classify hurricanes by their banding and central features in order to arrive at T number.3 Hurricanes are heat engines, and derive their intense energy from the latent heat of vaporization of water. To sustain a hurricane requires an environment with a high latent heat content and low friction. It has been empirically determined that these requirements are only filled over ocean basins with water temperature exceeding 26°C. Hurricanes quickly diminish in intensity when they move over cooler waters; when they make landfall; when they encounter dry or stable air; or when they move into an environment of vertical wind shear. It is the dependence on sea surface temperature that has led to speculation that global warming—specifically warming of the tropical ocean basins—could lead to changes in the frequency, distribution and intensity of hurricanes. Figure 2. Adopted from Webster et al [3], this plot shows trends in sea surface temperature for different ocean basins. However, it should be noted the other factors mentioned above may be just as important to hurricane climatology, but those factors are not as simply connected with global warming.4 Evidence for Changes in Hurricane Climatology Both the 2004 and 2005 hurricane seasons have been exceptional in the number of hurricanes affecting the United States and the damage caused by them, which as led to much speculation in both the print and broadcast media that there has been an increase in hurricane occurrence due to global warming. However, it is debatable whether such a trend exists. For example, Figure 3 is a plot of National Hurricane Center data of US hurricane strike by decade, with the data split into all hurricanes (categories 1 through 5) and major hurricanes (categories 3, 4, and 5). It is not at all clear from examining this plot that there is any trend whatsoever, either for all hurricanes or major hurricanes. This is a very limited data set, however, and it is advisable to examine more detailed studies of hurricane frequency and intensity, and especially studies that look at tropical basins other than the Atlantic/Caribbean/Gulf of Mexico. U.S. Hurricane Strikes by Decade0510152025301851-18601861-18701871-18801881-18901891-19001901-19101911-19201921-19301931-19401941-19501951-19601961-19701971-19801981-19901991-20002001-2004DecadeNumber of HurricanesTotalMajor Figure 3. U.S. Hurricane strikes by decade, for all hurricanes and just major hurricanes. Figure 4 shows another measure of hurricane climatology, the Accumulated Cyclone Energy (ACE) index. The ACE is a statistic that the National Oceanic and Atmospheric Administration (NOAA) uses to measure the total seasonal tropical cyclone activity. It is5 calculated by summing the squared values of a storm’s maximum wind speed (taken every six hours) for all Atlantic storms for a season. 0501001502002503001950195219541956195819601962196419661968197019721974197619781980198219841986198819901992199419961998200020022004 Figure 4. ACE index for the North Atlantic since 1950, showing values above average for most of the last decade. Webster et al [3] look at trends in tropical cyclone number, duration, and intensity for various tropical basins over the period 1970–2004. They find no global trend for a change in either the frequency or duration of tropical cyclones worldwide. They do find a statistically significant increase in the number and duration of North Atlantic hurricanes over the same period, but do not attribute this to ocean warming because the same trends are not seen in other oceans. This is shown in Figure 5. Although we can see the decadal oscillations in both the
View Full Document
Unlocking...