Global warming and the frequency of intense Atlantic hurricanes: model results

Could global warming increase wind shear over the Atlantic, potentially leading to a decrease in the frequency of Atlantic hurricanes? There is a growing consensus among hurricane scientists that this is indeed quite possible. Two recent studies, by Zhao et al. (2009), "Simulations of Global Hurricane Climatology, Interannual Variability, and Response to Global Warming Using a 50-km Resolution GCM", and by Knutson et al. (2008), "Simulated reduction in Atlantic hurricane frequency under twenty-first-century warming conditions", found that global warming might increase wind shear over the Atlantic by the end of the century, resulting in a decrease in the number of Atlantic hurricanes. For example, the second study took 18 relatively coarse (>60 km grid size) models used to formulate the 2007 IPCC climate report, and "downscaled" them using a higher-resolution (18 km grid size) model called ZETAC that was able to successfully simulate the frequencies of hurricanes over the past 50 years. When the 18 km ZETAC model was driven using the climate conditions we expect in 2100, as output by the 18 IPCC models, the authors found that a reduction of Atlantic tropical storms by 27% and hurricanes by 18% by the end of the century resulted. An important reason that their model predicted a decrease in the frequency of Atlantic hurricanes was due to a predicted increase in wind shear. As I explain in my wind shear tutorial, a large change of wind speed with height over a hurricane creates a shearing force that tends to tear the storm apart. The amount of wind shear is critical in determining whether a hurricane can form or survive.


Figure 1. Top: predicted change by 2100 in wind shear (in meters per second per degree C of warming--multiply by two to get mph) as predicted by summing the predictions of 18 climate models. Bottom: The number of models that predict the effect shown in the top image. The dots show the locations where tropical storms formed between 1981-2005. The box indicates a region of frequent hurricane formation where wind shear is not predicted to change much. Image credit: Geophysical Research Letters, "Increased Tropical Atlantic Wind Shear in Model Projections of Global Warming", by Vecchi and Soden, 2007.

Since the Knutson et al. study using the 18 km resolution ZETAC model was not detailed enough to look at what might happen to major Category 3 and stronger hurricanes, a new study using a higher resolution model was needed. This was done by a team of modelers led by Dr. Morris Bender of NOAA's GFDL laboratory, who published their results in Science in February. The authors used the GFDL hurricane model--the model that has been our best-performing operation hurricane track forecasting model over the past five years--to perform their study. The GFDL hurricane model runs at a resolution of 9 km, which is detailed enough to make accurate simulations of major hurricanes. The researchers did a double downscaling study, where they first took the forecast atmospheric and oceanic conditions at generated by the coarse (>60 km grid) IPCC models, used these data to initialize the finer resolution 18 km ZETAC model, then used the output from the ZETAC model to initialize the high-resolution GFDL hurricane model. The final results of this "double downscaling" study suggest that although the total number of hurricanes is expected to decrease by the end of the century, we should expect an increase of 81% in the number of Category 4 and 5 storms in the Atlantic. This trend should not be clearly detectable until about 60 years from now, given a scenario in which CO2 doubles by 2100. The authors say that their model predicts that there should already have been a 20% increase in the number of Category 4 and 5 storms since the 1940s, given the approximate 0.5°C warming of the tropical Atlantic during that period. This trend is too small to be detectable, given the high natural variability and the difficulty we've had accurately measuring the exact strength of intense hurricanes before the 1980s.The region of the Atlantic expected to see the greatest increase in Category 4 and 5 storms by the year 2100 is over the Bahama Islands (Figure 2), since wind shear is not expected to increase in this region, and sea surface temperatures and atmospheric instability are expected to increase there.

The net effect of a decrease in total number of hurricanes but an increase in the strongest hurricanes should cause an increase in U.S. hurricane damages of about 30% by the end of the century, the authors compute, assuming that hurricane damages behave as they did during the past century. Over the past century, Category 4 and 5 hurricanes made up only 6% of all U.S. landfalls, but accounted for 48% of all U.S. damage (if normalized to account for increases in U.S. population and wealth, Pielke et al., 2008.)


Figure 2. Expected change in Atlantic Category 4 and 5 hurricanes per decade expected by the year 2100, according to the Science paper by Bender et al. (2010).

Commentary
These results seem reasonable, since the models in question have been successfully been able to simulate the behavior of hurricanes over the past 50 years. However, the uncertainties are high and lot more research needs to be done before we can be confident of the results. Not all of the IPCC models predict an increase in wind shear over the tropical Atlantic by 2100, so the increase in Category 4 and 5 hurricanes could be much greater. Also, the GFDL model was observed to under-predict the strength of intense hurricanes in the current climate, so it may not be creating enough Category 4 and 5 hurricanes in the future climate of 2100. On the other hand, IPCC models such as the UKMO-HadCM3 predict a very large increase in wind shear, leading to a drastic reduction in all hurricanes in the Atlantic by 2100, including Category 4 and 5 storms. So Category 4 and 5 hurricane frequency could easily be much greater or much less than the 81% increase by 2100 found by Bender et al.

The estimates of a 30% increase in hurricane damages by 2100 may be considerably too low, since this estimate assumes that sea level rise will continue at the same pace as was observed in the 20th century. Sea level rise has accelerated since the 1990s, and it is likely that this century we will see much more than than the 7 inches of global sea level rise that was observed last century. Higher sea level rise rates will sharply increase the damages due to storm surge, which account for a large amount of the damage from intense Category 4 and 5 hurricanes.

Keep in mind that while a 30% in hurricane damage by the end of the century is significant, this will not be the main reason hurricane damages will increase this century. Hurricane damages are currently doubling every ten years, according to Pielke et al., 2008. This is primarily due to the increasing population along the coast and increased wealth of the population. The authors theorize that the Great Miami Hurricane of 1926, a Category 4 monster that made a direct hit on Miami Beach, would have caused about $150 billion in damage had it hit in 2005. By 2015, the authors expect the same hurricane would do $300 billion in damage. This number would increase to $600 billion by 2025 (though I think it is likely that the recent recession may delay this damage total a few years into the future.) It is essential that we limit coastal development in vulnerable coastal areas, particularly along barrier islands, to reduce some of the astronomical price tags hurricanes are going to be causing. Adoption and enforcement of strict building standards is also a must.

The authors of the GFDL hurricane model study have put together a nice web page with links to the paper and some detailed non-technical explanations of the paper.

References
Bender et al., 2010, "Modeled Impact of Anthropogenic Warming on the Frequency of Intense Atlantic Hurricanes", Science, 22 January 2010: Vol. 327. no. 5964, pp. 454 - 458 DOI: 10.1126/science.1180568.

Vecchi, G.A., B.J. Soden, A.T. Wittenberg, I.M. Held, A. Leetmaa, and M.J. Harrison, 2006, "Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing", Nature, 441(7089), 73-76.

Vecchi, G.A., and B.J. Soden, 2007, "Increased Tropical Atlantic Wind Shear in Model Projections of Global Warming", Geophysical Research Letters, 34, L08702, doi:10.1029/2006GL028905, 2007.

Jeff Masters