The atmosphere above the major hurricane formation and intensification region in the Atlantic ocean continues to evolve in a manner that is virtually opposite to the way it is projected to evolve by climate models run with increasing levels of atmospheric carbon dioxide and other anthropogenic emissions. This fact suggests that the role played by natural variability in the recent upswing in hurricane activity in the Atlantic Ocean (including storms striking the U. S. coastline) is likely large and significant. A just-published paper further adds to this evidence.
Researchers Gabriel Vecchi and Brian Soden (from GDFL and the University of Miami respectively) examined how anthropogenic global warming may, at least according to climate models, alter the environment of the tropical Atlantic and thus possibly tropical cyclone statistics there as well. Recall that it is the tropical cyclones that form in the Atlantic Ocean that are the ones that cause the greatest impact the United States. And also recall that there is still an on-going, sometimes rather heated, debate about whether natural or anthropogenic (as if somehow we aren’t part of nature) causes are thought to be responsible for the increase in storminess since the early-1970s, as well as whether or not a continued enhancement to the earth’s greenhouse effect will lead to more and/or more intense hurricanes.
Motivated by this uncertainty, Vecchi and Soden collected the output from a suite of 18 of the latest, greatest, climate models all run with a middle-of-the-road scenario for future emissions (IPCC SRES Scenario A1B for those keeping score). They then examined how the atmospheric quantities that are thought to be most related to hurricane formation and development evolve into the future. These quantities included vertical wind shear (how different the wind is blowing at different levels in the atmosphere—the less the difference, the better it is for hurricanes), low level vorticity (the amount of twisting in the air), low level relative humidity (a measure of moisture and temperature), and something called the maximum potential intensity (which combines surface temperatures, moisture, and upper air temperature). Of these four parameters, the focus of the paper was primarily on the vertical wind shear.
What Vecchi and Soden found was that climate models almost unanimously project that there will be a “robust increase” in the vertical shear during the hurricane season (in this case defined as June – November) in the areas in the Atlantic Ocean where hurricanes form and traverse (Figure 1). A “robust increase” in vertical wind shear acts to inhibit tropical cyclone formation and intensification. The authors do not claim that in the future overall hurricane intensity or frequency will decline in the Atlantic, because along with the increases in vertical wind shear, there will also be increases in sea surface temperatures (SST)—which lead to stronger storms. However, the increases in vertical wind shear are found to lessen, to a substantial degree, the impacts of rising SST.
Figure 1. (top) The 18-model average change in wind shear per degree C of warming and (bottom) the number of models (out of 18) that show increase in wind shear. (From Vecchi and Soden, 2007).
This finding by Vecchi and Soden ties into earlier findings by researchers Thomas Knutson and Robert Tuleya. In work that they published in 2004, Knutson and Tuleya found that in an enhanced greenhouse world, SST increases acted to make for more intense storms, but that this intensity increase was partially offset by increases in another hurricane-influencing parameter—vertical stability (as opposed to vertical wind shear that was the focus of Vecchi and Soden’s study). The suite of climate models examined by Knutson and Tuleya virtually unanimously projected that in a CO2-enhanced world, the middle and upper troposphere will warm at a faster rate than the surface, especially over the tropical oceans. More warming aloft than at the surface makes the atmosphere more stable and less conducive to storm formation. Thus, Knutson and Tuleya reported that the model-projected vertical stability increases in the future would temper (but not totally cancel out) the increase in storm intensity by rising SST. Knutson and Tuleya also looked for changes in vertical wind shear in the climate models, but could find no clear signal. But, as just reported by Vecchi and Soden, an updated suite of climate models does show a rather strong indication for increases in vertical wind shear.
Knutson and Tuleya concluded from their analysis that a more than doubling of atmospheric CO2 concentrations from current levels would lead to an increase in Atlantic Ocean temperatures of about 2ºC and an increase the average wind speed in tropical cyclones by about 6 percent. We would surmise that had the models used by Knutson and Tuleya demonstrated vertical wind shear increases as strongly as the newer models examined by Vecchi and Soden did, that they would have found that the future increases in maximum wind speeds in Atlantic tropical systems would have been even less and would have approached the level of undectability given the large degree of year-to-year natural variability (noise) that is present in the Atlantic tropical cyclone characteristics.
Thus, overwhelmingly, the world’s best climate models indicate that the future evolution of the tropical Atlantic environment resulting from human industrial emissions is not towards conditions that will lead to a dramatic increase in tropical cyclone intensity there. In fact, the models suggest that the projected intensity increase will be so modest as to be largely undetectable for decades (or longer) to come.
But what of the claims made that already the increase in tropical cyclone activity in the Atlantic Ocean observed since the early 1970s has been largely caused by human industrial emissions (e.g., Emanuel, 2005; Webster et al., 2005; Mann and Emanuel, 2006)?
Well to make this claim, you first have to throw out all of the climate model results, and for that matter, all of the actual atmospheric observations. You have to throw out all of the models because, as we have just seen, they project only very modest increases in hurricane intensity over the course of the next century or so as a result from increasing emissions, and yet we have already witnessed increases in storm activity during the past 30 years that greatly exceeded those meager projections for the coming century. Thus, the observed increase in storminess does not match model projections. You have to throw out all of the atmospheric observations because they indicate that the atmosphere over the tropical Atlantic in evolving in a manner that is in many ways opposite to how it should be evolving according to climate models. Instead of the atmosphere trending towards becoming more hostile to hurricanes (as the models suggest it should), it has been trending towards becoming more hospitable to hurricanes.
As luck (wink, wink) would have it, we are currently researching just exactly how atmospheric conditions in the tropical Atlantic have been changing over the past 50 years or so. In fact, Dr. Robert Davis is presenting some of our research results to the meeting of the Association of American Geographers which is being held in San Francisco this week (Dr. Davis’s presentation is scheduled for 4 p.m. Friday April 20, 2007). Figure 2 shows how tropical cyclone activity has changed in the Atlantic (study area: 5N to 25N, 20W to 90W, June through November) from 1957 through 2005. There has clearly been in increase since the early 1970s, primarily manifest in the hurricanes seasons beginning in 1995. Figure 3 shows the average vertical stability (small print: the vertical stability, or more precisely, moist static stability, is calculated as the equivalent potential temperature at 1000mb minus the equivalent potential temperature at 500mb). In Figure 3, higher values mean less stability which means better conditions for hurricane development and intensification. Recall that with increasing CO2 levels, the models overwhelmingly project that the vertical stability should increase (trend downwards in Figure 3). So obviously what has been happening isn’t what is supposed to be happening. Figure 4 shows more of the same. In this case, Figure 4 shows the history of the vertical wind shear (small print: absolute value of the winds at 850mb minus the winds at 200mb). Notice that the wind shear in the tropical Atlantic has been declining since the early 1970s. Less wind shear means better conditions for tropical cyclones. Again, the observed trend is opposite to the trend projected by climate models. (And just in case for some reason that you don’t believe our results, you can check out the results by Hoyos et al. 2006 and find the same thing.)
Figure 2. Tropical cyclone activity in the Atlantic (as measured by the Accumulated Cyclone Index (ACE)–the sum of the square of the 6-hrly wind speed observations for each storm) from 1957-2005.
Figure 3. The seasonal average vertical stability over the tropical Atlantic, 1957-2005 (see text for additional details).
Figure 4. The seasonal average vertical wind shear over the tropical Atlantic, 1957-2005 (see text for additional details).
So, combine the observed decline in vertical stability with the lessening of the vertical wind shear with the observed increase in sea surface temperatures and you get some pretty prime breeding grounds for hurricanes. The fact that two of these three parameters are trending contrary to enhanced greenhouse expectations leads one to conclude that natural factors must be involved in the observed upward trend in Atlantic tropical cyclone activity—just like many hurricane specialists have been saying is the case for nearly a decade now. They theorize that a natural, ongoing, multi-decadal oscillation in a combination of environmental conditions is responsible for the alternating quiet and active periods of hurricane behavior that is documented in the historical record of Atlantic tropical cyclone observations (see , for example, here and here and here, etc.). They see no need to involve anthropogenic global warming to explain the past and current patterns of storminess. And further, they suggest that future impacts of global warming will be minimal on Atlantic tropical cyclone characteristics. This outcome is also supported by general climate model projections (some climate models actually produce declines in hurricane activity).
Thus, to make claims that current conditions are largely the result of anthropogenic changes in the tropical environment and that future changes will make things dramatically worse, runs counter to both model projections and actual observations.
Davis, R.E., Knappenberger, P.C., Frauenfeld, O.W., Michaels, P.J., 2007. Observed changes in North Atlantic hurricane frequency and intensity using a multivariate model. 2007 Annual Meeting of the Association of American Geographers, San Francisco, CA, April 17-21, 2007.
Emanuel, K., 2005. Increasing destructiveness of tropical cyclones over the past 30 years. Nature, 436, 686-688.
Hoyos, C.D., et al., 2006. Deconvolution of the factors contributing to the increase in global hurricane intensity. Science, 312, 94-97.
Knutson, T.R. and R. E. Tuleya, 2004. Impact of CO2-Induced Warming on Simulated Hurricane Intensity and Precipitation: Sensitivity to the Choice of Climate Model and Convective Parameterization. Journal of Climate, 17, 3477-3495.
Mann, M.E., and K. A. Emanuel, 2006. Atlantic hurricane trends linked to climate change. Eos: Transactions of the American Geophysical Union, 87, 233-244.
Vecchi, G.A. and B. J. Soden, 2007. Increased tropical Atlantic wind shear in model projections of global warming. Geophysical Research Letters, L08702, doi:10.1029/2006GL028905.
Webster, P.J., et al., 2005. Changes in tropical cyclone number, duration, and intensity ion a warming environment. Science, 309, 1844-186.