March 21, 2007

Signal vs. Noise in Atlantic Hurricanes

Filed under: Climate Extremes, Hurricanes

A just-published Comment and Reply exchange concerning the relationship between sea surface temperatures and the intensity of Atlantic hurricanes shows that there is just too much noise in the natural system (or at least in the observational record) at this time to clearly be able to identify a signal from sea surface temperatures on the maximum wind speed in major hurricanes.

About a year ago we (Michaels, Knappenberger, and Davis) published a paper in the journal Geophysical Research Letters (GRL) in which described the results of our investigation into the role of local, rather than basin-wide, sea surface temperatures (SST) on tropical cyclone strength. After examining the data from the Atlantic Ocean collected during the period 1982-2005 (the period of available fine spatial scale SST data), we concluded that there is an overall positive relationship between observed SST and observed tropical cyclone intensity (that is, higher SSTs lead to generally stronger storms). BUT, and this is a big but, we found that this positive relationship breaks down at SSTs greater than about 28.25ºC—the critical value for becoming a major hurricane (a category 3, 4, or 5 storm). In other words, there is no telling how strong a tropical cyclone will become once it encounters SST exceeding 28.25ºC and increasing the SST doesn’t change this fact. Thus, if an anthropogenic enhancement to the world’s greenhouse effect leads to rising SSTs in the Atlantic Ocean, more tropical cyclones may encounter SSTs high enough to become major hurricanes, but the average intensity of storms once they encounter SSTs exceeding 28.25ºC will not change. See here for more detailed coverage of our findings.

M.I.T. professor and hurricane expert Dr. Kerry Emanuel didn’t fully agree with our results—well, not necessarily our results actually, but rather, our conclusions—and wrote a Comment to our original GRL article. He thought that the primary reason that we didn’t find a statistically significant positive relationship between SST and tropical cyclone intensity at SSTs exceeding 28.25ºC was that there simply were not enough occurrences of such events during our study period. In other words, our sample size was too small to reliably be able to detect the signal (stronger storms) from the noise (natural variation).

Instead of being limited to the actual observations (like we were), Emanuel used a model of hurricane frequency and intensity that he ran using climatological values of SST and other atmospheric variables characteristic of the Atlantic Ocean environment. Using this model, Emanuel simulated the occurrence of 3,000 tropical cyclones in the Atlantic (for reference sake, our analysis included 270 tropical cyclones, so Emanuel’s synthetic sample size was more than 10 times ours) and then, using a methodology similar to ours, analyzed his results for SST and storm intensity. In doing so, he found just what we had reported, no statistically significant relationship between SST and storm intensity at SSTs exceeding 28.25ºC. In Emanuel’s own words,

In either case, the correlation of storm intensity with potential intensity when the latter exceeds 120 knots [closely related to SSTs exceeding 28.25ºC] is statistically insignificant, in agreement with MKD [Michaels, Knappenberger, and Davis]. This lack of correlation, as stated by MKD, is owing to the large scatter of storm intensities for a given potential intensity, reflecting the influence of other environmental factors such as wind shear. [emphasis original]

But, in establishing this, Emanuel did not think it sufficient information to conclude that in a warmer world with higher SST that the strongest storms would not get stronger (as we had). Instead, to test that premise, he re-ran all 3,000 tropical cyclone swith SSTs elevated by a couple of degrees (in actuality his warmer runs had the potential intensity elevated by 10% but the effect is similar). In doing so, he found that in a warmer world, the average maximum wind speed (a measure of intensity) of the most intense storms increased by about 15% (from 129 to 148 knots). Thus, he concluded, that contrary to our original conclusions, increased SSTs would result in increases in the average intensity of major hurricanes.

As is established scientific protocol, GRL asked us whether we would like the opportunity to Reply to Emanuel’s Comment, and indeed we did. Again, instead of looking a collection of modeled storms, we preferred to look at the observations of real storms to see how SST effect the strongest of storms. This time, instead of looking at all tropical cyclones that encountered SSTs of at least 28.25ºC, we only looked at major hurricanes. Figure 1 below shows the observed relationship between SST and major hurricane intensity (as measured by maximum attained wind speed). We find that the relationship is on the verge of statistical significance, but falls just shy of it—again indicating that the noise is about as large, or even larger, than the signal we are looking for.

Figure 1. Scatterplot of maximum wind speed and the highest SST encountered prior to (or concurrent with) reaching the maximum wind speed for all Atlantic basin major (category 3, 4, 5) hurricanes. The two storms in the lower left (open circles) did not encounter the 28.25ºC threshold (vertical line) that is typically a prerequisite for major hurricane development. (from Michaels et al., 2007)

Combining our observational analysis with Emanuel’s model analysis provides some, but certainly not overwhelming, evidence that rising SSTs will lead, on average, to stronger intense hurricanes. However, it is obvious that other factors play large roles in determining the maximum strength of Atlantic tropical cyclones. The future evolution pathway of these other factors will ultimately determine whether the intensity of the strongest tropical cyclones increase in the future (and/or whether or not any increase is dectable). But as for now, in the observed record, a definitive association between SST and major hurricane intensity cannot unequivocally be established.

Here is the conclusion to our Reply to Emanuel’s Comment:

We thank Emanuel [2007] for analyses that provide further detail about the complex relationship between SST and tropical cyclone intensity in the Atlantic basin. In agreement with Michaels et al. [2006], Emanuel’s model-based findings confirm the lack of a strong, statistically significant SST signal in the average intensity of tropical cyclones encountering high SST (i.e., above the 28.25ºC threshold). There is, however, some indication that higher SST may lead to more intense strong storms. Currently, observations can neither confirm nor refute that relationship.

We do not contend that increasing SST has not, or will not, influence the intensity or frequency of Atlantic tropical cyclones. We have demonstrated in our initial findings [Michaels et al., 2006], and Emanuel has furthered in his comments, that a modest impact is likely. However, the overall weakness in the relationship between SST and storm intensity indicates that SST is only one of myriad determinants of tropical cyclone strength. This complicates clear attribution of observed or future changes. These results reaffirm our original conclusions that the full reason for the observed changes in tropical cyclone activity in the North Atlantic has yet to be established, and that “we therefore recommend a cautious approach to assigning an underlying cause in this complex system [Michaels et al., 2006].”


Emanuel, K. 2007. Comment on “Sea-surface temperatures and tropical cyclones in the Atlantic basin” by Patrick J. Michaels, Paul C. Knappenberger, and Robert E. Davis, Geophysical Research Letters, 34, L06702, doi:10.1029/2006GL026942.

Michaels, P. J., P. C. Knappenberger, and R. E. Davis, 2006. Sea-surface temperatures and tropical cyclones in the Atlantic basin, Geophysical Research Letters, 33, L09708, doi:10.1029/2006GL025757.

Michaels, P. J., P. C. Knappenberger, and R. E. Davis, 2007. Reply to comment by K. Emmanuel on “Sea-surface temperatures and tropical cyclones in the Atlantic basin”, Geophysical Research Letters, 33, L06703, doi:10.1029/2006GL025727

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress