What is the relation between greenhouse gas-induced climate warming and tropical cyclone intensities?  Are the recent active hurricane seasons in the Atlantic basin or other basins a response to climate warming?  In this seminar, I will present an overview of recent research at GFDL relevant to these questions, and discuss this research in the context of recent papers on trends in tropical cyclone intensity measures by Emanuel and Webster et al.
 
Sea surface temperatures in the tropical Atlantic's "main development region" for hurricanes appear to have a century-scale warming trend, similar to the global temperature trend and to trends in other tropical basins, but with additional multi-decadal variability superimposed.  For the Atlantic, no strong evidence has been presented to date for a similar century-scale trend in major hurricane activity or intensity measures.  The Atlantic basin-wide data going back to the 1940s has a more "cyclical" or "U-shaped" time series character, as opposed to a trend.  Emanuel maintains that a combined tropical cyclone power dissipation index for the North Pacific and Atlantic basins does show evidence for a trend (due to increased intensity and duration) over the past half century or so.  At GFDL, we have conducted several high-resolution modeling studies of the relation between greenhouse warming and hurricane intensity using the GFDL hurricane model in idealized mode.  Under warmer, high CO2 conditions, simulated hurricanes are more intense (and have higher precipitation rates) than under present-day conditions. The simulated sensitivity is roughly one-half category on the Saffir-Simpson scale (14% in terms of pressure fall, 6% in terms of maximum surface winds) for an 80-year build-up of CO2 at 1%/yr compounded.  Our results are broadly robust to the use of different climate models to define the high CO2 conditions, and to details of the treatment of moist convection in the hurricane model.  The SST changes due to increased CO2 in our experiments ranged from about +0.8 to +2.4 degrees C, which is substantially greater than the ~0.5 degrees C warming experienced in the tropical Atlantic and other basins during the 20th century.  Assuming historical lapse rate trends (per degree SST warming) are close to those from +1%/yr CO2 experiments, our results imply that any trends in tropical cyclone intensity in the various basins should be too small to be detectable at this time. Some possible reasons for the discrepancy between our modeling results and the observational results of Emanuel will be discussed, along with our new regional modeling efforts aimed at simulating Atlantic hurricane activity.

Print page