Understanding the Causes of Southern Hemisphere Tropical Expansion

Abstract:

Observations of the tropical edge derived directly from radiosondes show an expansion of 0.45±0.23° decade^-1 in the Southern Hemispheres (SH) over the period 1979-2010, much larger than the 0.1-0.2° decade^-1 rate deduced from climate models. Our objective is to build an understanding of the factors driving tropical expansion in the SH with a combination of statistical analysis and experiments using the NCAR Community Climate System Model 4 (CCSM4).

Using multiple linear regression, approximately 60% of the variance is explained by considering ENSO, global volcanic aerosol, SH and global average temperature and the size of the Antarctic ozone hole as predictors. These are observational proxy variables for the climate forcings examined in the simulations. Of these factors, ozone explains ~30-60% of the trend. Global or SH temperature, representing greenhouse gas forcing, accounts for the remaining 10-40% of the trend. Volcanic aerosol accounts for ~20% and ENSO about 10%; these two natural factors are short-lived and their apparent effect on the trend is a result of the period chosen.

With CCSM4, historical ‘full-forcing’ runs are compared to ‘single-forcing’ simulations using only ozone, greenhouse gas, anthropogenic aerosol or natural forcings to estimate the contribution of each of these factors to the overall trend. Three-member ensembles are used for each simulation and the period 1960-2005 is analysed. Ozone and greenhouse gases are the dominant drivers behind expansion, with a slightly larger effect ascribed to ozone. When considering only the period since 1979, natural forcing from volcanic aerosol plays a more significant role, reducing the contribution from greenhouse gas forcing.

Our results indicate that SH tropical expansion is attributable to a combination of factors, primarily ozone depletion and greenhouse gases. Natural factors also play a role; however, these factors are not ongoing climate forcings. Their effect on the trend is temporary and depends on the starting time of the analysis. Since 1979, natural factors have been of increased importance. Accounting for these factors is crucial for understanding the drivers of tropical expansion, for resolving the differences between model simulations and observations and for estimating its future impacts.