Projected changes in late 21st century tropical cyclone frequency in CMIP5 models

Changes in tropical cyclone (TC) frequency under anthropogenic climate change are examined for twenty global models from the Coupled Model Intercomparison Project Phase 5 (CMIP5), using the OWZP TC detection method developed by the authors in earlier papers. The advantage of this method is that it detects large-scale conditions within which TCs form. It was developed and tuned in atmospheric reanalysis data, and verified against individual storms. All adjustments since the first published description have been re-verified against individual observed TCs, and then applied without further change to the climate models to ensure model and detector independence. Changes in TC frequency are determined by comparing TC detections in the CMIP5 historical runs (1970—2000) with high emission scenario (representative concentration pathway 8.5, rcp85) future runs (2070—2100). The study builds on earlier work (Tory et al. 2013) that reported on projections from thirteen of the twenty models.

TC detections in twelve of the twenty models reproduce observed TC formation numbers and geographic distributions reasonably well, with annual numbers within ±50% of observed. TC detections in seven models were particularly low in number, ranging from 10% to 28% of global observed TC numbers, and detections in the remaining model were approximately double that observed. All but one of the twelve models with a reasonable TC climatology project decreases in global and Southern Hemisphere frequency varying between 7 and 28%. In the Northern Hemisphere, where there are larger inter-model variations, projections range from a 13% increase to a 25% decrease in TCs.

While these results are consistent with results from previous generations of climate models, they are at odds with two recent publications by Emanuel (2013) and Camargo (2013), which showed respectively a dominant increasing trend in TC frequency, and a mix of increasing and decreasing TC frequency. It is of concern that three independent studies of TC projection frequency in the same set of models could have such widely diverse findings. Given the level of public interest in this subject it is imperative that the differences in results be investigated further. It is now more important than ever to understand exactly what it is that the TC detectors detect, and to understand what is changing in the models that leads to the differences in detections between the future and historical scenarios.

In this presentation we will first attempt to reassure the audience that the OWZP TC detector does indeed detect TCs, and that projections made from using this technique are not particularly sensitive to the choice of thresholds, before summarizing the projection results discussed above. We anticipate that an investigation into differences in seasonal parameters (known to impact TC formation) and differences in TC precursor behavior between the historical and rcp85 scenarios will begin in the near future. Preliminary results from that study will also be presented.

References

Camargo, S. J., 2013: Global and regional aspects of tropical cyclone activity in the CMIP5 models. J. Clim. In press.

Emanuel, K. A., 2013: Downscaling CMIP5 climate models shows increased tropical cyclone activity over the 21st century. PNAS Early Edition. www.pnas.org/cgi/doi/10.1073/pnas.1301293110

Tory, K. J., S. S. Chand, J. L. McBride, H. Ye and R. A. Dare, 2013: Projected Changes in Late-twenty-First-Century Tropical Cyclone Frequency in 13 Coupled Climate Models from Phase 5 of the Coupled Model Intercomparison Project. J. Clim., 26, 9946—9959.