On the tropical origin of uncertainties in the global land precipitation response to global warming
Understanding the response of the global hydrological cycle to recent and future anthropogenic emissions of greenhouse gases and aerosols is a major challenge for the climate modelling community. Recent climate scenarios produced for the fourth assessment report of the Intergovernmental Panel on Climate Change are analysed here to explore the geographical origin of, and the possible reasons for, uncertainties in the hydrological model response to global warming. Using the twentieth century simulations and the SRES-A2 scenarios from eight different coupled ocean–atmosphere models, it is shown that the main uncertainties originate from the tropics, where even the sign of the zonal mean precipitation change remains uncertain over land. Given the large interannual fluctuations of tropical precipitation, it is then suggested that the El Niño Southern Ocillation (ENSO) variability can be used as a surrogate of climate change to better constrain the model reponse. While the simulated sensitivity of global land precipitation to global mean surface temperature indeed shows a remarkable similarity between the interannual and climate change timescales respectively, the model ability to capture the ENSO-precipitation relationship is not a major constraint on the global hydrological projections. Only the model that exhibits the highest precipitation sensitivity clearly appears as an outlier. Besides deficiencies in the simulation of the ENSO-tropical rainfall teleconnections, the study indicates that uncertainties in the twenty-first century evolution of these teleconnections represent an important contribution to the model spread, thus emphasizing the need for improving the simulation of the tropical Pacific variability to provide more reliable scenarios of the global hydrological cycle. It also suggests that validating the mean present-day climate is not sufficient to assess the reliability of climate projections, and that interannual variability is another suitable and possibly more useful candidate for constraining the model response. Finally, it is shown that uncertainties in precipitation change are, like precipitation itself, very unevenly distributed over the globe, the most vulnerable countries sometimes being those where the anticipated precipitation changes are the most uncertain.
KeywordsAnomaly Correlation Coefficient Interannual Timescale Global Hydrological Cycle Global Land Precipitation CNRM Model
The authors are very grateful to all IPCC4 participants and to the PCMDI for the build up of the IPCC4 data base. Thanks are also due to the anonymous reviewers, as well as to Dr. A. Gershunov, Dr. J-F. Royer and Dr. S. Conil for their helpful comments. This work has been supported by the European Commission Sixth Framework Program (ENSEMBLES contract GOCE-CT-2003-505539).
- Brooks N (2004) Drought in the African Sahel: long term perspectives and future prospects. Tyndall Centre working paper no 61: 31ppGoogle Scholar
- Douville H, Royer J-F, Polcher J, Cox P, Gedney N, Stephenson DB, Valdes PJ (2000) Impact of CO2 doubling on the Asian summer monsoon: robust versus model dependent responses. J Meteorol Soc Jpn 78:1–19Google Scholar
- Douville H (2005) Impact of regional SST anomalies on the Indian monsoon response to global warming in the CNRM climate model. J Climate (revised)Google Scholar
- IPCC (2001) Climate change 2001: the scientific basis. In: Houghton JT et al (eds) Cambridge University Press, CambridgeGoogle Scholar
- Jancovici J-M (2004) Energy and climate change: discussing two opposite evolutions. J de Phys (Proce) 121:171–184Google Scholar
- Liepert BG, Feichter J, Lohmann U, Roeckner E (2004) Can aerosols spin down the water cycle in a warmer and moister world? Geophys Res Lett 31: doi:10.1029/2003GL019060Google Scholar
- Rayner NA, Parker DE, Horton EB, Folland CK, Alexander LV, Rowell DP, Kent EC, Kaplan A (2003) Global analyses of sea surface temperature, sea ice and night maritime air temperature since the late nineteenth century. J Geophys Res 108: doi:10.1029/2002JD002670Google Scholar