1. r.p.allan@reading.ac.uk© University of Reading 20091 Improved understanding of how rainfall responds to a warming world Richard Allan Environmental Systems Science Centre, Reading

2. r.p.allan@reading.ac.uk© University of Reading 20092 Increased Precipitation More Intense Rainfall More droughts Wet regions get wetter, dry regions get drier Precipitation Change (%) CLIMATE MODEL PROJECTIONS IPCC WGI Precipitation Intensity Dry Days

3. r.p.allan@reading.ac.uk© University of Reading 20093 How should the water cycle respond to climate change? Precipitation Change (%) relative to 1961-1990: 2 scenarios, multi model (IPCC, 2001) See discussion in: Allen & Ingram (2002) Nature; Trenberth et al. (2003) BAMS

4. r.p.allan@reading.ac.uk© University of Reading 20094 Water vapour rises with temperature in models & observations in accordance with Clausius Clapeyron equation John et al. (2009) GRL models Water Vapour (mm)

5. r.p.allan@reading.ac.uk© University of Reading 20095 1979-2002 For a given precipitation event, more moisture would suggest more intense rainfall Can realism of model projections be assessed?

6. r.p.allan@reading.ac.uk© University of Reading 20096 Methodology Analyse daily rainfall over tropical oceans –SSM/I satellite data, use 1988-2005 –Climate model data (AMIP experiments) Create monthly PDFs of rainfall intensity Calculate changes in the frequency of events in each intensity bin Does frequency of most intense rainfall rise with atmospheric warming?

7. r.p.allan@reading.ac.uk© University of Reading 20097 Daily Satellite Microwave Observations over tropical ocean appear to confirm warmer months are associated with more frequent intense rainfall Allan and Soden (2008) Science More frequentLess frequent

8. r.p.allan@reading.ac.uk© University of Reading 20098 Frequency of rainfall intensities vary with SST in models and obs Frequency of intense rainfall increases with warming in models and satellite data Model scaling close to 7%/K expected from Clausius Clapeyron SSM/I satellite data suggest a greater response of intense rainfall to warming dP/dSST=7%/K Allan and Soden (2008) Science

9. r.p.allan@reading.ac.uk© University of Reading 20099 Trenberth et al. (2009) BAMS This is not the whole story…

10. r.p.allan@reading.ac.uk© University of Reading 200910 Models simulate robust response of clear-sky radiation to warming (~2-3 Wm -2 K -1 ) and a resulting increase in precipitation to balance (~3 %K -1 ) e.g. Allen and Ingram (2002) Nature, Stephens & Ellis (2008) J. Clim Radiative cooling, clear (Wm -2 K -1 ) Allan (2009) J Clim

11. r.p.allan@reading.ac.uk© University of Reading 200911 Contrasting precipitation response expected Precipitation Heavy rain follows moisture (~7%/K) Mean Precipitation linked to radiation balance (~3%/K) Light Precipitation (-?%/K) Temperature e.g.Held & Soden (2006) J. Clim; Trenberth et al. (2003) BAMS; Allen & Ingram (2002) Nature

12. r.p.allan@reading.ac.uk© University of Reading 200912 Contrasting precipitation response in wet and dry portions of the tropical circulation GPCP/NCEPModels ascent descent Precipitation change (mm/day) 4 %/dec -9 %/dec 0.6 %/dec -0.1 %/dec All Tropics: 0.3 %/dec All Tropics: ~1 %/dec Allan and Soden (2007) GRL; John et al. (2009) GRL

13. r.p.allan@reading.ac.uk© University of Reading 200913 Conclusions Low level moisture responses robust Explains enhanced frequency of intense rainfall with warming in models and obs –Differences between individual models/obs Mean and regional precipitation response: a tug of war –Slow rises in radiative cooling (~2-3 Wm -2 K -1 ) –Rises in low-level moisture (~7%/K) faster than precipitation (~3%/K) –Reduced frequency? Rich get richer? –Who cares about drought/flooding over the ocean? Precipitation Responses appear larger in observations than models. –Could aerosol be influencing decadal variability in the hydrological cycle? –Are observing systems up to monitoring changes in the water cycle? –Special issue of ERL on water cycle (2010) –New NERC project PREPARE with Met Office/ETH Zurich starting soon

14. r.p.allan@reading.ac.uk© University of Reading 200914 Extra Slides

15. r.p.allan@reading.ac.uk© University of Reading 200915 Large spread in the response of the heaviest precipitation to warming between models and compared with satellite data. But intense vertical motion and PDF of precipitation events in models are unrealistic: Wilcox and Donner (2007) J Clim; Field and Shutts (2009) QJ; OGorman and Schneider (2009) PNAS Models Change in Frequency of Precipitation (% per K warming) in Bins of Intensity Light RainHeavy Rain

16. r.p.allan@reading.ac.uk© University of Reading 200916 A=0.4(1-A)=0.6 dP w /dT=7%/KdP d /dT dP/dT=3%/K Assume wet region follows Clausius Clapeyron (7%/K) and mean precip follows radiation constraint (~3%/K) dP / dT = A( dP w / dT )+(1-A)( dP d / dT ) dP d = (dP-AdP w )/(1-A) P w =6 mm/dayP d =1 mm/day P=3 mm/day WetDry APwPw PdPd dP d /dTs (mm/day/K) (%/K) 0.4 0.2 6969 1 1.5 -0.1 -0.05 -10 -4.5 0.110.52.2+0.02+0.9 A is the wet region fractional area P is precipitation T is temperature

17. r.p.allan@reading.ac.uk© University of Reading 200917 Could changes in aerosol be imposing direct and indirect changes in the hydrological cycle? e.g. Wild et al. (2008) GRL Wielicki et al. (2002) Science; Wong et al. (2006) J. Clim; Loeb et al. (2007) J. Clim Mishchenko et al. (2007) Science

18. r.p.allan@reading.ac.uk© University of Reading 200918 Are the issues of cloud feedback and the water cycle linked? 2008 Allan et al. (2007) QJRMS

19. r.p.allan@reading.ac.uk© University of Reading 200919 Towards regional prediction of the water cycle…

20. r.p.allan@reading.ac.uk© University of Reading 200920 Conclusions Low level moisture responses robust –Less clear over land and at higher levels. –Inaccurate model mean state? Precipitation extremes linked to moisture –Moisture response at lowest level? –Changes in updraft velocity? –Differences between individual models/obs Mean and regional precipitation response: a tug of war –Slow rises in radiative cooling (~3 Wm -2 K -1 ) –Rises in low-level moisture (~7%/K) faster than precipitation (~3%/K) –Reduced frequency? Wet get wetter and dry get drier –Who cares about drought/flooding over the ocean? Recent Precipitation Responses appear larger in observations than models. –Could aerosol be influencing decadal variability in the hydrological cycle? –Are observing systems up to monitoring changes in the water cycle? Understanding changes in near surface conditions may be important

21. r.p.allan@reading.ac.uk© University of Reading 200921 Precipitation response depends upon the forcing and the feedback Andrews et al. (2009) J Climate Partitioning of energy between atmosphere and surface is crucial to the hydrological response

22. r.p.allan@reading.ac.uk© University of Reading 200922 TRMM ASCENT ALL DESCENT Contrasting ascent/descent response robust Magnitude of responses sensitive to time period/dataset John et al. (2009) GRL The descent region precipitation is particularly sensitive to inaccuracies in the reanalyses and how this varies with time.