The cloud radiative effect on the atmospheric energy budget and global mean precipitation F. Hugo Lambert 1, Mark J. Webb, Masa Yoshimori 3 and Tokuta Yokohata 1. University of Exeter. Met Office 3. Hokkaido University. NIES July, 1
Outline Slab model CO versus control HadSM3 (3-hourly hour snapshots), QUMP (annual) and MIROC3. (3-hourly) experiments. Introduction and hypotheses Selected results Discussion
Introduction (1) Try to understand the relatively robust nature of cloud impacts on atmospheric longwave flux changes during climate change. With applications to why modelled global mean precipitation changes are in the range 1 3 %K 1.
Introduction (): Radiative and turbulent flux changes Wm 1 8 6 (a) Latent heat release, L P CAM3 CCCma GFDL GISS E-R other CFMIP 6 T (K) (c) LW cloud net radiative cooling Wm 1 8 6 1 (b) Clear-sky net radiative cooling 6 T (K) (d) Net sensible cooling Clear-sky radiative cooling. (We think we understand that quite well?) Turbulent fluxes. Sensible versus latent. Wm - Wm 8 6 Cloud radiative cooling. (LW only.) -6-8 -1 6 T (K) 6 T (K) Source: (3).
Introduction (3): The clear-sky environment (a) TOA clear-sky flux (b) Surface clear-sky flux (c) Atmospheric clear-sky absorption 1 1 1 net downward radiative flux [Wm ] -1 net downward radiative flux [Wm ] -1 radiative absorption anomaly [Wm ] -1 - - - 1 3 1 3 1 3 All components Temperature Water vapour See also (; 1; )...
With a focus on the tropics, we ask... Why are changes in cloud radiative effect robust across models? Hypotheses and ideas: 1 The cloud masking effect. FAT 3 Subtropical low cloud fraction changes are unimportant to LW? Cloud bases to stay at the same pressure. Geographical changes in ITCZ unimportant?
Results: global means net downward radiative flux [Wm ] net downward radiative flux [Wm ] net downward radiative flux [Wm ] (a) TOA CRE - 1 3 (d) Surface CRE - 1 3 (g) Atmos. CRE - 1 3 net downward radiative flux [Wm ] net downward radiative flux [Wm ] net downward radiative flux [Wm ] (b) TOA cloud masking - 1 3 (e) Surface cloud masking - 1 3 (h) Atmos. cloud masking - 1 3 net downward radiative flux [Wm ] net downward radiative flux [Wm ] net downward radiative flux [Wm ] (c) TOA cloud PRP - 1 3 (f) Surface cloud PRP - 1 3 (i) Atmos. cloud PRP - 1 3 1 The cloud masking effect is indeed robust but it doesn t explain everything. The results are similar in the tropics.
Now focus on the tropics (a) TOA CRE (b) TOA cloud masking (c) TOA cloud PRP 1 1 1 cosθ. Wm -6-3 3 6 (d) Surface CRE cosθ. Wm -6-3 3 6 (e) Surface cloud masking cosθ. Wm -6-3 3 6 (f) Surface cloud PRP HadSM3 (solid) MIROC (dash) QUMP (shade) 1 1 1 cosθ. Wm cosθ. Wm cosθ. Wm cosθ. Wm 1-6 -3 3 6 (g) Atmos. CRE -6-3 3 6 cosθ. Wm -6-3 3 6 (h) Atmos. cloud masking 1-6 -3 3 6 cosθ. Wm 1-6 -3 3 6 (i) Atmos. cloud PRP -6-3 3 6 These series are weighted by cos θ, so you can see how each region contributes to global mean CRE.
Tropical cloud changes (as a function of ) (a) Tropical cloud change in HadSM3 (b) FAT hpa 6 6.6. FAT: FAT hypothesis of (). 8 8. hpa 1 3 1 1 3 (c) FATPAT 6 8 1 3 1 1 3 1 3 1 1 3 (d) a c 6 8 1 3 1 1 3....6 FATPAT: FAT vertical cloud means, but rearranged horizontally to mimic CO.
Tropical cloud changes (as a function of longitude) (a) Tropical cloud change in HadSM3 (b) FAT hpa 6 6.6. FAT: FAT hypothesis of (). 8 8. hpa 1 6 8 1 1 1 (c) FATPAT 1 1 longitude 1 6 8 1 1 1 (d) a c 1 1 longitude....6 FATPAT: FAT vertical cloud means, but rearranged horizontally to mimic CO.
Does FAT predict tropical PRP? (a) Tropical mean TOA cloud forcing and feedback forcing term [Wm ] forcing term [Wm ] forcing term [Wm ].. -. -1. -1... -. -1. -1... -. -1. -1. CRE Cloud masking Cloud PRP FATPAT -... 1. 1. feedback term [Wm K -1 ] (b) Tropical mean surface cloud forcing and feedback -1. -1. -... feedback term [Wm K -1 ] (c) Tropical mean atmospheric cloud forcing and feedback -... 1. 1. feedback term [Wm K -1 ] HadSM3 (big cross) MIROC (big diamond) QUMP (stars) Not bad, but not as well as it predicts cloud cover above hpa. Not always the right sign. (Notice also the other panels and the correlations between forcings and feedbacks.)
Are changes in subtropical low cloud important? (a) TOA SW CRE QUMP (b) Surface LW cloud PRP QUMP 1 8 6 Surface LW vs Surface SW (data available) (c) TOA SW CRE QUMP σ (e) TOA SW CRE MIROC (d) Surface LW cloud PRP QUMP σ (f) Surface LW cloud PRP MIROC - -6-8 3 Differences in SW flux changes are very large both in their magnitude and differences between regions. LW fluxes considerably less so outside of specific regions. -1
Evaluate the hypotheses and ideas... 1 The cloud masking effect. Yes it s quite robust. No, it doesn t dominate CRE. FAT Controls relevant mid-high cloud amount. Doesn t always match PRP satisfactorily (but our treatment is a little approximate. 3 Subtropical low cloud fraction changes are unimportant to LW? It appears that way. Supported by idealised work, but more probably needed. Cloud bases to stay at the same pressure. Geographical changes in ITCZ unimportant? We ve done some extra simulations. It is difficult to imagine GCMs that produce much larger precipitation increases per degree warming within our current understanding.
Perspective I have presented the forcings and feedbacks perspective. It s a helpful but arbitrary breakdown. An alternative dynamical perspective might be useful at TOA: FAT maintains cloud tops at almost the same temperature. Cloud OLR stays almost the same. Surface and atmospheric temperatures increase, increasing OLR. Weak positive CRE should be generated as clouds mask the increasing OLR. In our GCMs, TOA CRE is quite close to zero, but it s negative! FAT s slight overestimate of high cloud increase may be important or further unresolved issues for mid and low level cloud.
[1] R. A. Colman and B. J. McAvaney. On tropospheric adjustment to forcing and climate feedbacks. Clim. Dyn., 36:169 168, 11. [] D. L. Hartmann and K. Larson. An important constraint on tropical cloud-climate feedback. Geophys. Res. Lett., 9:191 19,. [3] F. H. Lambert and M. J. Webb. Dependency of global mean precipitation on surface temperature. Geophys. Res. Lett., 3:doi:1.19/8GL3838, 8. [] A. G. Pendergrass and D. L. Hartmann. The atmospheric energy constraint on global-mean precipitation change. J. Clim., 7:77 768, 1. [] M. Previdi. Radiative feedbacks on global precipitation. Environ. Res. Lett., :doi:1.188/178 936///11, 1.