Investigation of the Aerosol Indirect Effect on Ice Clouds and its Climatic Impact Using A-Train Satellite Data and a GCM Yu Gu 1 , Jonathan H. Jiang 2 , Hui Su 2 , and K. N. Liou 1 1 Department of Atmospheric and Oceanic Sciences 2 Jet Propulsion Laboratory, California Institute of Technology and Joint Institute for Regional Earth System Science and Engineering 4800 Oak Grove Drive, Pasadena, California, 91109 University of California, Los Angeles, Los Angeles, California UCLA AG CM PhysicalParam eterizations Planetary boundary layerprocesses:Suarez etal. (1983), Lietal. (1999, 2001) Cum ulus Convection:Prognostic A rakaw a-Schubert(Pan and Randall1998),w ith dow ndrafts(Cheng and A rakaw a 1997) Radiation:H arshvardhan etal.(1987)(Controlrun) Fu and Liou (1992,1993), G u etal. (2003) Prognostic Cloud W ater/Ice:K ohler(1999)+ Fractionalclouds/Cloud overlap (G u etal. 2003) G ravity W aterD rag:K im and A rakaw a (1995) Dynam ics H orizontalFinite D ifferenceSchem e:A rakaw a and Lam b (1981) Resolution:5 longitude x 4 latitude VerticalFinite DifferenceSchem e:Suarez and A rakaw a (1983) Resolution (top at1 hPa):15 layers Tim e integration:Leapfrog, M atsuno Surface C onditions Prescribed sea surface tem peratures (Rayneretal.1995),albedo,ground w etness, and surface roughness(D orm an and Sellars1989) Parameterizations for Ice Number Summary GCM Simulation Results Param eters Polluted (A O T=0.5) Clean (A O T=0.2) Polluted-Clean Precipitation (m m /day) 2.831 2.855 -0.024 O LR (W /m 2 ) 212.839 213.775 -0.936 Flux atSFC (W /m 2 ) 142.264 143.423 -1.159 Flux atTO A (W /m 2 ) 207.332 208.313 -0.981 Cloud Cover(% ) 63.470 63.213 0.257 SFC A ir Tem perature (K ) 281.440 281.746 -0.306 None of the De parameterizations accounted for the distinction between “polluted” and “clean” clouds A conventional approach has been to prescribe a mean effective ice crystal size in GCMs (e.g., Köhler 1999; Ho et al. 1998; Gu et al. 2003) Use IWC and/or temperature produced from GCMs to determine a mean effective ice crystal size (Kristjánsson et al. 2005; Gu and Liou 2006; Ou and Liou 1995; Ou et al. 1995; McFarquhar et al. 2003; Liou et al. 2008). Model Description Parameterizations for Re Relate ice nucleation and number to aerosol concentration on the basis of explicit microphysics modeling, laboratory studies, as well as theoretical considerations (e.g., Diehl and Mitra 1998; karcher and Lohmann, 2002, 2003; Riemer et al. 2004; Liu and Penner 2005; Karcher et al. 2006) Mean effective ice crystal size calculated from ice mass and number for radiation calculations large uncertainties in the parameterization of ice microphysics processes and requirement of significant computational efforts Empirical Re, IWC, and AOT Relation Using least-squares fitting, we obtained an empirical formula for Re as a function of IWC and AOT. This function broadly captures the variation of Re with IWC and AOT. Satellite Observations =1.436, =0.5858, =0.282, =8.09, Re 0 =56.0293, IWC 0 =1.3838, =1.11 Precipitatio n OLR Cloud Cover Surface Solar Flux Surface Air Temperature Global July Mean Aerosol First Indirect Effect Change in cloud droplet/ice particle numbers associated with increase in aerosol number concentrations Cloud particle mean effective size Cloud radiative forcing Use of A-Train Satellite Data Investigation of the aerosol indirect effect on ice clouds have been limited due primarily to the lack of accurate global-scale observations. New data from the NASA’s A-Train constellation makes it possible to examine the aerosol-cloud interaction in a more comprehensive way that can lead to improved physical understanding of this interaction. The A-Train is a constellation of 6 satellites spaced a few minutes apart from each other and so their collective observations can be used to construct high-definition three dimensional images of the earth’s atmosphere and surface. Polluted Clouds (Aerosol 1 st Indirect Effect): Less OLR Less solar radiation reaching the surface More reflected solar radiation at TOA Increased cloud cover Reduced precipitation GCM simulations show that the global averaged OLR and net solar flux at TOA are smaller in polluted case, illustrating more reflected solar flux and trapped OLR due to smaller De. Global radiative forcing produced by the aerosol 1 st indirect effect is about 0.94 W/m 2 for IR and - 0.98 W/m 2 for solar radiation. Surface solar flux is also reduced, resulting in colder surface air temperature. Reduced precipitation and increased cloud cover are found globally in response to the aerosol 1 st indirect effect. Changes in the precipitation pattern show that, due to the aerosol 1 st indirect effect, reduced precipitation is found in the regions Left Fig: Satellite observations indicate Re increases with IWC but decreases with AOT Right: an empirical model derived from fitting of satellite data, as described below. 1 AOT AOT 1 IWC IWC R R 0 0 e e β γ α This empirical relationship of Re with IWC and AOT can serve as a first-order parameterization of the first indirect effect of aerosols on ice clouds for application to climate models.