Mechanisms for the Influence from Ice Nucleus Aerosols on Clouds and their Indirect Effects: Cloud Modelling

Sammanfattning: The role of multiple groups of primary biological aerosol particles (PBAPs) as ice nucleating particles (INPs), and of ice formation processes such as time-dependent freezing of various INPs, and various secondary ice production(SIP) mechanisms in overall ice concentration has been evaluated in a range of cloud systems by simulating them numerically with the state-of-the-art ‘Aerosol-Cloud’ (AC) model in a 3D mesoscale domain. Also, the mechanismsof aerosol indirect effects (AIEs) arising from anthropogenic INPs, and the responses to these AIEs from time-dependent INP freezing and SIP processes are investigated in the simulated clouds. The cloud systems simulated with AC are: events of summertime deep convection observed over Oklahoma, USA during the Midlatitude Continental Convective Cloud Experiment (MC3E) in 2011 on 1) 11 May, and 2) 20 May, and wintertime 3) orographic clouds observed during the Atmospheric Radiation Measurement Cloud Aerosol Precipitation Experiment (ACAPEX) on 07 February 2015 over North California, and 4) supercooled layer clouds observed over Larkhill, UK, during the Aerosol Properties, Processes And Influences on the Earth’s climate (APPRAISE) campaign on 18 February 2009.AC uses the dynamical core of the Weather Research and Forecasting (WRF) model, modified Geophysical Fluid Dynamic Laboratory (GFDL) radiation scheme, and hybrid bin-bulk microphysics scheme. AC is validated adequately with the coincident aircraft, ground-based, and satellite observations for all four cases. AC forms secondary ice through the Hallett-Mossop (HM) process of rime-splintering, and fragmentation during ice-ice collisions, raindrop freezing, and sublimation of dendritic snow and graupel. A measure of SIP is defined using theterm ‘ice enhancement’ (IE) ratio which is the ratio between the number concentration of total ice particles and active INPs at cloud tops. For both cases in MC3E, overall, PBAPs have little effect (+1-6%) on the cloud-liquid (droplet mean sizes, number concentrations, and their water contents) properties, overall ice concentration, and on precipitation. AC predicts the activity of various INPs with an empirical parameterization (EP). The EP is modified to represent the time-dependent approach of INP freezing in light of our published laboratory observations. It is predicted that the time dependence of INP freezing is not the main cause for continuous ice nucleation and precipitation in all simulated cases. Rather, the main mechanism of precipitation formation is the combination of various SIP mechanisms (in convection) and recirculation-reactivation of dust particles (in APPRAISE layer cloud episode). Also, for all cases, the inclusion oftime dependence of INP freezing causes little increase (about 10-20%) in the total ice concentration and ice from all SIP.Regarding SIP, in young developing convective clouds of MC3E (11 May), with tops >−15oC, the initial explosive growth is from the fast HM process, creating IE ratios as high as 103. By contrast, in mature convective clouds (tops <−20oC), fragmentation in ice-ice collisions prevails, creating IE ratios of up to about 102-103. Regarding AIEs from INPs, increasing anthropogenic pollution is predicted to exert a net cooling in APPRAISE, and a strong net warmingin MC3E (11 May). Furthermore, these net AIEs are mainly from glaciated clouds. Overall, the contribution to the AIEs from ice formation processes, such as time-dependent INP freezing and SIP, shows a high sensitivity with respect to anthropogenic INPs (about 20-60% increase in net AIEs). Also, two new indirect effects associated with ice initiation mechanisms are proposed here. These are, 1) the ‘SIP’ indirect effect, and 2) the ‘time-dependent INP’ indirect effect. It is predicted that in APPRAISE and MC3E, both SIP and time-dependent INP indirect effects form less than 30%, and more than 50% of the net AIE, respectively.