A new empirical parameterization (EP) for multiple groups of primary biological aerosol particles (PBAPs) is implemented in the aerosol-cloud model (AC) to investigate their roles as ice nucleating particles (INPs). The EP describes the heterogeneous ice nucleation by (1) fungal spores, (2) bacteria, (3) pollen, (4) detritus of plants, animals, and viruses, and (5) algae. Each group includes fragments from the originally emitted particles. A high-resolution simulation of a midlatitude mesoscale squall line by AC is validated against airborne and ground observations.Sensitivity tests are carried out by varying the initial vertical profiles of the loadings of individual PBAP groups. The resulting changes in warm and ice cloud microphysical parameters are investigated. The changes in warm microphysical parameters, including liquid water content and cloud droplet number concentration, are minimal (<10 %). Overall, PBAPs have little effect on the ice number concentration (<6 %) in the convective region. In the stratiform region, increasing the initial PBAP loadings by a factor of 1000 resulted in less than 40 % change in ice number concentrations. The total ice concentration is mostly controlled by various mechanisms of secondary ice production (SIP). However, when SIP is intentionally shut down in sensitivity tests, increasing the PBAP loading by a factor of 100 has an effect of less than 3 % on the ice phase. Further sensitivity tests revealed that PBAPs have little effect on surface precipitation and on the shortwave and longwave flux (<4 %) for a 100-fold perturbation in PBAPs.

Ice fragments are generated by sublimation of ice particles in subsaturated conditions in natural clouds. Conceivably, such sublimational breakup would be expected to cause ice multiplication in natural clouds. Any fragment that survives will grow to become ice precipitation that may sublimate and fragment further.As a first step towards assessing this overlooked process, a formulation is proposed for the number of ice fragments from sublimation of ice particles for an atmospheric model. This is done by amalgamating laboratory observations from previously published studies. The concept of a ‘sublimated mass activity spectrum’ for the breakup is applied to the dataset. The number of ice fragments is determined by the relative humidity over ice and the initial size of the parent ice particles. The new formulation applies to dendritic crystals and heavily rimed particles only.Finally, a thought experiment is performed for an idealized scenario of subsaturation with in-cloud descent. Scaling analysis yields an estimate of an ice enhancement ratio of about 5 (50) within a weak deep convective downdraft of about 2 m s-1, for an initial monodisperse population of dendritic snow (graupel) particles of 3 L-1 and 2 mm . During descent, there is a dynamic equilibrium between continual emission of fragments and their depletion by sublimation. A simplified bin microphysics parcel model exhibits this dynamical quasi-equilibrium, consistent with the thought experiment. The fragments have average lifetimes of around 90 and 240 seconds for dendrites and graupel respectively. Sublimational breakup is predicted to cause significant secondary ice production.

&lt;p&gt;&amp;#160;Clouds are a fundamental aspect of the Earth&amp;#8217;s atmosphere. One of the major challenges in cloud-resolving models (CRM) is the formation and generation of new cloud ice particles from pre-existed ice and liquid. Based on the basic broad cloud types, it is helpful to distinguish between their fundamental microphysical properties. The four basic cloud types are defined as: (1) warm-based convective and stratiform clouds; and (2) cold-based convective and stratiform clouds. Recent studies of ice initiation in clouds have shown that most ice particles in the mixed-phase region of clouds are from secondary ice production (SIP) mechanisms but have generally concentrated on only one specific cloud system.&lt;/p&gt; &lt;p&gt;In this study, Aerosol-Cloud model (AC) is used. AC includes the four mechanisms of secondary ice production as follows: ice-ice collisional breakup, raindrop freezing fragmentation, Hallett-Mossop (HM) process and sublimational breakup. The intent is to generalize the contribution of each SIP mechanism among basic cloud types. The numerical simulations are performed using our AC for each cloud type and validated against in-situ cloud observations. The observational data is collected during four different cloud observational campaigns, each representing a contrasting cloud type than others.&lt;/p&gt; &lt;p&gt;Here, we study the contributions from each process of SIP (HM process, ice-ice collisional breakup, raindrop-freezing fragmentation and sublimational breakup) by performing control simulations of each basic cloud type. For the warm cloud convective clouds, the HM process prevails near freezing level and contributes significantly from 0 to -15&lt;sup&gt;o&lt;/sup&gt;C. In cold-based convective clouds, the ice-ice collisional breakup is the most dominating SIP mechanism in each cloud type. In warm-based stratiform clouds, the HM process dominates the contribution of ice in the -5 to -15&lt;sup&gt;o&lt;/sup&gt;C temperature range for updrafts up to 8 m/s. In the slightly warm-based convective clouds, the breakup due to ice-ice collision is the most dominating mechanism for the convective updrafts between -5&lt;sup&gt;o&lt;/sup&gt;C and cloud top temperatures.&amp;#160;&lt;/p&gt;

The role of time-dependent freezing of ice nucleating particles (INPs) is evaluated with the ‘Aerosol-Cloud’ (AC) model in: 1) deep convection observed over Oklahoma during the Midlatitude Continental Convective Cloud Experiment (MC3E), 2) orographic clouds observed over North California during the Atmospheric Radiation Measurement (ARM) Cloud Aerosol Precipitation Experiment (ACAPEX), and 3) supercooled, stratiform clouds over the UK, observed during the Aerosol Properties, Processes And Influences on the Earth’s climate (APPRAISE) campaign. AC uses the dynamical core of the WRF model and has hybrid bin/bulk microphysics and a 3D mesoscale domain. AC is validated against coincident aircraft, ground-based and satellite observations for all three cases. Filtered concentrations of ice (> 0.1 to 0.2 mm) agree with those observed at all sampled levels. AC forms ice heterogeneously through condensation, contact, deposition, and immersion freezing. AC predicts the INP activity of various types of aerosol particles with an empirical parameterization (EP), which follows a singular approach (no time dependence). Here, the EP is modified to represent time-dependent INP activity by a purely empirical approach, using our published laboratory observations of time-dependent INP activity. In all simulated clouds, the inclusion of time dependence increases the predicted INP activity of mineral dust particles by 0.5 to 1 order of magnitude. However, there is little impact on the cloud glaciation because the total ice is mostly (80-90%) from secondary ice production (SIP) at levels warmer than about −36°C. The Hallett-Mossop process and fragmentation in ice-ice collisions together initiate about 70% of the total ice, whereas fragmentation during both raindrop freezing and sublimation contributes < 10%. Overall, total ice concentrations and SIP are unaffected by time-dependent INP activity.