Rachel L. James

The accurate representation of secondary ice production (SIP) is essential for describing the microphysics of deep convective clouds, yet the exact mechanisms and efficiencies of SIP are still uncertain. In this study, we used the University of Manchester bin microphysics parcel model to investigate four SIP parameterisations: rime splintering, ice-ice collisional breakup, spherical freezing fragmentation of drops (mode 1), and fragmentation between supercooled droplets and more massive ice particles (mode 2). We simulated air trajectories through deep convective clouds observed during the Deep Convective Microphysics EXperiment (DCMEX) field campaign. Our results show that mode 2, as currently understood and represented in the model, is the key mechanism for explaining the high ice particle concentrations observed. We also present the first systematic study of how different entrainment conditions (homogeneous and inhomogeneous mixing) affect SIP mechanisms. In our simulations, homogeneous and inhomogeneous mixing with aerosol entrainment provide reasonable agreement with cloud-core and cloud-edge microphysical properties observed during DCMEX, respectively. In addition, the peak liquid water content values in the adiabatic simulations was approximately 300 % of that in the entraining simulations. The entrainment of external aerosols was also found to accelerate the collision-coalescence process under homogeneous mixing, leading to earlier ice enhancement. Our results show that SIP mechanisms, which are dependent on large droplets, such as mode 2, are highly sensitive to different entrainment conditions, emphasising the importance of representing entrainment correctly when including SIP processes in large-scale models.

Accurate representation of secondary ice production (SIP) is essential for describing the microphysics of deep convective clouds, yet the dominant mechanisms and their efficiencies remain uncertain. In this study, we use the University of Manchester bin microphysics parcel model to investigate four SIP parameterisations, including rime splintering, ice–ice collisional breakup, and two modes of droplet freezing fragmentation. Air parcel trajectories are simulated through deep convective clouds observed during the Deep Convective Microphysics Experiment (DCMEX) field campaign. The results show that fragmentation between supercooled droplets and more massive ice particles (mode 2) plays a key role in explaining the high ice particle concentrations observed. We further present a systematic assessment of how different entrainment representations, including adiabatic, homogeneous, and inhomogeneous mixing, influence secondary ice production. Homogeneous and inhomogeneous mixing with aerosol entrainment provide reasonable agreement with cloud-core and cloud-edge microphysical properties observed during DCMEX, respectively. The entrainment of external aerosols is found to accelerate the collision–coalescence process under homogeneous mixing, leading to earlier ice enhancement, while having little effect under inhomogeneous mixing.

Accurate representation of secondary ice production (SIP) is essential for describing the microphysics of deep convective clouds, yet the dominant mechanisms and their efficiencies remain uncertain. In this study, we use the University of Manchester bin microphysics parcel model to investigate four SIP parameterisations, including rime splintering, ice–ice collisional breakup, and two modes of droplet freezing fragmentation. Air parcel trajectories are simulated through deep convective clouds observed during the Deep Convective Microphysics Experiment (DCMEX) field campaign. The results show that fragmentation between supercooled droplets and more massive ice particles (mode 2) plays a key role in explaining the high ice particle concentrations observed. We further present a systematic assessment of how different entrainment representations, including adiabatic, homogeneous, and inhomogeneous mixing, influence secondary ice production. Homogeneous and inhomogeneous mixing with aerosol entrainment provide reasonable agreement with cloud-core and cloud-edge microphysical properties observed during DCMEX, respectively. The entrainment of external aerosols is found to accelerate the collision–coalescence process under homogeneous mixing, leading to earlier ice enhancement, while having little effect under inhomogeneous mixing.

The hydroxyl radical (OH) plays a key role in regulating gas-phase pollutant concentrations and particle formation and composition in the global troposphere. Globally, OH formation is dominated by production of O(1D) from photolysis of ozone and subsequent reaction with water vapour. However, under conditions characteristic of the polluted wintertime Arctic, this mechanism is inhibited by limited short-wave radiation and water abundance. A lack of observations means that our understanding of radical and oxidant sources under such conditions is lacking. The Alaskan Layered Pollution and Chemical Analysis (ALPACA) field campaign, made comprehensive measurements of trace gas and aerosol pollution chemistry and meteorology during January and February 2022 in the sub-Arctic city of Fairbanks, Alaska, USA. During the campaign, severe surface-based temperature inversions gave rise to several enhanced pollution events, interspersed with weakly stable periods of lower pollution levels and influence from the free troposphere. Here, we use the Dynamically Simple Model of Atmospheric Chemical Complexity (DSMACC) 0D box model incorporating the comprehensive Master Chemical Mechanism (MCM) v3.3.1 to quantify processes controlling the abundances of atmospheric oxidants in Fairbanks. We constrain the model using available measurements from the ALPACA campaign, and compare HOx radical sources and sinks during a strongly-stable heavily polluted period of the campaign with a less stable, cleaner period. OH formation from HONO photolysis is the major chemical source of HOx in the polluted, low-light environment of wintertime Fairbanks and formaldehyde is an important precursor for HO2. We find that the HOx budget is distinct between a polluted, strongly stable inversion event and a clean, weakly stable period, with influences of meteorology being important in regulating peroxynitric acid (PNA, HO2NO2) cycling. Our results help improve understanding of the unique atmospheric pollution chemistry that regulates oxidant abundances in cold and low-light conditions.

Accurate representation of secondary ice production (SIP) is essential for describing the microphysics of deep convective clouds, yet the dominant mechanisms and their efficiencies remain uncertain. In this study, we use the University of Manchester bin microphysics parcel model to investigate four SIP parameterisations, including rime splintering, ice–ice collisional breakup, and two modes of droplet freezing fragmentation. Air parcel trajectories are simulated through deep convective clouds observed during the Deep Convective Microphysics Experiment (DCMEX) field campaign. The results show that fragmentation between supercooled droplets and more massive ice particles (mode 2) plays a key role in explaining the high ice particle concentrations observed. We further present a systematic assessment of how different entrainment representations, including adiabatic, homogeneous, and inhomogeneous mixing, influence secondary ice production. Homogeneous and inhomogeneous mixing with aerosol entrainment provide reasonable agreement with cloud-core and cloud-edge microphysical properties observed during DCMEX, respectively. The entrainment of external aerosols is found to accelerate the collision–coalescence process under homogeneous mixing, leading to earlier ice enhancement, while having little effect under inhomogeneous mixing.

We describe a novel aspect of the structure of the hexagonal phase of crystalline water ice, using vacuum-ultraviolet absorption spectroscopy as a diagnostic tool. The key result is that hexagonal ice (I_h), prepared from condensation of pure water vapour, displays net molecular orientation. This reflects order in the disposition of the protons in ice I_h and our results require revision of the near century old conventional structure in which water molecular dipoles point in random directions in hexagonal ice. Orientation gives rise to spontaneous polarization and internal fields of ∼4 × 10⁶ V m⁻¹ in films of ice I_h tens of nanometres thick, condensed on an insulating MgF₂ substrate. Data are in the form of Stark shifts in absorption spectra, with no free charges or external electric fields introduced in the experiment.

Accurate representation of secondary ice production (SIP) is essential for describing the microphysics of deep convective clouds, yet the dominant mechanisms and their efficiencies remain uncertain. In this study, we use the University of Manchester bin microphysics parcel model to investigate four SIP parameterisations, including rime splintering, ice–ice collisional breakup, and two modes of droplet freezing fragmentation. Air parcel trajectories are simulated through deep convective clouds observed during the Deep Convective Microphysics Experiment (DCMEX) field campaign. The results show that fragmentation between supercooled droplets and more massive ice particles (mode 2) plays a key role in explaining the high ice particle concentrations observed. We further present a systematic assessment of how different entrainment representations, including adiabatic, homogeneous, and inhomogeneous mixing, influence secondary ice production. Homogeneous and inhomogeneous mixing with aerosol entrainment provide reasonable agreement with cloud-core and cloud-edge microphysical properties observed during DCMEX, respectively. The entrainment of external aerosols is found to accelerate the collision–coalescence process under homogeneous mixing, leading to earlier ice enhancement, while having little effect under inhomogeneous mixing.

Accurate representation of secondary ice production (SIP) is essential for describing the microphysics of deep convective clouds, yet the dominant mechanisms and their efficiencies remain uncertain. In this study, we use the University of Manchester bin microphysics parcel model to investigate four SIP parameterisations, including rime splintering, ice–ice collisional breakup, and two modes of droplet freezing fragmentation. Air parcel trajectories are simulated through deep convective clouds observed during the Deep Convective Microphysics Experiment (DCMEX) field campaign. The results show that fragmentation between supercooled droplets and more massive ice particles (mode 2) plays a key role in explaining the high ice particle concentrations observed. We further present a systematic assessment of how different entrainment representations, including adiabatic, homogeneous, and inhomogeneous mixing, influence secondary ice production. Homogeneous and inhomogeneous mixing with aerosol entrainment provide reasonable agreement with cloud-core and cloud-edge microphysical properties observed during DCMEX, respectively. The entrainment of external aerosols is found to accelerate the collision–coalescence process under homogeneous mixing, leading to earlier ice enhancement, while having little effect under inhomogeneous mixing.

As part of the CERTAINTY project, we present results from our laboratory study of secondary ice production (SIP) from collisions between supercooled raindrops and more massive ice particles, building on our previous proof-of-concept work (James et al., 2021) with a refined experimental setup for improved quantification.In mixed phase clouds, ice formation can occur via two pathways: primary ice formation, via ice nucleating particles, or secondary ice production (SIP). Observations in both shallow and deep convective clouds often show ice concentrations that exceed those predicted by primary ice nucleation by several orders of magnitude. However, parameterisations of SIP mechanisms remain poorly constrained due to limited laboratory data.One proposed SIP mechanism involves collisions between supercooled raindrops and more massive ice particles, where secondary drops may form during impact, with some freezing to create secondary ice. Our previous work (James et al., 2021) demonstrated the viability of this SIP mechanism, and showed that approximately 30 % of the secondary drops froze under a limited set of conditions. Building on this, we have refined our experimental setup to reduce uncertainties in the freezing fraction of secondary drops by elevating the ice particle to allow the splashing to occur freely, without interference of a flat surface used in our previous experiments. We also explore a broader range of supercooled water drop diameters, ice particle sizes, impact velocities and temperatures to better reflect cloud conditions.Finally, we incorporate our updated results into a parcel model with bin microphysics for idealised clouds, which previously used our earlier results (James et al. 2023), to demonstrate the impact this improved quantification in conjunction with other SIP mechanisms.ReferencesJames, R. L., Phillips, V. T. J., and Connolly, P. J. (2021), Secondary ice production during the break-up of freezing water drops on impact with ice particles, Atmos. Chem. Phys., 21, 18519–18530, https://doi.org/10.5194/acp-21-18519-2021.James, R. L., Crosier, J., and Connolly, P. J. (2023), A bin microphysics parcel model investigation of secondary ice formation in an idealised shallow convective cloud, Atmos. Chem. Phys., 23, 9099–9121, https://doi.org/10.5194/acp-23-9099-2023.