Aerosol - Cloud interactions in contrasted Marine Environments – ACME

Understanding the role of aerosols on the formation of cloud ice is addressed through explicit small-scale atmospheric modelling, which allows simulation of relevant processes in cloud dynamics and microphysics. Modelling also includes number distribution functions of aerosol particles, droplets, and ice particles as well as the mass distribution of the residual aerosol in drops and ice crystals using bin resolved microphysics as available in DESCAM (Detailed SCAvenging Model; Flossmann and Wobrock (2010)). Newly implemented processes in DESCAM are based on laboratory experiments for SIP mechanisms and on cutting-edge measurements for heterogeneous ice nucleation (such as the online PINE (Portable Ice Nucleation Experiment; Möhler et al. (2021)) chamber).
Also, measurements from different marine environments obtained during field experiments (EXAEDRE, HAIC, Sea2Coud …) are used to compare with DESCAM simulations. The deployed instruments provide, among others, observations of several microphysics parameters (such as, the liquid and ice particle size distributions and integrated water contents, aerosol properties, thermodynamics …).

The DESCAM model has been extended with an explicit representation of the three more frequent/important SIP mechanisms: rime-splintering (or Hallett-Mossop), frozen drop shattering and ice breakup. For each of the SIP mechanisms, the implementation was based on parameterisations developed via laboratory studies. These parameterisations, which were available in the literature, consider various assumptions regarding the properties of the newly-formed ice splinters and the thermodynamics conditions (such as in Cotton et al. (1986), Choularton et al. (1980), and Mansell and Ziegler (2013) for the Hallett-Mossop mechanism, and Phillips et al. (2018), Sullivan et al. (2018), and Lauber et al. (2021) for the frozen drop shattering). Regarding the ice breakup process, a new representation of this SIP mechanism as well as the size distribution of the newly formed ice splinters have been studied using lab experiments (Grzegorczyk et al., 2023). All these new SIP mechanisms have been tested in an idealised numerical testbed (which characterise a deep convective system) where thermodynamics conditions are initiated with a sounding observed during the HAIC campaign and the convection is triggered using a surface thermal bubble perturbation. The first results are indicating that ice breakup mechanism seems to be the most dominant SIP mechanism in the ice crystal formation on the temperatures range between 0°C and -35°C (whatever the assumptions used to represent the mechanisms). Moreover, considering all the SIP mechanisms, the simulated ice crystal concentrations are in better agreement with the mean HAIC observations. Also, the investigation of the ice-nucleating particles (INPs) relevant for ice crystal formation in mixed-phase clouds has been investigated using the recently acquired PINE mobile cloud chamber. The PINE instrument has been deployed at the Puy de Dôme (PdD) observatory (located at an altitude of 1,465 m, close to Clermont-Ferrand) and has permitted to identify the INPs ability of the aerosols and developed parameterisations of the heterogeneous ice nucleation process according to the origin of atmospheric air masses (not only for the oceanic air mass but also for all the atmospheric air conditions sampled at the PdD summit). The newly-developed parameterisations have been implemented in the DESCAM model using the same idealised framework as described before. Preliminary results indicate significant differences in the liquid/ice partitioning during the development stage of the cloud system according to the parameterisations. However, this difference in the liquid/ice partitioning seems to become negligible during the mature stage of the cloud.

In the following steps of the project, it is planned to use the new version of the DESCAM model to simulate a case study observed during each of the different campaigns (HAIC, EXAEDRE, etc.) in order to study the impact of the different ice formation processes (primary and secondary) in cloud systems formed under different thermodynamic and pollution conditions. Through the complementarity of observations and DESCAM simulations, the analysis of each case study will aim to discern the dominant mechanisms driving ice formation during the development and evolution of the studied cloud systems.
This study will thus provide a better understanding of aerosol-cloud interactions and their role in the precipitating and radiative properties of cloud systems observed in various marine environments.

All the scientific productions are available on the dedicated project website (https://acme.uca.fr/) and on the HAL platform.

Significant gaps exist in our understanding of aerosol-cloud interactions which are reflected in the uncertainties in weather forecasts and climate projections. One key aspect that is poorly understood is the role of the ice phase and the aerosol particles on which atmospheric ice crystals form, especially in mixed-phase clouds which are ubiquitous in the troposphere. Whether a cloud is predominantly composed of water or ice strongly influences cloud properties and feeds back into the cloud life cycle, precipitation formation, and radiative energy balance. Both the impacts of the aerosol particles on the formation of the ice crystals as well as on the ice crystal evolution, are suspected to play a dominant role in determining the properties of clouds.

Recent studies suggest that marine aerosols (sea-spray, DMS-derived sulfate particles) are not only good CCN (cloud condensation nuclei) but have also good INP (ice-nucleating particle) ability. Depending on the marine environment, the number concentration of aerosol particles can vary by several orders of magnitude, as is the case over the Mediterranean Sea and the North Atlantic Ocean (areas that can be more or less affected by continental sources), or can be low in an open ocean such as the South Pacific Ocean (where the lack of measurements also induces important biases in numerical simulations). This variability has important consequences on both the droplet size distribution and the formation and evolution of ice crystals.

The main objectives of ACME are to improve the understanding of the mechanisms of ice formation, the interaction between ice crystals and droplets as well as the role of the aerosol particles in the development and the evolution of clouds in contrasted marine environments (e.g. with different aerosol loadings and thermodynamical properties).

To reach these objectives, ACME will exploit the measurements from different marine environments: over the Mediterranean Sea during the EXAEDRE campaign (autumn 2018, Corsica), over the North Atlantic Ocean during the HAIC campaign (May 2015, French Guiana), and over the South Pacific Ocean during the Sea2Cloud campaign (March 2020, New-Zealand), in order to better constrain the representation of the primary and secondary ice nucleation mechanisms in bin cloud-scale modelling (i.e. with the DEtailed SCAvenging Model (DESCAM) using sub-kilometre resolution), providing then a unique tool for aerosol-cloud interaction studies.

ACME will improve our knowledge of primary and secondary ice formation, the interaction between ice crystals and droplets, and the role of aerosol particles in cloud development and evolution, in contrasted marine environments. ACME will also reveal the dominant ice formation mechanism in different types of clouds, and improve the phase partitioning of the mixed-phase clouds. ACME will furthermore contribute to the development of new parameterisations on ice formation processes under different atmospheric thermodynamical conditions for larger scale models.