Displays

Lagrangian cloud models (LCMs) are considered the future of cloud microphysical modeling. However, LCMs are computationally expensive due to the typically high number of simulation particles (SIPs) necessary to represent microphysical processes such as collection/aggregation successfully. In this study, the representation of collection/aggregation is explored in one-dimensional column simulations, allowing for the explicit consideration of sedimentation, complementing the authors' previous study on zero-dimensional collection in a single grid box. Two variants of the Lagrangian probabilistic all-or-nothing (AON) collection algorithm are tested that mainly differ in the assumed spatial distribution of the droplet ensemble: The first variant assumes the droplet ensemble to be well-mixed in a predefined three-dimensional grid box (WM3D), while the second variant considers explicitly the vertical coordinate of the SIPs, reducing the well-mixed assumption to a two-dimensional, horizontal plane (WM2D). Both variants are compared to established Eulerian bin model solutions. Generally, all methods approach the same solutions, and agree well if the methods are applied with sufficiently high accuracy (foremost the number of SIPs, timestep, vertical grid spacing). However, it is found that the rate of convergence depends on the applied model variant.  Most importantly, the study highlights that results generally require a smaller number of SIPs per grid box for convergence than previous box simulations indicated. The reason is the ability of sedimenting SIPs to interact with an effectively larger ensemble of particles when they are not restricted to a single grid box. Since sedimentation is considered in most commonly applied three-dimensional models, the results indicate smaller computational requirements for successful simulations than previously assumed, encouraging a wider use of LCMs in the future.

How to cite: Unterstrasser, S., Hoffmann, F., and Lerch, M.: Collection/Aggregation in a Lagrangian cloud microphysical model: Insights from column model applications, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-1997, https://doi.org/10.5194/egusphere-egu2020-1997, 2020.

In this presentation, we summarize the main results of Shima et al. (2019). The super-droplet method (SDM) is a particle-based numerical algorithm that enables accurate cloud microphysics simulation with lower computational demand than multi-dimensional bin schemes. Using SDM, we developed a detailed numerical model of mixed-phase clouds in which ice morphologies are explicitly predicted without assuming ice categories or mass-dimension relationships. Ice particles are approximated as porous spheroids. The elementary cloud microphysics processes considered are advection and sedimentation; immersion/condensation and homogeneous freezing; melting; condensation and evaporation including cloud condensation nuclei activation and deactivation; deposition and sublimation; collision-coalescence, -riming, and -aggregation. To evaluate the model's performance, we conducted a 2D large-eddy simulation of a cumulonimbus. The results well capture characteristics of a real cumulonimbus. The mass-dimension and velocity-dimension relationships the model predicted show a reasonable agreement with existing formulas. Numerical convergence is achieved at a super-particle number concentration as low as 128/cell, which consumes 30 times more computational time than a two-moment bulk model. Although the model still has room for improvement, these results strongly support the efficacy of the particle-based modeling methodology to simulate mixed-phase clouds. 

Shima, S., Sato, Y., Hashimoto, A., and Misumi, R.: Predicting the morphology of ice particles in deep convection using the super-droplet method: development and evaluation of SCALE-SDM 0.2.5-2.2.0/2.2.1, Geosci. Model Dev. Discuss., https://doi.org/10.5194/gmd-2019-294, 1-83, 2019.

How to cite: Shima, S., Sato, Y., Hashimoto, A., and Misumi, R.: Predicting the morphology of ice particles in deep convection using the super-droplet method, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-2097, https://doi.org/10.5194/egusphere-egu2020-2097, 2020.

The Lagrangian Cirrus Module (LCM) is a Lagrangian (also known as particle-based) ice microphysics code that is fully coupled to the large-eddy simulation (LES) code EULAG. The ice phase is described by a large number of simulation particles (order 10⁶ to10⁹) which act as surrogates for the real ice crystals. The simulation particles (SIPs) are advected and microphysical processes like deposition/sublimation and sedimentation are solved for each individual SIP. More specifically, LCM treats ice nucleation, crystal growth, sedimentation, aggregation, latent heat release, radiative impact on crystal growth, and turbulent dispersion. The aerosol module comprises an explicit representation of size-resolved non-equilibrium aerosol microphysical processes for supercooled solution droplets and insoluble ice nuclei.

First, an general introduction to particle-based microphysics coupled to a grid-based (Eulerian) LES model is given.
In the following, emphasis is put on highlighting the benefits of the Lagrangian approach by presenting a variety of simulation examples.

How to cite: Unterstrasser, S.: The Lagrangian ice microphysics code LCM: Introduction, current developments and benefits, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-2307, https://doi.org/10.5194/egusphere-egu2020-2307, 2020.

This paper discusses a comparison of simulations applying either a traditional Eulerian bin microphysics or a novel particle-based Lagrangian approach to represent CCN activation and cloud droplet growth. The Eulerian microphysics solve the evolution equation for the spectral density function, whereas the Lagrangian approach follows computational particles referred to as super-droplets. Each super-droplet represents a multiplicity of natural droplets that makes the Lagrangian approach computationally feasible. The two schemes apply identical representation of CCN activation and use the same droplet growth equation; these make direct comparison between the two schemes practical. The comparison, the first of its kind, applies an idealized simulation setup motivated by laboratory experiments with the Pi Chamber and previous model simulations of the Pi Chamber dynamics and microphysics. The Pi Chamber laboratory apparatus considers interactions between turbulence, CCN activation, and cloud droplet growth in moist Rayleigh-Bénard convection. Simulated steady-state droplet spectra averaged over the entire chamber are similar, with the mean droplet concentration, mean radius, and spectral width close in Eulerian and Lagrangian simulations. Small differences that do exist are explained by the inherent differences between the two schemes and their numerical implementation. The local droplet spectra differ substantially, again in agreement with the inherent limitations of the theoretical foundation behind each approach. There is a general agreement between simulations and Pi Chamber observations, with simplifications of the CCN activation and droplet growth equation used in the simulations likely explaining specific differences.

How to cite: Grabowski, W. W.: Comparison of Eulerian bin and Lagrangian particle-based schemes in simulations of Pi Chamber dynamics and microphysics, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-2472, https://doi.org/10.5194/egusphere-egu2020-2472, 2020.

In the poster, we will present a new open-source cloud microphysics simulation package PySDM (https://github.com/atmos-cloud-sim-uj/PySDM). The package core is a Pythonic implementation of the Super-Droplet Method (SDM) Monte-Carlo algorithm for representing aerosol/cloud/rain collisional growth.

PySDM design features separation of a backend layer responsible for number-crunching tasks. The developed backend implementations based on Numba, Pythran and ThrustRTC leverage three different Python acceleration techniques dubbed just-in-time, ahead-of-time and runtime compilation, respectively. As a result, PySDM offers high performance with little trade-offs with respect to such advantages of the Python language as succinct and readable source code and portability (seamless interoperability between Windows, OSX and Linux). We will exemplify further advantages that result from embracement of the Jupyter platform which allowed us to equip PySDM with interactive examples and tutorials swiftly executable via web browser through cloud-computing platforms.

Example simulations of the warm-rain process in a kinematic two-dimensional framework mimicking stratoculumus deck will be presented and used as a basis for scalability analysis and discussion of parallelisation nuances of the SDM algorithm.

How to cite: Bartman, P., Olesik, M., Arabas, S., and Shima, S.: PySDM: Pythonic particle-based cloud microphysics package, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-3940, https://doi.org/10.5194/egusphere-egu2020-3940, 2020.

While the use of Lagrangian cloud microphysical models dates back as far as the 1950s, the integration of this framework into fully-coupled, three-dimensional dynamical models is only possible for about 10 years. In addition to the highly accurate and detailed representation of cloud microphysical processes, these so-called Lagrangian Cloud Models (LCMs) also allow for new ways of representing subgrid-scale dynamical processes and their effects on the microphysical development of clouds, typically neglected or only crudely parameterized due to computational constraints.

In this talk, I will present a new approach in which supersaturation fluctuations on the subgrid-scale of a large-eddy simulation (LES) model are represented by an economical, one-dimensional model that represents turbulent compression and folding. With a resolution comparable to direct numerical simulation (DNS), inhomogeneous and finite rate mixing processes are explicitly resolved. Applications of this modeling approach for warm-phase shallow cumuli and stratocumuli, and first applications for mixed-phase clouds will be discussed. Generally, clouds susceptible to inhomogeneous mixing show a reduction in the droplet number concentration and stronger droplet growth, in agreement with theory. Stratocumulus entrainment rates tend to be lower using the new approach compared to simulations without it, indicating a more appropriate representation of the entrainment-mixing process. Finally, the Wegner-Bergeron-Findeisen-Process, leading to a rapid ice formation in mixed-phase clouds, is decelerated.

All in all, this new modeling framework is capable of bridging the gap between LES and DNS, i.e., it enables representing all scales relevant to cloud physics, from entire cloud fields to the smallest turbulent fluctuations, in a single model, allowing to study their interactions explicitly and granting new insights.

How to cite: Hoffmann, F.: Beyond Cloud Microphysics: Representing Subgrid-Scale Processes in Lagrangian Cloud Models, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-4245, https://doi.org/10.5194/egusphere-egu2020-4245, 2020.

Owing to its key role in determining both the droplet collision probabilities and the radiative-transfer-relevant spectrum characteristics, the evolution of droplet spectral width has long been the focus of cloud modelling studies. Cloud simulations with detailed treatment of droplet microphysics face a twofold challenge in prognosing the droplet spectrum width. First, it is challenging to model and numerically represent the subtleties of condensational growth, even more so when considering the interplay between particle population dynamics and supersaturation fluctuations. Second, the discretisation strategies employed in representing the particle size spectrum and its evolution are characterised by inherent limitations. 

In the poster, we will present results of both Eulerian and Lagrangian numerical representations of spectrum width evolution. In the case of Lagrangian approach, we will discuss the differences in numerical integration procedures between (a) the sophisticated solvers typically used in parcel-model frameworks with moving-sectional spectrum representation and (b) the simpler solvers typically used in mathematically-analogous particle-based (super-droplet) microphysics representations used in multi-dimensional models.

In the case of Eulerian (bin microphysics) approach, we will present condensational growth simulations performed with the MPDATA numerical scheme using the newly developed MPyDATA package (http://github.com/atmos-cloud-sim-uj/MPyDATA/). The MPDATA family of numerical schemes for solving advective transport problems has been in continuous development for almost four decades. MPDATA features a variety of options allowing to pick an algorithm variant appropriate to the problem at hand. We will focus on the importance the MPDATA algorithm variant choice and the grid setup for the resultant numerical diffusion.

In the case of Lagrangian approach, we will present simulations performed using the newly developed PySDM package (https://github.com/atmos-cloud-sim-uj/PySDM) that features a set of cloud microphysics algorithms including condensational growth solvers. In the discussion, we will focus on: (a) the numerical realisability of the Ostwald ripening process (i.e. the growth of larger particles at the expense of water content of the smaller ones) and (b) the numerical approaches available for integrating stochastic fluctuations of ambient thermodynamic properties that drive the water vapour saturation.

How to cite: Olesik, M., Bartman, P., Arabas, S., Abade, G., Baumgartner, M., and Unterstrasser, S.: On the droplet spectral broadening numerics, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-5016, https://doi.org/10.5194/egusphere-egu2020-5016, 2020.

Atmospheric clouds play a very important role in the evolution of global atmosphere and climate through various interactive physical processes dynamically active over a huge range of scales [Devenish et al. QJRMS 2012, Grabowski and Wang. ARFM 2013]. However, many of such processed are yet to be understood; and in such context, we attempt to understand such a scientific question: whether large precipitating cloud drops can generate secondary droplets in it’s wake. Motivated by experimental investigation of large sedimenting cloud droplets [∼ mm radius] which showed presence of secondary cloud droplets in it’s wake [Prabhakaran et al. PRL 2017, ArXiv 2019]; we conduct direct numerical simulations of such precipitating hydrometeors using Lattice-Boltzmann method (LBM) to simulate cloud like ambient solving the evolution of the supersaturation field in the wake of the hydrometeor, and to investigate it’s impact on the nucleation of cloud aerosols. In our simulation results, we found various flow regimes based on the Reynolds number (Re = Droplet Diameter * Droplet Velocity / Kinematic Viscosity) in compliance with past researches. Steady axisymmetric wake for Re up to ∼ 220, after that steady oblique wake up to Re ∼ 280, then a transient oscillating nature of the wake up to Re ∼ 350, and beyond that Re, the wake is observed to become chaotic and turbulent. Comparison of drag coefficient, recirculation length and separation angles for fluid velocity at various Re shows good agreement with existing numerical and experimental simulations. The temperature profiles also fit well with other researches for similar Prandtl number (ratio of kinematic viscosity to thermal diffusivity). Evolution of the density of water vapor is similar to the temperature field, since both the equations show similar structure and the mass diffusivity of water vapor is almost same to the thermal diffusivity for atmospheric clouds. Distribution of the supersaturation field is computed using Clausius-Clapeyron Equation which gives saturation vapor pressure depending on temperature. In such simulations with background flow at -15^o C temperature with 60% relative humidity (RH) and with the hydrometeor as a warm cloud droplet at 4^o C temperature and 100% RH at it’s surface, the wake shows symmetric regions of supersaturation in the near vicinity of the hydrometeor at Re = 200. Whereas, at Re = 273, the wake is observed to become oblique, so the supersaturated region. Small pockets of supersaturated warm air parcels are observed to travel in the downstream direction when the hydrometeor started shedding vortices at higher Re. However, while traveling downstream, such supersaturated pockets also lost its’ excess of water vapor depending on the ambient cloud conditions. Due to higher supersaturation at the near vicinity of the warm hydrometeor, the cloud aerosols trapped inside the wake can be activated. However, whether such activated aerosols can become a drizzle drop, or may evaporate its liquid water content in subsaturated region, is to be understood by Lagrangian tracking of such aerosol tracers.

How to cite: Bhowmick, T., Wang, Y., Bagheri, G., and Bodenschatz, E.: Can Large Precipitating Cloud Hydrometeors Generate Secondary Cloud Droplets in its Wake?, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-17034, https://doi.org/10.5194/egusphere-egu2020-17034, 2020.

Sea-salt aerosols with radii exceeding 1 μm have been observed over the oceans. Cloud droplets formed on these giant aerosols can quickly grow to drizzle sizes through condensation of water vapor. Therefore giant aerosols, although not numerous, have been speculated to increase the amount of precipitation produced in clouds. Testing this hypothesis in LES simulations has been difficult, because Eulerian microphysics models are not well suited to model growth of droplets on giant aerosols. On the contrary, Lagrangian microphysics models, which are an emerging alternative to the Eulerian bin microphysics models, can model giant aerosols in a straightforward manner.

LES simulations performed using the University of Warsaw Lagrangian Cloud Model (UWLCM) will be presented. In UWLCM, the Lagrangian super-droplet microphysics model is used. We will assess how giant aerosols affect precipitation formation in marine cumulus (setup based on the RICO campaign) and stratocumulus clouds (setup based on the research flight 2 of the DYCOMS campaign). It will be discussed how the impact of giant aerosols changes with the concentrations of giant and regular aerosols. The results are of importance also for cloud seeding experiments, in which giant sea-salt aerosols can be released into a cloud.

How to cite: Dziekan, P., Jensen, J., Grabowski, W., and Pawłowska, H.: Giant aerosols increase precipitation in marine cumulus and stratocumulus clouds, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-18460, https://doi.org/10.5194/egusphere-egu2020-18460, 2020.

Modelling of small-scale turbulence in the atmosphere play a significant role in improving our understanding of cloud processes, thereby contributing to the development of better parameterization of climate models. One of the important problems is related to the transport of cloud particles, their activation and growth, which are influenced by small-scale turbulence motions. The idea presented in this work is to use fractal interpolation to reconstruct structures which are typically not resolved in the Large Eddy Simulations (LES) of clouds. Known filtered values of velocities on LES are basis of the reconstruction. The reconstructed small scales depend on the stretching parameter d, which is related to the fractal dimension of the signal. In many previous studies, the stretching parameter values were assumed to be constant in space and time. We modify this approach by treating the stretching parameter as a random variable with a prescribed probability density function (pdf). This function can be determined from a priori analysis of numerical or experimental data and within a certain range of wavenumbers it has a universal form, independent of the Reynolds number. We show, such modification leads to improvement in terms of reconstruction of two-point statistics of turbulent velocities. Preliminary results of simulations with Lagrangian particles (superdroplets) in the reconstructed field show the fractal model properly mimics the turbulent mixing processes at subgrid scales.

How to cite: Akinlabi, E., Waclawczyk, M., and Malinowski, S.: Fractal reconstruction of the subgrid scales in turbulence models in applications to cloud microphysics, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-18503, https://doi.org/10.5194/egusphere-egu2020-18503, 2020.

Turbulent clouds are challenging to model and simulate due to uncertainties in microphysical processes occurring at unresolved subgrid scales (SGS). These processes include the transport of cloud particles, supersaturation fluctuations, turbulent mixing, and the resulting stochastic droplet activation and growth by condensation. In this work, we apply two different Lagrangian stochastic schemes to model SGS cloud microphysics. Collision and coalescence of droplets are not considered. Cloud droplets and unactivated cloud condensation nuclei (CCN) are described by Lagrangian particles (superdroplets). The first microphysical scheme directly models the supersaturation fluctuations experienced by each Lagrangian superdroplet as it moves with the air flow. Supersaturation fluctuations are driven by turbulent fluctuations of the droplet vertical velocity through the adiabatic cooling/warming effect. The second, more elaborate scheme uses both temperature and vapor mixing ratio as stochastic attributes attached to each superdroplet. It is based on the probability density function formalism that provides a consistent Eulerian-Lagrangian formulation of scalar transport in a turbulent flow. Both stochastic microphysical schemes are tested in a synthetic turbulent-like cloud flow that mimics a stratocumulus topped boundary layer. It is shown that SGS turbulence plays a key role in broadening the droplet-size distribution towards larger sizes. Also, the feedback on water vapor of stochastically activated droplets buffers the variations of the mean supersaturation driven the resolved transport. This extends the distance over which entrained CNN are activated inside the cloud layer and produces multimodal droplet-size distributions.

How to cite: Abade, G., Waclawczyk, M., Grabowski, W. W., and Pawlowska, H.: Lagrangian stochastic microphysics at unresolved scales in turbulent cloud simulations, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-20398, https://doi.org/10.5194/egusphere-egu2020-20398, 2020.

This work focuses on the turbulent shearless mixing structure of a cloud/clear air interface with physical parameters typical of cumulus warm clouds. We investigate the effect of turbulence on the droplet size distribution, in particular, we focus on the distribution's broadening and on the collision kernel. We performed numerical experiments via Direct Numerical Simulations(DNS) of turbulent interfaces subject to density stratification and vapor density  fluctuation. Specifically, an initial supersaturation around 2 % and a dissipation rate of turbulent kinetic energy of 100 cm²/s³ are set in the DNSs. Taylor's Reynolds number is between 150 and 300. The total number of particles is around 5-10 millions, matching an initial liquid water content of 0.8 g/m³. Through these experiments, we provide a measure of the collision kernel and compare it with literature models [Saffman & Turner,1955], which is then included in a drops Population Balance Equation model (PBE). The PBE includes both processes of drops growth by condensation/evaporation and aggregation.

How to cite: Golshan, M., Tomatis, M., Abdunabiev, S., Fraternale, F., Vanni, M., and Tordella, D.: Impact of turbulence on cloud microphysics of water droplets population, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-22539, https://doi.org/10.5194/egusphere-egu2020-22539, 2020.