The response of snow cloud bands to the increase in sea surface temperatures (SSTs) over the Japan Sea was investigated. We focused on a typical snowfall event in Japan by intense cloud bands around a convergence zone on December 25, 2021. After confirming that a regional atmospheric model fairly reproduced the event, we conducted three sensitivity experiments replacing the initial and boundary values with air temperatures and/or SSTs uniformly increasing by 4 K. The results revealed that the model experiment with higher SSTs or lower air temperatures supplied more evaporation to the planetary boundary layer, which encouraged the higher cloud to along the convergence zone. This dominated the transversal mode (T-mode) of cloud bands in the east of the zone, diagnosed by a newly developed technique that discriminates it from the longitudinal mode (L-mode) by means of the absolute value of horizontal advection of hydrometers. In contrast, the experiment with lower SSTs or higher air temperature exhibited wider areas dominated by the L-mode cloud bands over the Japan Sea.

How to cite: Sato, K. and Inatsu, M.: Response of Snow Cloud Bands to Sea Surface Temperatures over Japan Sea, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1436, https://doi.org/10.5194/egusphere-egu24-1436, 2024.

Ice crystals play a crucial role in precipitation formation and radiation budget, with their various shapes influencing these processes differently. The shape of ice crystals is related to the environmental conditions (i.e. temperature) under which the ice crystal forms and the microphysical processes that ice crystal experiences. Therefore, ice crystal shape classification is important for understanding conditions and microphysical processes in cloud. However, current methods are mainly supervised learning algorithms like convolutional neural networks (CNNs), heavily relying on extensive manual labelling, which requires substantial labor. Moreover, the limitations in human’s knowledge of ice crystals and the bias of human subjectivity in classification hinder the generalization ability of these networks. In response to these challenges, we propose a semi-supervised algorithm for ice crystal classification. We use data from the 2019 Ny-Ålesund NASCENT campaign, collected by a holographic imager mounted on the balloon-borne platform HoloBalloon, which includes 18,864 ice crystal images. In our algorithm we initially extract key features from ice crystal images using an unsupervised learning network, prioritizing generalization rather than dependence on labelled data, which ensures unbiased feature identification. Subsequently, a small subset of images is manually labelled into nineteen categories based on a multi-label classification scheme that consider both basic habits and microphysical processes. The classification accuracy of our hybrid algorithm on nineteen categories is similar to the performance supervised learning algorithm. This hybrid algorithm not only reduces the labor needed for manual labelling but also incorporates physics-based constraints, which prevents the network from making unfounded assumptions, thus offering a robust and efficient framework for ice crystal classification.

How to cite: Chu, Y., Zhang, H., Li, X., and Henneberger, J.: Ice crystal images classification using semi-supervised contrastive learning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6010, https://doi.org/10.5194/egusphere-egu24-6010, 2024.

As part of the NWCSAF (Nowcasting Satellite Application Facility), the CNRM participates in the retrieval of cloud properties from geostationary satellite observations. These retrievals include the Cloud Mask and Cloud Types classification, thermodynamics properties at the macroscopic scales (Cloud Top Temperature and Height) as well as microphysical cloud properties (effective radius, optical thickness, liquid and ice water path). The cloud optical properties (including scattering, absorptions and emissions) are derived from cloud microphysical model in order to perform radiative transfer simulations. In this study, I combine cloud microphysical properties retrieved from DARDAR and in-situ observations with ERA-5 reanalysis to perform radiative transfer simulations with RTTOV. Hence, these simulation are compared with Meteosat Second Generation observations. Our goal is to identify the cloud properties that can affect the difference between observations and simulations in order to propose a new parameterization of the ice cloud scattering phase function in the radiative transfer model RTTOV (Radiative Transfer for TOVS).

How to cite: Joseph, R., Fontaine, E., and Vidot, J.: Improved ice cloud phase function for passive remote sensing, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5684, https://doi.org/10.5194/egusphere-egu24-5684, 2024.

Ice particles in cloud microphysics schemes are traditionally categorized as ice crystals, snow, graupel, and/or hail. Each category is defined by static parameters that determine density, diameter-mass relationship, and diameter-fall speed relationship. Several previous studies have reported considerable sensitivity in simulated precipitation systems based on these fixed parameters. This study introduces a prognostic approach for graupel density in the Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) microphysics scheme, based on the work of Milbrandt and Morrison (2013). This allows graupel density to vary from 100 to 900 [kg/m3]. The modified WDM6 is tested for idealized squall line and winter snowfall cases over the Korean Peninsula. In the idealized squall line case, simulation results reveal variant graupel density in time and space, according to the evolution of squall line. For winter snowfall cases, simulations using the modified WDM6 show improved statistical skill scores, such as the root mean square error and bias, compared to the original WDM6, mitigating the positive precipitation bias simulated in the original WDM6. The modified WDM6 increases surface graupel amounts and decreases graupel suspended in the atmosphere due to faster sedimentation of graupel. Therefore, the major microphysical processes that generate graupel are influenced, subsequently reducing surface snow and precipitation over mountainous regions. Importantly, the modified WDM6 adeptly captures the relationship between graupel density and fall velocity, as verified by 2D video disdrometer measurements. *This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (grant no. RS-2023-00272751).

How to cite: Park, S.-Y., Lim, K.-S. S., Kim, K., Lee, G., and Milbrandt, J. A.: Impact of Prognostic Graupel Density on Simulated Precipitating Convections, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7185, https://doi.org/10.5194/egusphere-egu24-7185, 2024.

The Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) microphysics scheme only predicts the number concentrations for CCN and liquid-phase hydrometeors such as cloud water and rain. Although the double-moment approach for the cloud ice is recently introduced in the WDM6 scheme by Park and Lim (2023), the single-moment approach, in which only mixing ratio is prognosed, is still employed for solid-phase precipitating hydrometeors such as snow and graupel. In this study, the double-moment approach is introduced to WDM6 for all hydrometeors by adding prognostic number concentration of snow and graupel. To evaluate the effects of prognostic snow and graupel number concentrations, simulated results between the new and original versions of WDM6 scheme are compared. The four summer-precipitating (cold-type and warm-type; Kim et al. 2019) and seven winter-precipitating convection cases (cold-low type and warm-low type; Ko et al. 2022) are selected to evaluate the new scheme. In comparison to the original WDM6 scheme, the new scheme exhibits increased snow mixing ratio, except for cold-type summer cases. Additionally, the new scheme reduces the graupel mixing ratio and rain number concentration for all cases. In the new scheme, the raindrop size becomes larger due to the reduced rain number concentration, which is more consistent results with the observation data from 2DVD. Furthermore, larger raindrop size in the new scheme makes the evaporation inefficient. Therefore, the new scheme produces more surface precipitation than the original one. Meanwhile, among total 11 cases, the new scheme improves the equitable treat score (ETS) for eight cases and probability of detection (POD) for seven cases.

*This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government. (MSIT) (RS-2023-00208394)

How to cite: Kwon, J., Park, S.-Y., Lim, K.-S. S., Kim, K., and Lee, G.: Double-moment approach for snow and graupel in the WDM6 scheme and its effects on simulated precipitation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7194, https://doi.org/10.5194/egusphere-egu24-7194, 2024.

Balloon sounding with the Compact Optical Backscatter Aerosol Detector (COBALD) and Frost Point hygrometers (FPs) provides in situ data for a better understanding of the vertical distribution of cirrus clouds. In this study, eight summer balloon-borne measurements in Kunming (2012, 2014, 2015, and 2017) and Lhasa (2013, 2016, 2018, and 2020) over the Tibetan Plateau were used to show the distribution characteristics of cirrus clouds. Differences of cirrus occurrence were compared by different indices: the backscatter ratio (BSR) at a 455 nm/940 nm wavelength (BSR455 > 1.2/BSR940 > 2), the color index (CI > 7), and the relative humidity with respect to ice (RHice > 70%). Analysis of the profiles indicated that BSR455 > 1.2 was the optimal criterion to identify the cirrus layer and depict the distribution of the CI and RHice within cirrus clouds. The results showed that the median CI (RHice) within the cirrus clouds at both sites was mostly in the 18–20 (90%–110%) range at pressures below 120 hPa. Furthermore, the balloon-borne measurements combined with Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) measurements indicated a high frequency of cirrus occurrence near the tropopause in Kunming and Lhasa. The top height of cirrus occurrence at both sites was above the cold point tropopause and the lapse rate tropopause. Both Kunming and Lhasa had the highest frequency of thin cirrus clouds in the 0–0.4 km vertical cirrus thickness range.

How to cite: Yang, Z., Li, D., Luo, J., Tian, W., Bai, Z., Li, Q., Zhang, J., Wang, H., Zheng, X., Vömel, H., Wienhold, F. G., Peter, T., Hurst, D., and Bian, J.: Determination of Cirrus Occurrence and Distribution Characteristics Over the Tibetan Plateau Based on the SWOP Campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7972, https://doi.org/10.5194/egusphere-egu24-7972, 2024.

Convective clouds serve as a primary mechanism for the transfer of thermal energy, moisture, and momentum through the troposphere. The lack of understanding of the convective updraft properties and their relationship to environmental factors limit our ability to represent deep convection and its feedbacks in large-scale circulation models. Satellites are the only viable means of efficiently sampling tropical convective clouds, predominantly found in ocean-covered regions.

The C2OMODO Project, proposed by CNES as contribution to the AOS NASA program scheduled for 2029, aims to target the vertical development of deep convective cells. The proposed concept is a tandem of identical microwave radiometers aboard two different satellites in the same orbit, separated by a small-time delay, between 1 and 2 minutes. Each radiometer will measure at 89 GHz, 183 GHz (6 Channels) and 325 GHz (6 Channels), with footprints of 10, 5, and 3 km, respectively. These observations inform about the vertical distribution of ice, thanks to the scattering of radiation (in the line of ICI, STERNA, SAPHIR instruments). The derivative-time measurements of the C2OMODO tandem will provide information on the updraft dynamics of growing convective cells. Furthermore, C2OMODO will contribute to enhance the understanding of the life cycle of convective systems and improve the representation of deep convection in both weather prediction and climate models.

The aim of the presented study is to introduce the inversion method developed to estimate convective mass flux of ice from C2OMODO measurements, based on the variational approach (1D-VAR). Assimilation approaches, based on Bayesian theory, are commonly applied to the inversion of satellite observations. To simulate C2OMODO measurements, the radiative transfer model, RTTOV, serves as the forward operator while the mesoscale model MESO-NH is used as nature-like representation for atmospheric state. Only growing convective cells are selected in this work. The general 1D-VAR approach is adapted to integrate derivative-time measurements, thereby directly incorporating the dynamical properties in the restitution process. In this presentation, we describe the ongoing development of the variational approach. Additionally, the restitution of vertical ice mass flux and the performance of the 1D-VAR be discussed.

The ongoing development of this method has yielded promising preliminary results, instilling optimism about the wealth of information that will be accessible through C2OMODO.

How to cite: Lefebvre, T., Brogniez, H., Hermozo, L., and Chevalier, F.: Measuring dynamical properties of atmospheric convection using C2OMODO: a tandem of microwave radiometers, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8065, https://doi.org/10.5194/egusphere-egu24-8065, 2024.

Aerosol-cloud-interactions are one of the major causes of uncertainty in the radiative forcing as presented in the IPCC report WG1 (2021). The aerosols in the atmosphere can act as cloud condensation nuclei (CCNs) or ice nucleating particles (INPs) and affect cloud properties. A quantification of the impact of aerosols on these properties is difficult since clouds are also strongly affected by synoptic conditions.

Volcanic eruptions are an ideal testbed as they cause a local perturbance of aerosol concentrations in the atmosphere. The emitted aerosols such as SO2 reacting to sulfuric acid or ash can act as CCNs or INPs respectively. By comparing a simulation with and without the volcanic eruption the impact can be quantified directly. The simulation with an eruption can be validated by comparison to observations, e.g. satellite data.

In the presented work, the eruption of the Raikoke volcano in 2019 is simulated using the ICOsahedral Nonhydrostatic model (ICON) and the module for Aerosols and Reactive Trace gases (ART). The model is run in limited area mode on a R2B10 grid with about 2.5 km horizontal resolution for a time span of 3 days. The volcanic plume is simulated using a setup provided by J. Bruckert. During this time, the plume overlaps with cloud systems associated with a low-pressure system east of the volcano. The simulations with and without the eruption are compared to observational data to improve the implemented interaction mechanisms between cloud particles and volcanic aerosols with a focus on ice nucleation due to volcanic ash particles. Additionally, a new parameterization for volcanic ash formulated by Umo et al. (2021) based on laboratory experiments is implemented and compared to the commonly used ice nucleation parameterizations for mineral dust by e.g. Ullrich et al. (2017). First results of an offline calculation of the ice nucleation active site show a decrease in the ice nucleating efficiency for the parameterization for volcanic ash.

These first results on the interactions between the volcanic ash plume and mixed-phase and ice clouds will be presented.

How to cite: Sebisch, M., Hoose, C., Bruckert, J., and Hoshyaripour, G.: The response of mixed-phase and ice clouds to volcanic eruptions- A model case study of the Raikoke eruption 2019, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8070, https://doi.org/10.5194/egusphere-egu24-8070, 2024.

Dusty cirrus clouds, a rare phenomenon occurring approximately a few times a year globally, are associated with desert dust plumes in the upper troposphere. The formation of these clouds involves a high-humidity layer above a mineral dust-rich layer. In the intermediate layer between these two layers, initially, a thin cirrus cloud forms heterogeneously on the mineral dust particles. The latent heat release caused by the ice nucleation and radiative cooling above the thin cirrus cloud layer cause instability and convection to occur. This convection uplifts mineral dust particles even higher until the humid layer is fully mixed with the mineral dust, resulting in the dusty cirrus covering the humid layer.
The objective of this study is to investigate the formation mechanisms of dusty cirrus clouds. Addressing the challenges highlighted by Seifert et al. (2023), current atmospheric models struggle to predict these events. This work aims to validate the hypothesis presented by Seifert and further advance the understanding of the formation mechanisms. The study involves a simulation study conducted using the UCLALES-SALSA large-eddy model (Tonttila et al., 2017). A case study is performed using the atmospheric conditions present during Saharan dust plumes over Europe in recent years.
The simulated dusty cirrus clouds show that the upward transport of the mineral dust is not as effective as in the regional model study by Seifert et al. (2023). This is because the mineral dust which gets uplifted initially sediments down back to the original mineral dust layer with the sedimenting ice crystals. Also, the predominant mechanism for the instabilization of the air in the initial stages of the cloud formation is the latent heat release caused by the ice nucleation and the growth of the ice crystals, rather than the radiative cooling suggested by Seifert et al. (2023).
In the future, simulations will be conducted using idealized cases to comprehensively understand the most relevant mechanisms involved in the formation of dusty cirrus clouds.

References

Seifert, A., Bachmann, V., Filipitsch, F., Förstner, J., Grams, C. M., Hoshyaripour, G. A., Quinting, J., Rohde, A., Vogel, H., Wagner, A., and Vogel, B. (2023) Aerosol–cloud–radiation interaction during Saharan dust episodes: the dusty cirrus puzzle, Atmos. Chem. Phys., 23, 6409–6430

Tonttila, J., Maalick, Z., Raatikainen, T., Kokkola, H., Kühn, T. and Romakkaniemi, S. (2017).
UCLALES-SALSA v1.0: a large-eddy model with interactive sectional microphysics for aerosol,
clouds and precipitation. Geosci. Model Dev., 10, 169-188

How to cite: Juurikkala, K., Raatikainen, T., and Laaksonen, A.: Patterns in dusty cirrus cloud formation mechanisms revealed by LES modeling study, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9194, https://doi.org/10.5194/egusphere-egu24-9194, 2024.

Experiments were conducted in the cold room of the wind tunnel laboratory at Johannes Gutenberg University Mainz, encompassing collisions between bare graupel-graupel, bare graupel-ice sphere, bare graupel-graupel with dendrites and bare graupel-snowflake. This study addresses the underrepresented domain of secondary ice processes in clouds, focusing on fragmentation due to ice-ice collisions and their role in augmenting ice particle concentration. For this study, graupels were created using a setup that simulates the natural rotation and tumbling motion of freely falling graupels. The first set of experiments aimed to recreate previous collision experiments by producing more realistic nature-like graupels, while also improving the ice crystal fragment detection and counting process. 2mm and 4mm sized graupels were chosen based on previous observational studies.

This research contributes vital preliminary data, including fragment number and size distribution, as well as their dependency on collision kinetic energy. For this purpose, new coefficients fitted on our experiments following the theoretical framework have also been proposed, which can be used to parameterize the number of fragments resulting from ice-ice collisions. Our study attempts to bridge the gap between laboratory observations and numerical simulations, advancing the accuracy of atmospheric models.

How to cite: Yadav, S., Grzegorczyk, P., Metten, L., Zanger, F., Kumar Mitra, S., Theis, A., and Szakáll, M.: Fragmentation of atmospheric ice particles due to collision, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10554, https://doi.org/10.5194/egusphere-egu24-10554, 2024.

Microphysical processes in the mixed-phase clouds play an important role in modulating the earth’s weather and climate. However, uncertainties in both observational data and model parameterization of microphysical properties (e.g., number concentrations of ice particles) constrains our ability to accurately simulate mixed-phase clouds and their impact with weather and climate models. Model configuration, such as one- or two-moment microphysical schemes, horizontal and vertical resolution of the model can affect the representation of cloud and precipitation processes and cloud radiative effects. For simplicity, many numerical models use 1-moment microphysical scheme to represent clouds. However, this scheme may not represent microphysical and precipitation processes accurately as it only predicts the mass or number mixing ratios of hydrometeors. To address this issue, the present study uses the Icosahedral Non-hydrostatic (ICON) model to assess the sensitivity of model configuration by comparing the predicted microphysical properties with the observations. In ICON, the one-moment microphysical scheme represents mass fractions of five cloud as well as precipitation particles such as: cloud water and ice, snow, graupel, and rain. Furthermore, the two-moment microphysical scheme in predicts both mass and number mixing ratios of hail and the five prognostic variables mentioned above. For the above discussed purpose, a case of observed mixed-phase clouds will be simulated with ICON. The profiles of the simulated cloud microphysical properties will be compared with the coincident aircraft and ground-based observations. Furthermore, various simulations will be performed by varying the vertical as well as horizontal resolution to analyse the changes in model predicted microphysical properties.

How to cite: Hoose, C., Waman, D., Keshtgar, B., Barthlott, C., and Wallentin, G.: Sensitivity of Microphysical Properties of Mixed-Phase Clouds on Model Resolution and Microphysics Scheme in ICON, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11022, https://doi.org/10.5194/egusphere-egu24-11022, 2024.

Radar Forward Operators (RFO) act as an important link between the physical properties of cloud and precipitation and the observed radar quantities. A major source of uncertainty in radar forward operators is identified in the scattering properties of frozen and mixed-phase hydrometeors. Appropriate modeling of the internal structures of complex-shaped hydrometeors plays a pivotal role in the simulation of their polarimetric scattering properties. When RFOs are applied to weather model output, it is also desirable to ensure the consistency between the properties of hydrometeors assumed in the weather model and those implemented in the scattering simulations. Failing to do so, would impede a correct interpretation of model-observation comparison studies. The present study aims to model the microphysical and scattering properties of realistic ice crystals and snowflakes using the snow particle aggregation and DDA scattering models. The aggregation model includes realistic monomer generators for various ice crystal shapes. The simulated scattering properties are implemented into the EMVORADO RFO of the ICON model. Simulated properties are primarily kept consistent with the ICON microphysical assumptions. The shapes of snowflakes and ice crystals (dendrites and plates) are generated from the aggregation model, and used as input to the DDA scattering model to compute multi-frequency polarimetric radar scattering properties. The derived scattering properties are expected to explain better the observed polarimetric radar signatures of ice crystals and snow aggregates. Nonetheless, when simulating the snowflake shapes, one must make some decisions regarding its monomer composition. This study also explores the use of the innovative Lagrangian-particle cloud model McSnow in combination with the snowflake aggregation simulator. McSnow is able to simulate the snowflake evolution based on the physical and thermodynamic profiles of clouds and thus informs the aggregation model about the snowflake composition in terms of monomer shapes, size, and number. The synergy of these models is expected to elucidate the link between ice cloud processes and the polarimetric properties of cold clouds.

How to cite: Dutta, S., Ori, D., Mendrok, J., and Blahak, U.: Scattering properties generated from real shaped ice crystals and snowflakes for ICON’s Radar Forward Operator EMVORADO, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12253, https://doi.org/10.5194/egusphere-egu24-12253, 2024.

As the most important greenhouse gas in the Earth's atmosphere, the presence of water vapor in the upper troposphere and lower stratosphere (UTLS) is essential for influencing global radiation patterns and surface climate conditions. Even minor changes in water vapor levels within the mostly dry lower stratosphere (LS) can impact the vertical water vapor gradient, making it a crucial factor in the decadal variability of surface temperature.
In condensed form, water holds significant importance for planetary radiation. Clouds play a dual role by reflecting incoming solar radiation into space and absorbing/emitting longwave radiation from the surface. Estimating the impact of cirrus clouds on the radiation budget is particularly challenging as it depends on a variety of factors, such as altitude, humidity and the microphysical properties of the cloud.
During the lifetime of a cirrus cloud, the radiative impact can even change from a warming to a cooling effect and vice versa. For the formation of cirrus clouds, ice supersaturated regions (ISSRs) play an important role. However, the required amount of supersaturation is dependent on the nucleation mechanism, with at least ∼ 45% supersaturation for homogeneous freezing and as low as ∼ 20% for heterogeneous freezing.
We present a statistical intercomparison of the In-service Aircraft for a Global Observing System (IAGOS) dataset with ERA5, the latest reanalysis product of the European Centre for Medium-Range Weather Forecasts (ECMWF). Furthermore, a machine learning based algorithm is developed to improve the accordance of relative humidity with respect to ice (RHi) of reanalysis data with in-situ measurements, enabling large scale analyses of water vapor in the UTLS region. 
With this tool, we build three-dimensional climatologies of RHi and ISSRs over the North Atlantic region and show their seasonal and regional variability. This will help foster a general understanding of the occurence of cirrus clouds and their impact on weather and climate.

How to cite: Brast, N., Li, Y., Rohs, S., Konjari, P., Rolf, C., Krämer, M., Petzold, A., Spichtinger, P., and Reutter, P.: Temporal and Spatial Patterns of Ice Supersaturation: A 3D climatology over the North Atlantic Region, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13078, https://doi.org/10.5194/egusphere-egu24-13078, 2024.

While prior research on Arctic clouds has predominantly focused on single-layer clouds, the presence of multi-layer clouds in the Arctic holds significance. In such complex atmospheric systems, upper-level clouds can exert influence on the phase of lower clouds. A notable scenario occurs when ice crystals descend from higher altitudes into supercooled liquid water clouds, triggering the formation of mixed-phase clouds.

Our project is dedicated to investigating the occurrence of multi-layer clouds and their seeding, employing a combination of radiosonde and cloud radar observations. We will share findings from various locations, including research stations in Ny Alesund and the ARM North Slope of Alaska site, as well as insights from research cruises in the Arctic. Data from several research cruises were utilized in this study, namely MOSAiC (2019/20), Arctic Ocean 2018, and the ACSE 2014 campaign.

In addition, for the MOSAiC campaign, we employ back trajectories from various cloud levels and clear sky regions above the clouds to gain deeper insights into the occurrence and formation of multi-layer clouds. Our focus extends to different seasons, particularly emphasizing the Arctic melt and freeze-up periods.

How to cite: Achtert, P., Seelig, T., Tesche, M., Wallentin, G., and Hoose, C.: Occurrence of multi-layer clouds and ice-crystal seeding in the Arctic observed by Radar and radiosondes, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15772, https://doi.org/10.5194/egusphere-egu24-15772, 2024.

Supercooled cloud water is the source of a meteorological phenomenon with significant societal challenges: icing. Icing occurs when supercooled water droplets freeze upon contact with a solid surface and is even more intense with larger drops, resulting in stronger accretion. Anticipating the icing risk is crucial to ensure aviation safety, as ice on the fuselage can lead to a loss of lift. Icing as well occurs on wind turbines and train catenaries, making it a concern for both energy and transport sectors.

Supercooled water is often underestimated in numerical models. Our objectives are first to assess the two-moment microphysical scheme LIMA (Vié et al., 2016), second to identify the physical processes which are responsible for the lesser supercooled water before improving them.

To this end, numerical simulations of the research model Meso-NH (Lac et al., 2018) are compared to the observations of the ICICLE measurements campaign (https://www.eol.ucar.edu/field_projects/icicle). This airborne campaign was launched in February 2019 from Rockford (USA) by the USA’s Federal Aviation Administration. During 29 flights, microphysical parameters as the mixing ratio and the size of liquid and icy hydrometeors have been measured. These observations form an exceptional data set for studying the microphysical behaviour of models.

19 days, including all the 23 research flights of the campaign, were simulated. An extensive evaluation of the simulations was carried out, both on a flight-by-flight basis using Meso-NH’s flight simulator, and statistically combining observations from all flights. During the campaign, several cases of classical freezing rain, and lake effect situations, were sampled, allowing for a robust evaluation of model performance in these situations.

For lake effect cases, supercooled liquid water is forecast down to −30 °C, and mixed phase clouds are present between 0 °C and −10 °C, but cloud are almost completely icy around −20 °C. In freezing rain events, the precipitation tends to freeze again below the warm part of the cloud. To identify the sources of supercooled liquid water underestimation, a detailed analysis of microphysical processes budgets is performed. The impact of aerosols on forecasts is also investigated, using in-situ aerosol observations and CAMS reanalyses.

How to cite: July Wormit, M., Vié, B., and Lac, C.: Supercooled liquid water representation with the LIMA 2-moment microphysical scheme during the ICICLE field campaign, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16983, https://doi.org/10.5194/egusphere-egu24-16983, 2024.

Complementarily to their formation mechanism, the origin of cirrus (liquid or in-situ) can substantially affect their microphysical and radiative properties. Liquid-origin cirrus, which stem from the freezing of water droplets at cirrus temperatures, are typically characterised by high ice crystal concentrations and associated with strong cooling effect. However, the global occurence and distribution of this cirrus type as well as the environmental conditions in which they originate. The role of aerosols on liquid-origin cirrus, through their influence on liquid clouds, is also not well understood but can be critical for understanding radiative forcings associated with aerosol-cloud interactions and in implications of potential geo-engineering strategies.

This study investigates cirrus by coupling satellite and reanalysis dataset. Observations from lidar-radar satellite instruments (DARDAR-Nice) provide detailed retrievals of cirrus microphysical properties, such as ice water content and crystal number concentration. To trace the origins of cirrus clouds, we employ Lagrangian air mass trajectories based on ERA5 reanalyses, using FLEXPART. The presence and role of aerosols during the formation phase of these clouds, either in mixed-phase or warm regions, are assessed by integrating these trajectories with aerosol reanalysis products, specifically from CAMS. 

This joint cloud-aerosol dataset from satellite and reanalysis tools is created for one year of satellite observations. The global occurence of liquid-origin cirrus is analysed. The role of aerosols on the formation of liquid-origin clouds is finally investigated, in particular to understand the relevance of low-level aerosols on cirrus properties. Associated radiative effects will also be explored.

How to cite: Sourdeval, O., Saiprakash, A., Coopman, Q., Bucci, S., and Müürsepp, T.: Investigating the Impact of Aerosols on Liquid-Origin Cirrus from Global Observations and Reanalysis Data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19522, https://doi.org/10.5194/egusphere-egu24-19522, 2024.

In the past, numerous airborne in-situ measurements of mixed-phase clouds have exhibited a clear discrepancy between the observed ice particle and ice nucleating particle (INP) number concentrations of up to four orders of magnitude, with the highest differences observed in marine clouds. This suggests that primary ice nucleation is not the dominant source of cloud ice and that secondary ice production (SIP) plays a significant role in governing the ice phase in these clouds. Based on laboratory and field observations a number of SIP mechanisms have been hypothesized. However, most of these mechanisms are not well quantified and, therefore, only a few SIP mechanisms are included in weather models so far.

In our research, we aim to spatially extend the observations from aircraft campaigns by linking them to satellite data. Here we will show the work done linking the albedo and brightness temperatures from the 16 available spectral bands of Himawari-8, ranging from 0.47 – 13.3 µm, with the ice particle number concentrations observed during the SOCRATES campaign in low-level boundary layer clouds over the Southern Ocean. Finally, we employed multiple linear regression machine learning techniques and also made use of the SOCRATES campaign lidar/radar onboard.

How to cite: Aboel Fetouh, Y., Cermak, J., Hoose, C., and Järvinen, E.: Secondary ice production over the Southern Atlantic Ocean: linking satellite data with aircraft observations, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20442, https://doi.org/10.5194/egusphere-egu24-20442, 2024.

Mixed-phase clouds are common in the Arctic atmospheric boundary layer and their representation is challenging for models. Recent studies suggest that the ECMWF Integrated Forecast System (IFS) shows too many cloudy periods in the Arctic in summer and generally misses the periods with clear skies, while in winter the cloudy state is underrepresented.
We use ground-based remote sensing data from the MOSAiC campaign to assess systematic errors in modelled liquid cloud water over the whole MOSAiC period and combine this with more detailed analyses of selected cases.
In addition, we perform sensitivity tests to identify ways to improve the parametrization for Arctic mixed-phase clouds in the IFS.
We run cases in the Single Column Model (SCM) version of the IFS and investigate the representativity of model sensitivities in the SCM for the 3D model.

How to cite: Schulte, L., Magnusson, L., Forbes, R., Day, J., Schemann, V., and Crewell, S.: Representation of Arctic mixed-phase clouds in the ECMWF Integrated Forecasting System during MOSAiC, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20507, https://doi.org/10.5194/egusphere-egu24-20507, 2024.

The Mediterranean basin is characterized by cyclonic activity that can often lead to adverse weather conditions. Lately, there is an increasing interest to specific types of cyclones, such as medicanes, due to their dynamic characteristics. However, these events can also lead to extreme precipitation, often resulting in flooding and causing severe damage, with potential human casualties. While there is continuous effort to understand the  dynamic evolution of these  systems, little is known about the underlying microphysical processes. Secondary Ice Production (SIP) processes are ice multiplication mechanisms that have been frequently linked to the onset of heavy precipitation and the generation of high concentrations of precipitation particles. In this study we investigate the impact of four SIP mechanisms (rime-splintering, collisional break-up, drop-shattering, sublimation break-up) on the evolution of medicane Qendresa using the Weather and Research Forecasting (WRF) model. Qendresa occurred in 2014 mainly in the vicinity of Italy and Malta, causing three fatalities and at least $250 million in damages in Italy.

How to cite: Sotiropoulou, G., Floka, F., and Patlakas, P.: Secondary Ice Processes during a Medicane Evolution, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-9345, https://doi.org/10.5194/egusphere-egu24-9345, 2024.