Vertically pointing Doppler cloud radars have been operated for over a decade by research networks around the globe, providing important insights into cloud microphysical processes and dynamics. In contrast, national weather radar networks mainly provide data at low-elevation angles to cover large areas and generally operate at longer wavelengths (typically at C-or S-band). Nonetheless, most polarimetric weather radars are also capable of collecting vertical profiling measurements, and such "birdbath" scans are already used routinely for radar calibration. Deutscher Wetterdienst (DWD)'s birdbath scan, for example, is repeated every 5 min within the operational scanning cycle and records the standard radar moments and full Doppler spectra. Analysis methods developed for cloud radars can be readily applied to C-band birdbath data, opening another avenue to study precipitation processes. Despite the lower sensitivity and coarser time resolution of C-band birdbath scans, our comparison of vertical C-band and Ka-band radar observations of winter and summer precipitation events shows remarkable agreement for light to moderate precipitation. One important advantage of a C-band system is the much weaker attenuation due to heavy precipitation so that the entire vertical structure of severe weather systems, such as hailstorms, can be resolved. As birdbath scans are already performed in many weather radar networks, these birdbath data should be leveraged to complement sparse cloud radar observations. Using this "new" data source, novel vertical cloud and precipitation products can be developed not only for validating model outputs and satellite observations but also for operational warning products. SIGNIFICANCE STATEMENT: A vertical profiler scan for calibration purposes is often part of the operational scanning in weather radar networks but usually not used for meteorological applications. We show remarkable agreement between weather radar and cloud radar measurements. In this paper, we argue for exploiting these data to study precipitation and cloud processes. Less sensitive to attenuation, a direct look into hail-producing thunderstorms becomes possible. Only a minor investment in software extension was needed at the German Meteorological Service to fully open up this fascinating data stream for research and new meteorological product development.

Snowflakes play a crucial role in weather and climate. A significant portion of precipitation that reaches the surface originates as ice, even when it ultimately falls as rain. Contrary to the popular image of symmetric, dendritic crystals, most large snowflakes are irregular aggregates formed through the collision of primary ice crystals, such as hexagonal plates, columns, and dendrites. These aggregates exhibit complex, fractal-like structures, particularly at large sizes. As a result of this structural complexity, each aggregate snowflake is unique, with properties that vary significantly around the mean-variability that is typically neglected in weather and climate models. Using a physically based aggregation model, we generate millions of synthetic snowflakes to investigate their geometric properties. The resulting dataset reveals that, for a given monomer number (cluster size) and mass, the maximum dimension follows approximately a lognormal distribution. We present a parameterization of aggregate geometry that captures key statistical properties, including maximum dimension, aspect ratio, cross-sectional area, and their joint correlations. This formulation enables a stochastic representation of aggregate snowflakes in Lagrangian particle models. Incorporating this variability in the Monte-Carlo super-particle model McSnow improves the realism of simulated fall velocities, enhances growth rates by aggregation, and broadens Doppler radar spectra in closer agreement with observations.

Abstract. The dendritic growth zone (DGZ) is associated with distinct polarimetric and multi-frequency radar signatures, yet the governing microphysical processes remain uncertain. We analyse characteristic DGZ observations showing a concurrent increase in dual-wavelength ratio (DWR), enhanced specific differential phase shift (KDP) and the maximum of the spectrally resolved ZDR (sZDRmax​), a pronounced reduction in mean Doppler velocity (MDV), and the occurrence of a secondary Doppler spectral mode near −15 °C. To investigate the governing processes, radar observations are combined with the Lagrangian particle-based Monte Carlo model McSnow, which includes an updated ice habit scheme and a new fragmentation parametrization. Forward radar simulations use a discrete dipole approximation scattering database. The simulations show that enhanced sZDRmax​ requires local formation of dendritic or plate-like crystals near −15 °C; sedimentation of pre-existing particles alone cannot reproduce the signal. The observed KDP enhancement is only reproduced when secondary ice production via collisional fragmentation is included, which also strengthens aggregation-related signatures. The reduction in mean Doppler velocity is explained by a habit change and aggregation of sedimenting columnar ice particles. Together, these signatures provide the most diagnostic constraints on DGZ microphysical processes identified so far. This study demonstrates that multi-frequency polarimetric radar observations combined with Monte Carlo Lagrangian particle simulations can disentangle competing ice microphysical processes in the DGZ. The results identify collisional fragmentation as a key unifying mechanism, with the DGZ radar fingerprint emerging from the interplay of depositional growth, aggregation, and secondary ice production.

As part of the international research programme TEAMx (multi-scale transport and exchange processes in the atmosphere over mountains - programme and experiment) a one-year long measurement campaign, the TEAMx Observational Campaign (TOC), was conducted between 2024 and 2025 in a north-south transect through the Alps. Building on the dense operational measurement network in the Alps, the TOC was designed to collect long-term atmospheric observations over the highly complex Alpine terrain. During two six-week long Extended Observational Periods, more than 40 research institutions came together to instrument about 30 sites in the four target areas of the TEAMx domain and study different transport processes, from gravity waves to orographic convection, thermally driven flows, and turbulent exchange. In addition to a suite of ground-based in-situ and remote-sensing instruments, observational activities included airborne measurements with up to three research aircraft and multiple UAS. This paper gives an overview of the science goals and the TOC design, together with preliminary analyses that highlight the potential of the collected dataset.

Abstract. Ground-based remote sensing instruments are often operated in a vertically pointing mode, producing time-height cross-section images (THIs) of the atmosphere. However, THIs are not Lagrangian observations: they cannot track the evolution of a single particle directly. Instead, several assumptions are required to derive the particle evolution from THI. We discuss these assumptions and show how their validity can be assessed using elevation scans. For demonstration, we analyze two intense riming cases. Since rimed particles exhibit enhanced sedimentation velocities, we first present a method to derive the vertical target velocity from scanning cloud radar observations. This method allows to study the spatial distribution of riming. The first case is comprised of several horizontally homogeneous, descending layers of rimed particles. Under these conditions, one can safely study the particle evolution in a "traditional" way by evaluating subsequent vertical profiles. The second case is more heterogeneous. For the analysis, we introduce the new spectral column vertical profile (SCVP) technique. SCVPs allow to trace the evolution of a single particle population's Doppler spectrum in space and time, thereby representing true Lagrangian observations. Our results demonstrate the value of scanning observations and show that downsides, for example regarding the use of Doppler velocity, can be overcome. Our results also raise the question whether the typically very high time resolution of THI is actually required for the common type of analyses performed on THI, and whether a combination of scanning and vertical observations could be the better observational strategy for ground-based remote sensing instruments.

The optical properties of atmospheric hydrometeors are a crucial component of any forward operator. These forward operators are essential data assimilation, and atmospheric model evaluations. Recent advances in microphysical modelling, such as Lagrangian super-particle models with habit prediction for ice particles, allow for the continuous evolution of particle properties in contrast to fixed hydrometer classes with fixed properties. This increasing complexity demands scattering databases capable of handling a wide range of particle properties.The discrete dipole approximation (DDA) is one of the most accurate and widely used methods for computing the scattering properties of irregular ice particles. However, its high computational cost typically limits either the diversity of particle shapes or the range of environmental parameters (e.g., frequency, temperature) represented in existing databases, constraining their applicability to models with highly variable microphysics.In this study, we present a new DDA-based database of optical properties at 5.6, 9.6, 35.6, and 94 GHz, specifically designed to accommodate the broad range of ice crystal morphologies predicted by habit-evolving schemes. The database contains 2627 individual ice crystals, including dendrites, plates, and columns, as well as 450 aggregates with varying degrees of riming and crystal types. The data are organized in three levels: Level 0 provides raw scattering matrices at individual orientations for a full range of scattering angles; level 1a summarizes Mueller and Amplitude matrix elements relevant for radar applications (at forward and backward scattering angles); and level 1b offers lookup tables of scattering properties that are relevant for polarimetric radars assuming azimuthally random orientations of the particles. These data allow for flexible treatment of the canting angle of the particles. The lookup tables are directly accessible via the McRadar simulator and can also be interfaced with other forward operators.The new database allows for a more consistent and realistic treatment of evolving ice particle properties in atmospheric models, improving the interpretation of radar observations and model-observation integration.

The 17 operational German C-band polarimetric weather radars routinely perform a vertical "birdbath" scan, which has so far primarily been used for calibration of differential moments. In this study, we transfer a retrieval algorithm for the rime fraction of snowflakes - originally developed for Ka-band cloud research radars - to the operational birdbath scan. This retrieval, which relies on the increase in detected mean Doppler velocity, serves as our benchmark. To validate the transfer of the retrieval, we apply it to a resampled birdbath dataset, constructed by downsampling cloud radar data to match the resolution of the operational birdbath scan. In addition, we present a new clutter filter and a melting layer detection algorithm for the operational birdbath scan. Finding good agreement between resampled and benchmark datasets, we apply the new retrieval to radar data recorded during the winters of 2021 to 2024. This results in a nationwide map of riming events in wintertime clouds. There is a north-south gradient in the riming distribution, which can be linked to Germany's precipitation climatology. Notably, we show that the occurrence of riming events correlates more strongly with precipitation intensity than with the total number of precipitation hours across sites. The temperature distribution associated with riming is consistently between -15 and 0 degrees C at all sites, except for the Feldberg site, which hints at a possible orographic effect. This study demonstrates that the operational birdbath scan of C-Band weather radars can be used for the retrieval of microphysical processes. Corresponding solutions, challenges and methods to transfer retrieval algorithms from research cloud radars to the operational weather radars are discussed.

A high accuracy of antenna beam pointing is essential for weather and cloud radars in order to precisely locate clouds and precipitation. It is also a critical requirement for estimating the horizontal wind field or retrieving particles' vertical motions.We present a general framework for radar pointing calibration using the Sun as a reference target. The workflow is structured into three steps: (i) measurement and analysis of individual Sun scans, (ii) estimation of scanner inaccuracies from a series of scans, and (iii) correction of these inaccuracies. Our approach is radar-agnostic and applicable to any instrument equipped with a two-axis pan-tilt scanner and a parabolic antenna. General recommendations for Sun scan implementation are given, and the full calibration process is demonstrated using a Mira-35 cloud radar. The method allows retrieval of a comprehensive set of parameters, including beamwidth in two orthogonal directions, pedestal tilt, axis misalignments, encoder offsets, gear backlash, and the receiver-scanner time offset. With this approach, absolute pointing accuracy better than 0.1 degrees can be achieved, and relative changes as small as 0.01 degrees can be detected. To facilitate automatic application, we provide the open-source Python library SunscanPy for radar pointing calibration. This toolset is especially valuable for stationary radars and radar networks, where it enables automatic monitoring of long-term pointing stability. Finally, we introduce a novel automatic pointing correction scheme based on inverse kinematics. Once the scanner inaccuracies are estimated, the required motor positions can be computed to compensate for the inaccuracies, without mechanical adjustments. Such functionality is particularly advantageous for mobile radars, research campaigns, or remote deployments, where frequent mechanical leveling is necessary but often difficult to perform.

AbstractTurbulence in clouds is known to enhance particle collision rates, as widely demonstrated for warm rain formation. A similar impact on ice growth processes is expected but a solid observational basis is missing. A statistical analysis of a 15‐month data set of cloud radar observations allows for the first time to quantify the impact of turbulence on ice aggregation and riming in Arctic low‐level mixed‐phase clouds. Increasing eddy dissipation rate (EDR), from below 10−4 to above 10−3 m2 s−3, yields larger ice aggregates, and higher particle concentration, likely caused by increasing fragmentation. In conditions more favorable to riming, higher EDR is associated with dramatically higher particle fall velocities (by up to 125%), under similar liquid water paths, indicative of markedly higher degrees of riming. Our findings thus reveal the key role of turbulence for cold precipitation formation, and highlight the need for an improved understanding of turbulence‐hydrometeor interactions in cold clouds.

Supplementary Data from Evaluation of a 1,4,7,10-Tetraazacyclododecane-1,4,7,10-Tetraacetic Acid–Conjugated Bombesin-Based Radioantagonist for the Labeling with Single-Photon Emission Computed Tomography, Positron Emission Tomography, and Therapeutic Radionuclides

Riming is a key process of precipitation formation in ice-containing clouds, but quantifying riming from observations is challenging, limiting our ability to evaluate the riming process in numerical weather models. One challenge for radar observations is that riming changes both the physical properties (mass, area cross-section) and scattering properties of ice particles. These changes need to be implemented consistently as a function of riming in radar forward operators, which are required for retrievals and model evaluation in observation space. In this study, mass-size, cross-section area-size, and backscattering cross-section relations are developed as a function of the normalized rime mass for aggregates composed of various monomer types (columns, dendrites, needles, plates, and rosettes). The proposed framework allows us to simulate scattering properties of aggregated ice particles consistently as a function of riming in retrievals and radar forward operators. The parameterizations are developed from a large data set of simulated rimed aggregates of different sizes and monomer crystal types. The backscattering cross-section parameterization (the "riming-dependent parameterization") is evaluated for radar frequencies of 35.6 and 94.0 GHz and is based on the Self-Similar Rayleigh-Gans approximation (SSRGA), which is increasingly used to calculate microwave scattering of ice crystals and snowflakes. Compared with parameterizations from the literature that do not consider riming, the riming-dependent parameterization leads to significantly smaller biases in terms of backscattering cross-section. When using the particle masses and scattering properties of the individual particles simulated by the aggregation and riming model as a reference, the bias of our parameterization is below 1 dB when integrating over an exponential particle size distribution with sizes from 0.1-10 mm. In this study, a large data set of simulated rimed aggregates of different sizes and monomer shapes is used to develop a parameterization of the Self-Similar Rayleigh-Gans approximation (SSRGA) parameter by the normalized rime mass M$$ M $$ (the riming-dependent parameterization). We analyse the backscattering cross-section bias in the Ka and W bands using the riming-dependent parameterization and present mass-size and cross-section area-size relations for the generated particles depending on M$$ M $$. Applying the riming-dependent parameterization in combination with the mass-size relations presented leads to biases below 1 dB for both frequencies and all riming levels, assuming exponential size distributions.image

We present a comprehensive quality-controlled 15-month dataset of remote sensing observations of low-level mixed-phase clouds (LLMPCs) taken at the high Arctic site of Ny-Ålesund, Svalbard, Norway. LLMPCs occur frequently in the Arctic region and extensively affect the energy budget. However, our understanding of the ice microphysical processes taking place in these clouds is incomplete. The dual-wavelength and polarimetric Doppler cloud radar observations, which are the cornerstones of the dataset, provide valuable fingerprints of ice microphysical processes, and the high number of cases included allows for the compiling of robust statistics for process studies. The radar data are complemented with thermodynamic retrievals from a microwave radiometer, liquid base height from a ceilometer, and wind fields from large-eddy simulations. All data are quality controlled, especially the cloud radar data, which are accurately calibrated, matched, and corrected for gas and liquid-hydrometeor attenuation, ground clutter, and range folding. We finally present an analysis of the temperature dependence of Doppler, dual-wavelength, and polarimetric radar variables, to illustrate how the dataset can be used for cloud microphysical studies. The dataset has been published in Chellini et al. (2023) and is freely available at: https://doi.org/10.5281/zenodo.7803064.

Riming, i.e., the accretion and freezing of supercooled liquid water (SLW) on ice particles in mixed-phase clouds, is an important pathway for precipitation formation. Detecting and quantifying riming using ground-based cloud radar observations is of great interest; however, approaches based on measurements of the mean Doppler velocity (MDV) are unfeasible in convective and orographically influenced cloud systems. Here, we show how artificial neural networks (ANNs) can be used to predict riming using ground-based, zenith-pointing cloud radar variables as input features. ANNs are a versatile means to extract relations from labeled data sets, which contain input features along with the expected target values. Training data are extracted from a data set acquired during winter 2014 in Finland, containing both Ka- and W-band cloud radar and in situ observations of snowfall by a Precipitation Imaging Package from which the rime mass fraction (FRPIP) is retrieved. ANNs are trained separately either on the Ka-band radar or the W-band radar data set to predict the rime fraction FRANN. We focus on two configurations of input variables. ANN 1 uses the equivalent radar reflectivity factor (Ze), MDV, the width from left to right edge of the spectrum above the noise floor (spectrum edge width – SEW), and the skewness as input features. ANN 2 only uses Ze, SEW, and skewness. The application of these two ANN configurations to case studies from different data sets demonstrates that both are able to predict strong riming (FRANN > 0.7) and yield low values (FRANN ≤ 0.4) for unrimed snow. In general, the predictions of ANN 1 and 2 are very similar, advocating the capability of predicting riming without the use of MDV. The predictions of both ANNs for a wintertime convective cloud fit with coinciding in situ observations extremely well, suggesting the possibility to predict riming even within convective systems. Application of ANN 2 to an orographic case yields high FRANN values coinciding with observations of solid graupel particles at the ground.

The dendritic growth layer (DGL), defined as the temperature region between −20 and −10 ∘C, plays an important role for ice depositional growth, aggregation and potentially secondary ice processes. The DGL has been found in the past to exhibit specific observational signatures in polarimetric and vertically pointing radar observations. However, consistent conclusions about their physical interpretation have often not been reached. In this study, we exploit a unique 3-months dataset of mid-latitude winter clouds observed with vertically pointing triple-frequency (X-, Ka-, W-band) and polarimetric W-band Doppler radars. In addition to standard radar moments, we also analyse the multi-wavelength and polarimetric Doppler spectra. New variables, such as the maximum of the spectral differential reflectivity (ZDR) (sZDRmax), allows us to analyse the ZDR signal of asymmetric ice particles independent of the presence of low ZDR producing aggregates. This unique dataset enables us to investigate correlations between enhanced aggregation and evolution of small ice particles in the DGL. For this, the multi-frequency observations are used to classify all profiles according to their maximum average aggregate size within the DGL. The strong correlation between aggregate class and specific differential phase shift (KDP) confirms the expected link between ice particle concentration and aggregation. Interestingly, no correlation between aggregation class and sZDRmax is visible. This indicates that aggregation is rather independent of the aspect ratio and density of ice crystals. A distinct reduction of mean Doppler velocity in the DGL is found to be strongest for cases with largest aggregate sizes. Analyses of spectral edge velocities suggest that the reduction is the combined result of the formation of new ice particles with low fall velocity and a weak updraft. It appears most likely that this updraft is the result of latent heat released by enhanced depositional growth. Clearly, the strongest correlations of aggregate class with other variables are found inside the DGL. Surprisingly, no correlation between aggregate class and concentration or aspect ratio of particles falling from above into the DGL could be found. Only a weak correlation between the mean particle size falling into the DGL and maximum aggregate size within the DGL is apparent. In addition to the correlation analysis, the dataset also allows study of the evolution of radar variables as a function of temperature. We find the ice particle concentration continuously increasing from −18 ∘C towards the bottom of the DGL. Aggregation increases more rapidly from −15 ∘C towards warmer temperatures. Surprisingly, KDP and sZDRmax are not reduced by the intensifying aggregation below −15 ∘C but rather reach their maximum values in the lower half of the DGL. Also below the DGL, KDP and sZDRmax remain enhanced until −4 ∘C. Only there, additional aggregation appears to deplete ice crystals and therefore reduce KDP and sZDRmax. The simultaneous increase of aggregation and particle concentration inside the DGL necessitates a source mechanism for new ice crystals. As primary ice nucleation is expected to decrease towards warmer temperatures, secondary ice processes are a likely explanation for the increase in ice particle concentration. Previous laboratory experiments strongly point towards ice collisional fragmentation as a possible mechanism for new particle generation. The presence of an updraft in the temperature region of maximum depositional growth might also suggest an important positive feedback mechanism between ice microphysics and dynamics which might further enhance ice particle growth in the DGL.

Low-level mixed-phase clouds (MPCs) occur widely and frequently in the Arctic, and on average introduce a strong positive radiative forcing. While precipitation is expected to affect radiative characteristics of Arctic MPCs, the relevancy of precipitation-formation processes, such as aggregation and riming, has been widely overlooked. An incomplete understanding of precipitation-formation processes in Arctic MPCs is likely to impact our ability to accurately simulate their evolution, macrophysical characteristics, and radiative effects. We employ a 3-year dataset of remote sensing observations from Ny-Ålesund, Svalbard, including two vertically-pointing Doppler radar systems, measuring at K- and W-band, to statistically assess the relevancy of aggregation and riming in Arctic low-level MPCs. We use the ratio of radar reflectivities measured at the two frequencies as a proxy for particle size, and match it with Doppler velocity information and temperature retrievals, to identify situations when the ice-particle growth is dominated by either aggregation or riming. We find observational evidence that large ice particles (mass median diameter > 1mm) mostly form when the mixed-phase layer of the low-level MPC is at temperatures compatible with dendritic growth (-15 to -10°C). Fall speeds of these larger particles are incompatible with significant riming. While mixed-phase layer temperatures between -15 and -10°C seem to be essential for the formation of large aggregates, these larger hydrometeors are not uniformly distributed across the cloud field. They are in fact observed in small pockets, suggesting that further dynamical processes might be needed to fully explain these signatures. Surprisingly, we find no evidence of enhanced aggregation at temperatures above -5°C in Arctic low-level MPCs. This is typically observed in mid-latitude clouds, and in deeper cloud systems in Ny-Ålesund as well. We hypothesize that ice particles sedimenting from higher levels might be an essential component needed to trigger enhanced aggregation above -5°C. We will discuss potential reasons for the absence of this feature, which are likely connected to the specific ice habits growing at these temperatures, as well as enhanced riming.

The Arctic shows an increased climate warming rate. There still exist uncertainties around the role that cloud feedback mechanisms play in this. In our study, we want to address these uncertainties. For that, we created a semi operational setup with daily cloud-resolving simulations over Svalbard. For these simulations with 600 m resolution, we use the ICOsahedral Non-hydrostatic model in the large eddy mode (ICON-LEM). The setup uses a two-moment microphysical parameterization and can handle heterogeneous surfaces. We apply the operational forecasts from the global ICON model as lateral boundary conditions. The advantage of this setup is that we can step away from focussing on single cases and instead look at many different types of large scale and local conditions. We created and evaluated several months of these simulations using observations from Ny-Ålesund (Svalbard) for comparison. The supersite “AWIPEV” is located there. It includes a microwave radiometer, daily radiosondes, a rain gauge and other remote sensing instruments. In addition, we were able to use the Cloudnet data set that provides a classification of the hydrometeors. We found that the model captures general features such as the wind flow, integrated water vapour, temperature and relative humidity profiles very well. The cloud occurrence was overestimated by the model but still lies in a climatologically realistic range. As the large scale dynamics are accurately simulated, this gives us the foundation to scrutinize the details of the cloud microphysical parameterization. As the next step, we investigate the shortcomings we could see in greater detail. One example is the more efficient production of cloud ice in the model than what was observed. For the analysis of specific microphysical processes, we are working on a software package which should enable the independent running of certain microphysical processes. This will make entangling the contributions of each process simpler and clearer while saving computational resources. Further, we show that for the Arctic, we must consider that standard nuclei concentrations, as used in models developed for the mid-latitudes do not represent the Arctic state. The goal in the long term is to improve the microphysical parameterization so that the model can better represent Arctic mixed-phase clouds.

This project is based on new multi-frequency radar observations collected during the AWARE field campaign in 2016 at McMurdo Station by the ARM Mobile facility. Three scientific goals have been achieved. We have developed ice and mixed-phase cloud microphysics retrievals from multi-wavelength radar observations and ancillary observations, generally applicable at ARM sites. These technique generally better characterize the size of the ice crystals, their equivalent water content and the degree of riming better than using single wavelength techniques. We have characterized ice microphysical properties and ice processes via multi-frequency signatures. We have identified a unique mode in the McMurdo dataset appearing at quite cold temperature that we have attributed to riming processes. Triple frequency radar measurements unfortunately were available only for a short period but our work lays the foundation for building climatological and testing their replicability in models. Simultaneously case studies from the AWARE campaign have been exploited to evaluate and improve the NASA GISS ModelE3 climate model, including selection of case studies for combined goals of (i) running side-by-side using large-eddy simulation (LES) and ModelE3 in single-column model (SCM) mode and (ii) fingerprinting cloud processes in multi-wavelength radar observations. The different components of this work have contributed to 15 published papers with two currently under review.

Low‐level mixed‐phase clouds (MPCs) occur extensively in the Arctic, and are known to play a key role for the energy budget. While their characteristic structure is nowadays well understood, the significance of different precipitation‐formation processes, such as aggregation and riming, is still unclear. Using a 3‐year data set of vertically pointing W‐band cloud radar and K‐band Micro Rain Radar (MRR) observations from Ny‐Ålesund, Svalbard, we statistically assess the relevance of aggregation in Arctic low‐level MPCs. Combining radar observations with thermodynamic profiling, we find that larger snowflakes (mass median diameter larger than 1 mm) are predominantly produced in low‐level MPCs whose mixed‐phase layer is at temperatures between −15 and −10°C. This coincides with the temperature regime known for favoring aggregation due to growth and subsequent mechanical entanglement of dendritic crystals. Doppler velocity information confirms that these signatures are likely due to enhanced ice particle growth by aggregation. Signatures indicative of enhanced aggregation are however not distributed uniformly across the cloud deck, and only observed in limited regions, suggesting a link with dynamical effects. Low Doppler velocity values further indicate that significant riming of large particles is unlikely at temperatures colder than −5°C. Surprisingly, we find no evidence of enhanced aggregation at temperatures warmer than −5°C, as is typically observed in deeper cloud systems. Possible reasons are discussed, likely connected to the ice habits that form at temperatures warmer than −10°C, increased riming, and lack of particle populations characterized by broader size distributions precipitating from higher altitudes.

Comparing the reflectivity flux at the top and bottom of the melting layer (ML) reveals the overall effect of the microphysical processes occurring within the ML on the particle population. If melting is the only process taking place and all particles scatter in the Rayleigh regime, the reflectivity flux increases in the ML by a constant factor given by the ratio of the dielectric factors. Deviations from this constant factor can indicate that either growth or shrinking processes (breakup, sublimation, and evaporation) dominate. However, inference of growth or shrinking dominance from the increase in reflectivity flux is only possible if other influences (e.g., vertical wind speed) are negligible or corrected. By analyzing radar Doppler spectra and multi‐frequency observations, we correct the reflectivity fluxes for vertical wind and categorize the height profiles by the riming degree at the ML top. We apply this reflectivity flux ratio (ZFR) approach to a multi‐month mid‐latitude winter data set that contains mostly stratiform clouds. The profiles of radar variables in the ML are found to be surprisingly similar for both unrimed and rimed profiles with slight differences, for example, in the absolute values of the reflectivity flux. Statistical analysis of the ZFR suggests that either microphysical processes other than melting are not important or strongly compensate for each other. The results seem to confirm that at least for moderately precipitating stratiform clouds, the melting‐only assumption applied in several retrievals and microphysical schemes is reasonable.