Abstract

Aim of the project is to exploit the synergy of two full polarimetric radars, the C-band POLDIRAD at DLR, Oberpfaffenhofen and the Ka-band miraMACS (Mira35) at LMU, Munich, to study convective initiation as well as ice particle growth and its role in precipitation formation. At a distance of 23 km between DLR and LMU the use of the two research radar systems allows targeted observations and coordinated scan patterns. In order to study life-cycles of convective cells or precipitation development (e.g. fall streaks), object tracking is applied using horizontal PPI and vertical RHI cross sections on or close to the line-of-sight. An ice particle size retrieval will be developed for the dual-wavelength radar measurements, which will be used to advance established polarimetric hydrometeor classifications. Other processes observable are drizzle formation, cloud glaciation, distinction between depositional growth of small ice particles and onset of quicker ice particle growth into precipitation sized particles by aggregation, and initiation of first precipitation at the surface. Timing of these processes and their spatial distribution are observed and compared to modeled processes. Using a numerical weather model (WRF) with a nested domain centered over Munich at high spatial resolution (∆x of around 100 m), it is analyzed how microphysical parameterizations of different levels of complexity compare to the observations.

Status 2023

Contribution of LMU:

Towards improving cloud microphysics in numerical weather prediction models, we have developed a setup to systematically characterize differences between numerical weather models (WRF, 400 m grid spacing) and polarimetric radar observations for convective weather situations. This setup is used to evaluate 5 different microphysical methods of varying complexity (Thompson 2-mom, Thompson aerosol-aware, Morrison 2-mom, spectral bin (SBM), and P3). Statistical comparison of simulated and observed radar signals shows that all schemes except the P3 scheme produce high reflectivities in the ice phase too frequently. The dual-wavelength signatures indicate that most schemes have problems correctly reproducing the size distribution of ice particles; graupel particles are produced too often, too large, or too dense. A comparison of polarimetric radar signatures shows that all schemes have problems reproducing the size distribution of raindrops. Restricting the analysis to high-impact weather events, i.e., hail and heavy precipitation, we find that compared to radar-observed precipitation, all schemes on average overestimate the area of heavy precipitation, but at the same time all schemes except the P3 scheme underestimate the area and frequency of hail and graupel events. The predicted frequency of heavy precipitation events is overestimated by three of the schemes (Thompson 2-mom, Thompson aerosol-aware, P3) but underestimated by the SBM and Morrison schemes. Combined with the simulated mass mixing ratios, we conclude that it is not the mass of raindrops that is the problem, but the simulated size distributions, with the Thompson 2-mom, Thompson aerosol-aware, and P3 schemes producing large raindrops too frequently, while the SBM scheme produces too few.

Contribution of DLR:

Within the DLR project in phase 1 of IcePolCKa, a novel approach combining the radar dual- frequency technique with slant-wise polarimetric observations to retrieve ice clouds microphysics was investigated. For this purpose, our two radars instruments were used to monitor precipitation over the Munich area. The weather radar POLDIRAD (5.5 GHz) and the cloud radar MIRA-35 (35.2 GHz), located at 23 km horizontal distance, performed coordinated scans towards each other, providing dual-frequency measurements (DWR) to infer information about the ice hydrometeors size. In combination to dual-frequency, the scanning radar POLDIRAD provided observations of radar reflectivity Z and differential radar reflectivity ZDR with sensitivity to the mass and shape of the ice hydrometeors, respectively. This radar setup was proven to successfully work during snowfall events, when some aspects were considered e.g., non-uniform beam filling or spatiotemporal mismatches.

Comparing the radar measurements to T-Matrix (e.g., Waterman, 1965, Mishchenko et al., 1996, Leinonen, 2014) scattering simulations for ice particles varying their aspect ratio (AR), the median mass diameter (Dm) and the ice water content (IWC) of their particle size distribution, a simple ice microphysics retrieval resolving the shape, the size and the mass of ice hydrometeors (Fig. 2a) was developed. The detected ice particles were represented by soft spheroids (e.g., Hogan et al., 2012) following well-established mass-size relations m(Dmax) and an exponential particle size distribution (PSD). Using this approach, microphysical information of the detected ice crystal was retrieved. The retrieval results for AR, Dm, and IWC in Fig. 2b could better explain the radar measurements when the ice spheroids were considered oblates which follow the aggregates mass-size relation (Yang et al., 2000) instead of the well-known Brown and Francis (Brown and Francis, 1995). The Brown and Francis mass-size relation suggests that larger ice particles have low density and the combination with the soft spheroid model, which considers reduced-density particles to represent realistic habits of the same mass and size, could not produce simulated ZDR that matched the radar observations. Among all the assumptions used in the retrieval for the unknown ice microphysics, i.e., m(Dmax), PSD, oblate or prolate spheroid shape, horizontal flutter of the spheroids, the mass- size relation was found to be the most crucial and thus, it needs to be better constrained. Adding the slant-wise polarimetric observations along with dual-wavelength ratio by using this novel radar setup, it was found that in some regions, e.g., above the Ka-band cloud radar, ZDR not only constrains the shape of the ice hydrometeors but also reduces the uncertainty for the retrieval of the size.

References

• Brown, P. R. A. and Francis, P. N., 1995: Improved Measurements of the Ice Water Content in Cirrus Using a Total-Water Probe, J. Atmos. Oceanic Technol., 12, 410–414, https://doi.org/10.1175/1520-0426(1995)012<0410:IMOTIW>2.0.CO;2.
• Fridlind, A. M., van Lier-Walqui, M., Collis, S., Giangrande, S. E., Jackson, R. C., Li, X., Matsui, T., Orville, R., Picel, M. H., Rosenfeld, D., Ryzhkov, A., Weitz, R., and Zhang, P.,2019: Use of polarimetric radar measurements to constrain simulated convective cell evolution: a pilot study with Lagrangian tracking, Atmos. Meas. Tech., 12, 2979–3000, https://doi.org/10.5194/amt-12-2979-2019.
• Hogan, Robin J., Lin Tian, Philip R. A. Brown, Christopher D. Westbrook, Andrew J. Heymsfield, and Jon D. Eastment, 2021: Radar Scattering from Ice Aggregates Using the Horizontally Aligned Oblate Spheroid Approximation, Journal of Applied Meteorology and Climatology 51, 3 (2012): 655-671, accessed Mar 31, 2021, https://doi.org/10.1175/JAMC-D-11-074.1.
• Leinonen, J., 2014: High-level interface to T-matrix scattering calculations: architecture, capabilities and limitations, Opt. Express, 22, 1655–1660, https://doi.org/10.1364/OE.22.001655.
• Morrison, Hugh, and Jason A. Milbrandt, 2021: Parameterization of Cloud Microphysics Based on the Prediction of Bulk Ice Particle Properties. Part I: Scheme Description and Idealized Tests“, Journal of the Atmospheric Sciences 72, 1 (2015): 287-311, accessed Jul 8, 2021, https://doi.org/10.1175/JAS-D-14-0065.1.
• Oue, M., Tatarevic, A., Kollias, P., Wang, D., Yu, K., and Vogelmann, A. M., 2021: The Cloud-resolving model Radar SIMulator (CR-SIM) Version 3.3: description and applications of a virtual observatory, Geosci. Model Dev., 13, 1975–1998, https://doi.org/10.5194/gmd-13-1975-2020.
• Shpund, J., Khain, A., Lynn, B., Fan, J., Han, B., Ryzhkov, A., et al., 2019: Simulating a mesoscale convective system using WRF with a new spectral bin microphysics: 1: Hail vs Graupel. Journal of Geophysical Research: Atmospheres, 2019; 124: 14072– 14101. https://doi.org/10.1029/2019JD030576.
• Thompson, Gregory, Paul R. Field, Roy M. Rasmussen, and William D. Hall., 2021: Explicit Forecasts of Winter Precipitation Using an Improved Bulk Microphysics Scheme. Part II: Implementation of a New Snow Parameterization, Monthly Weather Review 136, 12 (2008): 5095-5115, accessed Jul 8, 2021, https://doi.org/10.1175/2008MWR2387.1.
• Yang, P., Liou, K.N., Wyser, K., Mitchell, D., 2000: Parameterization of the scattering and absorption properties of individual ice crystals. J. Geophys. Res. Atmospheres 105, 4699–4718. https://doi.org/10.1029/1999JD900755.

Status 2021

Contribution of DLR:
Investigating the role of ice for the evolution of precipitation using multi-wavelength radar measurements

A novel combination of polarimetric, multi-wavelength radar measurements has been used to retrieve ice microphysics information during stratiform precipitation events. To that end, coordinated RHI scans (see Fig. 1, top) were performed along the 23 km cross-sectional area between the C-band Polarization Diversity Doppler Radar (POLDIRAD) at the German Aerospace Center (DLR), Oberpfaffenhofen and the Ka-band, MIllimeter-wave cloud RAdar of the Munich Aerosol Cloud Scanner (miraMACS) at Ludwig-Maximilians-Universität (LMU), Munich. The dual-wavelength ratio (DWR) obtained between both instruments provides information about the size of the detected ice particles. Radar reflectivity (Z_H) as well as differential radar reflectivity (ZDR) observations from POLDIRAD were used for the estimation of the mass and the apparent shape of these hydrometeors. PyTMatrix simulations (Leinonen 2014) for ice particles of varying size, aspect ratio and ice water content were calculated and saved in look-up tables. The ice spheroid approximation (e.g. Hogan et al., 2012) and different established mass-size relations m(D) have been used as ice crystal model in the scattering simulations. Matching the radar observations to these simulations, an ice microphysics retrieval (Fig. 1, bottom) for ice water content (IWC), median mass diameter (D_m) and shape (sphericity, S) was developed which also considers the total attenuation affecting the radar measurements. Moreover, the retrieval performance was studied by varying the assumed m(D) relation. Here, the m(D) relation of aggregates (Yang et al., 2000) was able to better explain the observed DWR-ZDR observations compared to the established m(D) relation of Brown and Francis (1995) which are too fluffy to produce enhanced ZDR. The strongest sensitivity to the m(D) choice was found for the retrieved IWC, especially when the m(D) choice is fixed throughout the whole cloud cross-section. Additional radar observations, i.e. Doppler velocity, could be exploited in the future here, to obtain supplementary information about the effective density of the ice hydrometeors. Hence, a combination of mass-size relations could be used instead of a fixed one which can adapt to the effective ice crystal density in a specific environment.

Contribution of LMU:
The life-cycle of cloud and precipitation microphysics in radar observation and numerical model

Parallel to the retrieval development, the multi-wavelength polarimetric measurements are also used as a benchmark for precipitation formation in NWP models. To this end, differences between model microphysics schemes and radar observations for convective weather situations have been systematically characterized. A high resolution (400 m) convection permitting regional weather model setup was implemented using 5 different microphysics schemes of varying complexity (Double-Moment, Spectral Bin (SBM) and Particle Property Prediction (P3)). By implementation of a polarimetric radar forward simulator (CR-SIM; Oue et al., 2020), hindcasts of all measurement days are compared to observations. A cell-tracking algorithm (TINT; Fridlind et al., 2019) applied to radar and model data facilitates comparison on a cell object basis. So far, the analyzed data set includes 24 convection days. Targeted dual-wavelength observations were performed on 5 of those, providing high resolution dual-wavelength profiles of about 200 convective cell observations. Statistical comparisons of convective characteristics on a cell object basis in radar observations and numerical weather models were performed. In general, simulations show too few weak and small-scale convective cells. Contoured frequency by altitude distributions (CFADs) of radar signatures (Reflectivity Z, Dual-Wavelength ratio DWR, Differential Reflectivity ZDR) have the potential for microphysical fingerprinting in observations as well as simulations (Fig. 2). Deviations in the ice phase reveal a bias to higher reflectivities by most schemes, except the P3 scheme (Morrison and Milbrandt, 2015). This hints at issues in correct representation of graupel and snow particle size distributions (PSDs). Polarimetric variables, e.g. differential reflectivity ZDR reveal deviations in rain, where most schemes produce PSDs that are too broad, with the exception of the SBM scheme (Shpund et al., 2019).

References

• Brown, P. R. A. and Francis, P. N., 1995: Improved Measurements of the Ice Water Content in Cirrus Using a Total-Water Probe, J. Atmos. Oceanic Technol., 12, 410–414, https://doi.org/10.1175/1520-0426(1995)012<0410:IMOTIW>2.0.CO;2.
• Fridlind, A. M., van Lier-Walqui, M., Collis, S., Giangrande, S. E., Jackson, R. C., Li, X., Matsui, T., Orville, R., Picel, M. H., Rosenfeld, D., Ryzhkov, A., Weitz, R., and Zhang, P.,2019: Use of polarimetric radar measurements to constrain simulated convective cell evolution: a pilot study with Lagrangian tracking, Atmos. Meas. Tech., 12, 2979–3000, https://doi.org/10.5194/amt-12-2979-2019.
• Hogan, Robin J., Lin Tian, Philip R. A. Brown, Christopher D. Westbrook, Andrew J. Heymsfield, and Jon D. Eastment, 2021: Radar Scattering from Ice Aggregates Using the Horizontally Aligned Oblate Spheroid Approximation, Journal of Applied Meteorology and Climatology 51, 3 (2012): 655-671, accessed Mar 31, 2021, https://doi.org/10.1175/JAMC-D-11-074.1.
• Leinonen, J., 2014: High-level interface to T-matrix scattering calculations: architecture, capabilities and limitations, Opt. Express, 22, 1655–1660, https://doi.org/10.1364/OE.22.001655.
• Morrison, Hugh, and Jason A. Milbrandt, 2021: Parameterization of Cloud Microphysics Based on the Prediction of Bulk Ice Particle Properties. Part I: Scheme Description and Idealized Tests”, Journal of the Atmospheric Sciences 72, 1 (2015): 287-311, accessed Jul 8, 2021, https://doi.org/10.1175/JAS-D-14-0065.1.
• Oue, M., Tatarevic, A., Kollias, P., Wang, D., Yu, K., and Vogelmann, A. M., 2021: The Cloud-resolving model Radar SIMulator (CR-SIM) Version 3.3: description and applications of a virtual observatory, Geosci. Model Dev., 13, 1975–1998, https://doi.org/10.5194/gmd-13-1975-2020.
• Shpund, J., Khain, A., Lynn, B., Fan, J., Han, B., Ryzhkov, A., et al., 2019: Simulating a mesoscale convective system using WRF with a new spectral bin microphysics: 1: Hail vs Graupel. Journal of Geophysical Research: Atmospheres, 2019; 124: 14072– 14101. https://doi.org/10.1029/2019JD030576.
• Thompson, Gregory, Paul R. Field, Roy M. Rasmussen, and William D. Hall., 2021: Explicit Forecasts of Winter Precipitation Using an Improved Bulk Microphysics Scheme. Part II: Implementation of a New Snow Parameterization, Monthly Weather Review 136, 12 (2008): 5095-5115, accessed Jul 8, 2021, https://doi.org/10.1175/2008MWR2387.1.
• Yang, P., Liou, K.N., Wyser, K., Mitchell, D., 2000: Parameterization of the scattering and absorption properties of individual ice crystals. J. Geophys. Res. Atmospheres 105, 4699–4718. https://doi.org/10.1029/1999JD900755.

Status 2019

Contribution of DLR:
Investigating the role of ice for the evolution of precipitation using multi-wavelength radar measurements

Logarithmic difference of radar reflectivity measurements from POLDIRAD (DLR Oberpfaffenhofen) and MIRA35 (miraMACS, LMU Munich), defined as Dual-Wavelength Ratio (DWR, e.g. Kneifel et al., 2011) provides information about the size of the observed hydrometeors. For the dual-wavelength measurements, two different scan strategies have been developed. During stratiform precipitation events, POLDIRAD and Mira35 performed RHI scans every 10 minutes covering the area between them (on-axis scans). However, for convective precipitation formation a setup has been developed focusing dual-wavelength RHI scans freely on interesting precipitation cells (off-axis scans). Both types of measurements produced a first dataset for retrieval of ice particle microphysical properties from DWR measurements (about 20 measurement days on-axis and 10 off-axis). Using snow events from this dataset in order to avoid strong attenuation effects in the Ka-band, a preliminary estimation of the size of the ice particles has been attempted. To this end, dual-wavelength measurements have been evaluated along with T-matrix scatter simulations for particles that follow Gamma function shaped size distributions, varying aspect ratio with size, and a constant density (Fig. 1).

Since July 2019 no further dual-wavelength measurements have been performed as POLDIRAD was dismounted and is currently located on Barbados for the EUREC4A research field campaign (Bony et al., 2017). Therefore, recently the focus was layed onto analysis and processing of the data collected before. Ewald et al., 2019 found an underestimation of ca. 4 dB in the measured radar reflectivity after calibration and characterization of the Ka-band cloud radar HAMP on board HALO aircraft (also a MIRA35). This correction now applied to the Mira35 data leading to results shown in Figure 2. In addition, in the Ka-band attenuation due to atmospheric gases and hydrometeors have to be considered for an accurate extraction of information from the observations and a realistic retrieval. Gaseous attenuation can be calculated using ITU-R P.676-11 line-by-line formulas (ITU, 2016) and atmospheric parameters from ECMWF database. For hydrometeor attenuation, different formulas from literature have been tested finally yielding satisfying results (also Fig. 2). However, experiments for the optimum calculation of the attenuation from different hydrometeors at Ka-band are currently ongoing. Attenuation effects at C-band (gaseous and hydrometeors) are neglected up to now.

Contribution of LMU:
The life-cycle of cloud and precipitation microphysics in radar observation and numerical model

At the LMU polarimetric and dual-wavelength measurements are used to analyze and improve the performance of microphysics schemes in WRF. A simulation setup has been established including a Europe-, a nested Germany- and a nested Munich-domain. The Munich domain covers the overlap area of our two radars Mira35 and PoldiRad with a horizontal resolution of 400 m. This allows the simulation of convective events, which is the focus of our interest. For each of our measurement days (10 days with about 100 cell observations), a WRF hindcast simulation is conducted with differing microphysics schemes. Up to now, the Thompson- (Thompson et al., 2008), Morrison- (Morrison et al., 2009) and Thompson-aerosol-aware (Thompson and Eidhammer, 2014) 2-moment bulk schemes were used. We plan on adding a P3 (Morrison and Milbrandt, 2015) and a spectral bin (Fan et al., 2012) scheme in the near future.
To compare the radar observations to the WRF model output, the radar forward-simulator CR-SIM (Oue et al., 2019) was implemented. Figure 3 shows a first example of one of the measurement days simulated with WRF/ CR-SIM. In this example, CR-SIM was simulating the DWD C-Band radar in Isen. Finally an analysis of the microphysics schemes’ performances will be performed statistically over a large dataset including all our measurement days and convective cells. To make such a statistical comparison between radar measurements and WRF simulations possible, currently an automated cell tracking algorithm is implemented that allows tracking of convective cells over their life-cycle in model and observation. At the moment the algorithm TINT (Fridlind et al, 2019) is used to this end, which was specifically developed for cell tracking over large datasets.

References

• Bony, S., Stevens, B., Ament, F. et al. EUREC4A, 2017: A Field Campaign to Elucidate the Couplings Between Clouds, Convection and Circulation. Surv. Geophys., 38, 1529–1568, doi:10.1007/s10712-017-9428-0.
  

• Ewald, F., Groß, S., Hagen, M., Hirsch, L., Delanoë, J., and Bauer-Pfundstein, M., 2019: Calibration of a 35 GHz airborne cloud radar: lessons learned and intercomparisons with 94 GHz cloud radars, Atmos. Meas. Tech., 12, 1815–1839.
  

• Fan, J., Leung, L.R., Li, Z., Morrison, H., Chen, H., Zhou, Y., Qian, Y. and Wang, Y., 2012: Aerosol impacts on clouds and precipitation in eastern China: Results from bin and bulk microphysics. Journal of Geophysical Research: Atmospheres, 117(D16).
  

• Fridlind, A.M., van Lier-Walqui, M., Collis, S., Giangrande, S.E., Jackson, R.C., Li, X., Matsui, T., Orville, R., Picel, M.H., Rosenfield, D. and Ryzhkov, A., 2019: Use of polarimetric radar measurements to constrain simulated convective cell evolution: a pilot study with Lagrangian tracking.
  

• ITU, 2016: Recommendation ITU-R P.676-11: Attenuation by atmospheric gases, International Telecommunications Union.
  

• Kneifel, S., Kulie, M. S., and Bennartz, R., 2011: A triple-frequency approach to retrieve microphysical snowfall parameters, Journal of Geophysical Research: Atmospheres, 116(D11), D11203.
  

• Morrison, H., Thompson, G. and Tatarskii, V., 2009: Impact of cloud microphysics on the development of trailing stratiform precipitation in a simulated squall line: Comparison of one-and two-moment schemes. Monthly weather review, 137(3), pp.991-1007.
  

• Morrison, H. and Milbrandt, J.A., 2015: Parameterization of cloud microphysics based on the prediction of bulk ice particle properties. Part I: Scheme description and idealized tests. Journal of the Atmospheric Sciences, 72(1), pp.287-311.
  

• Oue, M., Tatarevic, A., Kollias, P., Wang, D., Yu, K. and Vogelmann, A. M., 2019: The Cloud Resolving Model Radar Simulator (CR-SIM) Version 3.2: Description and Applications of a Virtual Observatory. Geosci Model Devel Discuss.
  

• Thompson, G., Field, P.R., Rasmussen, R.M. and Hall, W.D., 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Monthly Weather Review, 136(12), pp.5095-5115.
  

• Thompson, G. and Eidhammer, T., 2014: A study of aerosol impacts on clouds and precipitation development in a large winter cyclone. Journal of the atmospheric sciences, 71(10), pp.3636-3658.