the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Mar 2026
Development of the TCWA2 Bulk Cloud Microphysics Scheme and Its Integration with a Dual-Polarization Radar Operator for Forecasting Applications
Abstract. This study presents the development and evaluation of TCWA2, a double-moment bulk cloud microphysics scheme designed for weather forecasting that incorporates radar observations at the Taiwan Central Weather Administration. By simplifying the triple-moment NTU microphysics scheme, TCWA2 retains a gamma-type particle size distribution with variable spectral parameters, diagnoses hydrometeor-associated physical properties, revises number sinks due to evaporation loss, and implements theoretically based fall-speed formulations that account for particle density and aspect ratio. To connect bulk microphysics parameterizations with radar-based diagnostics, TCWA2 is coupled with a customized bulk dual-polarization radar operator derived from offline bin-based scattering calculations under the Rayleigh approximation. This integrated microphysics–radar system provides an internally consistent representation linking particle-size distribution characteristics, hydrometeor morphology, sedimentation processes, and bulk radar observables. The intrinsic behavior of TCWA2 is first examined through two-dimensional idealized squall-line simulations in the WRF model, which reveal realistic microphysical structures and coherent polarimetric radar signatures. The scheme is further assessed through a real-case simulation of an afternoon convective event using the MPAS model, with validation against observed dual-polarization radar data. The joint distributions of radar reflectivity and polarimetric variables show strong agreement with observations, with pattern correlations exceeding 0.9 across three altitude layers, indicating that TCWA2 effectively captures the dominant microphysical features in radar signatures. Therefore, TCWA2 offers a physically consistent and computationally efficient framework for integrating bulk cloud microphysics with dual-polarization radar operators across platforms, with potential for future radar-based forecasting.
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-1257', Anonymous Referee #1, 20 Apr 2026
-
RC2: 'Comment on egusphere-2026-1257', Toshi Matsui, 24 Apr 2026
Summary
The manuscript describes the development of the new TCWA2 bulk microphysics scheme, which introduces several microphysical parameterizations designed to better reproduce polarimetric radar signatures without substantially increasing the computational burden for operational weather forecasting in Taiwan. Several aspects of the development are innovative, detailed and impressive, especially the treatment of spectral shape parameters, particle shapes, density, fall speeds, and their coupling to the polarimetric radar operator. These improvements allow the model to generate realistic distributions of dual-polarization radar variables. My main concern is that the MPAS real-case simulation section is simultaneously too ambitious and too limited. It attempts to demonstrate the operational applicability of TCWA2, but the manuscript does not provide sufficient space to describe the case setup, verification methodology, physical interpretation, and comparison with other schemes in adequate detail. As a result, this section feels underdeveloped relative to the more carefully presented microphysics and idealized-simulation sections.
I therefore suggest that the authors consider removing the MPAS real-case simulation from the present manuscript and instead focus this paper on the development, formulation, and idealized evaluation of TCWA2 and its internally consistent radar operator. This should be enough to provide a very informative paper for community. A separate follow-up paper could then be devoted to real-case simulations and detailed operational evaluation. Thus, my recommendation is “major revision”. Please see my detailed comments below.
Major Comments
Section 2.3: I could not find a clear description of how elevation-angle dependence is treated in the radar operator. In both radar simulations and observations, the radar elevation angle affects the apparent particle shape relative to the radar beam direction, except in circular-polarization measurements. This effect can influence simulated Zdr and Kdp. The authors should clarify whether the fitted formulas already incorporate an assumed mean elevation angle when calculating Zdr and Kdp., or whether the radar variables are computed under a fixed viewing geometry. If a fixed or simplified geometry is assumed, the authors should discuss its potential impact on comparisons with PPI radar observations at different elevation angles.
Fig. 6 (RI): The bulk densities of rimed ice (RI) do not appear to be very large near the convective core, around X = 350–380 km, where riming should be most active based on the cloud droplet distribution shown in Fig. 3. Could the authors explain why the RI bulk density remains relatively low in this region? This point is important because active riming would generally be expected to produce denser particles, and later plot (Fig. 15) TCWA2 scheme underestimates 40dBZ elevation in comparison with observation. The authors should clarify whether this behavior results from the RI density parameterization, the diagnosed riming rate, particle size effects, melting fraction, or other assumptions in the TCWA2 scheme.
(If you consider removing Section 4: disregard below major comments)
Section 4. MPAS real-case simulations: This section requires substantially more detail to meet the standard expected for a case-study manuscript. Additional information is needed on 1) the synoptic/thermodynamic environment, 2) the temporal evolution of the storm system, such as horizontal composite reflectivity and echo-top height, 3) convective–stratiform separation for detailed analyses, 4) the relationship between simulated hydrometeor distributions and the corresponding polarimetric radar signatures upon additional case study (organized versus isolated deep convection). However, the manuscript is already quite long, and incorporating these details would likely make it even longer and less focused. In my view, the model formulation, parameterization updates, and idealized simulations are sufficient to support publication if presented as the central contribution of the paper. Otherwise, the real-case simulation section would require substantial expansion and improvement.
Also, is the use of MPAS with this variable-resolution mesh configuration intended to support future operational applications? If not, the rationale for this configuration should be clarified, because the setup appears to allocate substantial computational resources to a broad domain while reducing the finest model resolution from 1 km in the idealized WRF simulation to 2 km in the MPAS real-case simulation. This resolution difference may affect the representation of convective dynamics and, consequently, the simulated microphysical behavior and polarimetric radar signatures. The authors should explain why this mesh configuration was chosen and whether the coarser 2-km grid spacing is sufficient for evaluating TCWA2 in convective cores.
Figs 15, 16, 17: A CFAD is defined as a normalized frequency distribution at each height level; that is, the integral or sum of the normalized histogram should be unity at each altitude. Please revisit the original definition in Yuter and Houze. Under this definition, the CFADs of Z, Zdr, and Kdp should be presented in that form, as in Matsui et al. (2023, Fig. 7c), for example.The figures currently shown in this manuscript appear to be joint height–variable histograms rather than CFADs. I therefore suggest either revising the terminology to describe them as “joint height–Z,” “joint height–ZDR,” and “joint height–KDP” histograms, or modifying all related plots so that they correctly represent CFADs. This applies especially to Figs. 15, 16, and 17.
In addition, separating convective-core and stratiform regimes would clarify the underlying microphysical processes much more effectively than using lumped CFADs or joint height–variable histograms (again see Fig. 7 of Matsui et al. 2023). For example, in line 658, the authors refer to “a clear reflectivity core centered around 2–10 km altitude.” However, this signal is likely dominated by the much larger spatial coverage of stratiform precipitation, rather than by the convective core itself. In convective cores, the maximum reflectivity typically appears near the high-reflectivity envelope at each altitude. In the manuscript, 40 dBZ echoes appear to extend above 9 km in the WSM6 and Thompson simulations, whereas they remain below about 7 km in the TCWA2 simulation. This may indicate that TCWA2 underestimates the size, density, or concentration of rimed particles in convective cores. The authors should examine whether this behavior is related to the PSD parameterization, rimed-ice density treatment, or fall-speed formulation.
Matsui, T., D. B. Wolff, S. Lang, K. Mohr, M. Zhang, S. Xie, S. Tang, S. M. Saleeby, D. J. Posselt, S. A. Braun, J.-D. Chern, B. Dolan, J. L. Pippitt, and A. M. Loftus, (2023), Systematic validation of ensemble cloud-process simulations using polarimetric radar observations and simulator over the NASA Wallops Flight Facility. Journal of Geophysical Research: Atmospheres, 128, e2022JD038134. https://doi.org/10.1029/2022JD038134Finally, Fig. 15 shows clear melting-layer signatures in both the WSM6 and Thompson simulations, even though these schemes do not explicitly predict mixed-phase particles. The authors should clarify the assumptions used in the polarimetric radar simulator for these schemes. Were the PSD, particle density, and hydrometeor phase treated consistently with each microphysics scheme, or were additional melting assumptions applied within the radar operator?
Fig 16: In convective cores, strong turbulence generally promotes particle tumbling and random orientation, which tends to reduce Zdr and Kdp toward near-zero values, particularly for ice particles within convective cores. Exceptions may occur under strong electric fields associated with lightning activity, which can preferentially orient ice crystals and produce anomalous polarimetric signatures, including negative Zdr. Therefore, the interpretation in lines 697–699 should be reconsidered. The reduced or near-zero Zdr and Kdp signals in the convective core may not necessarily indicate improved microphysical representation; they may simply reflect assumptions about particle orientation or tumbling in the radar simulator. The authors should clarify how particle canting, tumbling, and possible electric-field effects are treated before attributing these signatures to improvements in the microphysics scheme.
Minor Comments
Line 41: Define WRF.
Line 44: Define MPAS.
Line 44: “radar data” should be “radar measurements (or observation)”
Line 48: “computationally efficient” Is it possible to provide any quantitative information?
Line 66-67: Probably you should add “Rutledge and Hobbs 198X” paper here.
Line 69: “like” should be “such as”
Line 76: “microphysical properties (Fan et al. 2017)” Please discuss/cite uncertainties of cloud dynamics/turbulence, too. For example, most of operational km-scale storm-resolving NWP still cannot resolve cumulus thermals, which are the main building block of convective microphysics (e.g., Hernandez-Deckers et al. 2021).
Hernandez-Deckers, D., T. Matsui, and A. M. Fridlind (2021), Updraft dynamics and microphysics: on the added value of the cumulus thermal reference frame in simulations of aerosol-deep convection interactions, Atmos. Chem. Phys., 22, 711–724, https://doi.org/10.5194/acp-22-711-2022, 2022.
Line 96-98: What are “external assumptions” or “empirical lookup tables to fill in missing variables”?
Line 146: “dual-pol” should be “dual-polarimetric”
Line 315: After the sentence, also discuss that most (nearly all) dual-polarimetric radar simulators must parameterize particle shape and tumbling rates, which significantly changes the simulated polarimetric signals.” Please check the results from Matsui et al. (2019, Table1) that compare three major assumptions.
Matsui, T., Dolan, B., Rutledge, S. A., Tao, W.‐K., Iguchi, T., Barnum, J., & Lang, S. E. (2019). POLARRIS: A POLArimetric Radar Retrieval and Instrument Simulator. Journal of Geophysical Research: Atmospheres, 124. https://doi.org/10.1029/2018JD028317Line 333: Need reference for this sentence.
Line 382-384: I do not see a clear description of how these formulations account for the mean particle orientation angle or the degree of tumbling. Are these parameters assumed to be constant regardless of precipitation regime, such as convective core versus stratiform regions? If so, this assumption should be clearly stated and its potential impact on simulated Zdr and Kdp should be discussed.
Fig. 10, 11, 12: What is the elevation angle of radar beam to calculate Zdr and Kdp?
Line 623: Does MPAS require lateral boundary conditions in this configuration? Since MPAS is a global model, only initial conditions should be required.
Line 629: What are the elevation angles for PPI?
Line 726-727: Again, these signatures are not controlled by microphysics alone. They are also strongly influenced by assumptions about particle tumbling and mean orientation angle, which can substantially alter simulated polarimetric variables such as Zdr and Kdp. These assumptions should be clearly described and considered before attributing differences solely to microphysical processes.
Toshi Matsui
Citation: https://doi.org/10.5194/egusphere-2026-1257-RC2
Viewed
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 526 | 0 | 3 | 529 | 0 | 0 |
- HTML: 526
- PDF: 0
- XML: 3
- Total: 529
- BibTeX: 0
- EndNote: 0
Viewed (geographical distribution)
Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of EGUSphere-2026-1257
This paper describes the development and evaluation of the new TCWA2 bulk microphysics scheme (BMS), which is based on the more detailed 3-moment NTU scheme. The new scheme is described, along with descriptions of how aspects of TCWA2 are “parameterized” based on NTU. The scheme also has a dual-polarization radar simulator component. The scheme and the forward model (radar operator) are described. Detailed illustrations of how the scheme works are shown through idealized 2D WRF simulations of a squall line. Finally, a real case convective system over Taiwan is performed using the MPAS model with comparisons to observations (precipitation and dual-pol radar) and to the two existing BMPs in MPAS.
Overall this is a solid paper. It is clearly written and presented and scientifically sound. The authors do a nice job in illustrating how the new scheme works in the simple idealized 2D framework before proceeding to show its behavior for a real-case 3D simulation. I do not have any fundamental problems or major comments, just a few minor comments. I will recommend minor revisions, but these should be straightforward.
SPECIFIC COMMENTS: