Next Article in Journal
Fault Diagnosis of Wind Turbine Gearbox Based on Mel Spectrogram and Improved ResNeXt50 Model
Previous Article in Journal
Solid Core Magnetic Gear Systems: A Comprehensive Review of Topologies, Core Materials, and Emerging Applications
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 15
Issue 15
10.3390/app15158562
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

An Observational Study of a Severe Squall Line Crossing Hong Kong on 15 March 2025 Based on Radar-Retrieved Three-Dimensional Winds and Flight Data

by
Pak-wai Chan
*,
Ying-wa Chan
,
Ping Cheung
and
Man-lok Chong
Hong Kong Observatory, Hong Kong, China
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(15), 8562; https://doi.org/10.3390/app15158562
Submission received: 28 May 2025 / Revised: 28 July 2025 / Accepted: 28 July 2025 / Published: 1 August 2025
(This article belongs to the Section Environmental Sciences)
Browse Figures
Versions Notes

Abstract

The present paper reports for the first time the comparison of radar-derived eddy dissipation rate (EDR) and vertical velocity with measurements from six aircraft for an intense squall line crossing Hong Kong. The study objectives are three-fold: (i) to characterise the structural dynamics of the intense squall line; (ii) to identify the dynamical change in EDR and vertical velocity during its eastward propagation across Hong Kong with a view to gaining insight into the intensity change of the squall line and the severity of its impact on aircraft flying near it; (iii) to carry out quantitative comparison of EDR and vertical velocity derived from remote sensing instruments, i.e., weather radars and in situ measurements from aircraft, so that the quality of the former dataset can be evaluated by the latter. During the passage of the squall line and taking reference of the radar reflectivity, vertical circulation and the subsiding flow at the rear, it appeared to be weakening in crossing over Hong Kong, possibly due to land friction by terrain and urban morphology. This is also consistent with the maximum gusts recorded by the dense network of ground-based anemometers in Hong Kong. However, from the EDR and the vertical velocity of the aircraft, the weakening trend was not very apparent, and rather severe turbulence was still recorded by the aircraft flying through the squall line into the region with stratiform precipitation when the latter reached the eastern coast of Hong Kong. In general, the radar-based and the aircraft-based EDRs are consistent with each other. The radar-retrieved maximum vertical velocity may be smaller in magnitude at times, possibly arising from the limited spatial and temporal resolutions of the aircraft data. The results of this paper could be a useful reference for the development of radar-based turbulence products for aviation applications.

1. Introduction

Squall lines are common in southern China in spring and summer time. They could be associated with cold fronts which mark the interface between the cooler continental air from the north and the warmer maritime airstream to the south over the south China coast. Squall lines are known to produce high surface winds and small tornadoes [1] and they can be damaging because of the occurrence of high surface gusts, sometimes in excess of 100 km/h. Timely alerting of the high winds is an essential service of the weather authorities in the region in these seasons.
A severe squall line moved across Hong Kong, China, from west to east in the early afternoon of 15 March 2025. Early alerting of the high gusts associated with the squall line had been provided by the Hong Kong Observatory (HKO) to the public a couple of hours before the arrival of the squall line. However, there was still an event of workers trapped in a swinging gondola in strong wind at a place at the northwestern part of Hong Kong (a news report can be found in https://www.scmp.com/news/hong-kong/health-environment/article/3302509/hong-kong-braces-squally-thunderstorms-violent-gusts-saturday-afternoon (accessed on 30 July 2025)). Fortunately, the workers managed to find a way to stabilize the gondola and there was no injury or death during the event, but the incident did raise general concern in Hong Kong about the safety of workers in high-wind situations during thunderstorms. It also shows the possible damaging powers of strong winds associated with squall lines. In fact, many parts of Hong Kong had registered wind gusts in excess of 25 m/s, with some places reaching 28 m/s. The paper serves to document this case of severe squall line for reference of the weather forecasters.
There have been quite a number of studies about the physical mechanisms of squall lines over southern China [2,3]. These studies are mostly based on radar observations and simulation of the events using numerical models. However, there is a lack of in-depth discussion about the change in vertical velocity and turbulence severity associated with the evolution of squall lines, not to mention the availability of valuable aircraft data to verify radar-based observations. In the present study, the squall line structure is studied using three-dimensional (3D) winds retrieved from multiple weather radars and the HKO has the unique advantage of having a dense network of weather radars in the vicinity of the Pearl River Estuary to enable derivation of 3D wind fields for evaluating the dynamic change of severe weather systems. The study is thus novel with the inclusion of aircraft data in the observational study, such as vertical wind velocity and eddy dissipation rate (EDR). The vertical winds and EDR are also compared with the values retrieved from the weather radars. The flight data provide first-hand information to verify the radar-based retrieved winds and the EDRs. The results of the study would be useful for the study of the quality of the latter data and would have useful applications such as alerting of turbulence to be encountered by aircraft in squall lines. A previous study with similar nature can be found in Chan et al. [4] based on one single flight only, but the present study involves multiple flights covering different locations and heights with respect to the main convective areas of the squall line. The vertical wind and EDR are derived from the quick access recorder (QAR) data of commercial aircraft using the algorithms of Haverdings and Chan [5] (“NLR”) and EDR2W of Kim et al. [6] (“Spectrum”).

2. Weather Radars, Wind Retrieval Algorithm and EDR Calculation

A network of four S-band weather radars with two installed at Tai Mo Shan (TMS) and Tate’s Cairn (TC), respectively, in Hong Kong, one at Qiu Yu Tan (QYT) in Shenzhen, China, and one in Zhuhai near Macao provide essential radar observations of the squall line that affected Hong Kong on 15 March 2025. Data from these S-band weather radars together with an X-band Phased Array Weather Radar (PAWR) installed at Sha Lo Wan (SLW) in Hong Kong are used for the retrieval of 3D winds associated with the squall line. The antenna heights of the weather radars at TMS, TC, QYT, Zhuhai and SLW are 968 m, 588 m, 328 m, 257 m and 34 m above sea level, respectively.
The radar-based wind retrieval was based on a pythonic library, PyDDA, using Doppler velocity data from the above five sets of weather radars and the 3D variational data assimilation (3DVAR) wind retrieval method as suggested by Jackson et al. [7], Gao et al. [8,9], Potvin et al. [10], Bell et al. [11] and North et al. [12]. For the calculation of EDR which is used as a quantitative measure of convective turbulence, it is based on the signal-to-noise ratio (SNR) and spectral width (W) data from the weather radars using a fuzzy logic approach and details of the algorithm can be found in Chan et al. [13].
The retrieval of 3D wind field using variational multiple Doppler radar analysis is a very useful method for diagnosing atmospheric wind flow and the dynamical features of weather systems. It basically combines radar velocity observations, initial analysed wind fields from the Weather Research and Forecasting (WRF) model and other optional constraints from different sources to produce a more accurate wind analysis. The optimal wind analysis is obtained through minimization of a cost function using the Limited Memory Broyden–Fletcher–Goldfarb–Shanno Bounded (L-BFGS-B) method [14] and details are described in [15]. The 3D wind field retrieval technique has been successfully applied for studying a number of severe weather events occurring in Hong Kong, including vertical motion and cloud electrification in convective systems [15], cross-mountain wavy wind flow caused by terrain-induced windshear [16], thunderstorm-induced windshear and microbursts [4] and waterspouts [17], etc. In 2025, the 3D wind retrieval technique was used by HKO on a real-time basis for operational forecasting of severe weather [18]. Low-level horizontal wind fields at 2 km and 3 km heights as well as the corresponding vertical velocities are made available at every 6 min for monitoring of low-altitude weather. The 3D wind fields show the potential for nowcasting of intense convective weather although there are uncertainties in the retrieval method [19] and further improvement efforts can be made on deriving the vertical velocity.
There have been developments of 3D wind retrieval that can account for orographic effects such as topographic forcing triggering convective activity [20,21,22,23]. Further improvement is made using the Spline Analysis at Mesoscale Utilizing Radar and Aircraft Instrumentation (SAMURAI) analysis technique in which the finite element method is used to handle complex boundary geometries directly without interpolation to a Cartesian grid and the method has demonstrated good results as evidenced by previous studies [11,24,25,26]. More recent studies deal with retrievals over complex terrain like the ones by Tai et al. [27] using the ghost cell immersed boundary method (GCIBM) and Cha and Bell [19] introducing the immersed boundary method (IBM) have gained positive results in advancing the knowledge of orographic precipitation dynamics and convective initiation with topographic forcing.

3. Synoptic and Mesoscale Features

A cold front advanced to the south over southern China in the afternoon of 15 March 2025. From the weather radars, a squall line appeared over the western coast of southern China in the morning and moved across Hong Kong in the early afternoon. In the mesoscale, moisture was transported to southern China in the lower levels (e.g., 850 hPa and 925 hPa) in association with southwesterly winds at 00 UTC on that day (Hong Kong time = UTC + 8 h). Perturbations could be found in the westerly winds at mid-levels of the troposphere such as 500 hPa and 700 hPa at that time while there was apparent cyclonic shear in the lower levels at 925 hPa and 850 hPa. Veering of winds within the 0–3 km layer was observed with a vertical wind shear of around 15 m/s. Some isolated areas of wind speed divergence could be found in the upper troposphere at 00 UTC, e.g., at 200 hPa, supporting convective developments. The mesoscale setup is typical for the occurrence of intense convection in association with a cold front.
At 06 UTC, the convective available potential energy (CAPE) reached about 630 J/kg. The K-index was about 34 degrees Celsius. The tropospheric was generally unstable, supporting the development of severe convection. Although the CAPE value is not particularly large, it has been recognized that low CAPE/strong shear is common for tropical squall-line environments [28] which is favourable for triggering severe convective systems.
From the surface wind observations, the westerly winds associated with the squall line first picked up at the western part of Hong Kong, e.g., at the Hong Kong International Airport (HKIA) as in Figure 1a. Figure 1b shows the gusts at the various stations around that time with many at HKIA exceeding 25 m/s. The squall line then moved eastwards across Hong Kong, as shown in the wind observations from Figure 1c,e. From the wind gust observations in the whole event, namely, from Figure 1b to Figure 1d and eventually Figure 1f, the maximum gust was in fact decreasing in the eastward movement of the squall line, suggesting gradual weakening of the squall line in crossing Hong Kong. This may be partly related to the frictional effect of the terrain and urban morphology of Hong Kong.

4. Evolution of the Squall Line from 3D Retrieved Wind

The 1 km Constant Altitude Plan Position Indicator (CAPPI) radar reflectivity and the radar-retrieved horizontal wind at a height of 3 km above sea level are shown in Figure 2a. The corresponding vertical wind at 3 km is shown in Figure 2b. A vertical cross section, with location given in Figure 2a, is shown in the inset of that figure, with the wind arrows showing the wind component projected on the cross-sectional plane with the colour shaded as the speed of that wind component. It is shown that, at the time of Figure 2, the squall line was severe and mature. Significant wind convergence occurred at the fore-front of the squall line with strong radar echo as well as high reflectivity gradient. Strong upward vertical motion with vertical velocity exceeding 10 m/s could be found in that region. At the northern part of the squall line, there was clear signature of a cyclonic vortex which can be perceived as a bookend vortex and typically appears at midlevels (3–7 km) behind the leading edge of active convection [2,29,30]. A bookend vortex is possibly generated through upward tilting of vortex lines that are originated from low-level baroclinically forced or ambient shear-forced horizontal vorticity [2]. Radar observations showed that the squall line exhibited features of an elongated bow echo evolving to become a comma echo as described by Fujita [31] with apparent curvature stretching a distance of over 80 km from north to south as it moved across Hong Kong from west to east. In the vertical cross section, complete vertical circulation could be found in the fore-front of the squall line. Behind it was subsiding flow and strong horizontal winds as high as 50 knots or around 25.7 m/s at a height of 3 km, which is consistent with radar wind profiler observations. This seemed to resemble the feature of Rear-Inflow Jet (RIJ) which suggested strong acceleration of midlevel flow, causing the amplification of concavity and giving rise to the bow-shaped echo [32]. The presence of rear inflow was signified by an area of weak radar echo in the stratiform region (the black ellipse in Figure 2a) which was likely caused by evaporation and melting [33,34,35]. This weak-echo channel is termed as “rear-inflow notch” (RIN) by Przybylinski [36].
About half an hour later, while the radar echo remained strong as in Figure 3a, the upward vertical motion area became less extensive as in Figure 3b. The complete circulation at the fore-front of the squall line was not so apparent in the inset of Figure 3a. This trend continued as in Figure 4a,b. In particular, there was just an extensive area of upward motion at the front of the squall line, without signature of vertical circulation (inset of Figure 4a). The subsiding motion was also not apparent in Figure 4b, consistent with the weaker surface gusts in Figure 1f. The whole radar picture sequence shows that, in the progress of time following the movement of the squall line across Hong Kong, the squall line shows signs of weakening with less pronounced vertical structure (vertical circulation), subsiding motion behind the fore-front part and the weakening of the surface gusts.

5. Comparison with Flight Data

Despite the approach and the passage of a squall line across Hong Kong, the Hong Kong International Airport (HKIA) continued to operate in the period. There were a number of flights passing through the squall line and the associated rain areas. The flights provide valuable first-hand information about the weather within the squall line. Data from a number of flights are obtained, and they are briefly analysed here.
The data for the first flight are shown in Figure 5. On the left hand side, they are the EDR, vertical wind, horizontal wind and altitude from the flight. On the right hand side, there is the radar-based EDR map and the radar-retrieved vertical wind. At the time of this flight, the squall line had just reached HKIA. The aircraft departed to the west and flew around the western edge of Lantau Island. The maximum EDR has reached around 0.6 m2/3s−1, i.e., severe turbulence [37]. The maximum vertical velocity is in the region of −10 m/s. The EDR value is consistent with the radar-derived value, and strong downdraft is indeed retrieved from the radar-based, three-dimensional winds.
The second, third and fourth flights are shown in Figure 6, Figure 7 and Figure 8, respectively. In that period of time, the squall line was passing the central and the eastern parts of Hong Kong. All these three flights had to conduct missed approach for the first time, and successfully landed at the second time. During the two landings, all the aircrafts encountered severe convection and highly turbulent flows were measured for both times. The maximum EDR could reach 0.8 to 1.0 m2/3s−1, i.e., highly turbulent flows. Both updrafts and downdrafts were encountered during the landings, generally reaching a magnitude of 10 m/s. For the third flight (Figure 8), the downdraft (positive value of vertical wind) could reach around 14 m/s during the first approach, with the corresponding EDR reaching the exceptionally high value of around 1.0 m2/3s−1. Interesting instances of the radar observations are shown in the bottom two panels of the figures (refer to Figure 5b,c for the case of Figure 6). The radar-derived EDRs are generally consistent with the actual observations, particularly the maximum value recorded onboard the aircraft. The vertical velocity, especially the maximum value, appears to be weaker in the radar-retrieved wind particularly for 1 km height or below. Since radar data are only available every 6 min and are pulse volume averages, they may not capture the maximum vertical winds from the point (spatially) and instantaneous (temporally) measurements of the aircraft. Another reason may be that the radar-based wind retrieval has not handled well the vertical motion in the lower atmosphere as a result of complex topographic effects [19,27]. There is also the possibility of artifacts involved in deriving the vertical velocity in the interpolation and smoothing of the polar radar data to a Cartesian coordinate system [19,38].
The last two flights are shown in Figure 9 and Figure 10. In that period, the squall line had reached the eastern coast of Hong Kong. The aircraft just flew through the squall line once, and successfully landed at HKIA. Even though in the previous discussion, the squall line may have appeared weaker with the loss of vertical circulation and strong subsiding flow, the maximum EDR and the maximum vertical velocity associated with the squall line are not particularly low. In fact, for the latter flight (Figure 10), the maximum EDR could still marginally reach 0.8 to 0.9 m2/3s−1, and the magnitude of maximum updraft could be in excess of 10 m/s. The turbulent flow appears to be more “spread-out” in the time series of EDR and vertical velocity, showing that it is still rather turbulent even in the stratiform rain area behind the fore-front, main body of the squall line. This observation from the aircraft data is confirmed from the radar-based EDR maps and the radar-derived vertical velocity on the right hand panels of Figure 9 and Figure 10. As a result, though the climax of the development of the squall line has passed, the associated rain areas could still be hazardous to the operation of the aircraft.
A table summarizing the comparison of EDR and vertical velocity for the various flights is given in Table 1. In general, the radar-derived EDR matches rather well the maximum EDR recorded in flight. In severe cases, the maximum EDR can increase to around 1.0 m2/3s−1 (encountered by flight nos. B and D) while the magnitude of vertical velocity can exceed 10 m/s (encountered by flight nos. A, B, C and D). There could be some discrepancies in the maximum vertical velocity, due to the transient nature of the vertical motion. The radar-derived quantities are “averages” over the radar pulse volume whereas the measurements onboard the aircraft are essentially point measurements. Despite such differences, the comparison is considered to be good and the radar-derived quantities could be useful for alerting the aircraft, e.g., possible occurrence of severe turbulence in the planned flight paths.

6. Conclusions

A novel study is conducted in this paper to compare the vertical winds and EDRs retrieved from the weather radars for an intense squall line over southern China with the point measurements from six aircraft. In general, the EDR maps based on radars are consistent with the aircraft data and they are verified, at least for this case, to be useful for timely alerting of severe turbulence for aviation applications. Some discrepancies in the vertical velocities are observed, and they may be related to the limited spatial and temporal resolutions of the weather radars particularly the sampling time of radar data can significantly affect the accuracy of retrieved vertical velocity as mentioned by Cha and Bell [19]. To the knowledge of the authors, such comparisons using actual commercial flight data are very limited in the literature, and the results in this paper could be useful for reference for the development of turbulence alerts for aircraft.
HKO is going to expand the network of ground-based X-band PAWR in Hong Kong, based on the success of a previous trial [15]. This network would provide data nearer the ground, so that more radar data near the sea surface would be available for retrieval of low-altitude winds. Moreover, the volume scan would be performed every 1 to 1.5 min, with a higher spatial resolution at the same time. All these may help reduce the discrepancies between the radar-derived products and the aircraft’s point measurements. Future observations and comparison studies would be reported in the literature later.

Author Contributions

Conceptualization, P.-w.C. and Y.-w.C.; methodology, P.-w.C. and Y.-w.C.; software, Y.-w.C. and P.C.; validation, Y.-w.C. and M.-l.C.; formal analysis, P.-w.C. and Y.-w.C.; investigation, Y.-w.C. and P.C.; resources, Y.-w.C. and P.C.; data curation, M.-l.C.; writing—original draft preparation, P.-w.C. and Y.-w.C.; writing—review and editing, P.C. and M.-l.C.; visualization, Y.-w.C. and M.-l.C.; supervision, P.-w.C.; project administration, P.-w.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The datasets presented in this article are not readily available because the study involves proprietary aircraft data.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Weisman, M.L. Bow Echoes: A Tribute to T.T. Fujita. Bull. Am. Meteorol. Soc. 2001, 82, 97–116. [Google Scholar] [CrossRef]
  2. Meng, Z.; Zhang, F.; Markowski, P.; Wu, D.; Zhao, K. A Modeling Study on the Development of a Bowing Structure and Associated Rear Inflow within a Squall Line over South China. J. Atmos. Sci. 2012, 69, 1182–1207. [Google Scholar] [CrossRef]
  3. Xu, X.; Ju, Y.; Liu, Q.; Zhao, K.; Xue, M.; Zhang, S.; Zhou, A.; Wang, Y.; Tang, Y.Z. Dynamics of Two Episodes of High Winds Produced by an Unusually Long-Lived Quasi-Linear Convective System in South China. J. Atmos. Sci. 2024, 81, 1449–1473. [Google Scholar] [CrossRef]
  4. Chan, Y.-W.; Chan, P.-W.; Cheung, P. Observation of Downburst Associated with Intense Thunderstorms Encountered by an Aircraft at Hong Kong International Airport. Appl. Sci. 2025, 15, 2223. [Google Scholar] [CrossRef]
  5. Haverdings, H.; Chan, P.W. Quick Access Recorder Data Analysis Software for Windshear and Turbulence Studies. J. Aircr. 2010, 47, 1443–1447. [Google Scholar] [CrossRef]
  6. Kim, S.-H.; Kim, J.; Kim, J.-H.; Chun, H.-Y. Characteristics of the derived energy dissipation rate using the 1 Hz commercial aircraft quick access recorder (QAR) data. Atmos. Meas. Tech. 2022, 15, 2277–2298. [Google Scholar] [CrossRef]
  7. Jackson, R.; Collis, S.; Lang, T.; Potvin, C.; Munson, T. PyDDA: A Pythonic direct data assimilation framework for wind retrievals. J. Open Res. Softw. 2020, 8, 20. [Google Scholar] [CrossRef]
  8. Gao, J.D.; Xue, M.; Shapiro, A.; Droegemeier, K.K. A variational method for the analysis of three-dimensional wind fields from two Doppler radars. Mon. Weather Rev. 1999, 127, 2128–2142. [Google Scholar] [CrossRef]
  9. Gao, J.D.; Xue, M.; Brewster, K.; Droegemeier, K.K. A three-dimensional Variational data analysis method with recursive filter for Doppler radars. J. Atmos. Ocean. Technol. 2004, 21, 457–469. [Google Scholar] [CrossRef]
  10. Potvin, C.K.; Shapiro, A.; Xue, M. Impact of a vertical vorticity constraint in variational Dual-Doppler wind analysis: Tests with real and simulated supercell data. J. Atmos. Ocean. Technol. 2012, 29, 32–49. [Google Scholar] [CrossRef]
  11. Bell, M.M.; Montgomery, M.T.; Emanuel, K.A. Air–sea enthalpy and momentum exchange at major hurricane wind speeds observed during CBLAST. J. Atmos. Sci. 2012, 69, 3197–3222. [Google Scholar] [CrossRef]
  12. North, K.W.; Oue, M.; Kollias, P.; Giangrande, S.E.; Collis, S.M.; Potvin, C.K. Vertical air motion retrievals in deep convective clouds using the ARM scanning radar network in Oklahoma during MC3E. Atmos. Meas. Tech. 2017, 10, 2785–2806. [Google Scholar] [CrossRef]
  13. Chan, P.W.; Zhang, Y.; Doviak, R.J. Calculation and application of eddy dissipation rate map based on spectrum width data of a S-band radar in Hong Kong. Mausam 2016, 67, 411–422. [Google Scholar] [CrossRef]
  14. Byrd, R.H.; Lu, P.; Nocedal, J.; Zhu, C. A Limited Memory Algorithm for Bound Constrained Optimization. SIAM J. Sci. Comput. 1995, 16, 1190–1208. [Google Scholar] [CrossRef]
  15. Yang, Z.; Chan, P.-W.; Chan, Y.-W.; Zhao, K.; Chen, H.; Chen, C.; Xu, Y. Application of three-dimensional wind fields and dual-polarization signals of an X-band phased array weather radar in diagnosing vertical motion and cloud electrification in convective storms. Adv. Atmos. Sci. 2005, 42, 968–980. [Google Scholar] [CrossRef]
  16. Chan, Y.W.; Lo, K.W.; Cheung, P.; Chan, P.W.; Lai, K.K. Observation and numerical simulation of cross-mountain airflow at the Hong Kong International Airport from Range Height Indicator Scans of Radar and LIDAR. Atmosphere 2024, 15, 1391. [Google Scholar] [CrossRef]
  17. Leung, Y.T.; Lam, K.F.; Lau, T.K.; Chan, Y.W.; Chan, P.W. Meteorological indices for potential occurrence of waterspout/tornado in Hong Kong. Meteorol. Z. 2025. [Google Scholar] [CrossRef]
  18. Lau, T.K.; Chan, Y.W. Retrieval of 3D wind field using variational multi-doppler wind analysis for studying high impact weather in Hong Kong. In Proceedings of the 37th Guangdong-Hong Kong-Macao Seminar on Meteorological Science and Technology, Guangzhou, China, 13–14 March 2025. [Google Scholar]
  19. Cha, T.Y.; Bell, M.M. Three-dimensional variational multi-Doppler wind retrieval over complex terrain. J. Atmos. Ocean. Technol. 2023, 40, 1381–1405. [Google Scholar] [CrossRef]
  20. Georgis, J.F.; Roux, F.; Hildebrand, P.H. Observation of precipitating systems over complex orography with meteorological Doppler radars: A feasibility study. Meteorol. Atmos. Phys. 2000, 72, 185–202. [Google Scholar] [CrossRef]
  21. Chong, M.; Georgis, J.-F.; Bousquet, O.; Brodzik, S.R.; Burghart, C.; Cosma, S.; Germann, U.; Gouget, V.; Houze, R.A.; James, C.N.; et al. Real-time wind synthesis from Doppler radar observations during the Mesoscale Alpine Programme. Bull. Am. Meteorol. Soc. 2000, 81, 2953–2962. [Google Scholar] [CrossRef]
  22. Chong, M.; Cosma, S. A formulation of the continuity equation of MUSCAT for either flat or complex terrain. J. Atmos. Ocean. Technol. 2000, 17, 1556–1565. [Google Scholar] [CrossRef]
  23. Liou, Y.-C.; Chang, S.-F.; Sun, J. An application of the immersed boundary method for recovering the three-dimensional wind fields over complex terrain using multiple-Doppler radar data. Mon. Weather. Rev. 2012, 140, 1603–1619. [Google Scholar] [CrossRef]
  24. Foerster, A.M.; Bell, M.M.; Harr, P.A.; Jones, S.C. Observations of the eyewall structure of Typhoon Sinlaku (2008) during the transformation stage of extratropical transition. Mon. Weather Rev. 2014, 142, 3372–3392. [Google Scholar] [CrossRef][Green Version]
  25. Martinez, J.; Bell, M.M.; Rogers, R.F.; Doyle, J.D. Axisymmetric potential vorticity evolution of Hurricane Patricia (2015). J. Atmos. Sci. 2019, 76, 2043–2063. [Google Scholar] [CrossRef]
  26. Cha, T.Y.; Bell, M.M.; Lee, W.-C.; DesRosiers, A.J. Polygonal eyewall asymmetries during the rapid intensification of Hurricane Michael (2018). Geophys. Res. Lett. 2020, 47, e2020GL087919. [Google Scholar] [CrossRef]
  27. Tai, S.-L.; Liou, Y.-C.; Sun, J.; Chang, S.F. The development of a terrain-resolving scheme for the forward model and its adjoint in the four-dimensional variational Doppler radar analysis system (VDRAS). Mon. Weather. Rev. 2017, 145, 289–306. [Google Scholar] [CrossRef]
  28. Jorgensen, D.P.; LeMone, M.A.; Trier, S.B. Structure and evolution of the 22 February 1993 TOGA COARE squall line: Aircraft observations of precipitation, circulation, and surface energy fluxes. J. Atmos. Sci. 1997, 54, 1961–1985. [Google Scholar] [CrossRef]
  29. Weisman, M.L. The genesis of severe, long-lived bow echoes. J. Atmos. Sci. 1993, 50, 645–670. [Google Scholar] [CrossRef]
  30. Atkins, N.T.; Arnott, J.M.; Przybylinski, R.W.; Wolf, R.A.; Ketcham, B.D. Vortex structure and evolution within bow echoes. Part I: Single-Doppler and damage analysis of the 29 June 1998 derecho. Mon. Weather Rev. 2004, 132, 2224–2242. [Google Scholar] [CrossRef]
  31. Fujita, T.T. Manual of Downburst Identification for Project Nimrod; Satellite and Mesometeorology Research Paper No. 156; University of Chicago: Chicago, IL, USA, 1978; p. 104. [Google Scholar]
  32. Schmid, W.; Schiesser, H.-H.; Furger, M.; Jenni, M. The origin of severe winds in a tornadic Bow-Echo storm over northern Switzerland. Mon. Weather Rev. 2000, 128, 192–207. [Google Scholar] [CrossRef]
  33. Przybylinski, R.W.; Gery, W.J. The reliability of the bow echo as an important severe weather signature. In Proceedings of the 13th Conference on Severe Local Storms, Tulsa, OK, USA, 17–20 October 1983; pp. 270–273. [Google Scholar]
  34. Smull, B.F.; Houze, R.A. A midlatitude squall line with a trailing region of stratiform rain: Radar and satellite observations. Mon. Weather Rev. 1985, 113, 117–133. [Google Scholar] [CrossRef]
  35. Smull, B.F.; Houze, R.A. Rear inflow in squall lines with trailing stratiform precipitation. Mon. Weather Rev. 1987, 115, 2869–2889. [Google Scholar] [CrossRef]
  36. Przybylinski, R.W. The bow echo: Observations, numerical simulations, and severe weather detection methods. Mon. Weather Rev. 1995, 10, 203–218. [Google Scholar] [CrossRef]
  37. International Civil Aviation Organization (ICAO). Annex 3—Meteorological Service for International Air Navigation, 20th ed.; ICAO: Montreal, QC, Canada, 2018; p. 144. [Google Scholar]
  38. Collis, S.; Protat, A.; Chung, K.-S. The effect of radial velocity gridding artifacts on variationally retrieved vertical velocities. J. Atmos. Ocean. Technol. 2010, 27, 1239–1246. [Google Scholar] [CrossRef]
Applsci 15 08562 g001
Figure 1. Wind and gust observations at 07:00 UTC ((a) and (b), respectively), 07:30 UTC ((c) and (d), respectively) and 08:00 UTC ((e) and (f), respectively) on 15 March 2025. “M” in (b,d,f) indicates gust observations not available from weather buoys or automatic weather stations under maintenance. “V” in (a,c,e) indicates the wind direction was variable.
Figure 1. Wind and gust observations at 07:00 UTC ((a) and (b), respectively), 07:30 UTC ((c) and (d), respectively) and 08:00 UTC ((e) and (f), respectively) on 15 March 2025. “M” in (b,d,f) indicates gust observations not available from weather buoys or automatic weather stations under maintenance. “V” in (a,c,e) indicates the wind direction was variable.
Applsci 15 08562 g001
Applsci 15 08562 g002aApplsci 15 08562 g002b
Figure 2. Horizontal winds at 3 km above sea level overlaying on 1 km CAPPI radar reflectivity (a) and vertical wind at 3 km height (b) at 06:48 UTC on 15 March 2025 (inset in (a,b) shows the vertical cross section of winds along the A–B line). Features of rear-inflow jet and rear-inflow notch associated with the squall line can be seen in (a).
Figure 2. Horizontal winds at 3 km above sea level overlaying on 1 km CAPPI radar reflectivity (a) and vertical wind at 3 km height (b) at 06:48 UTC on 15 March 2025 (inset in (a,b) shows the vertical cross section of winds along the A–B line). Features of rear-inflow jet and rear-inflow notch associated with the squall line can be seen in (a).
Applsci 15 08562 g002aApplsci 15 08562 g002b
Applsci 15 08562 g003aApplsci 15 08562 g003b
Figure 3. Horizontal winds at 3 km above sea level overlaying on 1 km CAPPI radar reflectivity (a) and vertical wind at 3 km height (b) at 07:06 UTC on 15 March 2025 (inset in (a,b) shows the vertical cross section of winds along the A–B line).
Figure 3. Horizontal winds at 3 km above sea level overlaying on 1 km CAPPI radar reflectivity (a) and vertical wind at 3 km height (b) at 07:06 UTC on 15 March 2025 (inset in (a,b) shows the vertical cross section of winds along the A–B line).
Applsci 15 08562 g003aApplsci 15 08562 g003b
Applsci 15 08562 g004
Figure 4. Horizontal winds at 3 km above sea level overlaying on 1 km CAPPI radar reflectivity (a) and vertical wind at 3 km height (b) at 08:00 UTC on 15 March 2025 (inset in (a,b) shows the vertical cross section of winds along the A–B line).
Figure 4. Horizontal winds at 3 km above sea level overlaying on 1 km CAPPI radar reflectivity (a) and vertical wind at 3 km height (b) at 08:00 UTC on 15 March 2025 (inset in (a,b) shows the vertical cross section of winds along the A–B line).
Applsci 15 08562 g004
Applsci 15 08562 g005aApplsci 15 08562 g005b
Figure 5. Flight data including eddy dissipation rate (EDR), vertical wind, horizontal wind and altitude from the quick access recorder (QAR) of flight no. A in the period of around 06:54 to 07:02 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:00 UTC on the same day. The blue circles indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Figure 5. Flight data including eddy dissipation rate (EDR), vertical wind, horizontal wind and altitude from the quick access recorder (QAR) of flight no. A in the period of around 06:54 to 07:02 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:00 UTC on the same day. The blue circles indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Applsci 15 08562 g005aApplsci 15 08562 g005b
Applsci 15 08562 g006
Figure 6. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. B in the period of around 06:53 to 07:31 UTC on 15 March 2025. The corresponding 1 km height composite EDR image from weather radars and 2 km height vertical wind derived from radar-based wind retrieval (at 07:00 UTC on the same day) are the same as Figure 5b,c.
Figure 6. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. B in the period of around 06:53 to 07:31 UTC on 15 March 2025. The corresponding 1 km height composite EDR image from weather radars and 2 km height vertical wind derived from radar-based wind retrieval (at 07:00 UTC on the same day) are the same as Figure 5b,c.
Applsci 15 08562 g006
Applsci 15 08562 g007aApplsci 15 08562 g007b
Figure 7. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. C in the period of around 06:50 to 07:37 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:30 UTC on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Figure 7. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. C in the period of around 06:50 to 07:37 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:30 UTC on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Applsci 15 08562 g007aApplsci 15 08562 g007b
Applsci 15 08562 g008aApplsci 15 08562 g008b
Figure 8. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. D in the period of around 06:53 to 07:44 UTC on 15 March 2025 (a); 2 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:00 UTC on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Figure 8. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. D in the period of around 06:53 to 07:44 UTC on 15 March 2025 (a); 2 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:00 UTC on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Applsci 15 08562 g008aApplsci 15 08562 g008b
Applsci 15 08562 g009aApplsci 15 08562 g009b
Figure 9. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. E in the period of around 07:25 to 07:34 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval at 07:30 UTC (c) on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Figure 9. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. E in the period of around 07:25 to 07:34 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval at 07:30 UTC (c) on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Applsci 15 08562 g009aApplsci 15 08562 g009b
Applsci 15 08562 g010aApplsci 15 08562 g010b
Figure 10. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. F in the period of around 07:25 to 07:36 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:36 UTC on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Figure 10. Flight data including EDR, vertical wind, horizontal wind and altitude from the QAR of flight no. F in the period of around 07:25 to 07:36 UTC on 15 March 2025 (a); 1 km height composite EDR image from weather radars (b) and 2 km height vertical wind derived from radar-based wind retrieval (c) at 07:36 UTC on the same day. The blue circles/ellipse indicate the magnitudes of EDR and vertical winds (a) and roughly the corresponding location of the aircraft when encountering turbulence (b,c).
Applsci 15 08562 g010aApplsci 15 08562 g010b
Table 1. Summary of comparison results between radar and flight data for EDR and vertical velocity.
Table 1. Summary of comparison results between radar and flight data for EDR and vertical velocity.
Flight No.Time
Period		Aircraft AltitudeEDR from
Aircraft Data
(m2/3s−1)		EDR Derived from Radar Spectrum Width Data
(m2/3s−1)		Vertical Velocity from
Aircraft Data		Vertical Velocity
Derived from Radar-Based Wind Retrieval
A06:54 to 06:56 UTC.Ascending from about 0.06 km to  1.3 km.About 0.6 at around 06:56 UTC when the aircraft was at a height of about
1.1 km.		About 0.6 to 0.7 at
1 km height.		About −10.6 m/s at around 06:56 UTC.Less than −10.0 m/s at height of 2 km. Wind retrieval method was unable to depict downward velocity at 1 km height and below.
B06:56 to 06:57 UTC
(missed approach ~0.09 km).		Descending from about 0.27 km to around 0.09 km.Between 0.6 to 0.7 at around 06:57 UTC when the aircraft was at a height of about 0.09.Between 0.6 to 0.7 at 1 km height.Max. between −12.0 to −13.0 m/s at around 06:57 UTC.
07:00 to 07:01 UTC.At a height of about 1.8 km.Between 0.6 to 0.7 at around 07:01 UTC when the aircraft was at a height of about 1.8 km.Between 0.6 to 0.7 at 2 km height.Max. about +10.0 m/s at around 07:01 UTC.Estimated −6.0 to −9.0 m/s at 07:00 UTC at 2 km height.
07:26 to 07:27 UTC.At a height of about 1.4 km. Max. about 1.0 between 07:26 and 07:27 UTC when the aircraft was at a height of about 1.4 km. Max. 0.3 to 0.35 at 1 km height and 0.08 to 0.1 at 2 km height.Fluctuating rapidly from less than −10.0 m/s to around +8.6 m/s.Estimated to fluctuate between −9.0 m/s and +9.0 m/s at 2 km height.
C06:57 to 07:00 UTC (missed approach ~0.27 km).Descending from about 0.46 km to around 0.27 km but then ascending to about 1 km.Between 0.6 to 0.7 at around 06:59 UTC when the aircraft was at a height of about 0.8 km.Between 0.6 to 0.7 at 1 km height.Max. −12.5 m/s at around 06:59 UTC.Less than −10.0 m/s at height of 2 km. Wind retrieval method was unable to depict downward velocity at 1 km height and below.
07:04 to 07:05 UTC.At a height of about 2 km.Between 0.6 to 0.7 at around 07:04 UTC when the aircraft was at a height of about 2 km.Between 0.6 to 0.7 at 2 km height.About −1.0 m/s at around 07:04 UTC.Estimated to fluctuate between −3.0 m/s and +3.0 m/s at 2 km height.
0730 to 07:32 UTC.At a height of about 1.4 km.Fluctuating from 0.1 to 0.5 in 07:30–07:32 UTC when the aircraft was at a height of about 1.4 kmVarying from 0.1 to 0.3 at 1 km height and 0.05 to 0.3 at   2 km height.Fluctuating rapidly from less than −10.0 m/s to around +7.0 m/s in 07:30–07:32 UTC.Estimated to fluctuate between −9.0 m/s and +9.0 m/s at 2 km height.
D07:01 to 07:03 UTC (missed approach ~0.55 km).Descending from about 0.67 km to around 0.55 km but then ascending to 1.5 km.Max. about 1.0 at around 07:03 UTC when the aircraft was at a height of about 1.5 km.0.6 to 0.8 at 1 km height and 0.7 to 0.8 at 2 km height. Fluctuating rapidly from about −14.0 m/s to around +7.0 m/s in 07:01–07:03 UTC.Estimated to fluctuate between −15.0 m/s and +15.0 m/s at 2 km height.
E07:26 to 07:28 UTC.Flying at a height of about 1.4 km. Max. about 0.7 at around 07:27 UTC when the aircraft was at a height of about 1.4 km.Varying from 0.1 to 0.3 at 1 km height and 0.05 to 0.3 at   2 km height.Fluctuating rapidly from about −7.0 m/s to around +10.0 m/s at around 07:27 UTC.Estimated to fluctuate between −3.0 m/s and +12.0 m/s at 2 km height.
F07:35 to 07:37 UTC.Descending from about 0.6 km to around 0.06 km.Fluctuating from 0.2 to 0.7 in 07:35–07:37 UTC during the descend of the aircraft.Varying from 0.3 to 0.8 at 1 km height.Fluctuating rapidly from about −7.5 m/s to around +4.5 m/s in 07:35–07:36 UTC.Estimated to fluctuate between −3.0 m/s and +3.0 m/s at 07:36 UTC at 2 km height.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Chan, P.-w.; Chan, Y.-w.; Cheung, P.; Chong, M.-l. An Observational Study of a Severe Squall Line Crossing Hong Kong on 15 March 2025 Based on Radar-Retrieved Three-Dimensional Winds and Flight Data. Appl. Sci. 2025, 15, 8562. https://doi.org/10.3390/app15158562

AMA Style

Chan P-w, Chan Y-w, Cheung P, Chong M-l. An Observational Study of a Severe Squall Line Crossing Hong Kong on 15 March 2025 Based on Radar-Retrieved Three-Dimensional Winds and Flight Data. Applied Sciences. 2025; 15(15):8562. https://doi.org/10.3390/app15158562

Chicago/Turabian Style

Chan, Pak-wai, Ying-wa Chan, Ping Cheung, and Man-lok Chong. 2025. "An Observational Study of a Severe Squall Line Crossing Hong Kong on 15 March 2025 Based on Radar-Retrieved Three-Dimensional Winds and Flight Data" Applied Sciences 15, no. 15: 8562. https://doi.org/10.3390/app15158562

APA Style

Chan, P.-w., Chan, Y.-w., Cheung, P., & Chong, M.-l. (2025). An Observational Study of a Severe Squall Line Crossing Hong Kong on 15 March 2025 Based on Radar-Retrieved Three-Dimensional Winds and Flight Data. Applied Sciences, 15(15), 8562. https://doi.org/10.3390/app15158562

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop