Next Article in Journal
Displacement Time Series Forecasting Using Sentinel-1 SBAS-InSAR Results in a Mining Subsidence Case Study—Evaluation of Machine Learning and Deep Learning Methods
Previous Article in Journal
Assessment of Instrument Performance of the FY3E/JTSIM/DARA Radiometer Through the Analysis of TSI Observations
Previous Article in Special Issue
Spatiotemporal Distribution, Sources, and Impact on Atmospheric Oxidation of Reactive Nitrogen Oxides in the North China Plain Agricultural Regions in Summer
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 17
Issue 23
10.3390/rs17233904
remotesensing-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

Vertical Characteristics of an Ozone Pollution Episode in Hong Kong Under the Typhoon Mawar—A Case Study

by
Libin Zhu
1,2,
Jie Wang
1,2,3,*,
Yiwei Xu
1,2,
Na Ma
2,
Xiaoquan Song
4,
Jie Qin
2,
Beibei Li
2,
Wilson B. C. Tsui
5,
Lihui Lv
1,6 and
Tianshu Zhang
2,6
1
Institutes of Physical Science and Information Technology, Anhui University, Hefei 230601, China
2
Institute of Environment, Hefei Comprehensive National Science Center, Hefei 230031, China
3
National Key Laboratory of Opto-Electronic Information Acquisition and Protection Technology, Anhui University, Hefei 230601, China
4
College of Marine Technology, Faculty of Information Science and Engineering, Ocean University of China, Qingdao 266100, China
5
PTC International Limited, Hong Kong, China
6
Key Laboratory of Environment Optics and Technology, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2025, 17(23), 3904; https://doi.org/10.3390/rs17233904
Submission received: 13 November 2025 / Revised: 29 November 2025 / Accepted: 29 November 2025 / Published: 1 December 2025
(This article belongs to the Special Issue Stereoscopic Remote Sensing of Air Pollutants: Emission, Formation, and Transport)
Browse Figures
Versions Notes

Highlights

What are the main findings?
  • Transportation and accumulation processes of ozone and aerosols were revealed by High-resolution Differential Absorption Lidar.
  • An ozone episode was jointly driven by the Western Pacific Subtropical High and the non-landfall Typhoon Mawar.
What are the implication of the main findings?
  • This study elucidates a coupled mechanism driving coastal ozone pollution: the synergy among meteorology, local photochemistry, and regional transport.
  • It provides key insights for regional air quality management in coastal cities of southeastern China.
  • It quantitatively analyzes the vertical distribution characteristics of ozone pollution in the lower troposphere during a specific pollution episode.

Abstract

This study investigates a typical ozone pollution episode in Hong Kong from May 29 to 31, 2023. Based on the observations of a Differential Absorption Lidar (DIAL) system, both ozone and aerosols accumulated below 1.5 km during the pollution episode. Ozone exhibited distinct formation and accumulation characteristics, with concentrations exceeding 200 μg m−3. Aerosols presented evident features of regional transport and local coupling, with extinction coefficients surpassing 1.1 km−1. During late spring to early summer, the northward extension of the Western Pacific Subtropical High (WPSH) established favorable conditions for ozone production. This background was amplified by Typhoon Mawar, whose peripheral circulation channeled pollutants from the Pearl River Delta into Hong Kong through horizontal and vertical pathways, significantly worsening near-surface air quality. The episode was eventually mitigated, as enhanced vertical mixing facilitated the dispersion and removal of accumulated pollutants. These results highlight the critical role of meteorological–chemical interactions in shaping this ozone pollution episode.

Graphical Abstract

1. Introduction

The tropospheric ozone is a common secondary pollutant mainly produced by photochemical reactions between volatile organic compounds (VOCs) and nitrogen oxides (NOX) in the presence of solar radiation, mainly consisting of the chain reactions of the NO–NO2–O3 system [1,2]. In recent years, toxicologic research has found that chronic exposure to elevated ozone levels can harm the human respiratory system, exacerbate asthma and chronic obstructive pulmonary disease (COPD), and increase the risk of cardiovascular diseases and diabetes [3]. Over the past two decades, global tropospheric ozone concentrations have been continuously increasing, with an average rise of approximately 4.5 μg m−3 per decade. Correspondingly, the number of premature deaths attributable to ozone exposure increased from about 290,000 in 2000 to 420,000 in 2019, representing a 46% rise [4,5,6]. However, the ecological impacts of ozone cannot be neglected. High levels of ozone can significantly reduce the photosynthetic activity in plants, leading to yield reductions of 3.5% and 6.1% in maize and soybean, respectively [7,8]. In China, ozone pollution control faces serious challenges. Although particulate matter levels (such as PM2.5) have improved year by year in key cities (e.g., 339 major prefecture-level cities), surface ozone concentrations have continued to rise rather than decline [9]. A previous study has indicated that the summer maximum daily 8 h average (MDA8) ozone concentration in China grew by an average of 3.72 μg m−3 per annum between 2013 and 2019, while the annual mean ozone concentration rose at a rate of 1.84 μg m−3 per year [10,11]. These findings highlight the urgent need to optimize ozone pollution control measures and to deepen the understanding of its underlying mechanisms.
Tropospheric ozone formation involves a complex interaction of chemical and physical processes that remain subject to substantial uncertainty. Central to this mechanism are photochemical reactions driven by radical chemistry, in which hydroxyl radical (OH), hydroperoxyl radical (HO2), and related radicals sustain chain reactions that control the production and removal of ozone [12,13,14,15,16,17,18]. During the daytime, photochemical reactions of ozone precursors like NOX cause rapid accumulation of ozone near the surface, but at night, ozone levels fall sharply due to attenuation of solar radiation and the titration effect [19,20,21,22]. In addition to local photochemical processes, regional-scale horizontal transport [23], vertical transport [24], and stratospheric intrusion [25] also play significant roles in the formation and evolution of urban ozone pollution. Studies have shown that ozone and its precursors transported on a regional scale can contribute to between 20% and 50% to the rise in surface ozone concentrations, with contributions occasionally reaching as high as 72% [26,27,28,29]. With the change in the boundary layer height, vertical mixing can entrain upper tropospheric air into the near-surface layer, creating a superimposing effect on daytime ozone concentrations on the urban surface. Studies have shown that vertical transport can contribute to between 24.0% and 63.6% increase in near-surface ozone concentrations [30,31,32]. Stratosphere–troposphere exchange (STE) and stratospheric intrusion (SI) under extreme conditions are other important pathways resulting in high surface ozone concentrations. According to Chen et al. [33], stratospheric intrusion to the surface (SITS) can rapidly elevate near-surface ozone by approximately 20 ppbv (39.3 μg m−3). The strongest response commonly appears within the first 8–10 h after the arrival of stratospheric air, during which O3 increases by 60–70% relative to the non-intrusion background. The interaction of these multiple mechanisms has increasingly positioned urban ozone pollution as a critical concern in contemporary atmospheric research.
The Pearl River Delta (PRD) and Hong Kong have become firmly established as one of China’s most ozone-prone regions. According to Li et al. [34], alongside growing anthropogenic emissions, ozone mixing ratios at urban sites in the PRD increased at a rate of 0.28–1.02 ppbv yr−1 from 2006 to 2019. This trend, marked by a springtime MDA8 O3 increase of about 1 ppbv yr−1, demonstrates that high ozone levels are no longer confined to midsummer but extend into spring [35]. These persistent increases underscore the growing importance of meteorological regulation, photochemical production, and regional transport processes in shaping ozone variability across the PRD, including Hong Kong. Hong Kong’s surface ozone levels are shaped by the interplay of regional transport, local photochemistry, and specific meteorological conditions. Modeling studies indicate that regional inflow from the Pearl River Delta is a dominant source, contributing to 40–70% of daytime and 50–90% of nighttime surface ozone in the city [28]. The local chemical environment that facilitates ozone production has intensified over time, as evidenced by a growing sensitivity of ozone formation to VOCs between 2005 and 2014, despite stable overall VOC levels [36]. This mechanism explains the paradoxical rise in Hong Kong’s ozone concentrations concurrent with substantial declines in local NOX and CO emissions, which Feng et al. attribute to an enhanced atmospheric photochemical oxidation capacity [37]. Furthermore, meteorological conditions, particularly those associated with typhoons—characterized by high temperatures, weak winds, and subsidence—are known to promote photochemical production and pollutant accumulation, exacerbating ozone episodes across southern China, including Hong Kong [38]. Overall, Hong Kong exemplifies a typical region within the PRD where meteorology, chemistry and regional transport jointly characterize the complex formation of costal city ozone pollution.
Focused on a typical ozone pollution episode during Hong Kong’s 2023 spring–summer transition under Typhoon Mawar’s influence, this study addresses a characteristic case resulting from synergistic seasonal and meteorological effects. By integrating ground-based monitoring, Lidar observations, and reanalysis datasets, this study aims to elucidate the mechanisms of coastal ozone pollution under complex transitional-season meteorology and to provide insights for its regulation and forecasting in similar coastal urban environments.

2. Materials and Methods

2.1. Lidar System

During the observation period from 00:00 on May 29 to 23:40 on May 31, 2023, an atmospheric ozone Lidar system was deployed near the Hong Kong Tai Mei Tuk Sea Activity Centre (22.47°N, 114.23°E, CASEO-O3-Lidar, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, China). The study area is shown in Figure 1a, and the specific location of the Lidar site within the Hong Kong Special Administrative Region is depicted in Figure 1b. The detailed Lidar location is shown in Figure 1b. This location was chosen because, being situated at a coastal location with open terrain and minimal obstruction from high-rise buildings, this site provides favorable conditions for reliable Lidar profiling.
The Lidar system consisted of four main components: a laser transmitter, optical receiver, data acquisition module, and control/processing unit. The laser transmitter was a semiconductor-pumped, all-solid-state, air-cooled system capable of emitting two ultraviolet (0.4 mJ at 290 nm, 0.2 mJ at 295 nm) and two visible wavelengths (3.3 mJ at 580 nm, 4 mJ at 590 nm). Specifically, atmospheric ozone concentrations were retrieved using the Differential Absorption Lidar (DIAL) technique, which exploits the difference in absorption cross-sections of ozone molecules at 290 nm and 295 nm by comparing their backscattered signal intensities. Similarly, the vertical distribution of atmospheric aerosols was obtained using the Mie scattering principle by measuring elastic backscatter signals at visible wavelengths (580 nm and 590 nm). The DIAL system employed a 200 mm diameter Cassegrain telescope as its optical receiver. This configuration utilizes a concave primary mirror and a convex secondary mirror to collect and focus the atmospheric backscattered light through a folded path onto the focal plane. Owing to its compact form and high light-gathering efficiency, this telescope design has been widely adopted in atmospheric Lidar and DIAL systems [39,40]. Following the optical collection, the backscattered signals were directed to the photodetection unit, which consists of a spectral separation module and a high-sensitivity photomultiplier tube (PMT, Hamamatsu R9800, Hamamatsu Photonics, Hamamatsu City, Shizuoka, Japan), enabling the precise acquisition and processing of multi-wavelength DIAL returns. During the campaign, the system operated at a sampling frequency of 5 kHz, providing a vertical spatial resolution of 7.5 m and a minimum temporal resolution of 3 s. Its effective detection height extended beyond 3 km, while the blind zone remained below 0.3 km.
The data quality and retrieval accuracy of this DIAL system have been validated through several previous experiments. A comparison at 495 m altitude with an ultraviolet (UV) photometric ozone analyzer on the Canton Tower demonstrated close agreement (R ≥ 0.84) [41,42]. Furthermore, intercomparison with an unmanned aerial vehicle (UAV)-borne ozone monitor confirmed the system’s accuracy, showing a mean statistical error of 6.4% within the 0.2–1.0 km range [42]. In this experiment, the ozone concentrations retrieved by the DIAL system at 300 m and within the 300–350 m altitude were compared with the in situ measurements from the Tai Po Air Quality Monitoring Station (AQMS) from 29–31 May 2023. The correlation coefficients were 0.79 (Figure 2a) and 0.86 (Figure 2b), respectively. These intercomparison results collectively demonstrate that the DIAL system provides highly reliable and accurate measurements for near-surface observations.

2.2. Other Supporting Datasets

To assess its spatial representativeness, we followed the methodology proposed by Baylon et al. [43], and hourly ozone and PM2.5 data from the Tai Po AQMS (Jan–Jun 2023) were compared with measurements from the easternmost (Tap Mun AQMS), westernmost (Tung Chung AQMS), southernmost (Southern AQMS), and northernmost (North AQMS) stations in Hong Kong [44,45]. The results showed strong correlations (R > 0.8 for ozone and R > 0.7 for PM2.5), indicating that the Tai Po AQMS is representative of the regional background conditions in Hong Kong (Table 1).
This study employed the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis products. Meteorological data were derived from the ECMWF Reanalysis v5 (ERA5) datasets, with a minimum temporal resolution of 1 h, a horizontal resolution of 0.25° × 0.25°, and a vertical coverage ranging from 1000 hPa to 1 hPa, obtained from the ECMWF Climate Data Store [46,47,48,49]. All analyses are based on the May 2025 data release. Considering the limited horizontal resolution of ERA5, meteorological data at the grid point (22.5°N, 114.25°E) were used to represent the atmospheric conditions above the DIAL observation site.
In addition to the aforementioned supporting datasets, the track and intensity data of Typhoon Mawar used in this study were obtained from the China Meteorological Administration (CMA) Tropical Cyclone Data Center [50].

2.3. Backward Trajectory and Cluster Analysis

National Oceanic and Atmospheric Administration (NOAA)’s Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model was used to calculate 48 h backward trajectories at 4 h intervals for May 29 to 31, 2023 [51]. A subsequent cluster analysis categorized the dominant transport pathways and assessed the relative contributions of various air mass origins.

3. Results

3.1. Background of the Ozone Pollution Episode

From May 20 to 28, as Typhoon Mawar moved westward, Hong Kong’s MDA8 ozone concentrations stayed below 120 μg m−3. A sharp increase occurred from May 29 to 31 when Mawar was positioned over the Philippine Sea, approximately 1100 km from Hong Kong. During this period, MDA8 ozone notably exceeded China’s Ambient Air Quality Standard Grade II threshold (160 μg m−3) and surpassed the World Health Organization (WHO) guideline (100 μg m−3) by more than 50%. From June 1 to June 3, as Typhoon Mawar moved northeastward and transitioned into an extratropical cyclone, the MDA8 ozone concentrations subsequently subsided (Figure 3a). Given the exceedance of both Chinese and WHO MDA8 ozone thresholds for three consecutive days (May 29–31), this period is defined as an ozone pollution episode and selected for detailed analysis.
A comparative analysis of the diurnal ozone trend (Figure 3b) reveals that daytime concentrations (09:00–18:00) during the pollution episode were substantially elevated compared to the pre-episode period. Exceedances of the Chinese ozone standard occurred consistently between 12:00 and 19:00, and elevated levels persisted even after the episode. Concurrently, PM2.5 concentrations (Figure 3c) also rose, indicating a concurrent increase in particulate pollution. Together, these results confirm that the event from May 29 to 31 significantly degraded air quality in Hong Kong.

3.2. Overview of the Ozone Pollution Episode

According to the hourly air pollutant data from May 29 to 31 at the Tai Po AQMS operated by the HKEPD, the temporal evolution and key characteristics of individual pollutants during the pollution episode can be clearly identified.
As illustrated in Figure 4a,b, PM2.5 and PM10 exhibited highly consistent temporal variations during the first two days of the pollution episode, showing a strong correlation (R = 0.88). Commencing at 00:00 on May 29, both concentrations progressively escalated and reached their respective daily peaks at 19:00, i.e., 35 μg m−3 for PM2.5 and 61 μg m−3 for PM10. On May 30, particulate matter concentrations initially showed a minor decrease after 00:00, which was succeeded by a synchronized rise starting at 07:00. PM2.5 reached its highest level during the entire pollution episode, peaking at 57 μg m−3 at 16:00. PM10 attained its peak slightly later than PM2.5, recording its maximum concentration of 79 μg m−3 at 21:00. After 22:00, both PM concentrations began gradually declining. The results indicate that both PM2.5 and PM10 followed a consistent pattern of progressive accumulation on May 29 and 30. Nonetheless, their intra-day variation patterns differed slightly. On May 29, both concentrations increased in the early hours, followed by a slight dip before rising again to reach the daily peak. Conversely, on May 30, both concentrations continuously increased throughout the whole day until reaching their respective maxima. The concentration patterns on May 31 significantly differed from those of the previous two days. Both PM2.5 and PM10 demonstrated a gradual decline starting from 00:00, reaching their daily minima at 18:00, i.e., 21 μg m−3 for PM2.5 and 28 μg m−3 for PM10. Despite a minor recovery being observed afterward, the overall trend remained downward, in clear contrast to the accumulation dynamics seen on May 29 and 30.
NO2 and NOX exhibited synchronized variation during the episode (Figure 4c,d). Such a pattern deviates from the typical diurnal features of traffic emissions, likely reflecting an enhanced regional background induced by large-scale pollutant accumulation. On May 29 and 30, both pollutants demonstrated increasing trends with two daily peaks coinciding with traffic rush hours. However, on May 31, both pollutants reached their minima at 14:00, i.e., 17 μg m−3 for NO2 and 19 μg m−3 for NOX, before surging to about 90 μg m−3 by 19:00. This pattern markedly diverged from the steady decline observed for particulate matter.
During the pollution episode, ozone concentrations exhibited a distinct diurnal pattern, with low values observed in the early morning, peak levels during the day, and a decline in the evening (Figure 4e). On May 29 and 30, ozone values remained consistently above 100 μg m−3 between 12:00 and 20:00, with the daily maxima exceeding 200 μg m−3 (Grade II 1 h ozone limit of China’s National Ambient Air Quality Standard). By contrast, May 31 recorded the lowest daily mean ozone concentration of the three days (84.67 μg m−3), and the duration of elevated ozone levels above 100 μg m−3 was not only shorter but also occurred earlier in the day (from 10:00 to 15:00), compared to the previous two days.
Overall, PM2.5, PM10, NO2, and NOX all exhibited an upward trend on May 29 and 30. However, PM2.5 and PM10 concentrations showed a persistent decline, with only a slight rebound observed in the evening on May 31. In contrast, NO2 and NOX levels dropped until 14:00, and then spiked abruptly, reaching around 90 μg m−3, signifying a trend markedly different from that of particulate matter. Ozone maintained its typical diurnal cycle but showed notable variation. On May 29 and 30, the duration of elevated concentrations (exceeding 100 μg m−3) was significantly longer, with daily peaks exceeding 200 μg m−3 on both days. However, by May 31, the period of elevated ozone was notably shorter, and its daily peak was much lower than that on the preceding days.

3.3. Lidar Observations of the Pollution Episode

3.3.1. Ozone Observation Results

Results from Lidar observation in Figure 5a indicate that, from May 29 to 31, ozone periodically accumulated in the lower troposphere over Hong Kong, predominantly distributed within the 0.3–1.5 km height layer. Pronounced ozone plumes appeared regularly between 16:00 and 22:00, with all peak concentrations exceeding 200 μg m−3.
From 0:00 to 12:00 on May 29, ozone concentrations remained low and stable. Within the 0.3–1.5 km height layer, vertical mean ozone concentrations stayed below 90 μg m−3. During this period, from 12:00 to 22:00, a steady increase in ozone concentrations was observed in the same layer, with the vertical mean concentrations exceeding 150 μg m−3. In the 0.3–0.6 km height layer, vertical mean concentrations also surpassed 150 μg m−3 and approached 200 μg m−3 between 17:30 and 18:00. In contrast, the increase in ozone in the 0.6–1.5 km layer occurred later, with vertical mean concentrations rising above 150 μg m−3 only after 16:00. The most substantial ozone enhancement was recorded between 19:00 and 19:30, during which values hovered around 170 μg m−3. This evolution agreed with surface observations, which revealed that the enhancement resulted from a combination of factors: although photochemical production had weakened, the near-surface layer had not yet experienced strong NO titration, and a significant reservoir of daytime ozone persisted in the residual layer.
On May 30, from 0:00 to 12:00, vertical mean ozone concentrations within the 0.3–1.5 km layer continuously decreased before 08:00, and then sharply increased, exceeding 120 μg m−3 at 11:50. The series of concentrations remained above 90 μg m−3 throughout this period, in agreement with the surface-based observation. Between 12:00 and 18:00, vertical mean concentrations within the 0.3–1.5 km layer exhibited a rise–fall pattern, surpassing 100 μg m−3 from 12:00 to 16:00, then drastically dropped below this threshold between 16:00 and 18:00. In the same period, peak concentrations in this layer exceeded 200 μg m−3, reaching up to 270 μg m−3 from 15:00–18:00. This divergence between declining mean values and surging peak levels indicates that ozone plumes were intensifying and descending in altitude, gradually accumulating below 0.9 km. From 18:00 to 22:00, vertical mean ozone concentrations within the 0.3–1.5 km height layer continuously decreased, while the vertical distribution height rebounded rapidly to approximately 1.2 km.
From 0:00 to 7:00 on May 31, vertical mean ozone concentrations within the 0.3–1.5 km layer remained relatively low. Their lowest value was observed at 06:00, measuring only 63.71 μg m−3. During the period, from 7:00 to 14:00, vertical mean ozone concentrations increased to 90–110 μg m−3. However, the maximum concentrations consistently appeared near 0.3 km. After 14:00, vertical mean ozone concentrations within the 0.3–1.5 km range remained above 100 μg m−3, while the altitude of maximum concentrations shifted upward to between 0.7 and 1.0 km. This transition indicates that ozone was produced near the surface, and it gradually diffused upward, accompanied by enhanced vertical mixing.

3.3.2. Aerosol Observation Results

As illustrated in Figure 5b, aerosol loading within 0.3–1.5 km remained low during the period from 0:00 to 15:30 on May 29, and vertical mean extinction coefficients stayed below 0.6 km−1. Following this period, vertical mean extinction coefficients within the same layer increased to above 0.6 km−1 between 15:30 and 22:00. The maximum extinction coefficient within this layer reached approximately 0.9 km−1, while its corresponding altitude gradually decreased from 1.5 km to 0.9 km over time, indicating a downward transport of aerosols. Notably, between 17:30 and 21:00, vertical mean extinction coefficients within the 0.3–0.6 km layer rose from 0.4 km−1 to 0.6 km−1, whereas within the 0.6–1.5 km layer, it decreased from 0.8 km−1 to 0.6 km−1. This phenomenon reflects a distinct vertical stratification of aerosols, with 0.6 km serving as a dividing height.
Aerosol extinction coefficients exhibited a clear accumulation pattern and were primarily distributed below 1.2 km on May 30. Vertical mean extinction coefficients within the 0.3–1.2 km increased from 0.5 km−1 to 0.7 km−1 between 0:00 and 12:00. During the same period, surface-based observation illustrated rising concentrations for both PM2.5 and PM10, which may be attributed to the combined influence of aerosols descending from upper levels on the previous day and local aerosol formation. During the same period, from 12:00 to 18:00, vertical mean extinction coefficients within the 0.3–1.2 km layer increased to about 0.9 km−1, with peak values exceeding 1.1 km−1. Meanwhile, surface PM2.5 concentrations increased from 36 μg m−3 to 55 μg m−3, which indicated that the coupled aerosols had a considerable impact on near-surface air quality. After 18:00, vertical extinction coefficients within the 0.3–1.2 km layer progressively declined from about 0.65 km−1 to 0.36 km−1, suggesting that aerosols were gradually being diluted or settling. In contrast, surface-based observations illustrated that PM10 concentrations remained around 70 μg m−3, and PM2.5 values stayed above 50 μg m−3. This result implied that the influence of elevated aerosol accumulation on surface air quality persisted.
On May 31, aerosol distribution over Hong Kong exhibited a notable vertical expansion, extending up to 1.5 km. In the early hours, vertical mean extinction coefficients within 0.3–1.5 km gradually increased from 0.3 km−1 to 0.55 km−1. After 12:00, the mean extinction coefficients remained relatively stable at around 0.6 km−1, while peak extinction coefficients within 0.3–1.5 km exceeded 1.1 km−1 after 22:00, with the highest values mainly occurring at around the 0.4 km altitude. During this period, surface PM2.5 concentrations declined from 35 μg m−3 to 27 μg m−3, and PM10 values dropped from 50 μg m−3 to 35 μg m−3. This reduction may be attributed to the upward vertical dispersion of aerosols, which strongly weakened the coupling effect with the surface layer.

3.4. Vertical Profiles During the Key Periods

To gain deeper insight into the formation and evolution mechanisms of ozone and aerosols, this study identified three key periods for detailed analysis. Each selected five-hour window was centered on the hourly ozone peak observed at the Tai Po AQMS, covering two hours before and after the peak time. Specifically, the selected intervals were as follows: 16:00–20:00 on May 29, 12:00–16:00 on May 30, and 11:00–15:00 on May 31. These three intervals are hereafter referred to as key periods I, II, and III, respectively (Figure 6).
During the key period I (Figure 6a), the ozone vertical profile revealed that the peak concentrations between 16:00 and 18:00 were consistently concentrated near 0.3 km, pointing to substantial ozone buildup close to the surface. By 19:00–20:00, the height of ozone peak shifted from 0.3 km to 1.16 km, while its concentrations within 0.3–0.6 km declined to about 150 μg m−3. This pattern coincided with a sharp drop in boundary layer height from 1.13 km to 0.14 km (Figure A1), which likely suppressed the upward dispersion of near-surface ozone. Together with the downward mixing of ozone aloft, this contributed to the surface ozone peak observed around 18:00 on May 29. A sudden drop in ozone concentrations below 0.6 km was also observed around 20:00, likely due to near-surface ozone titration at the observation site. Meanwhile, aerosols during the key period on May 29 displayed a vertically stratified structure (Figure 6b). The height of maximum extinction coefficient gradually decreased from 1.25 km at 16:00 to 1.13 km at 20:00, suggesting that the high-concentration aerosol plume remained aloft and was yet to descend to the surface.
During the key period II, the vertical ozone profile (Figure 6c) underwent pronounced structural changes. Between 14:00 and 16:00, the peak ozone concentration rose from 251.71 μg m−3 to 264.20 μg m−3, while the altitude of the ozone peak descended from 0.74 km to 0.5 km. Concurrently, the boundary layer height diminished from 0.92 km to 0.71 km. This pattern suggests that, as the boundary layer contracted, ozone that had been distributed within it was compressed downward, i.e., below 0.6 km, triggering a sharp increase in near-surface ozone concentrations. Meanwhile, the aerosol profile (Figure 6d) revealed a consistent rise in extinction coefficients below 0.6 km, along with a weakening of vertical gradients. This points to enhanced vertical mixing, enabling aerosols aloft to mix downwards and exert a notable impact on the surface layer, ultimately contributing to elevated concentrations of near-surface particulate matter.
During the key period III, the mean ozone concentrations within the 0.3–1.5 km layer, calculated from vertical profiles at 11:00, 13:00, and 15:00, were all approximately 60 μg m−3 (Figure 6e). Compared with the ozone concentrations on the previous two days, the ozone concentrations in this period were notably lower in both magnitude and vertical variation. The boundary layer height during this period was about 0.7 km (Figure A1), i.e., the highest recorded throughout the day. This favorable condition for vertical dispersion prevented the accumulation of ozone near the surface, resulting in overall lower concentrations. The mean extinction coefficients of all aerosol profiles in Figure 6f remained around 0.28 km−1, with noticeably lower intensity compared to that on the previous two days. Furthermore, the vertical variation was minimal, indicating a more uniform distribution.

4. Discussion

4.1. Synoptic Background Under the Influence of Typhoon Mawar

From late May to early June 2023, the northward shift in the Western Pacific Subtropical High (WPSH) led to clear-sky, low-cloud, and weak-wind conditions over the PRD, including Hong Kong. The evolution of the WPSH is illustrated in Figure 7a–c. This synoptic situation, as supported by Jiang et al. [52] who noted the substantial influence of WPSH variation on photochemical processes, established a meteorological environment highly conducive to ozone formation. Acting as a reinforcing mechanism, the non-landfall Typhoon Mawar further stabilized the atmosphere over the PRD, and its peripheral subsidence compounded the warm, sunny, and stagnant conditions initiated by the WPSH. This synergy exerted dual influences: it suppressed pollutant dispersion physically and it also stimulated biogenic volatile organic compound (BVOC) emissions and enhanced photochemical oxidation [53]. These interconnected mechanisms were instrumental in setting up the ozone pollution episode. Moreover, this synergy gave rise to the pronounced diurnal variations in ozone and PM2.5 observed during the pollution episode, as presented in Section 3.1.
In the late stage of the pollution episode (the afternoon of May 31), enhanced vertical convection developed over Hong Kong under the combined influence of the WPSH and the peripheral circulation of Typhoon Mawar. The strengthened boundary layer mixing first promoted the vertical dispersion of pollutants and also triggered local convective rainfall [54,55]. The precipitation further scavenged residual pollutants through washout and dilution. This cleansing process continued with the heavy rainfall on June 1, ultimately marking the termination of the ozone pollution episode.

4.2. Transport Pathways During Key Periods

According to the ERA5 data, surface winds over Hong Kong between 16:00 and 20:00 on May 29 (Key period I) were predominantly westerly to southwesterly, with speeds remaining below 3 m/s (Figure 8a). These weak winds favored the advection of ozone-rich, aerosol-poor marine air masses into Hong Kong [56,57,58]. At the 850 hPa level (approximately 1.5 km altitude), northerly winds prevailed, with wind speeds increasing from 4–5 m/s to 8–9 m/s (Figure 8b). These conditions promoted the southward transport of ozone, aerosols, and their precursors from northern cities such as Dongguan and Shenzhen southward into the Hong Kong region. The vertical cross-section of vertical velocity (Figure 9) further revealed persistent subsidence over Hong Kong within the 700–900 hPa layer (roughly 1–3 km altitude), with downward vertical velocities reaching about 0.2 Pa/s.
From 12:00 to 16:00 on May 30 (key period II), surface winds over Hong Kong were predominantly northwesterly to northerly, with speeds around 2 m/s (Figure 8c). Compared to the previous day, although surface wind speeds were weaker, these conditions still favored pollutant transport. Under such circumstances, pollutants originating from urban clusters in the Pearl River Delta, such as Guangzhou, Foshan, and Jiangmen, were more readily transported southward to Hong Kong, thereby exacerbating local air pollution. At the 850 hPa level, the wind field was dominated by northerly winds with speeds around 5 m/s (Figure 8d). The alignment of surface and upper-level winds indicates a vertically consistent transport pathway, transporting air masses from the northern regions across multiple lower atmosphere layers. According to the vertical cross-section of vertical velocity, between 12:00 and 16:00 on May 30, vertical motion in the 700–900 hPa layer shifted from upward to downward, with vertical velocity increasing from −0.15 Pa/s to 0.19 Pa/s (Figure 9). This transition indicates the onset of subsidence over Hong Kong, which likely facilitated the downward convergence of air masses and allowed pollutants transported from northern regions to descend and accumulate near the surface.
Between 11:00 and 15:00 on May 31, surface winds over Hong Kong gradually shifted from northeasterly to north-northwesterly, with speeds averaging around 2 m/s (Figure 8e). This phenomenon indicated that the influence of air masses from northern urban areas persisted. At the 850 hPa level, winds remained predominantly northerly but gradually weakened, with wind speeds decreasing from around 5 m/s to 3 m/s over time, reflecting a decline in upper-level transport strength (Figure 8f). In contrast to the previous two days, vertical air movement within the 700–900 hPa layer on May 31 was dominated by upward motion, with vertical velocities reaching about −0.5 Pa/s (Figure 9).
Hong Kong and the PRD were located outside the outer circulation of Typhoon Mawar throughout the pollution episode. Based on the analysis of transport pathways during the three key periods, we found that Mawar’s peripheral wind field established an efficient transport pathway that channeled pollutant air masses from the PRD urban cluster toward downwind Hong Kong. Concurrently, subsiding airflows within Mawar’s outer circulation hindered the dispersion of ozone and its precursors over Hong Kong, thereby intensifying local accumulation. Thus, these horizontal and vertical transport processes acted as key drivers of this pollution episode.

4.3. Transport Signatures from Lidar

Figure 10 illustrates the vertical profiles of the aerosol extinction coefficient at 16:00 and 20:00 on May 29 and 00:00 on May 30. The aerosol profile exhibited an extinction coefficient peak at 1.37 km with a value of 0.83 km−1 at 16:00 on May 29. By 20:00, the peak had shifted downward to 1.01 km accompanied by a peak value of 0.90 km−1. At 00:00 on May 30, the peak further descended to 0.5 km, with a corresponding value of 0.63 km−1. These evolving profile features indicate that aerosols were clearly transported toward the near-surface layer and coupled with locally generated particular matter, leading to a noticeable rise in near-surface concentrations during the night of May 29 and the daytime of May 30.
Transport pathways identified by backward trajectory clusters at 1500 m (Figure A2b) were from the northeast region to Hong Kong and underwent gradual subsidence. When aerosol descended below approximately 0.6 km, the influence of surface marine air masses likely led to partial dilution of this aerosol (Figure A2a), resulting in a noticeable decrease in extinction coefficients. Aerosol loading has been associated with air masses carrying ozone precursors [59]. In this case, the subsiding polluted air mass retained elevated levels of ozone precursors (NOX and VOCs) overnight within the residual layer. These precursors persisted until sunrise, then rapidly participated in photochemical reactions, triggering the sharp increase in near-surface ozone observed the following day.
The subsiding polluted air mass acted as a reservoir, introducing these precursors into the local environment and facilitating the rapid buildup of near-surface ozone on the following day.

4.4. Vertical Transport Flux Analysis

As illustrated in Figure 11, vertical ozone fluxes over Hong Kong at the 0.3 km, 1.0 km, and 1.5 km levels were calculated by combining Lidar-derived vertical ozone concentration profiles with vertical velocity data from ERA5. On May 29, the ozone flux at 0.3 km was −5.47 μg m−2 s−1, indicating upward vertical transport near the surface. In contrast, positive fluxes were recorded at 1.0 km and 1.5 km, reaching 35.10 μg m−2 s−1 and 15.55 μg m−2 s−1, respectively, signifying strong downward transport at higher altitudes. This vertical flux pattern favored ozone accumulation near the surface during this key period. On May 30, the ozone flux at 0.3 km dropped further to −9.72 μg m−2 s−1, reflecting stronger upward transport compared to the previous day. At 1.0 km, the flux was −14.82 μg m−2 s−1, while a weaker upward flux of −0.65 μg m−2 s−1 was observed at 1.5 km. These results indicate the presence of an upward transport channel at higher altitudes, although the pollutant layer remained vertically confined to around 1.5 km. Thus, the rapid rise in surface ozone concentrations on this day was likely driven by the combined effects of previously transported ozone-rich air masses and ongoing local emissions. On May 31, ozone fluxes at the 0.3 km, 1.0 km, and 1.5 km levels were −10.80 μg m−2 s−1, 28.65 μg m−2 s−1, and −10.29 μg m−2 s−1, respectively. These results indicated a more pronounced upward transport compared that on the previous days, which played an important role in diluting near-surface ozone concentrations. Notably, the flux variation at 1.0 km was the most pronounced, suggesting that this altitude served as a particularly dynamic zone for vertical ozone exchange within the lower troposphere.

4.5. Mechanism of This Episode

In summary, the ozone pollution episode in Hong Kong from May 29 to 31, 2023, was primarily driven by the northward extension of the WPSH, a typical feature of the late-spring–early-summer transition. This condition was further intensified by the peripheral circulation and subsiding airflows of Typhoon Mawar, which jointly enhanced local photochemical production and promoted the regional transport and accumulation of ozone and its precursors. The episode was ultimately terminated by convective rainfall that scavenged pollutants through enhanced vertical mixing and washout.
In the present case, the non-landfalling typhoon played an important role in modulating ozone pollution over Hong Kong. However, this mechanism, whereby remote typhoons amplify regional pollution, is not isolated. Research by Ouyang et al. confirms that the peripheral circulations of landfalling typhoons, such as Typhoon Mitag in 2019 [60], can also enhance the photochemical production and accumulation of ozone over the PRD. Another comparable case was reported by Huang et al., who investigated the non-landfalling Typhoon Danas that occurred in July 2019 [61]. The subsiding airflow on the periphery of the typhoon generated hot, dry, and stagnant conditions over Hong Kong. Concurrently, the synergy between vertical transport and local photochemical reactions ultimately led to a compound high-ozone and high-PM episode.
Although this episode shares features with previously reported cases, remote sensing revealed a unique interplay between regional transport and local photochemistry of ozone in this episode, thereby confirming the critical role of external transport in Hong Kong’s air quality. The associated vigorous vertical mixing and washout processes efficiently removed near-surface pollutants, bringing a definitive end to the ozone episode.

5. Conclusions

By combining atmospheric ozone DIAL observations, surface AQMS measurements, and ECMWF ERA5 data, this study provides a systematic analysis of a representative ozone pollution episode in Hong Kong, clarifying its spatiotemporal dynamics and underlying mechanisms.
The core findings of this study are summarized as follows: First, DIAL observations captured the complete evolution of near-surface ozone, following an accumulation and dispersion sequence, while the aerosol extinction coefficient exhibited a distinct transport and local coupling pattern. Lidar and Tai Po AQMS in situ measurements retrieved vertical structures and spatiotemporal evolution of pollutants. Second, combined with meteorological data from ERA5, the study results indicate that, under the northward shift in the WPSH during late spring and early summer, the peripheral circulation and subsiding airflow of the non-landfalling typhoon established both horizontal and vertical transport pathways for pollutants. Finally, the episode was terminated as enhanced vertical mixing associated with convective development facilitated the efficient dispersion and removal of pollutants. In summary, in this case, the ozone pollution over Hong Kong was jointly driven by local photochemical production, regional transport, and large-scale meteorological modulation.
The scientific significance of this study lies in the integration of multi-source observations and reanalysis data, which not only captured the spatiotemporal evolution of an ozone pollution episode influenced by a non-landfalling typhoon, but also provided a vivid and representative case study for understanding coastal ozone pollution induced by the interaction between the WPSH and typhoon system during late spring and early summer. Additionally, the results deliver critical observational evidence for refining regional air quality forecasting models, thereby supporting the development of coordinated pollution control strategies across the PRD and Hong Kong. Furthermore, this study offers a benchmark case for environmental management under complex meteorological conditions.
This study has certain limitations. As a single detailed case, it provides valuable insights into pollution mechanisms under specific meteorological conditions, but its general applicability remains limited. Future research should incorporate a larger number of non-landfalling typhoon cases to systematically evaluate their impacts on Hong Kong’s air quality and to clarify the associated meteorological–chemical interaction mechanisms.

Author Contributions

Conceptualization, J.W. and L.Z.; data curation, J.W., N.M., W.B.C.T., L.L., and T.Z.; formal analysis, L.Z., Y.X., N.M., J.Q., and B.L.; funding acquisition, J.W.; investigation, J.W., X.S., and W.B.C.T.; methodology, L.Z., Y.X., and J.W.; project administration, J.W.; software, L.Z., Y.X., and J.W.; supervision, J.W.; visualization, L.Z.; writing—original draft, L.Z.; writing—review and editing, L.Z. and J.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Study on Gaseous Air Pollutants Offshore Southeast China (AS 22026C, AHU-HK-202310), Provision of Consultancy Services for Environmental Protection Department of Hong Kong Special Administrative Region (21-07043, AHU-HK-202208-01), National Key Research and Development Program of China (2022YFC3700105, 2022YFC3700100), and Projects of Hefei Comprehensive National Science Center (2024KYHXXM001, 2024KYYQHZ005).

Data Availability Statement

Requests for access to the data should be directed to Jie Wang. In addition to the aforementioned non-public datasets, the ERA5 reanalysis data used in this study are publicly available from the ECMWF Climate Data Store (https://cds.climate.copernicus.eu/, accessed on 28 November 2025).

Acknowledgments

The authors would like to express their special gratitude to the Hong Kong Environmental Protection Department (HKEPD) for providing the air quality monitoring data. They also gratefully acknowledge the NOAA Air Resources Laboratory (ARL) for providing access to the HYSPLIT model. In addition, the authors sincerely thank Ambrose H. T. Chen and Qihua Li for their numerous valuable comments and constructive suggestions throughout this study.

Conflicts of Interest

Author Wilson B. C. Tsui is employed by the company PTC International Limited. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Abbreviations

The following abbreviations are used in this manuscript:
VOCsVolatile Organic Compounds
NOxNitrogen Oxides
COPDChronic Obstructive Pulmonary Disease
MDA8Maximum Daily 8 h Average
OHHydroxyl Radical
HO2Hydroperoxyl Radical
STEStratosphere–Troposphere Exchange
SIStratospheric Intrusion
SITSStratospheric Intrusion to the Surface
PRDPearl River Delta
DIALDifferential Absorption Lidar
UVUltraviolet
UAVUnmanned Aerial Vehicle
AQMSAir Quality Monitoring Station
HKEPDHong Kong Environmental Protection Department
ECMWFEuropean Centre for Medium-Range Weather Forecasts
CMAChina Meteorological Administration
NOAANational Oceanic and Atmospheric Administration
HYSPLITHybrid Single-Particle Lagrangian Integrated Trajectory
WHOWorld Health Organization
WPSHWestern Pacific Subtropical High
BVOCsBiogenic Volatile Organic Compounds

Appendix A

Appendix A.1. Supplementary Information on Vertical Profile Analysis

Remotesensing 17 03904 g0a1
Figure A1. Boundary layer height variations from ERA5 at the Lidar site during the pollution episode. Gray shading marks the three key periods.
Figure A1. Boundary layer height variations from ERA5 at the Lidar site during the pollution episode. Gray shading marks the three key periods.
Remotesensing 17 03904 g0a1
Remotesensing 17 03904 g0a2
Figure A2. Clustered backward trajectories at different heights: (a) 300 m and (b) 1500 m.
Figure A2. Clustered backward trajectories at different heights: (a) 300 m and (b) 1500 m.
Remotesensing 17 03904 g0a2

References

  1. Vazquez Santiago, J.; Hata, H.; Martinez-Noriega, E.J.; Inoue, K. Ozone Trends and Their Sensitivity in Global Megacities under the Warming Climate. Nat. Commun. 2024, 15, 10236. [Google Scholar] [CrossRef]
  2. Atkinson, R. Atmospheric Chemistry of VOCs and NOx. Atmos. Environ. 2000, 34, 2063–2101. [Google Scholar] [CrossRef]
  3. Vongelis, P.; Koulouris, N.G.; Bakakos, P.; Rovina, N. Air Pollution and Effects of Tropospheric Ozone (O3) on Public Health. Int. J. Environ. Res. Public Health 2025, 22, 709. [Google Scholar] [CrossRef]
  4. Donzelli, G.; Suarez-Varela, M.M. Tropospheric Ozone: A Critical Review of the Literature on Emissions, Exposure, and Health Effects. Atmosphere 2024, 15, 779. [Google Scholar] [CrossRef]
  5. Achebak, H.; Garatachea, R.; Pay, M.T.; Jorba, O.; Guevara, M.; Pérez García-Pando, C.; Ballester, J. Geographic Sources of Ozone Air Pollution and Mortality Burden in Europe. Nat. Med. 2024, 30, 1732–1738. [Google Scholar] [CrossRef] [PubMed]
  6. Wang, H.; Lu, X.; Jacob, D.J.; Cooper, O.R.; Chang, K.; Li, K.; Gao, M.; Liu, Y.; Sheng, B.; Wu, K.; et al. Global Tropospheric Ozone Trends, Attributions, and Radiative Impacts in 1995–2017: An Integrated Analysis Using Aircraft (IAGOS) Observations, Ozonesonde, and Multi-Decadal Chemical Model Simulations. Atmos. Chem. Phys. 2022, 22, 13753–13782. [Google Scholar] [CrossRef]
  7. Pei, J.; Liu, P.; Feng, Z.; Chang, M.; Wang, J.; Fang, H.; Wang, L.; Huang, B. Long-Term Trajectory of Ozone Impact on Maize and Soybean Yields in the United States: A 40-Year Spatial-Temporal Analysis. Environ. Pollut. 2024, 344, 123407. [Google Scholar] [CrossRef]
  8. Guarin, J.R.; Jägermeyr, J.; Ainsworth, E.A.; Oliveira, F.A.A.; Asseng, S.; Boote, K.; Elliott, J.; Emberson, L.; Foster, I.; Hoogenboom, G.; et al. Modeling the Effects of Tropospheric Ozone on the Growth and Yield of Global Staple Crops with DSSAT v4.8.0. Geosci. Model Dev. 2024, 17, 2547–2567. [Google Scholar] [CrossRef]
  9. Wang, Y.H.; Gao, W.; Wang, S.; Song, T.; Gong, Z.; Ji, D.; Wang, L.; Liu, Z.; Tang, G.; Huo, Y.; et al. Contrasting Trends of PM2.5 and Surface-Ozone Concentrations in China from 2013 to 2017. Natl. Sci. Rev. 2020, 7, 1331–1339. [Google Scholar] [CrossRef]
  10. Li, K.; Jacob, D.J.; Shen, L.; Lu, X.; De Smedt, I.; Liao, H. Increases in Surface Ozone Pollution in China from 2013 to 2019: Anthropogenic and Meteorological Influences. Atmos. Chem. Phys. 2020, 20, 11423–11433. [Google Scholar] [CrossRef]
  11. He, C.; Wu, Q.; Li, B.; Liu, J.; Gong, X.; Zhang, L. Surface Ozone Pollution in China: Trends, Exposure Risks, and Drivers. Front. Public Health 2023, 11, 1131753. [Google Scholar] [CrossRef] [PubMed]
  12. Warneck, P. On the Role of OH and HO2 Radicals in the Troposphere. Tellus 1974, 26, 39–46. [Google Scholar] [CrossRef]
  13. Levy, H. Photochemistry of the Lower Troposphere. Planet. Space Sci. 1972, 20, 919–935. [Google Scholar] [CrossRef]
  14. Archibald, A.T.; Neu, J.L.; Elshorbany, Y.F.; Cooper, O.R.; Young, P.J.; Akiyoshi, H.; Cox, R.A.; Coyle, M.; Derwent, R.G.; Deushi, M.; et al. Tropospheric Ozone Assessment Report: A Critical Review of Changes in the Tropospheric Ozone Burden and Budget from 1850 to 2100. Elem. Sci. Anthr. 2020, 8, 034. [Google Scholar] [CrossRef]
  15. Yang, X.; Wang, H.; Lu, K.; Ma, X.; Tan, Z.; Long, B.; Chen, X.; Li, C.; Zhai, T.; Li, Y.; et al. Reactive Aldehyde Chemistry Explains the Missing Source of Hydroxyl Radicals. Nat. Commun. 2024, 15, 1648. [Google Scholar] [CrossRef]
  16. Wang, W.; Li, X.; Cheng, Y.; Parrish, D.D.; Ni, R.; Tan, Z.; Liu, Y.; Lu, S.; Wu, Y.; Chen, S.; et al. Ozone Pollution Mitigation Strategy Informed by Long-Term Trends of Atmospheric Oxidation Capacity. Nat. Geosci. 2024, 17, 20–25. [Google Scholar] [CrossRef]
  17. Shao, M.; Ren, X.; Wang, H.; Zeng, L.; Zhang, Y.; Tang, X. Quantitative Relationships between the Production and Loss of OH and HO2 Radicals in Urban Atmosphere. Chin. Sci. Bull. 2004, 49, 1716–1721. [Google Scholar] [CrossRef]
  18. Zhang, Y.; Shao, K.; Tang, X.; Li, J. The Study of Urban Photochemical Smog Pollution in China. Acta Sci. Nat. Univ. Pekin. 1998, 34, 392–400. [Google Scholar]
  19. Zhang, R.; Lei, W.; Tie, X.; Hess, P. Industrial Emissions Cause Extreme Urban Ozone Diurnal Variability. Proc. Natl. Acad. Sci. USA 2004, 101, 6346–6350. [Google Scholar] [CrossRef]
  20. Shi, H.; Song, X.; Zeng, S. Impact of the Urban Heat Island Effect on Ozone Pollution in Chengdu City, China. Chin. Geogr. Sci. 2023, 33, 1017–1032. [Google Scholar] [CrossRef]
  21. Zeng, S.; Shi, H.; Song, X.; Jin, L. Impact of Urban Heat Island Effect on Ozone Pollution in Different Chinese Regions. Urban Clim. 2024, 56, 102037. [Google Scholar] [CrossRef]
  22. Jia, L.; Ge, M.; Xu, Y.; Du, L.; Zhuang, G.; Wang, D. Advances in Atmospheric Ozone Chemistry. Prog. Chem. 2006, 18, 1565–1574. [Google Scholar]
  23. Liu, X.; Zhao, C.; Niu, J.; Su, F.; Yao, D.; Xu, F.; Yan, J.; Shen, X.; Jin, T. Spatiotemporal Patterns and Regional Transport of Ground-Level Ozone in Major Urban Agglomerations in China. Atmosphere 2022, 13, 301. [Google Scholar] [CrossRef]
  24. Hu, X.; Klein, P.; Xue, M.; Zhang, F.; Doughty, D.C.; Forkel, R.; Joseph, E.; Fuentes, J.D. Impact of the Vertical Mixing Induced by Low-Level Jets on Boundary Layer Ozone Concentration. Atmos. Environ. 2013, 70, 123–130. [Google Scholar] [CrossRef]
  25. Zhang, Y.; Li, J.; Yang, W.; Du, H.; Tang, X.; Ye, Q.; Wang, Z.; Sun, Y.; Pan, X.; Zhu, L.; et al. Influences of Stratospheric Intrusions to High Summer Surface Ozone over a Heavily Industrialized Region in Northern China. Environ. Res. Lett. 2022, 17, 094023. [Google Scholar] [CrossRef]
  26. Kemball-Cook, S.; Parrish, D.; Ryerson, T.; Nopmongcol, U.; Johnson, J.; Tai, E.; Yarwood, G. Contributions of Regional Transport and Local Sources to Ozone Exceedances in Houston and Dallas: Comparison of Results from a Photochemical Grid Model to Aircraft and Surface Measurements. J. Geophys. Res. Atmos. 2009, 114, D00F02. [Google Scholar] [CrossRef]
  27. Shen, L.; Liu, J.; Zhao, T.; Xu, X.; Han, H.; Wang, H.; Shu, Z. Atmospheric Transport Drives Regional Interactions of Ozone Pollution in China. Sci. Total Environ. 2022, 830, 154634. [Google Scholar] [CrossRef]
  28. Wang, T.; Lam, K.; Xie, M.; Wang, X.; Carmichael, G.; Li, Y.S. Integrated Studies of a Photochemical Smog Episode in Hong Kong and Regional Transport in the Pearl River Delta of China. Tellus B 2006, 58, 31–40. [Google Scholar] [CrossRef]
  29. Li, J.; Yuan, B.; Yang, S.; Peng, Y.; Chen, W.; Xie, Q.; Wu, Y.; Huang, Z.; Zheng, J.; Wang, X.; et al. Quantifying the Contributions of Meteorology, Emissions, and Transport to Ground-Level Ozone in the Pearl River Delta, China. Sci. Total Environ. 2024, 932, 173011. [Google Scholar] [CrossRef]
  30. Qu, K.; Wang, X.; Cai, X.; Yan, Y.; Jin, X.; Vrekoussis, M.; Kanakidou, M.; Brasseur, G.P.; Shen, J.; Xiao, T.; et al. Rethinking the Role of Transport and Photochemistry in Regional Ozone Pollution: Insights from Ozone Concentration and Mass Budgets. Atmos. Chem. Phys. 2023, 23, 7653–7671. [Google Scholar] [CrossRef]
  31. Gu, Y.; Yan, F.; Xu, J.; Pan, L.; Yin, C.; Gao, W.; Liao, H. Observational Evidence of the Vertical Exchange of Ozone within the Urban Planetary Boundary Layer in Shanghai, China. Atmosphere 2024, 15, 248. [Google Scholar] [CrossRef]
  32. Liu, H.; Han, X.; Tang, G.; Zhang, J.; Xia, X.; Zhang, M.; Meng, L. Model Analysis of Vertical Exchange of Boundary Layer Ozone and Its Impact on Surface Air Quality over the North China Plain. Sci. Total Environ. 2022, 821, 153436. [Google Scholar] [CrossRef]
  33. Chen, Z.; Liu, J.; Qie, X.; Cheng, X.; Yang, M.; Shu, L.; Zang, Z. Stratospheric Influence on Surface Ozone Pollution in China. Nat. Commun. 2024, 15, 4064. [Google Scholar] [CrossRef]
  34. Li, X.B.; Yuan, B.; Parrish, D.D.; Chen, D.; Song, Y.; Yang, S.; Liu, Z.; Shao, M. Long-Term Trend of Ozone in Southern China Reveals Future Mitigation Strategy for Air Pollution. Atmos. Environ. 2022, 269, 118869. [Google Scholar] [CrossRef]
  35. Cao, T.; Wang, H.; Chen, X.; Li, L.; Lu, X.; Lu, K.; Fan, S. Rapid Increase in Spring Ozone in the Pearl River Delta, China during 2013–2022. npj Clim. Atmos. Sci. 2024, 7, 309. [Google Scholar] [CrossRef]
  36. Wang, Y.; Wang, H.; Guo, H.; Lyu, X.; Cheng, H.; Ling, Z.; Blake, D.R. Long-Term O3–Precursor Relationships in Hong Kong: Field Observation and Model Simulation. Atmos. Chem. Phys. 2017, 17, 10919–10935. [Google Scholar] [CrossRef]
  37. Feng, X.; Guo, J.; Wang, Z.; Gu, D.; Ho, K.F.; Chen, Y.; Liao, K.; Cheung, V.T.F.; Leung, K.K.M.; Yu, J.; et al. Investigation of the Multi-Year Trend of Surface Ozone and Ozone–Precursor Relationship in Hong Kong. Atmos. Environ. 2023, 315, 120139. [Google Scholar] [CrossRef]
  38. Chen, Z.; Liu, J.; Cheng, X.; Yang, M.; Wang, H. Positive and Negative Influences of Landfalling Typhoons on Tropospheric Ozone over Southern China. Atmos. Chem. Phys. 2021, 2021, 1–22. [Google Scholar] [CrossRef]
  39. Browell, E.V.; Carter, A.F.; Shipley, S.T.; Allen, R.J.; Butler, C.F.; Mayo, M.N.; Siviter, J.H.; Hall, W.M. NASA Multipurpose Airborne DIAL System and Measurements of Ozone and Aerosol Profiles. Appl. Opt. 1983, 22, 522–534. [Google Scholar] [CrossRef]
  40. Luo, J.; Liu, D.; Huang, Z.; Wang, B.; Bai, J.; Cheng, Z.; Zhang, Y.; Tang, P.; Yang, L.; Su, L. Polarization Properties of Receiving Telescopes in Atmospheric Remote Sensing Polarization Lidars. Appl. Opt. 2017, 56, 6837–6845. [Google Scholar] [CrossRef]
  41. Wang, X.; Zhang, T.; Pei, C.; Chen, D.; Lv, L.; Xiang, Y. Monitoring of Vertical Distribution of Ozone Using Differential Absorption Lidar in Guangzhou. Chin. J. Lasers 2019, 46, 279–287. [Google Scholar] [CrossRef]
  42. Fan, G.; Zhang, B.; Zhang, T.; Fu, Y.; Pei, C.; Lou, S.; Li, X.; Chen, Z.; Liu, W. Accuracy Evaluation of Differential Absorption Lidar for Ozone Detection and Intercomparisons with Other Instruments. Remote Sens. 2024, 16, 2369. [Google Scholar] [CrossRef]
  43. Baylon, P.M.; Jaffe, D.A.; Pierce, R.B.; Gustin, M.S. Interannual Variability in Baseline Ozone and Its Relationship to Surface Ozone in the Western U.S. Environ. Sci. Technol. 2016, 50, 2994–3001. [Google Scholar] [CrossRef]
  44. Environmental Protection Department, The Government of the Hong Kong SAR. Homepage. Available online: https://www.epd.gov.hk/epd/english/top.html (accessed on 6 November 2025).
  45. Environmental Protection Department, The Government of the Hong Kong SAR. Air Quality Monitoring Data Portal. Available online: https://cd.epic.epd.gov.hk/EPICDI/air/?lang=zh_cn (accessed on 6 November 2025).
  46. Copernicus Climate Data Store (CDS). Available online: https://cds.climate.copernicus.eu/ (accessed on 6 November 2025).
  47. Hersbach, H.; Bell, B.; Berrisford, P.; Biavati, G.; Horányi, A.; Muñoz Sabater, J.; Nicolas, J.; Peubey, C.; Radu, R.; Rozum, I.; et al. ERA5 Hourly Data on Pressure Levels from 1940 to Present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). 2023. Available online: https://cds.climate.copernicus.eu/datasets/reanalysis-era5-pressure-levels?tab=overview (accessed on 6 November 2025).
  48. Hersbach, H.; Bell, B.; Berrisford, P.; Biavati, G.; Horányi, A.; Muñoz Sabater, J.; Nicolas, J.; Peubey, C.; Radu, R.; Rozum, I.; et al. ERA5 Hourly Data on Single Levels from 1940 to Present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). 2023. Available online: https://cds.climate.copernicus.eu/datasets/reanalysis-era5-single-levels?tab=overview (accessed on 6 November 2025).
  49. Hersbach, H.; Comyn-Platt, E.; Bell, B.; Berrisford, P.; Biavati, G.; Horányi, A.; Muñoz Sabater, J.; Nicolas, J.; Peubey, C.; Radu, R.; et al. ERA5 Post-Processed Daily Statistics on Pressure Levels from 1940 to Present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). 2023. Available online: https://cds.climate.copernicus.eu/datasets/derived-era5-pressure-levels-daily-statistics?tab=overview (accessed on 6 November 2025).
  50. China Meteorological Administration Shanghai Typhoon Institute. Tropical Cyclone Data Center. Available online: https://tcdata.typhoon.org.cn/en/index.html (accessed on 6 November 2025).
  51. Stein, A.F.; Draxler, R.R.; Rolph, G.D.; Stunder, B.J.B.; Cohen, M.D.; Ngan, F. NOAA’s HYSPLIT Atmospheric Transport and Dispersion Modeling System. Bull. Am. Meteorol. Soc. 2015, 96, 2059–2077. [Google Scholar] [CrossRef]
  52. Jiang, Z.; Li, J.; Lu, X.; Gong, C.; Zhang, L.; Liao, H. Impact of the Western Pacific Subtropical High on Ozone Pollution over Eastern China. Atmos. Chem. Phys. 2021, 21, 2601–2613. [Google Scholar] [CrossRef]
  53. Xue, L.; Wang, T.; Louie, P.K.K.; Luk, C.W.Y.; Blake, D.R.; Xu, Z. Increasing External Effects Negate Local Efforts to Control Ozone Air Pollution: A Case Study of Hong Kong and Implications for Other Chinese Cities. Environ. Sci. Technol. 2014, 48, 10769–10775. [Google Scholar] [CrossRef]
  54. Hong Kong Observatory. Monthly Weather Summary: May 2023. Available online: https://www.hko.gov.hk/en/wxinfo/pastwx/mws2023/mws202305.htm (accessed on 6 November 2025).
  55. Hong Kong Observatory. Monthly Weather Summary: June 2023. Available online: https://www.hko.gov.hk/en/wxinfo/pastwx/mws2023/mws202306.htm (accessed on 6 November 2025).
  56. Zheng, Y.; Jiang, F.; Feng, S.; Shen, Y.; Liu, H.; Guo, H.; Lyu, X.; Jia, M.; Lou, C. Large-Scale Land–Sea Interactions Extend Ozone Pollution Duration in Coastal Cities along Northern China. Environ. Sci. Ecotechnol. 2024, 18, 100322. [Google Scholar] [CrossRef]
  57. Fu, X.; Chen, D.; Wang, X.; Li, Y.; Lang, J.; Zhou, Y.; Guo, X. The Impacts of Ship Emissions on Ozone in Eastern China. Sci. Total Environ. 2023, 903, 166252. [Google Scholar] [CrossRef]
  58. Zheng, S.; Jiang, F.; Feng, S.; Liu, H.; Wang, X.; Tian, X.; Ying, C.; Jia, M.; Shen, Y.; Lyu, X.; et al. Impact of Marine Shipping Emissions on Ozone Pollution during the Warm Seasons in China. J. Geophys. Res. Atmos. 2024, 129, e2024JD040864. [Google Scholar] [CrossRef]
  59. Ding, A.; Fu, C.; Yang, X.; Sun, J.; Petaja, T.; Kerminen, V.M.; Wang, T.; Xie, Y.; Herrmann, E.; Zheng, L.; et al. Intense Atmospheric Pollution Modifies Weather: A Case of Mixed Biomass Burning with Fossil Fuel Combustion Pollution in Eastern China. Atmos. Chem. Phys. 2013, 13, 10545–10554. [Google Scholar] [CrossRef]
  60. Ouyang, S.; Deng, T.; Liu, R.; Chen, J.; He, G.; Leung, J.C.H.; Wang, N.; Liu, S.C. Impact of a Subtropical High and a Typhoon on a Severe Ozone Pollution Episode in the Pearl River Delta, China. Atmos. Chem. Phys. 2022, 22, 10751–10767. [Google Scholar] [CrossRef]
  61. Huang, T.; Yang, Y.; O’Connor, E.J.; Lolli, S.; Haywood, J.; Osborne, M.; Cheng, J.C.H.; Guo, J.; Yim, S.H.L. Influence of a Weak Typhoon on the Vertical Distribution of Air Pollution in Hong Kong: A Perspective from a Doppler LiDAR Network. Environ. Pollut. 2021, 276, 116534. [Google Scholar] [CrossRef] [PubMed]
Remotesensing 17 03904 g001
Figure 1. The layout of observation sites: (a) the location of Hong Kong; (b) the elevation and observation position map of the Hong Kong Special Administrative Region, where “Lidar” stands for the DIAL system position, and “Tai Po” stands for the Tai Po AQMS; and (c) the view of Tai Po AQMS.
Figure 1. The layout of observation sites: (a) the location of Hong Kong; (b) the elevation and observation position map of the Hong Kong Special Administrative Region, where “Lidar” stands for the DIAL system position, and “Tai Po” stands for the Tai Po AQMS; and (c) the view of Tai Po AQMS.
Remotesensing 17 03904 g001
Remotesensing 17 03904 g002
Figure 2. The comparison between ozone concentrations retrieved by the DIAL and Tai Po AQMS from 29–31 May 2023: (a) 300 m altitude and (b) 350 m altitude.
Figure 2. The comparison between ozone concentrations retrieved by the DIAL and Tai Po AQMS from 29–31 May 2023: (a) 300 m altitude and (b) 350 m altitude.
Remotesensing 17 03904 g002
Remotesensing 17 03904 g003
Figure 3. (a) Daily MDA8 ozone concentration in Hong Kong (represented by bubble color) overlaid on the track and intensity of Typhoon Mawa from May 20 to June 3; (b) the diurnal variation in ozone before (May 20–28), during (May 29–31), and after (June 1–3) the pollution episode; (c) the diurnal variation in PM2.5.
Figure 3. (a) Daily MDA8 ozone concentration in Hong Kong (represented by bubble color) overlaid on the track and intensity of Typhoon Mawa from May 20 to June 3; (b) the diurnal variation in ozone before (May 20–28), during (May 29–31), and after (June 1–3) the pollution episode; (c) the diurnal variation in PM2.5.
Remotesensing 17 03904 g003
Remotesensing 17 03904 g004
Figure 4. Hourly concentrations recorded by the Tai Po AQMS from 29–31 May 2023 (the gray and white background bands are used to distinguish different days): (a) PM2.5; (b) PM10; (c) NO2; (d) NOX; (e) O3; and (f) SO2.
Figure 4. Hourly concentrations recorded by the Tai Po AQMS from 29–31 May 2023 (the gray and white background bands are used to distinguish different days): (a) PM2.5; (b) PM10; (c) NO2; (d) NOX; (e) O3; and (f) SO2.
Remotesensing 17 03904 g004
Remotesensing 17 03904 g005
Figure 5. The time–height profiles of Lidar observation results: (a) ozone concentration and (b) aerosol extinction coefficient.
Figure 5. The time–height profiles of Lidar observation results: (a) ozone concentration and (b) aerosol extinction coefficient.
Remotesensing 17 03904 g005
Remotesensing 17 03904 g006
Figure 6. The vertical profiles of ozone concentration and aerosol extinction coefficient during the three key periods with hourly averages. Error bars represent one-time standard deviation (1σ). Panels (a–e) are ozone profiles; panels (b), (d,f) are aerosol profiles. Key periods: I, 16:00–20:00 on May 29; II, 12:00–16:00 on May 30; and III, 11:00–15:00 on May 31.
Figure 6. The vertical profiles of ozone concentration and aerosol extinction coefficient during the three key periods with hourly averages. Error bars represent one-time standard deviation (1σ). Panels (a–e) are ozone profiles; panels (b), (d,f) are aerosol profiles. Key periods: I, 16:00–20:00 on May 29; II, 12:00–16:00 on May 30; and III, 11:00–15:00 on May 31.
Remotesensing 17 03904 g006
Remotesensing 17 03904 g007
Figure 7. Geopotential height (contour lines) and anomaly (color filled) at ERA5 500 hPa on different days: (a) May 20; (b) May 29; and (c) June 3. The magenta box represents the region of Hong Kong, and the red line (5880 gpm) denotes the Western Pacific Subtropical High (WPSH).
Figure 7. Geopotential height (contour lines) and anomaly (color filled) at ERA5 500 hPa on different days: (a) May 20; (b) May 29; and (c) June 3. The magenta box represents the region of Hong Kong, and the red line (5880 gpm) denotes the Western Pacific Subtropical High (WPSH).
Remotesensing 17 03904 g007
Remotesensing 17 03904 g008
Figure 8. The maps of horizontal winds at the surface level (left panels) and the 850 hPa level (right panels) at the daily ozone peak of pollution episode. Panels (a,b) correspond to 18:00 on May 29; panels (c,d) correspond to 14:00 on May 30; panels (e,f) correspond to 13:00 on May 31. The pseudo color indicates air temperature at 1000 hPa, and the contour lines denote the 1000 hPa isobars. The magenta box marks the location of Hong Kong, and the wind vectors in all subplots indicate both wind speed and wind direction.
Figure 8. The maps of horizontal winds at the surface level (left panels) and the 850 hPa level (right panels) at the daily ozone peak of pollution episode. Panels (a,b) correspond to 18:00 on May 29; panels (c,d) correspond to 14:00 on May 30; panels (e,f) correspond to 13:00 on May 31. The pseudo color indicates air temperature at 1000 hPa, and the contour lines denote the 1000 hPa isobars. The magenta box marks the location of Hong Kong, and the wind vectors in all subplots indicate both wind speed and wind direction.
Remotesensing 17 03904 g008
Remotesensing 17 03904 g009
Figure 9. The time–height distribution of vertical velocity and time series of boundary layer height (BLH, black line) from ERA5.
Figure 9. The time–height distribution of vertical velocity and time series of boundary layer height (BLH, black line) from ERA5.
Remotesensing 17 03904 g009
Remotesensing 17 03904 g010
Figure 10. The vertical profiles of aerosol extinction coefficient from 16:00 on May 29 to 0:00 on May 30: (a) 16:00 of May 29; (b) 20:00 of May 29; and (c) 0:00 of May 30. Black arrows indicate the transition of aerosol extinction coefficient peaks.
Figure 10. The vertical profiles of aerosol extinction coefficient from 16:00 on May 29 to 0:00 on May 30: (a) 16:00 of May 29; (b) 20:00 of May 29; and (c) 0:00 of May 30. Black arrows indicate the transition of aerosol extinction coefficient peaks.
Remotesensing 17 03904 g010
Remotesensing 17 03904 g011
Figure 11. Ozone transport fluxes during the key periods, evaluated at heights of 0.3 km, 1.0 km, and 1.5 km. Positive values indicate downward vertical transport (upper to lower), while negative values represent upward transport from the surface (lower to upper).
Figure 11. Ozone transport fluxes during the key periods, evaluated at heights of 0.3 km, 1.0 km, and 1.5 km. Positive values indicate downward vertical transport (upper to lower), while negative values represent upward transport from the surface (lower to upper).
Remotesensing 17 03904 g011
Table 1. The spatial correlation analysis between the Tai Po AQMS and the four peripheral AQMS in the HKEPD network.
Table 1. The spatial correlation analysis between the Tai Po AQMS and the four peripheral AQMS in the HKEPD network.
StationDistance (km) 1O3 R Value 2PM2.5 R Value 2
Southern (22.25°N, 114.16°E)22.580.82 0.75
North (22.50°N, 114.13°E)4.320.94 0.85
Tap Mun (22.47°N, 114.36°E)19.310.88 0.82
Tung Chung (22.29°N, 113.94°E)27.180.81 0.74
1 Distance (km): the straight-line distance from each station to the Tai Po station. 2 R: the correlation coefficient.
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

Zhu, L.; Wang, J.; Xu, Y.; Ma, N.; Song, X.; Qin, J.; Li, B.; Tsui, W.B.C.; Lv, L.; Zhang, T. Vertical Characteristics of an Ozone Pollution Episode in Hong Kong Under the Typhoon Mawar—A Case Study. Remote Sens. 2025, 17, 3904. https://doi.org/10.3390/rs17233904

AMA Style

Zhu L, Wang J, Xu Y, Ma N, Song X, Qin J, Li B, Tsui WBC, Lv L, Zhang T. Vertical Characteristics of an Ozone Pollution Episode in Hong Kong Under the Typhoon Mawar—A Case Study. Remote Sensing. 2025; 17(23):3904. https://doi.org/10.3390/rs17233904

Chicago/Turabian Style

Zhu, Libin, Jie Wang, Yiwei Xu, Na Ma, Xiaoquan Song, Jie Qin, Beibei Li, Wilson B. C. Tsui, Lihui Lv, and Tianshu Zhang. 2025. "Vertical Characteristics of an Ozone Pollution Episode in Hong Kong Under the Typhoon Mawar—A Case Study" Remote Sensing 17, no. 23: 3904. https://doi.org/10.3390/rs17233904

APA Style

Zhu, L., Wang, J., Xu, Y., Ma, N., Song, X., Qin, J., Li, B., Tsui, W. B. C., Lv, L., & Zhang, T. (2025). Vertical Characteristics of an Ozone Pollution Episode in Hong Kong Under the Typhoon Mawar—A Case Study. Remote Sensing, 17(23), 3904. https://doi.org/10.3390/rs17233904

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