Air pollution in Hong Kong

15 Feb 2017

Following is a question by the Hon Kenneth Leung and a written reply by the Secretary for Environment, Mr Wong Kam-sing, in the Legislative Council today (February 15):


On January 8 this year, smog shrouded Hong Kong, with the Air Quality Health Index recorded at some general air quality monitoring stations (AQMSs) reaching 8 or above, which indicated very high health risk and high concentration of key air pollutants such as nitrogen dioxide (NO2) and ozone (O3). It has been reported that some experts pointed out that local pollutants were the main causes for such smog. In this connection, will the Government inform this Council:

(1) of the respective annual average concentration of (i) NO2 and (ii) O3 in the air as recorded at AQMSs (both general and roadside ones) in each of the past five years;

(2) how the respective (i) concentration limit and (ii) numbers of exceedances allowed in respect of NO2 and O3 under Hong Kong’s Air Quality Objectives (AQOs) compare with those under the World Health Organization’s Air Quality Guidelines (AQGs); whether the annual average concentration of NO2 in 2016 met the targets under AQOs and AQGs;

(3) of the respective percentages of the NO2 emitted in Hong Kong in 2016 which originated from various air pollution sources (including motor vehicles, marine vessels, non-roadside machinery and power plants); the respective measures taken by the authorities in the past three years to reduce NO2 emission from such sources, and whether they have assessed if those measures were sufficient to enable Hong Kong to meet the targets under AQOs and AQGs;

(4) given that the annual average concentration of O3 in the air recorded in Hong Kong and the Pearl River Delta (PRD) Region both showed an upward trend in the past few years, whether the authorities have examined the contributions from the air pollution sources in Hong Kong towards the upward trend of O3 in the PRD Region; as high level of O3 in the air promotes the formation of NO2, whether they have studied the quantitative impacts of O3 on the annual average concentration of NO2 in Hong Kong in 2016; and

(5) of the concrete measures undertaken in the past three years and those to be taken in the next three years by the authorities in Hong Kong, Macau and the Guangdong Province to jointly curb the increase of O3 air pollutants in the PRD Region?



The air pollution episode on January 8, 2017 is typical of the influence of regional pollution in the Pearl River Delta (PRD) region on local air quality. In early January 2017, the weather in the PRD region was mainly sunny and calm, which was conducive to the accumulation of ozone (O3) and fine suspended particulates (FSP or PM2.5) in the PRD region. At that time, the prevailing wind in Hong Kong was easterly, which was relatively clean and blocked the polluting air mass in the PRD from affecting Hong Kong. In the morning of January 8, the easterly wind started to be taken over by westerly wind, which brought the polluted air in the PRD to Hong Kong, leading to deterioration in local air quality.

Before the wind turned westerly in the morning of January 8, 2017, the PM2.5 concentrations in Hong Kong were less than 40µg/m3 and the Air Quality Health Index was 4 (i.e. health risk of moderate category) or below. When the wind started turning westerly, the PM2.5 concentration at Tung Chung increased to 64µg/m3 at 10am, followed by a rise in O3 level. By 4pm, the PM2.5 and O3 levels thereat reached its peak on that day at 141µg/m3 and 188µg/m3 respectively. As the westerly wind extended its coverage, the adverse air quality impacts spread to other parts of Hong Kong, except the eastern side. A rise in the PM2.5 and O3 levels were also recorded at other stations like Tsuen Wan station and Central/Western station.

Our replies to the questions are as follows:

(1) The annual average concentrations of nitrogen dioxide (NO2) and O3 recorded at the general and roadside air quality monitoring stations for the past five years are in Table 1. Among the two pollutants, roadside NO2 and O3 in general air are more of a concern. There is a clear decreasing trend with NO2 in the past five years, particularly roadside NO2 which dropped by 31 per cent. Roadside O3 levels however have slightly increased over the same time period because of the reduced vehicular emissions of nitrogen oxides (NOx) which reacts with O3 to form NO2.

(2) The Air Quality Guidelines (AQG) of the World Health Organization (WHO) has recommended that countries in setting their air quality standards should strike a balance between public health and local circumstances, and take into account practical circumstances, such as health risk due to air pollution, latest technological developments, and economic, political and social considerations. Recognising that its AQG levels are very stringent, WHO has also recommended interim targets (ITs) so that countries could progressively improve their air quality with a view to achieving the AQGs in the long run. As far as we know, no country has fully adopted the AQGs as its air quality standard.

The current Air Quality Objectives (AQO) of Hong Kong was drawn up in light of the afore-mentioned recommendations of WHO and the practices of environmentally advanced countries. About half of the Hong Kong’s AQO parameters are already set at WHO AQG levels with the rest pitching at ITs of WHO. Specifically for NO2, our AQO limits are already set at AQG levels; whereas for O3, the AQO limits are set at the IT level of WHO AQG because our O3 level is subject to strong influence of the PRD. The relevant AQO and WHO AQG for NO2 and O3 are in Table 2.

WHO sees it necessary to specify allowable numbers of exceedance for the air pollutant concentration levels and leaves it for countries to do so. When setting the allowable numbers of exceedance of the current AQOs, we made reference to the practices of the European Union (EU). The same number of exceedance as EU, which is 18, is allowed for our hourly limit of NO2 (i.e. 200µg/m3). In the case of O3, EU allows the 8-hour average limit (i.e. 120µg/m3) to be exceeded 25 times while we allow nine times with our limit (i.e. 160µg/m3).

In 2016, three out of the 16 air quality monitoring stations (including 13 general stations and three roadside stations) met the annual average concentration of NO2, which are set at the AQG of WHO. Tseung Kwan O and Tap Mun stations did not provide sufficient data for the comparison because the former started operation only in March 2016 while the latter was closed for renovation works for two months in the year.

(3) It takes time to gather data to compile and validate emission inventories. We are now finalising the inventory of 2015 and have started collecting data for compiling the 2016 emission inventory. The contributions to the local NOx emissions from major sectors based on the 2014 emission inventory are in Table 3.

As seen from Table 3, road transport, electricity generation, navigation and non-road mobile machinery are the main emission sources of NOx, accounting for over 90 per cent of Hong Kong’s total NOx emission. The Government has undertaken a host of measures in past years, focusing on these four main sources. The key measures are as follows:

Road Transport

(a) We launched in March 2014 an incentive-cum-regulatory scheme to phase out some 82 000 pre-Euro IV diesel commercial vehicles by the end of 2019 with $11.4 billion set aside to assist the affected vehicle owners. As at the end of 2016, the scheme retired more than 60 per cent of them;

(b) Starting on September 1, 2014, we have a strengthened vehicle emission control regime for liquefied petroleum gas (LPG) and petrol vehicles, under which roadside remote sensing equipment is deployed to screen out LPG and petrol vehicles with excessive emissions. The vehicle caught will have to pass within 12 working days an advanced emission test to confirm the rectification of the excessive emission problem. Otherwise, its vehicle licence will be cancelled;

(c) We have subsidised franchised bus companies to retrofit their Euro II and III franchised buses with selective catalytic reduction devices (SCRs) to upgrade their emission performance to Euro IV or above levels; and

(d) We have set up from December 31, 2015 franchised bus low emission zones in busy corridors in Causeway Bay, Central and Mong Kok such that the franchised buses in the zones have to be low emission buses (i.e. buses meeting Euro IV or higher emission standards or the SCR-retrofitted Euro II and III buses).

Electricity Generation

To control emissions from power plants, we have been imposing statutory emission caps on their emissions of sulphur dioxide (SO2), NOx and respirable suspended particulates (RSP) through the issuance of Technical Memorandums (TM) since 2008. So far, we have promulgated six TMs with progressively tightened emission caps. The latest one, which is the sixth TM, was promulgated in November 2016 to tighten the emission caps of power plants from 2021 onwards by 72 per cent, 52 per cent and 56 per cent for SO2, NOx and RSP respectively as compared with the first TM.


The control measures for marine vessels mainly focus on the emissions of SO2 and RSP. We have capped the sulphur content of locally supplied marine light diesel at 0.05 per cent since April 2014. Since July 2015, we have also required ocean going vessels to switch to low sulphur marine fuel (with sulphur content not exceeding 0.5 per cent) while at berth.

Non-road Mobile Machinery

A new regulation was introduced in June 2015 requiring newly imported non-road mobile machinery to comply with statutory emission standards.

The above measures have borne fruits. From 2012 to 2016, the roadside RSP, FSP, NO2 and SO2 concentrations have a clear decreasing trend of 28 per cent, 28 per cent, 31 per cent and 30 per cent respectively. Over the same period, the ambient concentrations of RSP, FSP, NO2 and SO2 have dropped by 21 per cent, 21 per cent, 8 per cent and 18 per cent respectively, while the O3 level in general air has a slight decline of 3 per cent.

With the implementation of the committed air quality improvement measures, the Government targets to attain broadly the current AQOs by 2020. Meanwhile, we are reviewing the AQOs with a view to identifying new practicable air quality improvement measures and assessing the scope of tightening the AQOs made possible by their implementation. The review will be completed in 2018.

(4) Under sunlight, NOx reacts with volatile organic compounds (VOC) to form O3, which in turn helps the formation of fine particulates (commonly called photochemical smog). O3 pollution is a regional problem. A recent research result (Note 1) concluded after analysing the local O3 data from 2002 to 2013 that, on average, regional O3 accounted for 70 per cent of the O3 in Hong Kong while the rest was locally produced. An observation of the study is that locally produced O3 had been reduced but the reduction had been more than offset by the increase in regional O3, thereby leading to a rise in O3 levels in Hong Kong. There are some initial signs of a reversal of the past increasing trend in O3 in general during 2012 to 2016 (Table 1), which could be due to our collaboration with the Guangdong Provincial Government in reducing the emissions of NOx and VOC, the precursors leading to the formation of ozone via photochemical reactions. Further observations are needed to affirm the reduction trend while we continue our collaboration with Guangdong in reducing the emissions of NOx and VOC, among other key air pollutants, in the whole PRD region.

O3 reacts with NOx (e.g. from vehicular emissions) to form NO2. Whenever Hong Kong is under the influence of regional O3, the NO2 level here, particularly at the roadside, will also increase. Hence, it is imperative to reduce the O3 level both locally and in the PRD region for tackling the roadside NO2 pollution, even though we have not quantified the impacts of O3 on the annual average NO2 concentrations in Hong Kong.

(5) Collaboration with Guangdong is imperative to tackle the O3 pollution.  Specifically, the emissions of NOx and VOC in the PRD region have to be reduced.  To this end, the Hong Kong Special Administrative Region and Guangdong Provincial Governments agreed in November 2012 to a set of regional emission reduction targets and reduction ranges for NOx and VOC, alongside SO2 and RSP, for 2015 and 2020 respectively. Both sides are now conducting a mid-term review to assess the attainment of the emission reduction targets for 2015 and to finalise the emission reduction targets for 2020. The review will be completed in the first half of this year.

To meet the reduction targets, Hong Kong has been reducing our emissions of NOx by measures outlined in part (3) of this reply. As for VOC, we have been regulating the VOC contents of paints, consumer products, printing inks, adhesives and sealants, which are the major contributor to VOC emissions. The replacement of old vehicles by new ones will also help reduce vehicular VOC emissions, which account for about 18 per cent of the local VOC emissions. In the next three years, we, apart from on-going efforts to reduce VOC emissions, will extend the VOC control to fountain solution and cleansing solution of the printing industry and explore the feasibility of tightening the VOC content of some architectural paints to further reduce our VOC emissions.

As for Guangdong, it has also been targeting the control of major emission sources, including industries, power plants, vessels and vehicles to reduce their NOx and VOC emissions.

While Macau is not a party to the regional emission reduction targets, Guangdong, Hong Kong and Macau have been jointly monitoring the regional air quality; conducting the Joint Regional PM2.5 Study; and fostering exchanges and training on air pollution forecasting technology. To better manage the regional O3 problem, the three Governments are in preparation to include VOC monitoring in the regional air quality monitoring network in phases, starting from 2018. These efforts will provide a scientific base for formulating policies to tackle the regional O3 problem.

Note 1: The research is carried out by Professor Wang Tao of the Hong Kong Polytechnic University and co-researchers.