This chapter describes a range of methods for monitoring ambient air quality and outlines the level of precision required for different monitoring purposes. Required methods for ambient air quality standards and recommended methods for ambient air quality guidelines are discussed. The chapter also includes a brief section on meteorological monitoring.
4.1 Using appropriate monitoring methods
A wide variety of methods are available for measuring contaminants in ambient air, with an equally wide variation in cost and precision. Specific monitoring methods should be chosen taking into consideration the purpose, objectives and budget of the monitoring programme.
Monitoring for the purposes of national environmental standards can only be carried out with the high-precision instrumental methods in accordance with Schedule 2 of the regulations. High-precision instrumental methods would also be generally used for research studies or other specific investigations, where there is a need to understand the ways in which contaminant levels fluctuate over short time periods (hours or days).
Ambient air quality guidelines carry recommended monitoring methods that should be used for the assessment of the contaminants covered by the guidelines. These are based on Australian / New Zealand and USEPA standards.
There is still a place for monitoring methods that fall outside the specifications for standards and guideline assessments. Methods that involve lower resolution instruments can be used for an initial screening survey, where a detailed study is not warranted. These may be used as a prelude to more detailed monitoring. If contaminant levels are found to be low, then the same method(s) could be used for repeat surveys over subsequent years. Low resolution methods are also useful for prioritising a number of different areas that have already been selected for detailed monitoring programmes.
It should be stressed that screening methods cannot be used to determine compliance with national standards or guidelines. A screening method is any non-standard method that is used on an exploratory basis and provides indicative data for a particular contaminant. Despite this, the method still needs to have a level of accuracy and precision suitable for the purpose of monitoring (eg, Occupational Health and Safety equipment is not suitable for ambient air quality monitoring).
Screening methods are purely indicative methods and any results must be treated with caution. For example, a one-day-in-three monitoring regime of PM10 may indicate that there have been no breaches of the ambient air standard. However, this does not necessarily mean that exceedences were not occurring during the two days that were not sampled. A number of exceedences detected by this method would indicate detailed monitoring for national environmental standards is warranted.
Recommendation 1: Screening methods
Screening methods cannot be used to determine compliance with the NES for air quality or to establish whether an airshed meets the ambient air quality guidelines. They may, however, be used to provide indicative data for other monitoring purposes.
A screening method’s level of accuracy and precision must be suitable for the purpose of monitoring (eg, occupational health and safety equipment is not suitable for ambient air quality monitoring).
4.2 Using existing monitoring for national environmental standards
Regional monitoring programmes conducted since the introduction of the RMA should, at least in the case of PM10, provide a good indication of where national environmental standards are likely to be breached. However care must be taken when considering whether such ambient air quality monitoring can be used for the regulatory requirements of national environmental standards. The following points should be considered:
Is the data capture sufficient and is there an appropriate time average?
Are monitoring methods appropriate for standards?
Are the appropriate contaminants being monitored?
Is it at an appropriate location?
Earlier PM10 monitoring in New Zealand traditionally adopted a one-day-in-three or one-day-in-six approach. This approach is not suitable for national environmental standards because continuous monitoring of 24-hour averages from midnight to midnight is required. It may be found that some analysers that require manual filter changes are not practical for midnight changeovers, unless these analysers can have automated samplers retrofitted. Regional councils have shifted to continuous monitors since the NES for air quality came into effect.
It is not only necessary to choose analysers that are compliant with the relevant Australian / New Zealand standards or USEPA standard; it is also necessary to operate the analyser in accordance with that standard. Simply ensuring an analyser’s specifications are compliant with the standard is not sufficient.
A large proportion of PM10 monitoring in New Zealand is conducted at residential neighbourhood sites (see section 6.2 on site classifications). While these may not exactly align with the ‘worst location’ requirement in the NES for air quality, careful consideration should be made concerning relocation of the site. There may be little to gain by relocating a current site within the same general vicinity of an airshed if it will end a useful record of long-term data.
4.3 Choosing appropriate monitoring equipment
Choosing the appropriate monitoring equipment is important for achieving the aims of the monitoring programme. Equipment that does not measure the contaminant in the required measurement range clearly will not provide useful data.
The following factors should be considered before purchasing monitoring equipment.
Purpose of monitoring: ie, screening, compliance monitoring or research. Different sensitivities may apply to different types of monitors. For example, compliance monitoring may require a higher level of sensitivity and resolution than a monitor used for screening purposes.
Duration of deployment: if the instrument is to be used for short-term screening surveys then portability, size, weight and robustness may be factors to consider.
Detection limit, precision and measurement range: will the instrument be able to measure within the required range, based on the monitoring objective?
Consumables: how frequently do parts need to be changed or replaced?
Ability to conform to relevant monitoring standards: does the instrument conform to monitoring standards such as those required in the NES for air quality?
Frequency of calibration: the time taken to complete automatic zero and span checks, and operational and multipoint calibrations. Instruments that spend significant time on an automatic zero and span checks will reduce your percentage data recovery. An instrument that requires frequent calibration in order to comply with a particular standard will impose costs on an organisation (time and resources).
Ease of use: some instruments can be extremely difficult to operate, which imposes costs in terms of training and person-hours spent resolving problems.
Communication: ideally instruments should have ethernet/IP ports, analogue and RS232 connectivity to allow connection to dataloggers and telemetry.
Ability to interface remotely: most modern instruments will allow an operator to remotely operate an instrument.
Environmental requirements: does the instrument require air conditioning, temperature or humidity control?
Cost: consider both the cost of the instrument and the cost of consumables.
Reliability: find out who else has operated similar instruments and discuss the pros and cons with other organisations before committing to a purchase.
Memory or on-board storage capacity: determine if memory capacity is sufficient for your current and future needs (eg, memory expansion features).
Instrument parameters that may be recorded in addition to output concentration: can the instrument record any other parameters (eg, temperature)?
Do your homework before making a purchase. For organisations that make decisions primarily based on price, consider imposing instrument specifications and/or standards as conditions of purchase.
4.4 Methodology types
Monitoring methodologies can be divided into three categories according to cost and the level of accuracy and precision.
Continuous monitoring methods
These are high-resolution methods that provide continuous records of contaminant levels. They can operate over extended periods (weeks or months) with minimal operator intervention. Remote communication is possible by telemetry. They have a high degree of measurement precision, and have detection levels around one order of magnitude or more below typical background levels. As might be expected, these are the most expensive monitoring methods. A high standard of maintenance, calibration, and operational and quality control procedures are required for good data quality.
Gravimetric particulate methods
In the past, gravimetric particulate methods have formed the mainstay of particulate monitoring in New Zealand. The implementation of the NES for air quality, however, has shifted the use of these methods to the analysis of airborne lead, co-location studies and screening surveys.
Monitoring starts when a known volume of air is pumped through a pre-weighed filter for a known length of time (typically 24 hours). The filter is reweighed after exposure and a concentration determined. Most systems used in New Zealand require manual changes of the sampling filters between each sample, although a number of semi-automated systems are also available. This can be done on consecutive days. Manually changing filters at midnight, however, is operationally impractical when compliance monitoring (the regulations by definition require filter changes to take place at midnight; see Appendix A).
- Passive monitoring methods (diffusion tubes and badges)
Diffusion tubes work when a contaminant is diffused into a tube containing either an adsorbent or reactive material. Analysis of the tubes following a known exposure time (typically two to four weeks) will provide a time-averaged contaminant concentration. Badges work in a similar way, the difference being the sampler configuration. Badges typically have higher uptake rates and are used more widely in New Zealand.
Because these methods are simple and cheap, they can provide a good picture of spatial variation over a large area. They are particularly useful in screening surveys and during the initial stages of an air quality monitoring programme.
Though a cheap screening tool, there are a number of limitations to this method, such as lower accuracy and no indication of peak levels. Quality control and assurance during laboratory analysis must be of the highest standard to attain consistent results. The results from passive samplers can be used in conjunction with high-resolution instruments to determine spatial variation across an airshed over a relevant averaging period. This method can also be useful for comparison with annual guidelines.
Historically, wet chemical methods were used to monitor levels of gaseous contaminants. These methods are no longer recommended due to interferences and the low resolution of data. High-resolution instrumental methods are now recommended.
Low-level instrumental methods using optical sensors are commonly used for monitoring occupational exposure. While such analysers may occasionally be useful for incident investigations, they should not be used for routine compliance monitoring due to their low sensitivity.
4.5 Mandatory ambient air monitoring methods
The following are mandatory monitoring methods under the NES for air quality. Note that these Australian / New Zealand and USEPA standards apply to monitoring methods. It is not sufficient to simply use an analyser that conforms to a standard: it is also necessary to operate the analyser in accordance with the operational requirements of that standard.
4.5.1 Carbon monoxide (CO)
CO monitoring instruments are predominantly gas filter correlation infrared (GFC-IR) absorption analysers. This is the recommended CO monitoring method, although AS 3580.7.1 also allows for a non-dispersive infrared gas chromatograph with flame ionisation detector, or electrochemical sensor systems. These alternative methods, however, suffer from a variety of interfering species and are considered less robust than GFC-IR analysers.
In a GFC-IR analyser, ambient air is continuously sampled using a pump unit and the CO concentration in the sample air is measured by the absorption of infrared radiation at 4.5 to 4.9 nanometers (nm) wavelength. A reference detection system is used to alternately measure absorption due to CO in the ambient air stream and absorption by interfering species. An infrared detector and amplification system produce output voltages proportional to the CO concentration. The concentration is derived from the Beer–Lambert relation:
I1 = I0 e-alc
where the sample passes through a cell tube of length ‘l’. The analyser alternately measures the absorption I0 of the air path with no CO present and the absorption I1 of the ambient sample, with ‘a’ being the absorption coefficient, to provide the CO concentration, ‘c’.
GFC instruments use a filter wheel to allow alternate measurement of total IR absorption. The analyser continually displays current CO concentrations, and, depending on the make and model of analyser, other parameters can be selected as necessary.
NES for air quality: Mandatory method for CO
Australian Standard AS 3580.7.1–1992, Methods for sampling and analysis of ambient air – Determination of carbon monoxide – Direct-reading instrumental method.
4.5.2 Nitrogen dioxide (NO2)
Nitric oxide (NO) in the sample air stream reacts with ozone (O3) in an evacuated chamber to produce activated NO2:
NO + O3 → NO2* + O2 → NO2 + O2 + hν
The intensity of the chemiluminescent radiation (hv) produced is measured using a photomultiplier tube (PMT) or photodiode detector. The detector output voltage is proportional to the NO concentration. The ambient air sample is divided into two streams; in one, ambient NO2 is reduced to NO using a molybdenum catalyst before reaction. The molybdenum converter should be at least 95 per cent efficient at converting NO2 to NO. This gas stream gives total NOx. The second stream measures NO directly by not passing through the molybdenum converter.
Separate measurements are made of total oxides of nitrogen NOx (= NO + NO2) and NO. The ambient NO2 concentration is calculated from the difference (NO2 = NOx – NO). This is an important point to remember, because the contaminant of interest (NO2) is actually measured by inference rather than directly, and the efficiency of the molybdenum converter should be checked on a regular basis.
In a chemiluminescent analyser, ambient air is drawn through the system via a pump and permapure drier unit. NOx analysers are equipped with either a single or a double reaction chamber and PMT system. A solenoid valve is used to alternately switch between NO and NOx measurements, typically at 15-second intervals. The analyser continuously displays current NO, NO2 and NOx concentrations, and, depending on the make and model of analyser, other parameters can be selected as necessary.
NES for air quality: Mandatory method for NO2
Australian Standard AS 3580.5.1–1993, Methods for sampling and analysis of ambient air – Determination of oxides of nitrogen – Chemiluminescence method.
4.5.3 Ozone (O3)
In an O3 analyser, ambient air is continuously sampled using a pump unit. O3 concentrations are calculated from the absorption of ultraviolet (UV) light at 254 nanometres (nm) wavelength. The absorption is measured using a UV detector. An O3-removing scrubber is used to provide a zero reference intensity. The concentration is calculated using the Beer–Lambert equation:
I1 = I0 e -alc
where the sample passes through a cell tube of length ‘l’, and the analyser alternately measures the absorption I0 of the air with no O3 present and the absorption I1 of the ambient sample, with ‘a’ being the absorption coefficient (at 254 nm), to provide the O3 concentration, ‘c’.
The analyser continually displays current O3 concentrations, and, depending on the make and model of analyser, other parameters can be selected as necessary.
NES for air quality: Mandatory method for O3
Australian Standard AS 3580.6.1–1990, Methods for sampling and analysis of ambient air – Determination of ozone – Direct-reading instrumental method.
The USEPA standard is described as a reference method (eg, gravimetric); 40 CFR Part 53 (www.epa.gov/ttn/amtic/criteria.html) contains a full list of equivalent methods. The most commonly used methods for the measurement of PM10 in New Zealand are discussed further in chapter 5.
NES for air quality: Mandatory methods for PM10
Australian/New Zealand Standard AS/NZS 3580.9.6:2003, Methods for sampling and analysis of ambient air – Determination of suspended particulate matter – PM10high-volume sampler with size selective inlet – Gravimetric method.
United States Code of Federal Regulations, Title 40 – Protection of Environment, Volume 2, Part 50, Appendix J – Reference method for the determination of particulate matter as PM10in the atmosphere.
Note: The following Australian / New Zealand standards were released after 2000 for the continuous monitoring of PM10.
Continuous sampling in accordance with AS 3580.9.8-2008, Methods for sampling and analysis of ambient air – Determination of suspended particulate matter – PM10continuous direct mass method using a tapered element oscillating microbalance analyser.
Australian / New Zealand Standard AS/NZS 3580.9.11:2008, Methods for sampling and analysis of ambient air – Determination of suspended particulate matter – PM10beta attenuation monitors.
4.5.5 Sulphur dioxide (SO2)
SO2 monitoring instruments are predominantly molecular UV fluorescence analysers. This is the recommended SO2 monitoring method. AS 3580.4.1 also allows flame photometric detector and electrochemical sensor systems.
UV fluorescence systems operate on the principle that an ambient air sample stream exposed to UV light excites SO2 molecules in the sample to higher, but unstable, excited states. These excited states decay, giving rise to the emission of secondary (fluorescent) radiation:
SO2 + hν → SO2* → SO2 + hν (fluorescence)
The fluorescent radiation is detected by a PMT, causing an output voltage proportional to the SO2 concentration. A permeable membrane ‘kicker’ is used to remove interfering hydrocarbons (aromatic hydrocarbons also fluoresce) before reaction. Ambient air is drawn through the system via a pump unit, and the analyser continuously displays current SO2 concentrations. Depending on the make and model of analyser, other parameters can be selected as necessary.
NES for air quality: Mandatory methods for SO2
Australian standard AS 3580.4.1–2008, Methods of sampling and analysis of ambient air – Determination of sulphur dioxide – Direct reading instrumental method.
The above standard supersedes Australian standard AS 3580.4.1-1990, Methods for sampling and analysis of ambient air – Determination of sulphur dioxide – Direct-reading instrumental method.
4.6 Recommended ambient air monitoring methods
The following methods are recommended for the measurement of contaminants for comparison with the New Zealand ambient air quality guidelines.
4.6.1 Hydrogen sulphide (H2S)
The recommended method is based on the Australian standard method for sulphur dioxide, with the addition of a catalyst to convert H2S to SO2.
Recommendation 2: Hydrogen sulphide
The recommended method for hydrogen sulphide is fluorescence monitoring, in accordance with AS3580.4.1–2008, Methods of sampling and analysis of ambient air – Determination of sulphur dioxide – Direct reading instrumental method.
4.6.2 Lead content of PM10
Recommendation 3: Lead content of PM10
The recommended method for lead content of PM10 is high-volume gravimetric sampling in accordance with United States Code of Federal Regulations, Title 40 – Protection of Environment, Volume 2, Part 50, Appendix J and Appendix G.
4.6.3 Benzene and 1,3-butadiene
Recommendation 4: Benzene and 1,3-butadiene
The recommended methods for benzene and 1,3-butadiene are:
USEPA method TO-1 – Method for the determination of VOCs in ambient air using Tenax®adsorption and gas chromatography / mass spectrometry (GC/MS)
USEPA method TO-14A – Determination of VOCs in air using specially prepared canisters with subsequent analysis by gas chromatography
USEPA method TO-15 – Determination of VOCs in air collected in specially prepared canisters and analysed by gas chromatography / mass spectrometry (GC/MS)
USEPA method TO-17 – Determination of VOCs in air using active sampling onto sorbent tubes
BS EN 14662-1:2005 – Ambient air quality – Standard method for measurement of benzene concentrations – Pumped sampling followed by thermal desorption and gas chromatography
BS EN 14662-2:2005 – Ambient air quality – Standard method for measurement of benzene concentrations – Pumped sampling followed by solvent desorption and gas chromatography
BS EN 14662-3:2005 – Ambient air quality – Standard method for measurement of benzene concentrations – Automated pumped sampling with in situ gas chromatography
BS EN 14662-4:2005 – Ambient air quality – Standard method for measurement of benzene concentrations – Diffusive sampling followed by thermal desorption and gas chromatography
BS EN 14662-5:2005 – Ambient air quality – Standard method for measurement of benzene concentrations – Diffusive sampling followed by solvent desorption and gas chromatography.
4.6.4 Formaldehyde and acetaldehyde
Recommendation 5: Formaldehyde and acetaldehyde
The recommended method for formaldehyde and acetaldehyde is USEPA method TO‑11A – Determination of formaldehyde in ambient air using adsorbent cartridge followed by high performance liquid chromatography (HPLP).
4.6.5 Benzo(a)pyrene (BaP)
Recommendation 6: Benzo(a)pyrene
The recommended methods for BaP are:
USEPA method TO-13A – Determination of polycyclic aromatic hydrocarbons (PAHs) in ambient air using gas chromatography / mass spectrometry (GC/MS)
BS EN 15549:2008 – Air quality – Standard method for the measurement of the concentration of benzo(a)pyrene in ambient air.
4.6.6 Mercury, chromium and arsenic
Recommendation 7: Mercury, chromium and arsenic
The recommended method for mercury, chromium and arsenic is:
PM10 sampling in accordance with 40CFR Part 50, Appendix J, followed by analysis using atomic absorption spectroscopy or an equivalent method.
Method IO-5 (Sampling and analysis for vapour and particle phase mercury in ambient air utilising cold vapour atomic fluorescence spectrophotometry)
BS EN 15852 – Ambient air quality – Standard method for the determination of total gaseous mercury.
The monitoring methods for the above air contaminants are based on procedures recommended by Standards Australia, the USEPA and the British Standards Institution (BSI). Detailed specifications for these methods can be obtained from the following websites:
Standards Australia publications site (http://www.standards.org.au)
USEPA site (http://www.epa.gov/ttn/amtic)
BSI site (http://www.bsi-global.com).
More information on recommended monitoring methods for hazardous air contaminants can also be found in the reports prepared for the review of the AAQG. These reports are available from the Ministry’s website (http://www.mfe.govt.nz/publications/air/11-hazardous-air-oct00.pdf).
4.7 Open-path monitoring systems
Open-path monitoring systems measure a range of contaminants based on absorption of a light beam transmitted over distances of up to several kilometres. As such, they are totally different from most other monitoring methods in common use. The main difference is that the open-path system records the average concentration simultaneously for a number of contaminants over the full measured distance rather than at a specific point. The measured results will therefore be lower than those at some points along the path and higher than at others. This method is particularly suitable for measuring along site boundaries of industrial processes, but is not often used for measuring ambient air quality at discrete points. In New Zealand, these methods are usually only used for research purposes.
One of the main attractions of the open-path systems is that they can be used for a wide variety of different contaminants, including most of the volatile organics. The main disadvantage is their cost, which is typically three to five times the cost of any of the more traditional instruments. Furthermore the concentration of a particular contaminant is averaged over the beam length, which can underestimate the ground-level concentration where there are one or more point sources of contamination present.
4.8 Meteorological monitoring
As mentioned in section 3.2, it is important to monitor meteorological conditions at the air quality monitoring site since weather is a significant factor which influences air contaminant concentrations. Measurements of wind speed, wind direction and air temperature are the minimum meteorological parameters to be monitored. Additional measurements that would provide an improved picture of weather conditions during monitoring are: relative humidity, solar radiation, rainfall, and a temperature profile at two heights.
Wind direction, by convention, is the direction the wind is blowing from and is quoted with reference to true north (not magnetic north). An exception to this is meteorological data collected for oceanographic monitoring purposes. In this case, the wind is recorded in the direction it is blowing towards. Care should be taken to determine the meteorological wind convention when using data collected from the marine environment.
Wind speeds are often quoted in different units. The preferred reporting unit is metres per second (m/s). A wind speed conversion table is show in Appendix B.
Recommendation 8: Meteorological monitoring
The minimum monitoring required is as follows:
mast, 6 m minimum, 10 m preferable
wind speed (resolution 0.1 m/s, accuracy ± 0.2 m/s, start-up 0.2 m/s)
wind direction (resolution 1°, accuracy ± 2°, referenced to true north)
air temperature (resolution 0.1°C, accuracy 0.2°C)
automated logging system, reliable power, with battery back-up.
The use of the Cartesian coordinate system is recommended, whereby data is converted to its x and y components. This data can then be accumulated in a vector form. This solves averaging and unweighted direction problems. Results may subsequently be converted to polar coordinates, if required.
Desirable measurements are:
humidity (or dew point) (resolution 1% relative humidity (rh), accuracy ± 5% rh)
solar radiation (for stability estimates) (resolution 1 W/m2, accuracy 10 W/m2)
rainfall (resolution 1 mm)
temperature profile (T at two heights, 1.5 m and 10 m, needs 0.1°C accuracy) using identical sensors at both heights.
Specific siting requirements:
Must be free of influence of trees, buildings, structures – should be at least two times the height away from the obstacle, and for wind sensors it should be at least 10 times the height away from obstacles (refer to Part I, sections 5.9.2 and 6.2 of the Guide to Meteorological Instruments and Methods of Observation (World Meteorological Organization, 1996; Oke TR, 2006).
Required time resolution:
data should be collected at the same minimum time resolution as air quality data
resolution should be at least hourly.
Period of monitoring:
For atmospheric modelling and trend analysis, a minimum of one year’s data is recommended.
4.9 Monitoring agencies and training
4.9.1 Monitoring agencies
Air monitoring services are provided by a number of environmental consultancies, Crown research institutes and universities. A number of regional councils also carry out monitoring in their own regions, as do many industries. Other organisations have taken the option of purchasing their own equipment but contracting consultants to run it. Equipment can also be leased from a number of suppliers. Names of suppliers and consultants can often be found in journals and magazines or by talking to other councils.
It is strongly recommended that agencies/firms undertaking ambient air quality monitoring for assessing compliance with the NES be accredited by an independent and approved accreditation organisation such as International Accreditation New Zealand (IANZ) or the Joint Accreditation System of Australia and New Zealand (JAS-ANZ). As a minimum, accredited staff should be used to audit and verify the data collection, validation and quality assurance processes at least every two years. For more information on the accreditation process, visit www.ianz.govt.nz or www.jasanz.com.au
Local training courses in ambient air quality monitoring are periodically offered in New Zealand by the Clean Air Society of Australia and New Zealand (CASANZ). More frequent training is held in Australia. Information on upcoming courses can be obtained from CASANZ (www.casanz.org.au). If equipment is being purchased, the manufacturer should provide initial training in its use. Conferences also provide a useful opportunity to view and discuss monitoring equipment with suppliers.
The National Air Quality Working Group (NAQWG) convenes biannually to discuss air quality issues from a regional council perspective. Presentations are given on various initiatives being undertaken by regional councils in air quality management and research.
Recommendation 9: Independent accreditation
Accreditation of agencies/firms undertaking air quality monitoring by an independent and approved accreditation organisation is strongly recommended.