How does BLANkET measure air quality?
The BLANkET data are measured by an optical scattering method. In essence, air is drawn into a cavity illuminated by a laser beam. The more particles in the air, the more the laser light is scattered in the cavity. A similar effect is familiar to us when driving at night in fog - the fog scatters the car headlights back towards the car. The thicker the fog, the more light is returned towards the car.
The instrument we are using for BLANkET is called a DRX DustTrak.
This optical method does not directly measure the mass concentration (which is the mass in a given volume) of the particles, but instead it measures a quantity related to the number of particles in a given volume of air. We have a calibration to convert from the number of particles to the total mass, based on assumed (usual) particle properties. However, because we are not measuring mass directly, the data remain indicative of the likely mass concentrations at the time so are not reference data. We have found that, with care, they give reliable results.
How have the BLANkET sites been selected?
The initial 15 sites were chosen to try to balance between having as broad a cover of as much of the state as possible, but not so broad as to spread the stations too thinly. Subsequently, station locations have been chosen due to several factors, including population density, strategic location, and likely winter-time smoke levels
The BLANkET network complements the major air quality stations in Hobart, Launceston and Devonport.
What do you mean by indicative air quality data?
The BLANkET data are indicative data, not reference data.
The BLANkET indicative data cannot be used to determine if an air quality standard has been breached. The data will however provide a good indication of air quality at any given time.
Reference air quality data, collected in accordance with National and International standards, are obtained at the Hobart, Launceston and Devonport stations. Reference data are used to assess whether exceedances of air quality standards have occurred.
Why don't you use a reference method for BLANkET?
The advantages of the scattering instrument are that it is of relatively low cost, small in size, has a low power consumption, and operates almost entirely without needing human intervention. This means we can deploy many more such instruments than if we used reference air samplers. Additionally we obtain a real-time measure of air quality, rather than needing to pre-weigh, deploy, recover, and post-weigh a filter.
How good are these indicative data?
We have been comparing and cross-checking the performance of the BLANkET instrument against our reference instruments and will continue to do so. The graph below shows one such comparison, obtained at the New Town (Hobart) air station on 16 April 2009. The solid line shows data from the TEOM (which can be operated as a reference instrument), while the dotted line with the small points shows data from the BLANkET dustTrak instrument. The indicative data shows good agreement with the TEOM measurements. A long series of tests, extending over several weeks, were performed prior to selecting the DRX dustTrak for use in BLANkET. The agreement between the TEOM and BLANkET instrument shown in the figure is representative of the test findings.
A comparison plot between TEOM PM10 and DRX dustTrak PM10 data for 16 April 2009 at Hobart
The figure below shows a comparison of day-averaged DRX data and reference data from low-volume air samplers at New Town station for April to September 2010. The top panel shows PM10, the lower panel shows PM2.5. The blue triangles show the reference air sampler data, the red squares show the day-averaged DRX data. The agreement is generally very good, particularly for PM2.5.
A comparison of reference low-volume air sampler (blue triangles) and day-averaged BLANkET (red squares) air quality data, for New Town station, winter 2010.
Can you tell the difference between smoke and other particles?
Smoke particles are small, and will be included in both the PM10 and PM2.5 measure. If both PM10 and PM2.5 are high, with PM10 only slightly larger than PM2.5, the instrument is probably measuring smoke. If PM10 is high and PM2.5 is low it means there are few fine particles in the current air sample, and the instrument may be measuring dust or sea salt aerosols.
To check for dust it is worth looking at wind speeds, as high winds can raise dust and keep it airborne. Note that it is possible for dusty air to be carried long distances by strong upper level winds even when surface winds are low, though this is less common in Tasmania than on the mainland of Australia or elsewhere.
The graph below compares the signatures from dust (high PM10, relatively low PM2.5) and smoke (high PM2.5, with PM10 slightly higher than PM2.5) as seen at Scottsdale in September 2009. The peak of the dust event occurred late on the 12th of September. Around midday on the 14th of September some local (agricultural) burning took place near the BLANkET station.
BLANkET - comparing dust and smoke
On the reports page you can find a description of this dust-deposition event over Tasmania that occurred on the 12th of September 2009 - see
BLANkET Technical Report 5.
Naturally occurring aerosols composed of sea-salt crystals can also give rise to high PM10 with very little increase in the PM2.5 level. These aerosols are liberated into the atmosphere during stormy weather at sea, and can be transported over land by wind systems. On the reports page you can find a description of an event lasting several days in May 2009 when moderate north-easterly winds brought aerosol laden air over Tasmania - see
BLANkET Technical Report 4. Sea-salt aerosols can contribute to atmospheric haze. The BLANkET instruments can distinguish between smoke and sea-salt aerosols from a comparison of the PM10 and PM2.5 levels.
Can you tell the differences between smoke from planned burns, bushfires, and domestic wood heaters?
There are likely to be small differences in particle type and chemical composition in smoke from a planned burn or bushfire compared to a domestic heater. The BLANkET instruments will not be able to measure this.
However there are other ways to address this question. When a relatively small amount of smoke is produced locally (such as from a nearby wood heater, or from a number of wood heaters), and is detected at a nearby station, the smoke is not well mixed with the air, but shows a lot of variation in the measured concentration. Conversely, if a large plume of smoke has travelled a number of kilometres from the source before reaching the station it has had time to become well mixed. The graphs below compare the two cases. The top panel shows two days of PM2.5 data from Lilydale BLANkET station in July 2009. There is significant variation between each measurement taken 10 minutes apart. In contrast the lower panel shows two days of data from Lilydale station in November 2009 when a smoke cloud was seen to move from west to east across the north-east Tasmanian BLANkET stations. In this case the graph is much smoother. Satellite imagery can also be used to provide additional information.
A plot comparing signatures of smoke from local and remote sources, from PM2.5 data collected at the Lilydale BLANkET station in July (top) and November (lower) 2009.
On the reports page you can find an analysis of the November 2009 smoke event which was detected at the air stations in the Tamar Valley as well as across the north-east of the State - see
BLANkET Technical Report 6. The source appears to have been a 20 ha bushfire near Barrington in central northern Tasmania.
There are now a number of other reports on the
BLANkET report page on smoke impacts from planned burns.