Overview of thesis
Determining speciated, ground level concentration of particulate matter in large, sparsely populated, arid rural areas using remote sensing
Elevated concentrations of particulate matter (PM) impact human health. This has become an issue of international concern with heightened exposure to numerous sources and prominence in airsheds leading to increased impacts on human health. Understanding the impacts of PM concentrations (PM2.5 and PM10) on health exposure in regional areas, such as Western Australia’s Pilbara region, is limited by the lack of monitoring of ground-level concentrations. Regional areas experience numerous sporadic fires and dust storms whilst in urban areas, combustion from vehicles and boilers dominate. The lack of monitoring and different regional sources types leads to uncertainty regarding health impacts of affected populations in regional areas and undermines exposure management of these areas. Remote sensing provides the spatial range for monitoring over vast regions and removes the uncertainty of plume location since the plume is indicated by its spectral response. The spectral response, or optical density, is a measure of the amount of particulate matter reflecting or absorbing solar energy.
This proposal has identified a number of limitations in the current methods and proposes solutions to improve the methodology.
- Daily snapshots from polar-orbiting satellites are insufficient to capture the temporal resolution from short-term events such as fires and dust storms. The temporal resolution requires geostationary satellites;
- Pixel averaging produces a coarse product of insufficient spatial resolution required for air quality studies;
- Assumptions of aerosol characterisation influence Aerosol Optical Density (AOD) computations and require local verification;
- Air quality studies require ground level PM concentrations (GLC) to determine health impacts, not a dimensionless quantity of total column optical density;
This study aims to determine whether the accuracy of regional ground-level concentrations of PM2.5 and PM10, classified by aerosol type, can be improved using ten-minute spectral data from the Himawari satellite. It is proposed to determine plume location and optical density from spectral data. Source type will be inferred from spectral characteristics, geography, and meteorology. Dispersion modelling will be used to convert aerosol optical density to ground level PM concentrations.
This work will have global impacts for air quality management in determining background PM contributions from natural sources in regional areas. It will resolve modelling input data constraints and wind field errors and will improve the temporal resolution from the current daily (MODIS) and hourly (dispersion modelling) to ten minutes.
- MSc (chemistry) with distinction, University of Pretoria South Africa, 1992-1995
- BSc Hons (Chemistry), University of Natal Pietermaritzburg 1986
- BSc (Chemistry & physics), University of Natal Pietermaritzburg 1982-1986
Awards and Recognition
- CASANZ 2017, most innovative presentation award
- CASANZ, Certified Air Quality Practitioner
- Australia Postgraduate award 2015
- Remote sensing
- Photochemical modelling
- Dispersion modelling
- Emission inventories
Past Research employment history
- 2004-2008: CSIR, Senior Atmospheric Scientist, Durban South Africa. “Photochemistry in South Africa”
- 1987-1997: Atomic Energy Corporation of South Africa, Senior analytical scientist “GC-MS analytical methods”
- 2012-2015: Environ, Senior Atmospheric Scientist, Perth
- 2008-2012: SKM/Jacobs, Senior Atmospheric Scientist, Perth
- 2004-2008: CSIR, Senior Atmospheric Scientist, Durban South Africa
- 1997-2004: EMS, Dispersion model programmer, Pretoria South Africa
- 1987-1997: Atomic Energy Corporation of South Africa, Senior analytical scientist
Recent Publications (within the last five years)
- Sowden, M., Mueller, U., Blake, D. (2018). Review of surface particulate monitoring of dust events using geostationary satellite remote sensing. Atmospheric Environment. 183, 154-164.
Conference Publications/ Presentations
- IAMG 2017, Remote sensing how fast is fast enough?
- CASANZ 2017, Monitoring Regional Aerosol Using Himawari Geostationary Data
- ECU Research week 2017, Determining Ground Level Concentration of Particulate Matter across the Pilbara region using Himawari-8 Remote Sensing Data
Associate Professor Andrea Hinwood (EPA Victoria)
Centre for Ecosystem Management
School of Science
Mobile: 0405 347 726