How artificial intelligence will help give early warnings of dangerous increases in air pollution
Air pollution is now considered to be one of the world's largest environmental health threats.
It's also a serious health concern for vulnerable members of the community, who need to know ahead of time if the air they breathe is becoming less safe.
University of Southern Queensland PhD candidate Ekta Sharma has designed novel artificial intelligence early-warning systems that help solve the problem by accurately predicting future levels of critical pollutants.
Her novel AI early-warning systems can forecast the levels of particulate matter (PM), atmospheric visibility, as well as other pollutants such as greenhouse gases.
"PM is tiny particles that include dust, metals, chemicals, or mould spores," she said.
"We all have seen dirty bathrooms with black mould growing on their surfaces, or cars covered in a yellow blanket of pollen in the spring.
"These are visible sources of PM.
"Smaller PM easily gets into the lungs and bloodstream, triggering asthma or allergic reactions.
"Larger PM may irritate the eyes, nose, and throat, leading to persistent coughing, phlegm, and difficulty breathing."
Dr Sharma expected AI systems such as those designed in her research would be utilised globally to help respond to the effects of climate change and natural disasters such as bushfires. She designed the technology to help vulnerable people such as elderly people or children with respiratory concerns to help manage their exposure to pollutants that are a risk to their health.
Her latest research paper demonstrated the power of these AI predictive systems by deploying them in four air pollution hotspots across Australia – Mount Isa (zinc, coal mines), Hunter Valley (coal mines), Port Pirie (lead smelter) and Latrobe Valley (power stations).
"The research outcomes signify enormous benefits to any nation's atmospheric environment and its impacts on public health, climate modelling, general risk assessments and ensuring our overall well-being," she said.
Dr Sharma will graduate from UniSQ this week with her PhD awarded with Excellence in Doctoral Research. She completed Australian Postgraduate Research Internship with the Department of Defence, which was mentored by her UniSQ's PhD supervisor Professor Ravinesh Deo.
Of only five projects awarded Australia-wide, Dr Sharma was the only Queenslander to have been awarded a coveted fellowship by the Australian Government's Office of National Intelligence.
She was also awarded the UniSQ Vice Chancellors Postdoctoral Fellowship for Women in STEMM discipline within the School of Mathematics, Physics and Computing.
"I will work on developing the next-generation data security technologies to help resolve communication problems in the context of space satellite challenges," she said.
Dr Sharma said the fellowship supported UniSQ's commitment to developing career pathways for women as part of the Science in Australia Gender Equity Athena Swan Action Plan.
