Health effects of bushfire smoke in the Australian monsoon tropics

Abstract

Introduction
There is increasing scientific consensus about the adverse health effects of outdoor particulate air pollution. Most evidence has come from studies set in large urban centres where airborne particulate matter is primarily derived from fossil fuel combustion by industrial plants and motor vehicles. The impact of ambient pollution originating from biomass combustion, particularly vegetation or forest fires, is less certain. Despite this uncertainty, land managers in Australia and overseas are increasingly required to meet strict air quality targets in their planned fuel reduction burns and address criticism from communities affected by fire disasters about the adequacy of these preventive operations. This thesis contributes to the evidence-base about the health effects of biomass smoke, particularly at the lower levels that occur in Darwin, northern Australia where approximately 95% of outdoor particulate matter comes from low-intensity seasonal vegetation fires.

Methods
I conducted a prospective panel study of the relationship between daily fluctuations in atmospheric particulate matter and daily symptoms, medication use and heath care attendances in a cohort of people with asthma. I also conducted case-crossover and timeseries studies of daily levels of particulate matter and hospital admissions for respiratory and cardiovascular conditions, and a case-crossover study of emergency department attendances.

Results
Particulate loadings generally remained low with a mean of PM10 of 18 μg/m3 over the dry seasons of 2000, 2004 and 2005 and a mean PM2.5 of 10 μg/m3 over the dry seasons of 2004 and 2005. In the panel study, both PM10 and PM2.5 were positively associated with the proportion of people with asthma experiencing symptoms and commencing steroid tablets. In the time-series and case-crossover studies, PM10 showed borderline positive associations with respiratory conditions in non-Indigenous people and significant positive associations in iv Indigenous people. There were mostly inverse or absent associations for cardiovascular admissions. While no associations were found between either PM10 or PM2.5 and emergency department attendances, the available data for this outcome were limited and the study lacked adequate statistical power.

Conclusions
Adverse health respiratory outcomes were measurable at particulate pollution levels well below Australia’s current air quality standards. Managers of deliberate burns, the primary intervention for preventing severe wildfires, therefore need to manage the health risks associated with smoke from these fires. While these risks might be justifiable when compared with the many harms associated with uncontrolled fires and extreme pollution episodes, it is important that the amount of smoke affecting urban areas is minimised, the impact on air quality in urban areas is evaluated and that advance public advisories are provided to allow people at higher risk from exposure to smoke to take appropriate action.

Date of Award	2008
Original language	English
Supervisor	Ross Stewart Bailie (Supervisor)