Health impact assessments of environmental exposures are currently hampered by the absence of integrated datasets, workflows and metadata/provenance requirements. The Scientific workflow system for environmental health impact assessments (Air-Health SWS) project will solve the difficulties associated with merging environmental and health datasets using existing computational and data infrastructure in such a way that no technical skills in database systems, network-based remote access or coding will be required by users.

The project will develop a scientific workflow system for health impact assessments of air pollution that will enable users to build and extend analyses by linking data acquisition, data transformation, mathematical operations, graphing, statistical analysis and outputs. The tools will streamline desktop and web-based research processes to translate research into policy, and will be built on open software and cloud services.

The platform will use existing application programming interfaces to access environmental and health data from several sources, including state government air pollution monitors, satellites from space agencies, the Bureau of Meteorology, land use/planning agencies, Australian Bureau of Statistics and the Australian Transport Research Cloud.

Start date 1 March 2021
Expected completion date 31 December 2021
Investment by ARDC $163,460
Lead node
1 Policy stakeholder workshops
Workshops with key policy stakeholders and platform users to determine a set of core policy questions and HIA scenarios. They will discuss current urban infrastructure planning proposals with potential impacts on air pollution, touching on bushfire management, energy and transport objectives, and regulatory arrangements with operators of polluting industries.
2 Academic and user stakeholder workshops
Academic stakeholders will bring skills and experience in the scientific aspects of air pollution modeling for the identified policy scenarios. For example, to inform bushfire management strategies, computational frameworks for data from air pollution monitors, records of bushfire and smoke haze days and corresponding meteorological observations will be discussed with a focus on the required assumptions, health impact functions and epidemiological methods. Discussions with key users will establish features and design elements that facilitate wide usability of the platform.
3 Scientific Workflow System
Health impact assessments will be automated using the Apache Airflow workflow system. The system will support users in the selection of study areas, exposure types (particulate matter, nitrogen dioxide and ozone, etc), time periods (historical and future) and desired health impact assessment outputs. Users will then be guided through decisions about spatiotemporal considerations and exposure data types and sources to support rigorous epidemiological analysis. The SWS will identify and access required exposure and health datasets and will perform analysis steps that combine these to calculate health burden and impact.
4 Dynamic data and models
Airflow schedulers that have been adapted to dynamically store and update air pollution observations and predictors, and population and health data will be provided in an interface on which users can select areas and times of interest, air pollution types, and health related variables to model policy outcomes.

Core features

Simple interfaces
Complex analyses can be performed by users without technical skills in database systems, network-based remote access or coding.
Evidence-based policy
Assists health decision makers, researchers and public sector users to translate research into policy.

Who is this project for?

  • Researchers
  • Research organisations
  • Public sector agencies
  • Government (state and commonwealth)
  • Polluting industries
  • Health sector

What does this project enable?

This project will solve the difficulties associated with merging environmental and health datasets using existing computational and data infrastructure. No technical skills in database systems, network-based remote access or coding will be required by users.

Handy resources

University of QueenslandVisit
Environment Protection Authority Victoria Visit
The University of Sydney Visit