Air dispersion model: Cadia Valley Operations
We engaged an air quality modelling expert, Jacobs, to prepare an independent atmospheric dispersion model that simulates the dispersion of particulate (PM10 and PM2.5) emissions from Cadia Valley Operations (CVO).
The EPA commissioned the study following community concerns raised regarding the accuracy of Cadia Valley Operation’s air quality data (annual average PM2.5 and PM10) included in the human health risk assessment.
The human health risk assessment was required by the EPA and Cadia Valley Operations has made the report available on its website: Human Health Risk Assessment.
Read the full Air quality impact assessment independent review (PDF 10MB) and report appendices (PDF 41MB).
Airborne particles are a suspended mixture of solid and liquid particles in the air. PM2.5 particles have diameters of less than or equal to 2.5 micrometres (µm) and include combustion particles, organic compounds and metals. PM10 particles have diameters less than or equal to 10 µm and include dust and pollen as well as PM2.5.
Read about particulate matter and health impacts.
To understand the sources of PM10 and PM2.5 emissions at CVO, Jacobs, accompanied by NSW EPA conducted a site visit to view all aspects of operations at CVO including the north and south tailings dams. The site visit included the meteorological and ambient air quality monitoring stations.
The modelling carried out by Jacobs used a ground-up approach of independently creating the model files. The data inputs for the model files were sourced from the Bureau of Meteorology and monitoring data and information on activities and equipment from CVO which were independently verified. This data and information were then used to develop an independent atmospheric dispersion model.
The results of the independent model were compared to actual measured PM10 and PM2.5 concentrations from monitors around the mine. The model results were found to predict slightly higher levels than were found at the monitoring sites (Bundarra, Triangle Flat and Meribah). This means the Jacobs independent modelling setup is considered ‘more conservative’ than actual impacts and provides a worst-case estimate of PM10 and PM2.5 concentrations in the ambient air surrounding CVO.
The Jacobs independent model under-predicts measured concentrations of PM10 and PM2.5 at the Woodville station. The report identifies this could be due to the complex terrain in that area or additional local sources such as wood smoke emissions not accounted for in the model.
The Jacobs modelling predicted similar annual averages of PM10 or PM2.5 to CVO’s model. Importantly, no exceedances were found of annual average PM10 or PM2.5 impact assessment criteria.
Annual average impacts for PM10 and PM2.5 from CVO’s model were used as inputs into the HHRA.
The Jacobs independent model showed some differences in the location of the highest predicted 24-hour average particle levels. Jacob’s independent model predicted exceedances of the PM10 24-hour criteria at sensitive receptors to the north, north-east and north-west of the CVO boundary. In the southern part of the study area, Jacob’s independent model predicted lower concentrations than CVO’s model.
Both CVO’s and Jacobs independent models highlight that the impacts from CVO should continue to be monitored and managed. Jacobs notes that the current locations of the existing monitoring locations might not capture areas where peak impacts are predicted.
Jacobs recommended the following to ensure accuracy and transparency of future air quality models and outcomes:
- Future air dispersion models should include more information and transparency on the emission estimation assumptions and development.
- An additional real-time dust monitor should be established south-west of the tailings dams to improve management and future modelling.
- A review of existing monitor locations.
- An additional meteorological weather station (minimum of wind speed and direction) to be located south of CVO to improve future modelling.
The EPA will implement all Jacobs recommendations.