Modelling of Air Quality in a Rural Area
of Unconventional Oil and Gas Development
for Health and Justice

Modelling of Air Quality in a Rural Area of Unconventional Oil and Gas Development for Health and Justice

Background

  • Canada is a top producer of natural gas, with a large amount coming from the Montney Shale Gas Play in northeastern British Columbia.
  • Extraction of natural gas can impact air, water and soil quality, leading to negative human health outcomes.
  • Due to limited air quality monitoring in rural BC, additional techniques are needed to explore connections between unconventional oil and gas extraction, air quality, and health.

Objectives

  1. Develop novel methods to assess current and historical exposures to air pollution in a region experiencing rapid unconventional oil and gas development with sparse air quality monitoring.

  2. Assess the environmental justice implications of unconventional oil and gas development in a rural area of Western Canada.

Project #1: Modelling spatial & temporal variability of air pollution in an area of unconventional natural gas operations

Key Findings

  1. We were able to harness the available air quality data to build models capable of predicting at unmeasured locations using oil and gas activity data to explain the variations in concentration.

  2. The models were used to predict concentrations of 12 different gases and particles at the home location of health study participants and across the study area.

  3. Minimal monitoring of pollutants with strong associations to fracking activities & several health impacts, such as BTEX compounds, highlighted the need for additional monitoring of these species closer to areas of residence.

Citation: Please refer to the Environmental Pollution paper “Modelling Spatial & temporal variability of air pollution in an area of unconventional natural gas operations” for more details. 

Project #2: Eighteen years of daily PM2.5 predictions (2005 – 2022) for a region of Western Canada: Machine learning and satellite inputs for applications in rural health

Key Findings

  1. We demonstrate the application of machine learning to the satellite-based estimates of aerosols and meteorology to estimate daily PM2.5 at a spatial and temporal resolution relevant for ongoing health studies.

  2. Our model was trained and validated on available air quality monitoring data from 2013 to 2022 and then used to predict and backcast concentrations from 2005-2022.

  3. We find that an increasing number of days are exceeding daily health guidelines, and these exceedances tend to occur during wildfire season.

Citation: Please refer to the Atmospheric Environment paper “Eighteen years of daily PM2.5 predictions (2005–2022) for a region of western Canada: Machine learning and satellite inputs for applications in rural health” for more details. 

Project #3: Inequity of exposure to unconventional natural gas development in northeastern British Columbia, Canada – an environmental justice analysis

Key Findings

  1. We find higher levels of modelled air pollution concentrations in areas with higher proportions of Indigenous population.

  2. We find that as an area’s socio-economic vulnerability increases, modelled concentrations of air pollutants, some facility emissions and counts of active hydraulically fractured wells also increase.

  3. The study develops a composite index of socioeconomic vulnerability that highlights rural deprivation. We also demonstrate different ways of assessing exposure when conducting environmental justice analysis.

Citation: Please refer to the Environmental Science & Technology paper “Inequity of Exposure to Unconventional Natural Gas Development in Northeastern British Columbia, Canada ─ An Environmental Justice Analysis” for more details.