10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Prevalence and Timing of Indoor PM Emission Events Observed in a Small Cohort of Homes Using Low-Cost Dust Sensors
Ian Longley, Gustavo Olivares, Ayushi Kachhara, Sam Edwards, GUY COULSON, National Inst of Water and Atmospheric Research, New Zealand
Abstract Number: 943 Working Group: Indoor Aerosols
Abstract Home occupants are exposed to airborne particulate matter from both indoor sources and the infiltration of outdoor sources. Mostly this differentiation has been achieved through compositional analysis. We conducted a pilot study using time series analysis of continuous data from a low-cost unobtrusive sensor package that would be suitable for mass deployment.
Eight homes were equipped with three custom-built air quality monitors for a 2-week period in winter 2016. Two monitors (PACMAN) were placed in two rooms indoors and one (ODIN) outdoors on the same property. All homes but one used wood-burning appliances as the primary source of heat and were all located in a town where wood-burning dominates both heating and PM emissions. Householders were invited to keep time-activity diaries focusing on cooking, heating and other activities which may impact PM levels in the home, and fire-lighting was monitored with temperature sensors.
Distinct particle ‘events’ were detectable in every home in which PM levels rapidly increased then decayed. Events were attributed to indoor or outdoor sources depending on whether a corresponding event was observed outdoors or not. The average prevalence of indoor sources was 1.2 per day (range 0.8 – 1.5). Events were attributed to outdoor sources in 6 out of 8 houses with an average prevalence of 0.4 events per day. Indoor events produced peak concentrations 3 times greater than outdoor events on average. Indoor source events were most common between 7 am and 8 am and between 5 pm and 6 pm, but with variations between homes. Outdoor source events appeared to have no temporal pattern. Completion of time-activity diaries, and the detail provided was highly variable. However, there was no clear correlation between events described and observed particle levels. Particle events did not, in general, correspond to the lighting of wood-burning fires. Data was visualised for presentation to, and discussion with, householders.