Investigating methods to quantifying uncertainty in PM2.5 emission rates from cooking by toasting bread

Abstract

Exposure to airborne fine particulate matter (PM2.5) is linked to multiple negative health effects and indoor sources are important contributors to personal exposure. Cooking is a common indoor source, but reported emission rates have high variability. Methods to quantify uncertainty in PM2.5 cooking emission rates are investigated so that they can be used in probabilistic exposure models to evaluate interventions. Controlled tests were conducted to measure emission rates from the toasting of bread because it is simple and repeatable. Two methods were compared: residential kitchen field tests and large chamber tests. The theoretical peak calculation method was used to determine emission rates from time-resolved PM2.5 concentration measurements. The large chamber tests produced more consistent results than the residential field tests, with a coefficient of variance almost an order of magnitude lower due to the improved control of variables. Then, the emission rates were normally distributed with mean 0.23mg/min and standard deviation 0.067mg/min. However, this distribution may be less representative of normal behaviour. The resulting dataset can be combined with other sources to represent housing stock exposures in probabilistic models, enabling the exploration of exposure uncertainties and interventions. More generally, key recommendations when measuring PM2.5 emission rates include: high temporal resolution measurements; custom calibration factors; identifying periods for emissions, mixing, and decay; constant ventilation rates; quantifying mixing conditions; and ensuring high quality decay data.