Comparison Between Light Scattering and Gravimetric Devices for Sampling PM10 Mass Concentration in Livestock Houses

Abstract

Further research is needed to validate real-time continuous devices for particulate matter (PM) against gravimetric samplers in livestock houses to obtain accurate absolute values. The objective of this study was to compare co-located real-time light scattering samplers and the equivalent gravimetric sampler in different poultry and pig livestock houses for PM10 mass concentration and to develop livestock-specific calibration factors for light scattering samplers. Paired DustTrak light scattering device (DustTrak aerosol monitor, TSI, U.S.) and PM10 gravimetric cyclone sampler (URG Corp., U.S.) were used for measuring PM10 mass concentrations during 24 h periods (from noon to noon) inside livestock houses. Sampling was conducted in 32 livestock houses in the Netherlands, including broilers, broiler breeders, layers in floor and in aviary system, turkeys, piglets, growing-finishing pigs in traditional and low emission housing with dry and liquid feed, and sows in individual and group housing. A total of 119 pairs of 24 h measurements (55 for poultry and 64 for pigs) were recorded and analyzed using linear regression analysis. In poultry, cyclone sampler and DustTrak data fitted well to a linear regression, with a regression coefficient equal to 0.41, an intercept of 0.16 mg/m3 and a correlation coefficient of 0.91 (excluding turkeys). In pigs, we found a regression coefficient equal to 0.61, an intercept of 0.05 mg/m3 and a correlation coefficient of 0.84. Measured PM10 concentrations using DustTraks were clearly underestimated (approx. by a factor 2) in poultry as in pigs compared with cyclone pre-separators. DustTrak light scattering devices should be self-calibrated to investigate PM10 mass concentrations accurately in livestock houses.