Joint measurements of PM 2. 5 and light-absorptive PM in woodsmoke-dominated ambient and plume environments

Abstract. DC, also referred to as Delta-C, measures enhanced light absorption of particulate matter (PM) samples at the near-ultraviolet (UV) range relative to the near-infrared range, which has been proposed previously as a woodsmoke marker due to the presence of enhanced UV light-absorbing materials from wood combustion. In this paper, we further evaluated the applications and limitations of using DC as both a qualitative and semi-quantitative woodsmoke marker via joint continuous measurements of PM 2. 5 (by nephelometer pDR-1500) and light-absorptive PM (by 2-wavelength and 7-wavelength Aethalometer ® ) in three northeastern US cities/towns including Rutland, VT; Saranac Lake, NY and Ithaca, NY. Residential wood combustion has shown to be the predominant source of wintertime primary PM 2. 5 emissions in both Rutland and Saranac Lake, where we conducted ambient measurements. In Ithaca, we performed woodsmoke plume measurements. We compared the pDR-1500 against a FEM PM 2. 5 sampler (BAM 1020), and identified a close agreement between the two instruments in a woodsmoke-dominated ambient environment. The analysis of seasonal and diurnal trends of DC, black carbon (BC, 880 nm) and PM 2. 5 concentrations supports the use of DC as an adequate qualitative marker. The strong linear relationships between PM 2. 5 and DC in both woodsmoke-dominated ambient and plume environments suggest that DC can reasonably serve as a semi-quantitative woodsmoke marker. We propose a DC-based indicator for woodsmoke emission, which has shown to exhibit a relatively strong linear relationship with heating demand. While we observed reproducible PM 2. 5 –DC relationships in similar woodsmoke-dominated ambient environments, those relationships differ significantly with different environments, and among individual woodsmoke sources. Our analysis also indicates the potential for PM 2. 5 –DC relationships to be utilized to distinguish different combustion and operating conditions of woodsmoke sources, and that DC–heating-demand relationships could be adopted to estimate woodsmoke emissions. However, future studies are needed to elucidate those relationships.