Residential wood combustion, particulate matter 2.5 sampling project, Whitehorse: chapter 1

• 1.1 Background
• 1.2 Current Study

1. Introduction

1.1 Background

Residential wood combustion (RWC) in wood stoves, fireplaces and fireplace inserts is a commonly used method for heating homes in Canada and is a significant source of fine particulate matter (PM_2.5) emissions. RWC accounts for 8% of all PM_2.5 emissions in Canada, or 36% when open sources are excluded (Environment Canada, 2009). In communities where RWC is used as a major heating source, this fraction can be much higher. PM_2.5 is an air pollutant of concern for human health due to its ability to penetrate deeply into the lungs, and exposure has been linked to reduced cardiovascular and lung function, increased hospitalization, and increased mortality. Many studies have investigated potential health effects associated specifically with wood smoke particles, and a number of comprehensive review papers on this topic have been well summarized recently (E Risk Sciences, 2009).

Whitehorse is the largest city in the Yukon Territory, having a population of 22 898 in the census metropolitan area (Statistics Canada, 2006). As in many other communities in Canada, due to the abundance of wood as an inexpensive fuel source for heating, RWC is common in Whitehorse. A 2006 emission inventory for Whitehorse (Senes, 2008), estimated that, on an annual basis, heating contributes 84% to PM_2.5 emissions, followed by fugitive dust (~9%), mobile (both on and off road) sources (~4%) and industrial point sources (~3%). Commercial buildings primarily rely on fuel oil or propane for heating while residences utilize a combination of fuels. No information on the detailed breakdown of emissions from residential heating in Whitehorse is currently available; however, the 2007 National Pollutant Release Inventory attributes 97% of PM_2.5 emissions from commercial and residential fuel burning in the Yukon to residential wood combustion (Environment Canada, 2009). Similarly, the 2006 Whitehorse emission inventory attributes more than 98% of PM_2.5 emissions from heating to wood fuel, although the estimate is based on a study of communities in northern B.C. (Senes, 2008).

In addition to the abundance of RWC in Whitehorse, meteorology and topography also affect ambient concentrations of particulate matter in the winter. Ambient concentrations of PM_2.5 on any given day are a function of emission levels and the degree of atmospheric dispersion present. Atmospheric dispersion is a function of both the wind speed and mixing height, with higher wind speeds and mixing heights allowing for more efficient dispersion and dilution of pollutants. In the case of RWC in Whitehorse, emission levels are expected to be inversely correlated to ambient temperature over a certain range, as more wood fuel is burned as temperatures decrease. Whitehorse, which is situated in the Yukon River Valley, has a continental arctic climate, leading to episodes of cold temperatures and surface-level inversions, which inhibit the dispersion of pollutants. The subdivision of Riverdale is particularly susceptible to stagnation due to its location in a bowl, partially encircled by an escarpment. Due to the proximity of Whitehorse to the Gulf of Alaska, however, cold stagnation episodes can be tempered by inflow of warmer maritime air masses.

The impact of RWC emissions in Whitehorse has been a concern for several decades. Poor air quality in Whitehorse in the early 1980s, particularly in the subdivision of Riverdale, prompted Environment Canada’s Environmental Protection Service to commission several studies. These included a wood burning survey (Orecklin and McCandless, 1983), monitoring studies and associated data analysis over four heating seasons from 1981 to 1984 at several sites in Riverdale (Senes, 1983; McCandeless, 1984), and a legislative review and recommendations paper (Hatfield, 1984). Monitoring studies included the measurement of total suspended particulate (TSP), carbon monoxide (CO), nitrogen oxides (NO_x), polycyclic aromatic hydrocarbons (PAHs) and meteorology. High concentrations of TSP (max. = 553 µg/m³) were observed during the study, with average concentrations above the Canadian 24-hour average objective of 120 µg/m³ at one site in Riverdale for two of the heating seasons. An inverse correlation was found between TSP concentrations and wind speeds at the airport. Analysis of a subset of samples showed that 95% of the TSP was made up of PM_2.5, and that RWC constituted the bulk of the particulate mass (McCandeless, 1984).

Recent data from the National Air Pollution Surveillance (NAPS) station in Whitehorse are shown in Figure 1. Hourly PM_2.5 data from 2001 to 2008 measured using a Tapered Element Oscillating Microbalance (TEOM), averaged over 24 hours and segregated by month, show that median values are <3 µg/m³ throughout the year, with the large peaks in the summer months attributable to forest fires in 2003 and 2004. Despite the relatively low values measured in the winter months in recent years, local government agencies often field complaints from residents regarding wood smoke, and there is anecdotal evidence that levels may be higher in certain parts of the city, such as Riverdale, particularly during stagnation events.

 

1.2 Current Study

An air quality project working group, including representatives from the Pacific and Yukon Region of Environment Canada, the City of Whitehorse, Yukon Environment, and Health and Social Services Yukon was convened in 2008 to discuss options for determining current impacts of RWC emissions in Whitehorse, including the subdivision of Riverdale. The objective was the development of a baseline from which to better understand the contribution of wood smoke to PM_2.5 levels in the city. Such a baseline would also help assess the effectiveness of any future intervention that may be considered for this community, such as an appliance change-out program. The purpose of this study is to assess PM_2.5 levels in Whitehorse during the winter months and to quantify the contribution of wood smoke. To assess the contribution of wood smoke to PM_2.5 using a limited number of samples obtained during this short-term study, three markers for wood smoke were employed: levoglucosan, ¹⁴C (carbon 14) isotope and an aethalometer, which measures black carbon.

Levoglucosan is a compound that is emitted in large concentrations during the burning of cellulose. It has no other significant sources and is relatively stable in the atmosphere after emission, making it a useful marker of a biomass burning contribution in particulate matter (Simoneit et al., 1999). Several studies have made use of ambient levoglucosan concentration ratios with particulate matter or organic carbon to identify biomass burning plumes (e.g. Leithead et al., 2006; Jordan et al., 2006a; Puxbaum et al., 2007); however, emission profiles can vary due to different burning conditions and fuel types, therefore reliance on ambient levoglucosan concentration ratios alone to estimate biomass burning contribution to PM can be subject to uncertainty (Hedberg et al., 2006). However, levoglucosan data when used in combination with other methods of source apportionment can provide an estimate of relative levels of wood smoke contribution.

The measurement of the radioactive isotope ¹⁴C has become an increasingly common technique used to apportion particulate matter to biomass combustion (e.g. Lewis et al., 2004; Jordan et al., 2006b; Szidat et al., 2006; Ward et al., 2006). ¹⁴C decays over time (half life = 5730 yrs), therefore determining its concentration in ambient particulate matter can be used to calculate the contribution of “contemporary” or “modern” carbon versus “fossil carbon.” Fossil fuels contain no ¹⁴C and thus are termed “fossil carbon” sources, while biomass has a much higher content due to recent incorporation of ¹⁴C from the atmosphere and are thus termed “modern” sources. The ¹⁴C content of the particulate matter, which can be expressed as a fraction of modern carbon (f_m), can therefore be used to determine the extent to which the particulate matter carbon originated from fossil fuel combustion versus biomass burning combustion.

The third method used to distinguish particulate emitted by RWC was a dual wavelength aethalometer. The aethalometer measures the attenuation of light at two different wavelengths by particulate that have been collected on a filter tape. Absorption at 880 nm is calibrated to give a concentration of black carbon, while increased absorption at the wavelength of 370 nm (UV) relative to 800 nm, due to certain aerosol components of wood smoke, has been shown to provide a qualitative indicator of the presence of wood smoke (Allen, 2004; Jeong et al., 2004; Park et al., 2006).