Source apportionment of speciated PM2.5 over Halifax, Nova Scotia, during BORTAS-B, using pragmatic mass closure and principal component analysis

During BORTAS-B, 42 days of contiguous PM2.5 filter samples were collected during the summer of 2011 in Halifax, Nova Scotia. The aim of the PM2.5 filter sampling was to apportion the source contribution to the total PM2.5 mass concentration in Halifax to inform and validate other surface measurements and chemical transport models related to BORTAS-B. Sampling was conducted on the roof of a Dalhousie University building at a height of 15 m. The building is located in a residential area of Halifax. Continuous black carbon (BC) was measured using a Magee AE-42 aethalometer. Continuous organic carbon was measured using an Aerodyne, Aerosol Chemical Speciation Monitor. Daily teflon filter samples were collected for the determination of fine particulate with a median aerodynamic diameter less than or equal to 2.5 microns (PM2.5). An additional, daily, nylon filter was used for the determination of PM2.5 cations and anions by IC. The PM2.5 teflon filter was analysed for 33 metals by XRF and 10 trace metals by ICP-MS. The biomass burning marker levoglucosan was analysed by GC-MS following derivatization. Excellent agreement (R2 = 0.88) was observed between continuous and filter based measurements with a gradient of 2.76. The median (min : max) PM2.5 mass concentration during BORTAS-B = 3.9 (0.08 : 13.7) μg-m3. The median (min : max) continuous BC = 0.39 (0.12 : 1.03); SO4 = 0.47 (0.14 : 5.59); NO3 = 0.067 (0.007 : 0.64); OC = 0.77 (0.18 : 2.77); NH4 = 0.15 (0:003 : 1.45); Cl = 0.011 (0.0019 : 0.32); Fe = 0.018 (0.0011 : 0.097); Al = 0.011 (0.0091 : 0.086); Si = 0.03 (0.0044 : 0.29); V = 0.0026 (0.0016 : 0.017) and Ni = 0.0007 (0.0005 : 0.0037) μg-m3 respectively. Absolute principal component scores (APCS) and pragmatic mass closure (PMC) will be used to identify the sources driving the observed PM2.5 variability over Halifax, during BORTAS-B. A comparison of APCS and PMC PM2.5 receptor model output results will be presented. These model data will provide further insight into the source contribution to summertime surface PM2.5 mass in Halifax, Nova Scotia, Canada.