Use of GIS to Characterize Pedestrian Exposure to Street Level Air Pollution–Cusco, Peru

Abstract

P-661 Introduction: Cusco, Peru, the capital of the historic Inca Empire, is located at 3350 meters in the Andes Mountains of South America, and currently has a population of approximately 325,000. Frequently referred to as the “Gateway to Machu Picchu”, Cusco has a vibrant tourist industry with nearly 1 million visitors per year. Objectives: Investigate the temporal and spatial variation of select air pollutants, potential sources, and the exposed in Cusco, Peru. Methods: First, a fixed monitoring station was established to measure carbon monoxide (CO), time-integrated particulate matter less than or equal to 2.5μm (PM2.5) (Harvard Impactor/24-hr), nitrogen dioxide (NO2) (Palmes tubes/72-hr) and select volatile organic compounds (VOCs) (tenax tubes/72-hr). Second, 46 potential “hot spots” were selected and monitored for real-time PM2.5, particulate matter less than or equal to 10 μm (PM10) (DustTrak), and CO (Langan) for three 15-min intervals during a 1-day period. At the same time and location of this monitoring, a site survey recording number of vehicles, pedestrians, presence or absence of industry, and other exposure-related variables was also conducted. Third, 75 stratified random samples were monitored (same as “hot spot”) for one 15-min interval between 6am-7pm. GIS-based spatial analyst and geostatistical applications were implemented to examine geographic patterns, interpolate areal concentrations, and identify areas of probabilistic concern. Results: Fixed site twenty-four hour average (± std dev) ambient concentrations were: PM2.5 = 45.2±16μg/m3; CO = 2.17±0.4ppm. “Hot spot” ambient concentrations were: morning PM2.5 = 233±151μg/m3; PM10 = 294±195μg/m3. CO = 3.1±2.3ppm; afternoon PM2.5 = 84±48μg/m3; PM10 = 109±60μg/m3, CO = 2.1±1.9ppm; evening PM2.5 = 164±118μg/m3; PM10 = 191±126μg/m3, CO = 3.3±2.4ppm. Random ambient concentrations were: PM2.5 = 98±76μg/m3; PM10 = 105±85μg/m3 CO = 1.6±0.8ppm. DustTrak particulate samples reflect unadjusted values and are likely 3X higher than actual mass based on discrepancies identified in scientific literature between co-located DustTrak aerosol and gravimetric mass measurements. Traffic density reflects moderate positive correlation with PM2.5 (r =.34), PM10 (r =.31) and CO (r =.45). Elevation reveals a negative correlation with PM2.5 (r =−.50), PM10 (r =−.47) and CO (r =−.36). Conclusion: The highest levels of monitored pollutants were recorded during the morning commute in downtown Cusco. Although weak, correlation results suggest that traffic and elevation are influencing street-level pollution. These results help provide a better understanding of the temporal and spatial variation associated with community and street-level air pollution exposure. Additionally, results can possibly aid in future epidemiologic research or be useful to policy makers in planning remedial measures.