Riley Erin A, Schaal LaNae, Sasakura Miyoko, Crampton Robert, Gould Timothy R, Hartin Kris, Sheppard Lianne, Larson Timothy, Simpson Christopher D, Yost Michael G
University of Washington Department of Environmental and Occupational Health Sciences, Box 357234, Seattle, WA, 98198. +1 (206) 543-3222.
University of Washington Department of Biostatistics, Box 357232 Seattle, WA, 98198.
Atmos Environ (1994). 2016 May;132:229-239. doi: 10.1016/j.atmosenv.2016.03.001.
Mobile monitoring has provided a means for broad spatial measurements of air pollutants that are otherwise impractical to measure with multiple fixed site sampling strategies. However, the larger the mobile monitoring route the less temporally dense measurements become, which may limit the usefulness of short-term mobile monitoring for applications that require long-term averages. To investigate the stationarity of short-term mobile monitoring measurements, we calculated long term medians derived from a mobile monitoring campaign that also employed 2-week integrated passive sampler detectors (PSD) for NO, Ozone, and nine volatile organic compounds at 43 intersections distributed across the entire city of Baltimore, MD. This is one of the largest mobile monitoring campaigns in terms of spatial extent undertaken at this time. The mobile platform made repeat measurements every third day at each intersection for 6-10 minutes at a resolution of 10 s. In two-week periods in both summer and winter seasons, each site was visited 3-4 times, and a temporal adjustment was applied to each dataset. We present the correlations between eight species measured using mobile monitoring and the 2-week PSD data and observe correlations between mobile NO measurements and PSD NO measurements in both summer and winter (Pearson's r = 0.84 and 0.48, respectively). The summer season exhibited the strongest correlations between multiple pollutants, whereas the winter had comparatively few statistically significant correlations. In the summer CO was correlated with PSD pentanes (r = 0.81), and PSD NO was correlated with mobile measurements of black carbon (r = 0.83), two ultrafine particle count measures (r =0.8), and intermodal (1-3 μm) particle counts (r = 0.73). Principal Component Analysis of the combined PSD and mobile monitoring data revealed multipollutant features consistent with light duty vehicle traffic, diesel exhaust and crankcase blow by. These features were more consistent with published source profiles traffic-related air pollutants than features based on the PSD data alone. Short-term mobile monitoring shows promise for capturing long-term spatial patterns of traffic-related air pollution, and is complementary to PSD sampling strategies.
移动监测为广泛空间尺度上的空气污染物测量提供了一种手段，而采用多个固定站点采样策略来进行此类测量则不太可行。然而，移动监测路线越长，测量的时间密度就越低，这可能会限制短期移动监测对于需要长期平均值的应用的有用性。为了研究短期移动监测测量的稳定性，我们计算了从一次移动监测活动中得出的长期中位数，该活动还在马里兰州巴尔的摩市整个区域分布的43个十字路口，针对一氧化氮、臭氧和九种挥发性有机化合物，采用了为期两周的集成被动采样探测器（PSD）。就当时进行的移动监测活动的空间范围而言，这是规模最大的活动之一。移动平台每隔三天在每个十字路口重复测量6 - 10分钟，分辨率为10秒。在夏季和冬季的两周时间段内，每个站点被访问3 - 4次，并且对每个数据集进行了时间调整。我们展示了使用移动监测测量的八种物质与为期两周的PSD数据之间的相关性，并观察到夏季和冬季移动一氧化氮测量值与PSD一氧化氮测量值之间的相关性（皮尔逊相关系数分别为0.84和0.48）。夏季多种污染物之间的相关性最强，而冬季具有统计学意义的相关性相对较少。在夏季，一氧化碳与PSD戊烷相关（r = 0.81），PSD一氧化氮与黑碳的移动测量值相关（r = 0.83）、与两种超细颗粒计数测量值相关（r = 0.8）以及与多峰（1 - 3微米）颗粒计数相关（r = 0.73）。对PSD和移动监测数据进行主成分分析，揭示了与轻型车辆交通、柴油尾气和曲轴箱窜气一致的多污染物特征。这些特征与已发表的与交通相关的空气污染物源谱比仅基于PSD数据的特征更为一致。短期移动监测在捕捉与交通相关的空气污染的长期空间模式方面显示出前景，并且是对PSD采样策略的补充。
