Morning peak of air pollutant concentrations in urban areas: effect of time lag between emissions and turbulence

Primary air pollutants in urban environments, like carbon monoxide (CO) and particulate matter (PM10), show a distinctive diurnal cycle in concentration, characterized by maximum values in the morning and in the evening. These peaks occur during the times of the day when high emission rates combine with relatively low turbulence in the atmospheric boundary layer (ABL). Moreover, during the morning and evening transition hours, both the emission rates and the ABL turbulence are highly transient. Therefore, we expect that the timing and the amplitude of the concentration peaks are controlled by the time evolution of the emission rates and the turbulence intensity, as well as by the relative time phase between emissions and turbulence. The latter effect is studied in the present work by considering changes in the diurnal cycles of concentrations for periods before and after Official Time (OT) changes due to Daylight Saving Time (DST) observance. Anthropogenic processes tied to OT, like school and job traffic, have an abrupt 1-hour shift in their daily emission patterns at the DST weekend of fall and spring. We show how this time shift in emissions has an effect on the concentration diurnal cycles, especially in the morning peaks.

The problem is addressed through data analysis and modeling. CO and PM10 data for Santiago City (Chile) is analyzed. The city has a severe air pollution problem, especially in terms of PM10 concentrations during the cold season. It has a 7-stations network measuring hourly concentrations of several criteria pollutants and basic meteorology, since the mid '90s. The large number of years in the data series allows the application of a statistical procedure aimed at extracting the DST effect on concentrations, beyond the noise introduced by the day-to-day variability of a single year. The results show that the DST effect is present at most of the monitoring stations, although it is most clear in the morning concentration peaks at the stations located in residential areas far from downtown. In the fall DST change, for example, the morning peaks of CO and PM10 occur later and have smaller magnitudes in the days after the OT change, despite the transition to generally worse dispersion conditions. These results are explained with a simple box model. Finally, the data analysis results are compared to those obtained with a 3-D simulation using the WRF-Chem model and a local emission inventory for Santiago City.