Characterizing the seasonal cycle and vertical structure of ozone in Paris, France using four years of ground based LIDAR measurements in the lowermost troposphere

Abstract

Systematic ozone LIDAR measurements were completed during a 4 year period (2011–2014) in Paris, France to study the seasonal variability of the vertical structure of ozone in the urban boundary layer. In addition, we use in-situ measurements from the surface air quality network that is located in Paris (AIRPARIF). Specifically, we use ozone and NO2 measurements made at two urban stations: Paris13 (60 m ASL) and the Eiffel Tower (310 m ASL) to validate and interpret the LIDAR profiles. Remote sensed tropospheric NO2 integrated columns from the SAOZ instrument located in Paris are also used to interpret ozone measurements. Comparison between ozone LIDAR measurements averaged from 250 m to 500 m and the Eiffel Tower in-situ measurements shows that the accuracy of the LIDAR (originally ±14 μg·m−3) is significantly improved (±7 μg·m−3) when a small telescope with a wide angular aperture is used. Results for the seasonal cycle of the ozone vertical gradient are found to be similar using two methods: (1) measured differences between AIRPARIF stations with measurements at 60 m ASL and 310 m ASL and (2) using LIDAR profiles from 300 m to the top of the Planetary Boundary Layer (PBL). Ozone concentrations measured by the LIDAR increase with altitude within the PBL, with a steeper gradient in winter (60 μg·m−3·km−1) and a less strong gradient in summer (20 μg·m−3·km−1). Results show that in winter, there is a sharp positive gradient of ozone at the surface, which is explained by ozone titration by NO combined with increased atmospheric stability in winter. In the afternoon during summer, photochemistry and vertical mixing are large enough to compensate for ozone titration near the surface, where NOx is emitted, and there is no gradient in ozone observed. In contrast, in the summer during the morning, ozone has a sharper positive vertical gradient similar to the winter values. Comparison of the vertically averaged ozone concentrations up to (0–3 km) and urban layer (0–310 m) ozone concentrations shows that the ratio between these two quantities is the largest in summer (86%) and the lowest in winter (49%). We conclude that satellite measurements that represent the 0–3 km integrated ozone column are not necessarily a good proxy for surface ozone and may lead to incorrect conclusions about the surface ozone seasonal variability. The ratio between the urban layer NO2 average concentration and the boundary layer NO2 average concentration obtained from SAOZ NO2 tropospheric columns is always less than 50%, meaning NO2 does not decrease linearly in the PBL, but with a sharper decrease close to the surface.