Comparison of OMI and ground‐based in situ and MAX‐DOAS measurements of tropospheric nitrogen dioxide in an urban area

Abstract

In this paper we compare retrieved tropospheric vertical column densities (VCDs) of nitrogen dioxide (NO 2 ) from the Ozone Monitoring Instrument (OMI) to coincident tropospheric columns retrieved from the Concurrent Multiaxis Differential Optical Absorption Spectroscopy (CMAX‐DOAS) instrument, installed at the University of Leicester (52.38°N, 1.08°W), and in situ near‐surface measurements from chemiluminescence detectors. The results show that tropospheric NO 2 columns retrieved from CMAX‐DOAS and OMI correlate well ( r = 0.64) when cloud clearing has been applied, and only those pixels that sample at least 90% of the Leicester area were included in the analyses. The correlation of OMI tropospheric VCDs with near‐surface measurements for cloud‐free days in 2005 and 2006 initially showed a strong positive bias in the near‐surface NO 2 measurements and scattered points. This was interpreted as being due to the satellite footprint of OMI sampling the NO 2 sources from the surrounding area of Leicester as well as emissions from the city. A field‐of‐view (FOV) weighted estimate for the OMI‐equivalent urban NO 2 was calculated for each coincidence by including background concentrations from a nearby in situ monitor, situated in a rural area. The subsequent agreement between the OMI tropospheric VCDs and FOV weighted near‐surface measurements is very good for spring ( r = 0.83) and summer ( r = 0.64) months. Finally, seasonal and weekly cycles of NO 2 are produced which show that OMI may be underestimating the amount of NO 2 during the winter months. However, all sets of data show expected weekly cycles, with lower values on a Sunday.