Retrieval of tropospheric NO2 columns over Berlin from high-resolution airborne observations with the spectrolite breadboard instrument

Tim Vlemmix, Xinrui (Jerry) Ge, Bryan T. G. de Goeij, Len F. van der Wal, Gerard C. J. Otter, Piet Stammes, Ping Wang, Alexis Merlaud, Dirk Schüttemeyer, Andreas C. Meier, J. Pepijn Veefkind, Pieternel F. Levelt
2017 Atmospheric Measurement Techniques Discussions  
This paper presents the retrieval method that was developed to derive tropospheric NO<sub>2</sub> columns from UV/VIS spectral measurements obtained with the Spectrolite Breadboard Instrument during the AROMAPEX campaign in Berlin (April 2016). A typical DOAS retrieval approach is followed. For the calculation of air mass factors this study specifically focuses on the impact of the surface reflectance, which varies considerably from pixel to pixel over this urban region. Ground-based aerosol
more » ... ical thickness measurements are used as prior information. It is shown that retrieved surface reflectance shows good agreement with those derived from Landsat 8 measurements performed on the same day. Furthermore we demonstrate that tropospheric NO<sub>2</sub> columns retrieved for pairs of adjacent pixels are self-consistent in the sense that they do not show a substantial systematic dependence on surface reflectance, in contrast to differential slant column densities. Also some cases are identified to illustrate this on a pixel-by-pixel level. An error budget is provided to quantify the impact of various assumptions on the accuracy of the retrieval of surface reflectance and tropospheric NO<sub>2</sub> columns. Both in the morning and afternoon flight a NO<sub>2</sub> plume is observed stretching out over Berlin from West to East. Peak values between 15&amp;thinsp;&amp;times;&amp;thinsp;10<sup>15</sup> and 20&amp;thinsp;&amp;times;&amp;thinsp;10<sup>15</sup>&amp;thinsp;molec/cm<sup>2</sup> are detected, whereas &amp;ndash; at much lower spatial resolution &amp;ndash; OMI detects peak values between 9&amp;thinsp;&amp;times;&amp;thinsp;10<sup>15</sup> (first overpass) and 4&amp;thinsp;&amp;times;&amp;thinsp;10<sup>15</sup>&amp;thinsp;molec/cm<sup>2</sup> (second overpass).
doi:10.5194/amt-2017-257 fatcat:kp6v4c6bxbbbffkzcpsmvnrg2e