Improved aerosol correction for OMI tropospheric NO<sub>2</sub> retrieval over East Asia: constraint from CALIOP aerosol vertical profile

Mengyao Liu,Jintai Lin,K. Folkert Boersma,Gaia Pinardi,Yang Wang,Julien Chimot,Thomas Wagner,Pinhua Xie,Henk Eskes,Michel Van Roozendael,François Hendrick,Pucai Wang,Yingying Yan
DOI: https://doi.org/10.5194/amt-2018-34
2018-01-01
Abstract:Abstract. Satellite retrieval of vertical column densities (VCDs) of tropospheric nitrogen dioxide (NO2) is critical for NOx pollution and impact evaluation. For regions with high aerosol loadings, the retrieval accuracy is greatly affected by whether aerosol optical effects are treated implicitly (as additional effective clouds) or explicitly, among other factors. Our previous POMINO algorithm explicitly accounts for aerosol effects to improve the retrieval especially in polluted situations over China, by using aerosol information from GEOS-Chem simulations with further monthly constraint by MODIS/Aqua AOD data. This study updates the retrieval algorithm to POMINO v1.1, by constructing a monthly climatological dataset of aerosol extinction profiles, based on Level-2 CALIOP/CALIPSO data over 2007–2015, to better constrain the modeled aerosol profiles. We find that GEOS-Chem captures the month-to-month variation of CALIOP aerosol layer height but with a systematic underestimate by about 300–600 m (season and location dependent), due to a too strong vertical gradient of extinction above 1 km. Correcting the model aerosol extinction profiles results in small changes in retrieved cloud fraction, increases in cloud top pressure (within 2–6 % in most cases), and increases in tropospheric NO2 VCD by 4–16 % over China on a monthly basis in 2012. The improved NO2 VCDs (in POMINO v1.1) are more consistent with independent ground-based MAX-DOAS observations (R2 = 0.80, NMB = −3.4 %) than POMINO (R2 = 0.80, NMB = −9.6 %) and DOMINO v2 (R2 = 0.68, NMB = −2.1 %) are. Especially on haze days, R2 reaches 0.76 for POMINO v1.1, much higher than that for POMINO (0.68) and DOMINO v2 (0.38). Furthermore, the increase in cloud pressure likely reveals a more realistic vertical relationship between cloud and aerosol layers, with aerosols situated above the clouds in certain months instead of always below the clouds. Our POMINO v1.1 algorithm will be applied to the recently launched TropOMI sensor.
What problem does this paper attempt to address?