Tropospheric nitrogen dioxide column retrieval from ground-based zenith–sky DOAS observations

Abstract

Abstract. We present an algorithm for retrieving tropospheric nitrogen dioxide (NO 2 ) vertical column densities (VCDs) from ground-based zenith-sky (ZS) measurements of scattered sunlight. The method is based on a four-step approach consisting of (1) the differential optical absorption spectroscopy (DOAS) analysis of ZS radiance spectra using a fixed reference spectrum corresponding to low NO 2 absorption, (2) the determination of the residual amount in the reference spectrum using a Langley-plot-type method, (3) the removal of the stratospheric content from the daytime total measured slant column based on stratospheric VCDs measured at sunrise and sunset, and simulation of the rapid NO 2 diurnal variation, (4) the retrieval of tropospheric VCDs by dividing the resulting tropospheric slant columns by appropriate air mass factors (AMFs). These steps are fully characterized and recommendations are given for each of them. The retrieval algorithm is applied on a ZS data set acquired with a multi-axis (MAX-) DOAS instrument during the Cabauw (51.97 • N, 4.93 • E, sea level) Intercomparison campaign for Nitrogen Dioxide measuring Instruments (CINDI) held from 10 June to 21 July 2009 in the Netherlands. A median value of 7.9 × 10 15 molec cm −2 is found for the retrieved tropospheric NO 2 VCDs, with maxima up to 6.0 × 10 16 molec cm −2 . The error budget assessment indicates that the overall error σ TVCD on the column values is less than 28 %. In the case of low tropospheric contribution, σ TVCD is estimated to be around 39 % and is dominated by uncertainties in the determination of the residual amount in the reference spectrum. For strong tropospheric pollution events, σ TVCD drops to approximately 22 % with the largest uncertainties on the determination of the stratospheric NO 2 abundance and tropospheric AMFs. The tropospheric VCD amounts derived from ZS observations are compared to VCDs retrieved from off-axis and direct-sun measurements of the same MAX-DOAS instrument as well as to data from a co-located Système d'Analyse par Observations Zénithales (SAOZ) spectrometer. The retrieved tropospheric VCDs are in good agreement with the different data sets with correlation coefficients and slopes close to or larger than 0.9. The potential of the presented ZS retrieval algorithm is further demonstrated by its successful application on a 2-year data set, acquired at the NDACC (Network for the Detection of Atmospheric Composition Change) station Observatoire de Haute Provence (OHP; Southern France).

High-resolution mapping of the NO<sub>2</sub> spatial distribution over Belgian urban areas based on airborne APEX remote sensing

“…Note that this is approximately 1 order of magnitude higher than for the fixed ground-based stations, e.g. the BIRA-IASB MAX-DOAS instrument (Tack et al, 2015), having a higher SNR and better spectral performance. Whereas σ DSCD i is a rather minor error source in the case of ground-based stations, it becomes a significant contributor to the total error in the case of an airborne imager.…”

“…The diurnal increase of the NO 2 stratospheric column between 80 • SZA sunrise and sunset is estimated to be approximately 1 × 10 14 molec cm −2 per hour (Tack et al, 2015) and thus much smaller than the retrieved VCDs. The latter have mean values of 1.7 × 10 16 and 7.7 × 10 15 molec cm −2 for the Antwerp and Brussels data sets.…”

Shipborne MAX-DOAS measurements for validation of TROPOMI NO<sub>2</sub> products

“…Because the stratospheric NO 2 VCDs have a diurnal cycle, we cannot interpolate the stratospheric NO 2 VCD directly at the TROPOMI overpass time using the MAX-DOAS morning and evening values. The interpolation has to be done using a chemistry model as presented by Tack et al (2015).…”

Shipborne MAX-DOAS measurements for validation of TROPOMI NO<sub>2</sub> products

“…• and 80 • of the SZA with an uncertainty of 10%. Using the uncertainty estimation method presented by Tack et al (2015), in total, we estimate that the uncertainty of the stratospheric NO 2 VCD is about 4 × 10 14 molec cm −2 and 1 × 10 14 molec cm −2 at SZA of 20 • and 80 • , respectively.…”

“…The interpolation has to be done using a chemistry model as presented by Tack et al (2015) A scatter plot of TROPOMI versus MAX-DOAS stratospheric NO 2 VCDs for all the cruises is shown in Fig. 7.…”