Preparation for Sentinel S5P and Air Quality 2

In Europe, establishing (long term) trends in pollutant emissions, and concentrations is a key part of evaluating the impact of policies. Traditionally, air pollutant concentrations  are monitored using in-situ measurement, while emissions are estimated on annual basis. Unfortunately, in both cases the methodology used are not consistent across Europe. In the past decade, studies using spaceborne instruments have illustrated that the tropospheric column of nitrogen dioxide contains valuable information about its sources, transport and sinks. Consequently, inverse modelling using satellite observations of NO₂ columns to estimate  anthropogenic NOx emissions has been extensively used.

Various studies have focused on estimating the trends in NOx emissions using the tropospheric NO₂ columns from OMI measurements, as it reaches global coverage on a daily basis with a 13 × 24 km² footprint at nadir. Across Europe, a significant negative trend of 5–6%  per year in highly industrialized areas was identified.  However, since all these studies make use of the same instrument, no information is provided on the impact of the instrument and its overpass time on the derived trends amplitude. In this study, we aim to quantify the discrepancies in the derived trends amplitude and provide significant information on the optimal overpass time to monitor the impact of the mitigation strategies across Europe. To this end, we compare trends in the anthropogenic NOx emissions derived from OMI and GOME-2 observations.

In this presentation, first, the sensitivity of each instrument to relevant emission source sectors across Europe will be shown. The NO₂ column sensitivity to source categories will allow us to better understand the observed trends.  Second, the trends derived across Western Europe from OMI and GOME-2 observations will be presented and, the discrepancies will be investigated. Finally, an attempt to reconcile the trend estimates from both instruments across Europe will be presented.

In 2016, the Tropospheric Monitoring Instrument (TROPOMI) will be launched on board of the Sentinel 5 Precursor (S5-P) mission. For this mission, the highly efficient Shortwave Infrared Carbon Monoxide (CO) Retrieval algorithm (SICOR) was developed to meet the requirements of an operational data processing. In this contribution, we present the application of the SICOR algorithm to measurements of ESA’s Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY) instrument. Both, SCIAMACHY and TROPOMI cover the 2.3 μm spectral range with the same spectral resolution but TROPOMI is distinguished by having an improved radiometric performance and a better spatial resolution. Using the same retrieval approach for both satellite instruments will ensure the comparability of the CO data sets of both missions. In the perspective of long-term data analysis, we derived a full-mission 10-year data set of CO vertical columns (2003-2012) from the short wave infrared (SWIR) measurements of SCIAMACHY. To account for SCIAMACHY's instrument degradation we used clear sky measurement over the Sahara region as a natural calibration target to determine spectral calibration, a spectral radiometric offset, the width of the instrument spectral response function for the entire mission. Finally, the 10 years CO dataset is validated with on-ground CO column measurements at several sites of the TCCON network.

The retrieval of tropospheric column densities of NO₂ requires the subtraction of the stratospheric fraction from the total columns derived by DOAS. Here we present a modified reference sector method, which estimates the stratosphere over "clean" regions, as well as over clouded scenarios in which the tropospheric column is shielded.
The selection of "clean" pixels is realized gradually by assingning weighting factors to the individual ground pixels, instead of applying binary flags. Global stratospheric fields are then compiled by "weighted convolution". In a second iteration, unphysical negative tropospheric residues are suppressed by adjusting the weights respectively. This algorithm is foreseen as "verification algorithm" for the upcoming TROPOMI on S5p.
We show the resulting stratospheric estimates and tropospheric residues for a test data set based on OMI observations. The dependencies on the a-priori settings (definition of weighting factors and convolution kernels) are discussed, and the results are compared to other products, in particular to DOMINO v.2 (based on assimilation, similar to the TROPOMI prototype algorithm) and the NASA standard product (based on a similar reference-region-type approach).

 

Nitrogen dioxide (NO₂) is on of the most prominent air pollutants and is of particular concern in densely populated urban areas. Here we present a study investigating recent trends in tropospheric NO₂ columns over 66 large urban agglomerations worldwide. These were derived using data from the SCanning Imaging Absorption spectroMeter for Atmospheric CHartographY (SCIAMACHY) instrument onboard the Envisat platform for the period August 2002 to March 2012. A seasonal model including a linear trend was fitted to the satellite-based time series over each site. The results indicate distinct spatial patterns in trends. While agglomerations in Europe, North America, and some locations in East Asia/Oceania show decreasing tropospheric NO₂ levels on the order of −5 % yr^-1, rapidly increasing levels of tropospheric NO₂ are found for agglomerations in large parts of Asia, Africa, and South America. The site with the most rapidly increasing absolute levels of tropospheric NO₂ was found to be Tianjin in China with a trend of 3.04 (±0.47) × 10¹⁵ molecules cm² yr^-1, whereas the site with the most rapidly increasing relative trend was Kabul in Afghanistan with 14.3 (±2.2) % yr^-1. In total, 34 sites exhibited increasing trends of tropospheric NO₂ throughout the study period, 24 of which were found to be statistically significant. A total of 32 sites showed decreasing levels of tropospheric NO₂ during the study period, of which 20 sites did so at statistically significant magnitudes. Our study further examines the impact of the recent economic crisis on NO₂ time series and investigates the relationship between urban NO₂ trends and changes in population growth. We find that this relationship is subject to substantial regional differences as well as influenced by economic and demographic factors.

In addition we present a methodology for directly validating satellite-based trends in NO₂ using the growing set of long-term Multi-Axis Differential Optical Absorption Spectroscopy (MAX-DOAS) stations worldwide. More specifically, MAX-DOAS instruments located in southern France and China (operated by BIRA) as well as multiple stations throughout Asia provided by the JAMSTEC network are used to provide reference information at various pollution levels. While the corresponding time series currently are still relatively short, several stations now have continuous time series of more than five years length and thus show potential to be used for direct trend validation. First results indicate that the method is feasible and provides similar trends from both data sources when the number of months with valid data is on the order of 50 or more. For stations at which this number is significantly lower, the trend uncertainties are currently too high to use them as a reference. However, these uncertainties in trends are bound to reduce rapidly with increasing time series length and it is anticipated that only 1-2 additional years of data are required to obtain statistically significant results in polluted regions. Within the next few years there are also likely to be more stations worldwide which provide operational long-term MAX-DOAS datasets with suitable record lengths for trend analysis. Once the MAX-DOAS time series are long enough for a larger number of stations worldwide it is anticipated that such as network will provide a method for directly validating the tropospheric NO₂ trends obtained from satellite-based platforms without the need for indirect validation using models or similar techniques, which introduce significant amounts of additional uncertainty.

A new version of the NO2 retrieval for OMI - the KNMI DOMINO retrieval code, version 3 - has been recently developed. The modifications in this algorithm compared to the current DOMINO version 2 product will be discussed in our contribution. The spectral fitting to derive the DOAS slant column for NO2 has been revised as explained in the paper by van Geffen et al., AMTD 2014, and the main modifications will be reviewed. The TM4 chemistry-transport model used for DOMINO-2 is upgraded to the latest TM5 release, and the resolution is increased from 3x2 degree to 1x1 degree. Further aspects which were updated include the temperature dependence of the O2-O2 cloud retrieval, the error estimation and terrain height treatment, and newly developed air-mass factor lookup tables. The estimate of the stratospheric NO2 column by means of data assimilation in TM5 is improved. Various aspects of these developments have been verified through intercomparisons with other retrieval approaches within the TROPOMI verification team and the European QA4ECV project. Finally we will discuss the future application of DOMINO-3 for TROPOMI.

The TROPOspheric Monitoring Instrument (TROPOMI) will be launched in 2016 onboard the ESA Sentinel-5 Precursor (S-5P) satellite. It will provide global and daily observations of atmospheric trace gases, with an unprecedented spatial resolution of 7x7 km². Formaldehyde (HCHO) is a central molecule of tropospheric chemistry. Its observation allows for the determination of non-methane hydrocarbon emissions, which are precursors of tropospheric ozone and biogenic aerosols and influence the oxidizing capacity of the troposphere. Measurements from TROPOMI will improve the monitoring capability of HCHO from anthropogenic and natural emissions, and will extend the long-term datasets from past and existing UV sensors (GOME, SCIAMACHY, OMI, GOME-2, OMPS).

Within the S-5P project, BIRA-IASB is in charge of the development of the prototype HCHO level-2 algorithm, while IUP-Bremen leads verification tasks.  We present the algorithm baseline and results from its application to the complete series of OMI and GOME-2 measurements. This includes the validation of the resulting data products using MAX-DOAS measurements available in China, Burundi and Europe. We also present a global trend analysis of the tropospheric HCHO columns between 2004 and 2014. Furthermore the specificities of the TROPOMI algorithm are discussed in more details, in particular the impact of using state-of-the-art TM5 profiles at the spatial resolution of 1°x1° as a priori in the air mass factor calculation. Finally results from the scientific verification based on synthetic spectra and on OMI test data are discussed.