DOMINO (NIVR projectnumber 53412KN. October 2004 - December 2005)
DOMINO-2 (NIVR projectnumber 53603KN, July 2006 - June 2007)

Ruud Dirksen, Folkert Boersma, Henk Eskes, Pepijn Veefkind, Pieternel Levelt, Ronald van der A

The final report is available here (April 2008).

The purpose of the DOMINO(-2) project, (Derivation of OMI tropospheric NO2), is to develop a processing system that retrieves worldwide concentrations of nitrogen dioxide (NO2) in the troposphere (0-12 km altitude) using the measurements of the OMI satellite instrument and makes the NO2 observations available on the internet www.temis.nl within a few hours after the OMI overpass at 13:40 hrs local time.

Funded by the National User Support Programme 2001-2005 (NUSP), executed by the Netherlands Agency for Aerospace Programmes (NIVR)

Four days of OMI tropospheric NO2 column observations over Europe. Cloudy areas have been masked (grey parts in the plot). Note the reduced concentrations on Sunday 16 October 2005.

The rationale for investigating tropospheric NO2 is its important role in the chemistry of the atmosphere and in particular in its effects on air pollution. NO2 is a precursor for tropospheric ozone (smog) via photochemical reactions with hydrocarbons. Furthermore, elevated concentrations of NO2 contribute directly (it is a toxic gas) and indirectly (through particle formation) to adverse health effects including lung damage, increased susceptibility to respiratory illnesses, and aggravation of heart and lung conditions. For these reasons it is important to monitor tropospheric NO2 concentrations, to identify their sources, and to assess world wide transport processes. The main sources of tropospheric NO2 are anthropogenic activities, including industry, traffic, and biomass burning. In addition there are natural sources of troposheric NO2, including soils, lightning discharges and forest fires. All of these NOx (NO+NO2) sources have considerable uncertainties in terms of their quantities, locations, timing, and trends, and this inhibits a better understanding of air pollution and its management.

The Ozone Monitoring Instrument (OMI) is an excellent tool to monitor worldwide NO2 concentrations, as it achieves coverage of the entire Earths atmosphere in a single day with ground pixel sizes as small as 24x13 km2 at nadir. This small pixel size approaches the size of large cities, enabling the identification of individual major sources. The daily global coverage allows for a day-to-day observation of emission sources, as well as the detection and tracking of transport phenomena. Compared to its predecessors like GOME and SCIAMACHY, the improved spatial resolution of OMI together with its daily global coverage is a huge step forward and is it safe to state that OMI has set a new standard in Earth observation.

The DOMINO project has achieved daily retrievals of tropospheric NO2 in near-real time mode (within 3 h of the actual OMI measurement at 13:40). The retrieval is based on the combined retrieval-assimilation-modelling approach developed at KNMI for off-line tropospheric NO2 from the GOME and SCIAMACHY satellite instruments. We have adapted the off-line system such that the required a priori information – profile shapes and stratospheric background NO2 – is now immediately available upon arrival (within 80 min of observation) of the OMI NO2 slant columns and cloud data at KNMI. DOMINO-2 has consolidated the near-real time data delivery started during DOMINO, and improved the retrieval system. In addition to the near-real time data delivery, special attention has been paid to generating an 'offline' data product based on improved information on spacecraft parameters that becomes available after one day of measurement. Furthermore, after one day, most orbital data is typically available, so that the offline data product is the more complete of the two. The generation of an offline OMI tropospheric NO2 dataset from October 2004 until today is an important milestone as it comprises a unique, consistent long-term dataset in combination with the existing data sets that have been retrieved at KNMI with very similar algorithms from GOME (1996-2003), and SCIAMACHY (2002-today).

The DOMINO tropospheric NO2 columns have been validated versus independent measurement during various campaigns. During the INTEX-B campaign in March 2006 over the southern United States, Mexico, and the Gulf of Mexico, DOMINO tropospheric NO2 column measurements compared favourably to NO2 columns derived from coinciding in situ aircraft measurements on vertical spirals. Good correlation with no significant bias (r-square=0.67, slope = 0.99 ± 0.17, n = 12) was found for the ensemble of comparisons when the aircraft could spiral sufficiently low to sample most of the NO2 column. During the DANDELIONS campaign in May-June 2005, and September 2006 at Cabauw, the Netherlands, OMI data was validated versus various ground-based measurements. NO2 from OMI showed good agreement with two independent ground-based techniques that provide tropospheric and total-column NO2 measurements.

A comparison of OMI (13:30 hrs overpass) tropospheric NO2 columns with SCIAMACHY (10:00 hrs overpass) revealed overall consistency between the two measurement sets. OMI and SCIAMACHY observe very similar distributions of NO2 and correlate well. But the comparison also showed that over the polluted regions of the world, for cloud-free situations, SCIAMACHY observes 30% higher tropospheric NO2 columns at 10:00 hrs local time than OMI at 13:30 hrs. In contrast, over biomass burning areas in the tropics, SCIAMACHY observes lower tropospheric NO2 columns than OMI. In both cases, these differences can be explained by the differences between instrument overpass times. Over the industrial regions, SCIAMACHY detects relatively high NO2 concentrations because its 10:00 hrs overpass time ensures the sampling of relatively fresh morning rush-hour NOx emissions. OMI detects lower NO2 concentrations at 13:30 hrs largely due to photochemical NO2-loss that is strongest during midday. Over the biomass burning regions, OMI detects higher NO2 concentrations because agricultural practices there lead to fires that burn most frequently and intensely in the (early) afternoon. Before DOMINO, there were strong doubts whether satellite retrievals would ever be accurate enough to detect boundary-layer chemistry on timescales of a few hours. With the DOMINO dataset, making use of the uniquely consistent SCIAMACHY and OMI retrieval algorithms, the DOMINO-team has now shown that this is in fact possible.

An important contribution to the success of DOMINO is the fact that data and images are freely available to scientists and to the public on internet. It was decided to link the DOMINO data sets to the TEMIS website, www.temis.nl. This highly successful TEMIS data portal has largely increased the visibility and impact of the DOMINO NO2 data sets. Traffic logs of this website show that there is a huge, worldwide interest in both the data and the images. This easy availability of the data has made the tropospheric NO2 product a true eye-catcher of the OMI project.

A cost benefit analysis for the DOMINO project demonstrates that DOMINO provides indirect but nevertheless potentially enormous benefits related to DOMINO contributing to a stronger presence of the air pollution problem in the public debate. The cost benefit analysis demonstrates that even very small changes in air pollution concentrations, whether they result from political measures or voluntary changes in citizen-behaviour, lead to dramatic economic savings because of reduced premature mortality. Our estimates are derived from literature- based estimates of the direct beneficial health effects of reducing air pollution in terms of reduced premature mortality and the economic value of a 'human life lost' (on the order of 1-2 million Euro). Clearly these estimates are speculative at best. However, they illustrate that even minor effects of DOMINO can lead to benefits that easily exceed the 160 kEuro project costs of DOMINO-2 by one or two orders of magnitude.