Atmospheric Applications / Data Assimilation 1

Satellite sulfur dioxide (SO₂) measurements from the Ozone Monitoring Instrument (OMI) satellite sensor processed with the new Principal Component Analysis (PCA) algorithm, averaged over a period of several years, are used to detect large point emission sources. The total of about 500 continuously emitting point sources releasing from about 30 kT y^-1 to more than 4000 kT y^-1 of SO₂ per year have been identified and grouped according to the country and the source origin: volcanic, power plants, smelters, and sources related to the oil and gas industry. The annual emissions from these sources were estimated using a new algorithm that combines satellite measurements with wind data. For volcanic sources, the study focused on continuous degassing and emissions from explosive eruptions were excluded. Emissions from degassing volcanic sources account for about 30% of total emissions, but that fraction has increased in recent years. Global cumulative emissions from power plants and smelters are declining, while emissions from oil and gas industry remained nearly constant. Anthropogenic emissions from the USA declined by 80% as well as emissions from Western and Central Europe, while emissions from India nearly doubled, and emissions from other largest SO₂-emittimg regions (South Africa, Russia, Mexico, and Middle East) remained fairly constant. In total, OMI-based estimates account for about a half of total reported anthropogenic emissions, the remaining half is likely related to sources emitting less than 30 kT y^-1 and not seen by OMI. The obtained emission information can be used to improve available emissions inventories, since some of the sources seen by OMI are not included in the inventories.  The catalogue could be particularly useful for cross-validation of polar-orbiting satellite instruments such as TROPOMI which is planned for launch on ESA’s Sentinel 5 Precursor satellite in 2016, with the data from three new generation geostationary satellites over North America, Europe, and Asia.

Emission inventories are developed for use in scientific applications and as input in urban, regional, continental or global scale air quality models. Furthermore, emission estimates are used by policy makers in order to evaluate progress towards emission abatement measures and to decide on future strategies.
Within the GlobEmission project (part of the Data User Element programme of ESA) emission estimates are developed from satellite observations of air constituents. The main advantage of space-based emission estimates are the spatial consistency, high temporal resolution and the rapid availability of these estimates to the user. The emission estimates are developed for specific applications of users that are involved in the project. A few examples of these applications are air quality modeling over China, monitoring emissions of the oil industry in the Middle East, isoprene emissions from the biosphere and the impact of fires. In this presentation examples will be given of those applications.

The El Niño/Southern Oscillation (ENSO) phenomenon is an important pacemaker for interannual variability in the Earth's atmosphere. ENSO impacts on trace gases have been observed and modelled for the stratosphere and the troposphere. However, unambiguous attribution is often difficult due to the limited length of homogenous observational records and thus long-term (decadal) trends are sometimes difficult to detect. Generally ENSO impacts in low latitudes are easier to detect, because the response emerges close (temporally and spatially) to the forcing. Moving from low to high latitudes it becomes increasingly difficult to isolate ENSO driven variability, due to time-lags involved and many other modes of variability playing a role as well. Here, we use a nudged version of the EMAC chemistry-climate model to evaluate ENSO impacts on trace gases over the last 35 years (a so-called Ref-C1SD integration) and contrast the nudged model with its free running counterpart. We use water vapour and ozone observations from the MIPAS instrument on ENVISAT from 2002 to 2012 to test the model performance. Using lagged correlations for the longer model time-series we trace the ENSO signal from the tropical lower troposphere to the polar lower and middle stratosphere and provide a framework for simple attribution of the ENSO signal in trace gases. This concise characterisation of the ENSO impact on trace gases aids improved trend detection in temporally limited time series.

Carbon monoxide (CO) is an important trace gas for understanding air quality and atmospheric composition. It is a good tracer of pollution plumes and atmospheric dynamics. In this presentation we use both the IASI and the MOPITT data to study the global and regional CO distributions as seen from space in the thermal infrared spectral range. With two IASI instruments flying on MetOp-A and MetOp-B, any location on Earth is now observed at least four times per day. All cloud free observations are analysed in near real time mode. 

 IASI and MOPITT data are jointly assimilated in the Copernicus Atmospheric Monitoring Service to generate CO pollution forecasts. We will discuss differences at the hemispheric and regional scales, and study trends over the overlapping period. Local pollution events will also be presented, and the sensitivity of the IASI observations at surface level will be discussed. Advices on how to access and how to optimize the use of this huge dataset will be provided.