Troposphere / Air Quality 1

Satellite measurements of atmospheric carbon monoxide (CO) provide a signature for biomass burning and anthropogenic combustion-related pollution emissions. CO plays an important role in both air quality and climate as a precursor for tropospheric ozone and as a major sink of OH, the atmospheric “detergent” that affects the lifetime of methane and other pollutants. Worden et al., [2013] showed decreasing global CO values in time series of satellite total column CO measurements over the past decade. All of the satellite instruments that measure CO in the thermal infrared showed consistent inter-annual variability due to fires and possibly the global recession in late 2008. Observed decreases in CO over N. America and Europe were consistent with expected decreases in CO emissions inventories [Granier et al., 2011], however, the decrease is not uniform globally. In particular, Africa shows regions with negligible trends and potentially increasing trends in CO concentration.

Here we examine 13-year time series for surface and total column CO concentrations in Africa over 2002-2014 using MOPITT V5J data. Our hypothesis is that temporal changes in CO will have different signatures related to anthropogenic and biomass burning emissions. We use singular value decomposition (SVD) with time series from different regions based on vegetation type and population density to diagnose the dominant drivers of  inter-annual variability.

To fully exploit the 10 years observation of ESA’s Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY) from 2002 until April 2012, we discuss the potential to infer carbon monoxide (CO) total column densities from cloudy observations over land and ocean scenes. About 80% of all SCIAMACHY observations are affect by clouds and so restricting the SCIAMACHY CO data product to clear sky observations represents a major constraint. Therefore, we employ the SICOR algorithm to retrieve cloud information from CH₄ absorption together with the CO vertical column densities using the 2310-2338 nm SCIAMACHY reflectance measurements. Here we assume that the atmospheric concentration of methane is known a priori using dedicated CH₄ model fields from a chemical transport model. In comparison with SCIAMACHY's clear-sky only retrievals, the validation with ground-based FTIR stations near the coast of the NDACC and TCCON network shows significantly improved, monthly averaged CO when including scenes with low clouds because of a better spatiotemporal coverage. Moreover, considering monthly CO data ensembles with clouds at different altitudes, information on the averaged CO abundance in the upper and lower troposphere can be extracted. This approach is demonstrated using SCIAMACHY data for selected regions. But the study results also mean a real data test of the SICOR algorithm, which is developed for the TROPOMI data processing.

In 2016, the Sentinel 5 Precursor mission will be launched with the TROPOMI instrument as its single payload. It will deliver daily global measurements of the atmospheric composition for air quality and climate application as part of the Copernicus atmospheric services.  In this presentation, we focus on the measurements of the shortwave infrared (SWIR) spectral range providing global distributions of CH4 and CO. We discuss the operational data processing of the SWIR trace gases including an estimate of the data quality based on simulated measurements. The main challenge is to account for scattering by water clouds, cirrus and tropospheric aerosols without exceeding the computational constraints of the processing facility. For this purpose we have developed the full-physics retrieval algorithms RemoTeC for CH4 and SICOR for CO. To demonstrate the maturity and heritage of the algorithms, we show applications to GOSAT and SCIAMACHY measurements and the verification with on-ground measurements.

The growth of mega-cities leads to air quality problems directly affecting the citizens. With satellite measurements becoming of higher quality and quantity, satellite instruments can more accurately retrieve the enhanced air pollutant concentrations over large cities. The aim of the project is to investigate the use of satellite retrievals of carbon monoxide to quantify point sources, such as large cities and industrial centres and estimate emission trends. The currently operational MOPITT instrument has been measuring carbon monoxide for more than 15 years. Previous analyses provided a proof of concept regarding the space borne detection of large point sources and estimation of emissions (Pommier et al., 2013). As a starting point we repeated this study (Pommier et al., 2013) with the latest MOPITT version 6 data and found significant differences, especially in the trend analyses over time. To improve the reliability of the emission estimation, we simulate MOPITT retrievals using the Weather Research and Forecast model with chemistry core (WRF-chem). The difference between model and measurements is used in a second step to optimize the emissions in WRF-chem. As a first case study, the method has been applied to the city of Madrid, Spain. The WRF-chem simulations show a reasonable agreement between the yearly averaged model output and satellite measurements (R²=0.778). Optimization, using Brent’s method to find the minimum cost function, increases the emission estimation of Madrid by 22%-43%.

Overall our study shows that caution is needed when analysing trends in emission or estimating sources of CO that are near the detection limit of MOPITT, especially in areas with seasonally varying cloud cover. When considering this, however, and applying the right corrections, we think it is possible to make a reliable estimation of city emissions based on satellite observations. As soon as data from the Sentinel 5P Tropomi instrument are available, we will apply our estimation techniques to the new data and it is expected that, due to the higher spatial resolution of the Tropomi data, emissions estimates can become even more precise.

Pommier, M., McLinden, C. A. and Deeter, M. N.: Relative changes in CO emissions over megacities based on observations from space, Geophys. Res. Lett., 40(14), 3766–3771, doi:10.1002/grl.50704, 2013.

In the last three decades the capabilities of measuring the atmospheric composition from space did grow tremendously with ESA’s ENVISAT and NASA’s Eos-Aura satellite programmes. The potential to operationally monitor the atmospheric composition, like the meteorological community is doing for the physical parameters, is now within reach. At the same time, the importance for society of operational environmental monitoring, related to the ozone layer, air quality and climate change, became apparent.

The Ozone Monitoring Instrument (OMI), launched on board of NASA’s EOS-Aura spacecraft in on July 15, 2004, provides unique contributions to air quality monitoring from Space. The combination of urban scale resolution (13 x 24 km² in nadir) and daily global coverage proved to be key features for the air quality community. The OMI data is currently used operationally for improving the air quality forecasts, for inverting high-resolution emission maps, UV forecast and volcanic plume warning systems for aviation. Due to its 12 year continuous operation OMI provides the longest NO₂ record from space, which is essential to understand the changes to emissions globally. Due to the extremely low degradation of the instrument, it provides an unique data record that is excellent for trend analyses and combination with other satellite instruments.

In 2016 Tropospheric Monitoring Instrument (TROPOMI), the successor of OMI, will be launched on board ESA’s Sentinel 5 Precursor satellite. TROPOMI will have a spatial resolution of 7x7 km² in nadir; a more than 6 times improvement over OMI. The high spatial resolution serves two goals: (1) emissions sources can be detected with even better accuracy and (2) the number of cloud-free ground pixels will increase substantially. In addition TROPOMI adds also addition spectral channels that allow for better cloud corrections, as well as the retrieval of carbon monoxide and methane. TROPOMI will be an important satellite mission for the Copernicus atmosphere service. TROPOMI will play a key role in the Air Quality Constellation, being the polar instruments that can link the 3 GEO UVN instruments, Sentinel 4, TEMPO and GEMS.

An overview of air quality applications and emission inventories, with an emphasis on trend analyses will be given, based on the excellent 12 years OMI data record. An outlook will be presented on the potentials of the TROPOMI instrument in the air quality domain.