Title: Evaluation of Copernicus Atmosphere Monitoring Service methane products
Publisher: Publications Office of the European Union
Publication Year: 2018
JRC N°: JRC112816
ISBN: 978-92-79-93409-4 (online),978-92-79-93410-0 (print)
ISSN: 1831-9424 (online),1018-5593 (print)
Other Identifiers: EUR 29349 EN
OP KJ-NA-29349-EN-N (online),KJ-NA-29349-EN-C (print)
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC112816
DOI: 10.2760/906932
Type: EUR - Scientific and Technical Research Reports
Abstract: The Copernicus Atmosphere Monitoring Service (CAMS) provides continuous data and information on atmospheric composition in an operational mode. The CAMS products include analyses/re-analyses of the greenhouse gases (carbon dioxide, methane, and nitrous oxide) for recent years. In this report, we evaluate the quality of the CAMS methane (CH4) products, focussing on the "near real time analyses" of the atmospheric CH4 concentrations (from the ECMWF IFS assimilation system) and the re-analyses of CH4 concentrations and fluxes (from the TM5-4DVAR inverse modelling system, provided by TNO / SRON). The CAMS CH4 products are compared to comprehensive independent observational data sets (from surface monitoring stations, ship cruises, various aircraft programmes, AirCore balloon soundings up to the middle stratosphere, and measurements of column averaged CH4 mole fractions) during 2010-2017. Furthermore, CH4 flux inversions from the JRC TM5-4DVAR system (which was used as prototype of the operational CAMS inversion system) are included in the analysis, providing a benchmark to evaluate the CAMS CH4 flux inversion products. Overall, the CAMS and JRC inversions show very similar performance and compare well to observations over remote regions near the surface and within the free troposphere, confirming that in general CH4 mole fractions in the background troposphere far from CH4 emissions are realistically simulated. Due to the relatively coarse horizontal resolution of 3o (longitude) x 2o (latitude), however, both CAMS and JRC inversions show clear limitations in simulating regional surface monitoring stations (which are influenced by regional CH4 emissions), in most cases underestimating measured CH4 mole fractions in these areas. Furthermore, the inversions show large differences to the observed CH4 mole fractions in the lower to middle stratosphere at mid to high latitudes, most likely due to shortcomings in simulating the transport and/or chemistry in the stratosphere and the stratospheric-tropospheric exchange. In contrast to the flux inversions, the CAMS "near real time analyses" show generally large biases (varying in space and time) in the simulated CH4 mole fractions compared to observations in the background troposphere. These large biases are probably mainly due to the assimilation strategy of including only satellite retrievals (but no surface observations) and potential biases in the assimilated satellite products, while the flux inversions assimilate satellite retrievals and surface observations simultaneously and thus correct for biases in the satellite data (along with potential biases of the models to simulate the stratosphere). The surface CH4 fluxes derived from the CAMS inversion system are in general similar to the JRC estimates. However, the latitudinal gradients of the fluxes are slightly different between the two inversion systems, probably in part due to the two different convection schemes applied.
JRC Directorate:Energy, Transport and Climate

Files in This Item:
File Description SizeFormat 
jrc112816_copernicus_ch4_validation_report_v15.pdf14.51 MBAdobe PDFView/Open

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.