ALBERT

Description: [1]  The causes of renewed growth in the atmospheric CH 4 burden since 2007 are still poorly understood and the subject of intensive scientific discussion. Here, we present a reanalysis of global CH 4 emissions during the 2000s, based on the TM5-4DVAR inverse modeling system. We use high-accuracy surface observations from the NOAA Earth System Research Laboratory global cooperative air sampling network for 2000–2010, together with retrievals of column-averaged CH 4 mole fractions from the Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY) instrument onboard ENVISAT from 2003 onwards. [2]  Using climatological OH fields, derived global total emissions for 2007–2010 are 16–20 Tg CH 4 /yr higher than for 2003–2005. Most of the inferred increase was located in the tropics (9–14 Tg CH 4 /yr) and mid-latitudes of the northern hemisphere (6–8 Tg CH 4 /yr), while no significant trend was derived for Arctic latitudes. The atmospheric increase can be attributed mainly to an increase of anthropogenic emissions. However, the derived trend in anthropogenic emissions is significantly smaller than the one estimated in the EDGARv4.2 emission inventory. Superimposed on the increasing trend in anthropogenic CH 4 emissions are significant inter-annual variations (IAV) of CH 4 emissions from wetlands (up to ±10 Tg CH 4 /yr), and biomass burning (up to ±7 Tg CH 4 /yr). [3]  Various sensitivity experiments have been performed to investigate the impact of the SCIAMACHY observations (compared to inversions using only surface observations), of the OH fields used, and of a priori emission inventories on the derived CH 4 emission trends and their inter-annual variability. Despite significant differences among these sensitivity experiments in their latitudinal attribution of IAV of CH 4 emissions, they show a reasonably consistent picture regarding the IAV aggregated on larger latitude bands. Furthermore, all sensitivity experiments show very similar performance against a comprehensive independent dataset of observations used for validation, including NOAA ship and aircraft profile samples, HIPPO aircraft transects, and CARIBIC aircraft data. Comparison of model simulations with BARCA aircraft measurements, however, show significantly better agreement in the free troposphere over the Amazon for the inversions using the SCIAMACHY and NOAA surface observations compared to inversions using only the surface observations, demonstrating the usefulness of the satellite measurements to better constrain tropical CH 4 emissions.