High-resolution inversion of methane fluxes using in-situ and satellite measurements over Africa

<p><span>Due to the growth of atmospheric methane and the remarkable uncertainties in its budget, there is a need to constrain the sources and sinks of this atmospheric tracer. Among the different regions emitting methane to the atmosphere, Africa has a considerable contribution to the global methane budget and has one of the largest associated uncertainties. Surface-based measurements could help to substantially diminish these uncertainties, but given the dearth of in-situ observations in most areas of Africa, satellite retrievals can be helpful to fill this gap.&#160;We use ground-based observations along with the SRON&#8217;s proxy and full physics XCH</span><span>₄</span><span> products from the Greenhouse Gases Observing Satellite (GOSAT), and an inversion system consisting of the global 3-dimensional transport model TM3 as well as the high-resolution regional Stochastic Time-Inverted Lagrangian Transport model STILT to spatially resolve the regional emissions in Africa with a focus on the natural wetlands and also the oil and gas industries in Nigeria and Angola as the top oil-producing African countries. In this study, the global inversion is performed for 2009-2015 on a coarse resolution, ~ 3.83&#730; x 5&#730;. The nominal spatial resolution of the regional inversion is 0.25&#730; x 0.25&#730; for 2011-2012. Prior fluxes for the nested run include the Emission Database for Global Atmospheric Research (EDGARv4.3.2) for anthropogenic sources including agriculture, and waste and energy, WetCHARTs wetlands ensemble, the Global Fire Assimilation System (GFAS) for biomass burning, Sanderson&#8217;s global database of termite methane emissions, and the soil Methanotrophy Model (MeMo) for the methane soil uptake.&#160;The significance of the petroleum industries&#8217; emissions is examined by high-resolution inversions. Additionally, the consistency of the optimised fluxed derived from different types of measurements (e.g. surface-only, surface+satellite, satellite-only) is investigated and discussed.</span></p>