Observing the Distribution of Atmospheric Methane from Space

Methane (CH4) is a potent greenhouse gas with a radiative forcing efficiency 21 times greater than that of carbon dioxide (CO2) and an atmospheric lifetime of approximately 12 years. Although the annual global source strength of CH4 is fairly well constrained, the temporal and spatial variability of individual sources and sinks is currently less well quantified. In order to constrain CH4 emission estimates, inversion models require satellite retrievals of XCH4 with an accuracy of < 1-2%. However, satellite retrievals of XCH4 in the shortwave infrared (SWIR) are often hindered by the presence of atmospheric aerosols and/or thin ice (cirrus) clouds which can lead to biases in the resulting trace gas total column of comparable magnitude. This thesis aims to quantify the magnitude of retrieval errors caused by aerosol and cirrus cloud induced scattering for the Full Spectral Initiation Weighting Function Modified Differential Optical Absorption Spectroscopy (FSI WFM-DOAS) retrieval algorithm. A series of sensitivity tests have been performed which reveal that a) for scenes of high optical depth, accurate aerosol a priori data is required to reduce retrieval errors, b) retrieval errors due to aerosol and ice cloud scattering are highly dependent on surface albedo, SZA and the altitude at which scattering occurs and c) errors induced in global retrievals by the presence of ice clouds (up to ~ 35%) are significantly greater than those owing to aerosols (~ 1-2%). Cloud filtering is therefore important even when employing proxy methods. Furthermore, the original FSI WFM-DOAS V2 algorithm (OFSI) has been successfully modified with improved a priori albedo and aerosol, resulting in two new versions of the retrieval: MFSI and GFSI. Initial comparison of OFSI, MFSI and GFSI retrievals of XCH4 over North America show minor improvements in retrieval error, however further comparison over regions of high optical depth are required.