摘要：This work presents the latest release (v9.0) of the University of Leicester GOSAT Proxy XCH4 dataset. Since the launch of the GOSAT satellite in 2009, these data have been produced by the UK National Centre for Earth Observation (NCEO) as part of the ESA Greenhouse Gas Climate Change Initiative (GHG-CCI) and Copernicus Climate Change Services (C3S) projects. With now over a decade of observations, we outline the many scientific studies achieved using past versions of these data in order to highlight how this latest versionmay be used in the future.We describe in detail how the data are generated, providing information and statistics for the entire processingchain from the L1B spectral data through to the final quality-filtered column-averaged dry-air mole fraction(XCH4) data. We show that out of the 19.5 million observations made between April 2009 and December 2019,we determine that 7.3 million of these are sufficiently cloud-free (37.6 %) to process further and ultimately obtain4.6 million (23.5 %) high-quality XCH4 observations. We separate these totals by observation mode (land andocean sun glint) and by month, to provide data users with the expected data coverage, including highlightingperiods with reduced observations due to instrumental issues.We perform extensive validation of the data against the Total Carbon Column Observing Network (TCCON),comparing to ground-based observations at 22 locations worldwide. We find excellent agreement with TCCON,with an overall correlation coefficient of 0.92 for the 88 345 co-located measurements. The single-measurementprecision is found to be 13.72 ppb, and an overall global bias of 9.06 ppb is determined and removed from theProxy XCH4 data. Additionally, we validate the separate components of the Proxy (namely the modelled XCO2and the XCH4/XCO2 ratio) and find these to be in excellent agreement with TCCON.In order to show the utility of the data for future studies, we compare against simulated XCH4 from the TM5model. We find a high degree of consistency between the model and observations throughout both space andtime. When focusing on specific regions, we find average differences ranging from just 3.9 to 15.4 ppb. Wefind the phase and magnitude of the seasonal cycle to be in excellent agreement, with an average correlationcoefficient of 0.93 and a mean seasonal cycle amplitude difference across all regions of −0.84 ppb.These data are available at https://doi.org/10.5285/18ef8247f52a4cb6a14013f8235cc1eb (Parker and Boesch,2020).
