By land, sea and air (and space): Verifying UK methane emissions at a range of scales by integrating multiple measurement platforms

Abstract:

The Greenhouse gAs Uk and Global Emissions (GAUGE) programme and Department of Energy and Climate Change (DECC) network aim to quantify the magnitude and uncertainty of UK greenhouse gas (GHG) emissions at a resolution and accuracy higher than has previously been possible. The on going DECC tall tower network consists of three sites, and an eastern background site in Ireland. The GAUGE project includes instruments at two additional tall tower sites, a high-density measurement network over agricultural land in eastern England, a ferry that performs near-daily transects along the east coast of the UK, and a research aircraft that has been deployed on a campaign basis. Together with data collected by the GOSAT satellite, these data represent the GAUGE/DECC GHG measurement network that is being used to quantify UK GHG fluxes. As part of the wider GAUGE modelling efforts, we have derived methane flux estimates for the UK and northwest Europe using the UK Met Office NAME atmospheric transport model and a novel hierarchical Bayesian “trans-dimensional” inversion framework. We will show that our estimated fluxes for the UK as a whole are largely consistent between individual measurement platforms, albeit with very different uncertainties. Our novel inversion approach uses the data to objectively determine the extent to which we can further refine our national estimates to the level of large urban areas, major hotspots or larger sub-national regions. In this talk, we will outline some initial findings of the GAUGE project, tackling questions such as: At what spatial scale can we effectively derive greenhouse gas fluxes with a dense, multi-platform national network? Can we resolve individual metropolitan areas or major hotspots? What is relative impact of individual stations, platforms and network configurations on flux estimates for a country of the size of the UK? How can we effectively use multi-platform observations to cross-validate flux estimates and determine likely errors in model transport?