The storage of CO2 in deep geological formations is one of the chief technological means of reducing anthropogenic emissions of CO2 to the atmosphere. The process requires capturing CO2 at source (e.g. coal-fired power plants), transporting CO2 to the injection site, and pumping liquefied CO2 into kilometre deep, porous reservoirs that are typically initially saturated in saline water or previously contained oil or gas. Initially, buoyant CO2 tends to rise through the porous reservoir until it is trapped by an impermeable horizon, in the same way that oil or gas has been trapped over millennia. Subsequently, buoyant CO2 may be more securely trapped by dissolving CO2 into water (carbonated water is more dense than non-carbonated water and will sink), or by capillary forces acting to hold the CO2 in the small confines of the pore space. Any risk of buoyant CO2 migrating through the overburden is therefore reduced by these trapping processes. Constraining the rates of dissolution and capillary trapping in realistic geological overburden is a key component of strategies to quantify and reduce the risks of leakage. The UK is geologically well placed to implement offshore CO2 storage, with many potential reservoirs in the North Sea. This proposal will improve our understanding of the risks of leakage through the overburden by quantifying trapping rates in faults and heterogeneous strata typical of the overburden of North Sea reservoirs, and by quantifying our ability to seismically detect any CO2 in the overburden. CO2 is less viscous than water and will finger along more permeable layers. Sedimentary strata exhibit large variations in permeability on all scales that will substantially increase the rates at which CO2 dissolves in the formation waters. The analysis, while general in scope and resultant techniques, is applied to the Goldeneye field, a target for CO2 storage and a candidate for the Government’s CCS commercialisation competition. Our approach is to geologically characterise the relevant geological heterogeneity within the overburden, and to map the structure and propensity for fluid flow within faults in that locality. Drill core provides samples of rock (5×20 cm) that can then be interrogated in the laboratory. We will directly image, at conditions typical of the overburden, the rates of fluid flow, dissolution, and capillary trapping both at the scale of individual pores within the rock (microns) and over the length of the core (centimetres). Geochemical analysis of the fluids will allow us to measure in situ dissolution and precipitation rates in our core flooding experiments. In order to determine how rates of flow and trapping may be applied at the scale of the reservoir and overburden the results must be interpreted in light of flow through 1-100 centimetre scale geological heterogeneities and along faults. To assess the impact of heterogeneities on the rates of trapping we will construct simplified models of flow along predominantly layered strata, or along cross-cutting faults, along with laboratory analogue experiments in which we can optically assess trapping rates and thereby provide a firm benchmark for our predictions. Finally, at larger scales, we will image flow up chimney structures in existing CO2 experiments (eg Sleipner in the North Sea) and thus provide quantitative estimates of our ability to seismically resolve leakage pathways in the storage overburden. Our proposal will develop tools needed to geologically characterise the North Sea overburden, provide quantitative estimates of trapping rates in geologically complex overburden and fault complexes, and demonstrate the ability to seismically resolve fluid flow pathways. To date geological CO2 storage has been demonstrated at relatively safe storage sites. This work would greatly expand the potential for geological CO2 storage by quantifying the potential risks associated with leakage in more geologically complex storage sites.