Output details
12 - Aeronautical, Mechanical, Chemical and Manufacturing Engineering
Imperial College London
A geostatistical and probabilistic spectral image processing methodology for monitoring potential CO2 leakages on the surface
Methodology for detecting CO2 leakages at terrestrial surfaces developed in a €5M EU Project in collaboration with BGS-UK, OGS-Italy and BGR-Germany. Uses airborne spectral data and was extended to spaceborne spectral data assessment [ENERGY PROCEDIA, 2011, Vol.4,3421-3427]. The geostatistical image filtering methodology combined with Independent Component Analysis and parametric Reed-Xiaoli anomaly detection indicates seepage points at the Latera and Laacher See natural analogue sites [ENERGY PROCEDIA, 2013, 37, 4057-4064]. Exploiting both spatial and spectral domain correlations improves accuracy significantly and provides for a robust CO2 leakage monitoring methodology. It led to new collaboration with JGI Inc. Japan on InSAR research.