When reservoirs leak, varying quantities of oil and gas migrate to the surface as macroseeps, which are visible, and microseeps, which are invisible. This review article describes the mechanisms of seepage and the resulting surface manifestations in relation to optical highβresolution remote sensing data. Oil pools and tar deposits (macroseeps) often can be detected directly by remote sensing. Microseeps are more difficult to study using remote sensing, but they give rise to vegetation stress, and cause geochemical alterations in soil and rocks, which can be studied indirectly using hyperspectral sensors. An integrated methodology is presented to combine various geoscience and remote sensing datasets for seepage detection. A combination of redβedge modelling algorithms and spectral matching tools is identified that provides a validated technique for onshore microseepage detection. These remote sensing tools are not only important for petroleum exploration, but also have environmental implications because seeps emit greenhouse gases. A statistical data integration approach was developed based on Bayesian assumptions, which can be used to integrate hyperspectral remote sensing data, other satellite remote sensing data, and ancillary field geological and geochemical datasets, for modelling microseepage. In a case study from the Ventura basin (Santa Barbara) in Southern California, Probeβ1 data from the 1998 Geosat Group Shoot are integrated with field and subsurface geological and geochemical data to predict possible sites of hydrocarbon microseepage.