MULTI-COMPONENT NEAR-SURFACE SEISMIC REFLECTION EXPERIMENTS IN A GLACIGENIC RESERVOIR SYSTEM

Near-surface, glacigenic clastic depositional systems comprise one of the most spatially variable assemblages of geologic materials present in shallow crustal regions. The natural sedimentologic variability of glacigenic systems, combined with sparse and uncertain borehole data, limits both geostatistical and interpretation-based framework model accuracy. We assess the utility of seismically imaging a near-surface (<100m) glacigenic reservoir system at four locations within a complex interlobate region of Northern Indiana. Here we report the results of experiments conducted at one of the research sites. Using a hammer source, we acquire 2D high-resolution, multi-component data along a 250m profile with 12-fold, common mid-point (CMP) acquisition geometry. On field records and pre-stack gathers, we observe surface wave interference with primary reflection events that limits our ability to obtain an accurate velocity model for time-to-depth (TD) image conversion. We obtain an improved 1D velocity model by conducting high-resolution multi-channel vertical seismic profile (VSP) experiments at a 30m deep borehole tied to the 250m 2D profile. We observe anomalous arrivals, apparent coherent noise, and reverberations in the VSP wave field that we are unable to easily interpret. There are at least three possible explanations for these anomalous seismic events. One possibility is that the glacigenic deposits display poroelastic behavior. It is also possible that anomalies are associated with borehole tube waves or casing-guided waves originating at the surface. A third explanation is that near-surface conditions within the shallow unsaturated zone enhance or contribute to coherent noise reverberation. To further evaluate the root cause of VSP data anomalies, we are separating the upgoing and downgoing VSP wavefields and constructing synthetic seismograms using forward models. Despite these current challenges, we have constructed a preliminary higher-resolution velocity model with accurate depth registration that has significantly improved the interpretative value of our original 2D survey.