Estimating reservoir parameters from surface data is a challenging problem, particularly when the reservoir is seated beneath a complex overburden whose properties are also unknown. Here, we present a new metric for detecting and quantifying errors in subsurface models that is well-suited for target-oriented inversion. We refer to this metric as an ''interferometric misfit'' because it relies on wavefield extrapolation from both convolution-and correlation-type reciprocity used in seismic interferometry. When the source point is outside the volume, both forward-or reverse-time extrapolation produce the same field. Although their results are the same, the physical interactions between the components of the data and of the extrapolators are different in forward or reverse time. Because of this, the forward-and reverse-time extrapolated fields are only equal when the model used is consistent with the real subsurface model. We thus use the difference between the forward-and reverse-time extrapolated fields to define a subsurface-domain metric that quantifies model errors. This misfit relies on full-waveform information from data and is nonlinear on the medium parameters. Because it is designed to sense medium parameters only inside a target volume of interest, our approach comprises a metric for target-oriented inversion in the subsurface domain.
|Original language||English (US)|
|Title of host publication||76th European Association of Geoscientists and Engineers Conference and Exhibition 2014: Experience the Energy - Incorporating SPE EUROPEC 2014|
|Publisher||EAGE Publishing BVfirstbreak@eage.org|
|Number of pages||5|
|State||Published - Jan 1 2014|