Imaging Below Gas Clouds Through Non-linear Full Waveform Inversion of the Reflection Response – The 2D Case

Imaging below anomalies like gas clouds suffers from the imprint of the complex wave propagation through such an overburden. Traditional imaging method do not have the capability to correct for such imprint. Therefore, we aim at estimating multi-dimensional full waveform transmission operators for propagation through the gas cloud, obtained from an effective model of the gas cloud by inversion of the gas cloud reflection response. For the 1,5D case, it was demonstrated that a non-linear full waveform inversion process can indeed provide an effective model of the gas cloud such that the transmission imprint on the reflection response can be removed. In this paper we show the extension to the full 2D case of the non-linear full waveform inversion to multi-dimensional inversion using global optimization process and multi-dimensional transmission deconvolution results. From the estimated contrasts that effectively describe the reflection response of the gas cloud, transmission correction operators can be derived by forward modeling. When using these transmission operators in a multi-dimensional, full waveform deconvolution process, true amplitude imaging of the reflections below this overburden can be obtained. The method is demonstrated for a 2D synthetic dataset with a realistic gas cloud model.