GNGTS 2016 - Atti del 35° Convegno Nazionale

GNGTS 2016 S essione 3.3 583 A Sensitivity Study of different marine-CSEM acquisition geometries G. Bernasconi 1 , G. Gentili 1 , L. Speranza 2 , B.Garcea 2 , J.Suffert 3 1 Politecnico di Milano, Italy 2 Edison, Milano, Italy 3 Edison Norge, Stavanger, Norway Introduction . EM geophysical methods are becoming more important in reducing interpretation ambiguities of seismic images, by adding information about the conductivity of the subsurface. These methods can be divided into many different types based on different criteria, such as the source or array configuration, the EM waveform, or the interpretative parameter. As a first general subdivision, we can distinguish between EM based on natural or on artificial sources. EM methods that uses natural sources falls under the category of Magnetotellurics (MT). The electromagnetic methods using an antenna as source of the electromagnetic field can be considered “Controlled Source Electro-Magnetic” methods (CSEM). In the marine CSEM (mCSEM) the transmitting antenna (TX), pulled by a vessel, sends a very low frequency EM in the subsea sediments, and a network of receivers (RX) collects the scattered field, revealing the presence of resistive bodies (Constable, 2010). In shallow water (< 500m) two mCSEM data acquisition systems are commonly used: the SeaBed Logging (SBL) deploys the receivers on the sea bottom while the newer Towed Streamer Electromagnetic (TSEM) system uses a line of receivers that are pulled behind the transmitting antenna. For local analysis also fixed vertical transmitters and receivers dipoles are deployed (Vertical Electric Dipoles or VED CSEM). This paper presents an experimental study on the sensitivity of the different acquisition layouts with respect to the polarization of the antennas and the position of an anomaly in the subsurface, for a realistic scenario. The study is useful to understand the potential and the limitations of each methods technology. In addition, it has been used to design innovative and effective strategies of resistivity inversion: we propose here a layer stripping inversion approach for the towed streamer (TSEM) geometry. Model . Recently more than 11000 km 2 of Towed Streamer ElectroMagnetic data has been acquired in new regions in SE Barents Sea. Bathymetry is very stable around 280 m, sea water conductivity is 3.2 S/m, sea bed conductivity is around 0.07 S/m. We use this shallow water average model for the case study analysis, as it is compatible with all the described acquisition geometries. We simulate a SBL geometry with receivers on the seabed and the transmitting antenna travelling at 250 m from the sea level (30 m above the sea bottom). For the TSEM geometry the transmitter and the 72 receivers are placed respectively at 10m and at 100m from sea level, in accordance with the service company (Petroleum Geo-Services) acquisition configuration (Anderson et al., 2010). Streamer length is 8 km. Method . We use a 2.5D FEM (Finite Element Method) modeling code to compute the EM field at the receivers for the different geometries and a constant conductivity value (0.07 S/m) of the sediments below the sea bottom (background reference model). We evaluate the reference field at several frequencies from 0.2 Hz up to 10 Hz. We simulate then, on every point P of a grid within the sediments, a small resistive anomaly at 0.01 S/m, that mimics the presence of hydrocarbons bearing rocks. The electrical field map of the “perturbed” model can be written as the superposition of the primary reference field E 1 (computed with air, sea and seabed), and the scattered field E s (computed with air, sea and seabed), E tot =E 1 +E s =G(J 1 )+G(J S ) , Js ≅ (σ body - σ seabed )E 1 , Js is the current density induced by the primary field at the anomaly and G is the Green function (numerical or analytical). We define sensitivity the strength (log of the magnitude) of the scattered field as a function of transmitter, receiver and scatter point positions ( X TX , X RX , X P ), Sensitivity (X TX , X RX , X P ) = log 10 |E s |.