Broadband seismic: The ultimate input for quantitative interpretation?

Abstract

Seismic frequency bandwidth has always been and still is in the reservoir geophysics community an important point of discussion especially, when geoscientists are trying to estimate reservoir properties from the seismic with a high degree of accuracy away from the calibration point: the wells. Seismic reservoir characterization or quantitative interpretation is the workflow that endeavors to extract elastic attributes from the seismic amplitudes with the support of well data, petrophysics, rock physics and seismic interpretation plus any prior knowledge of the geology. Thus, the importance of the seismic information in presence of only a few wells is of key importance for the derivation / estimation of the reservoir properties away from the well control. The limited frequency bandwidth of the conventional marine seismic data, where the sea‐surface ghost acts like a band‐pass filter on both the source and the receiver, is forcing the geoscientist to increase the amount of a priori knowledge to estimate the reservoir properties from seismic. This, consequently biases the results, as we are potentially analyzing a result driven by strong a priori information in the first place. In 2007 (Tenghamn et al. 2007 and Carlson et al. 2007), a new towed streamer was presented to the public combining two sensors (hydrophone and vertical velocity sensor). This new streamer allows us to remove the sea‐surface ghosting filter on the receiver side, resulting in an extended seismic bandwidth on both the low and the high side of the amplitude spectra, but with the source ghost still present. This significant enhancement of the frequency bandwidth results in improved resolution of the geological layers as well as improved accuracy in the pre‐stack extraction of elastic attributes and reservoir properties extraction. In addition, the dual‐sensor streamer is towed at about 20 m below the sea‐surface resulting in a significant reduction of noise. This paper presents two case studies where the qualitative benefits of the dual‐sensor streamer will be presented firstly through an improved AVA Analysis (Thompson et al. 2010), and secondly through a more quantitative analysis by estimating the reservoir properties.