Introduction
The study of wave attenuation in partially saturated porous rocks over a broad frequency range provides valuable information about the fluid system of reservoirs, which are inherently multiple phase fluid system. Until now, not much laboratory data have been collected in the seismically relevant low frequency range and existing literature data on partially saturated rock are very limited. The main goal of our work is to experimentally measure the bulk seismic attenuation on fluid-bearing rocks, using natural rock samples in an efficient way at in situ conditions. We are currently fine-tuning our attenuation measurement prototype. Preliminary bench-top results are promising and show consistency with reported experimental data with dry, partially and fully fluid saturated rocks. Measurements with the machine are accurate and precise.
Low-frequency wave propagation in partially saturated rocks is still not well understood because of the lack of precise and reliable experimental measurements. An experimental approach to measure attenuation at low frequencies is therefore not only a scientific problem, but it is also of significant interest for the petroleum industry. Biot theory is the most commonly used theory to study wave propagation in saturated porous rocks. However, this theory estimates considerable attenuation only for frequencies higher than seismic frequencies. Biot theory has been extended to model attenuation in the low frequency range, with the patchy-saturation models (White, 1975; Dutta and Odé, 1979a; Carcione et al., 2003; Toms et al., 2006). Patchy-saturation causes high attenuation in low frequency range for realistic material properties (Picotti and Carcione, 2006; Quintal et al., 2009). The so-called doubleporosity models also explain attenuation in the low frequency-range (Pride et al., 2004). However, few laboratory experiments have been performed to investigate real low-frequency effects in partially rocks. Such effects could be used to discriminate water-, gas-, and oil-saturated rocks (Korneev et al., 2004; Goloshubin et al., 2006) and/or focus on difficult petrophysical parameters such as permeability (Masson and Pride, 2007). The work done to date can be divided broadly in two classes. The first class deals with high-temperature and high-pressure attenuation and aims to reproduce attenuation that occurs in the deep earth (e.g. Jackson et al., 2002). These studies were conducted at low frequencies (10-2 to 10 Hz) to reproduce the attenuation of teleseismic waves, using the forced oscillation technique in torsion. The second class concerns porous rocks. Data available in the literature, especially on porous, permeable sandstone, do not homogeneously cover the low frequency range. The importance of gathering data at high resolution in the frequency domain is imperative since the details of a single mechanistic Kramers-Kronig cycle will likely manifest over a frequency band of about one decade. So far, few laboratories were able to gather accurate data over seismically relevant frequencies between 10-1-102 Hz at appropriate strain amplitudes <10-6 (e.g. Batzle et al., 2006; Moerig and Burkhard, 1989; Paffenholz and Burkhardt, 1989). Additionally, the most common laboratory techniques have relied on point measurements with strain gauges at one or more locations on a sample.
