Assessment of dynamic material properties of intact rocks using seismic wave attenuation: an experimental study

Abstract

The mechanical properties of any substance are essential facts to understand its behaviour and make the maximum use of the particular substance. Rocks are indeed an important substance, as they are of significant use in the energy industry, specifically for fossil fuels and geothermal energy. Attenuation of seismic waves is a non-destructive technique to investigate mechanical properties of reservoir rocks under different conditions. The attenuation characteristics of five different rock types, siltstone, shale, Australian sandstone, Indian sandstone and granite, were investigated in the laboratory using ultrasonic and acoustic emission instruments in a frequency range of 0.1–1 MHz. The pulse transmission technique and spectral ratios were used to calculate the attenuation coefficient (α) and quality factor (Q) values for the five selected rock types for both primary (P) and secondary (S) waves, relative to the reference steel sample. For all the rock types, the attenuation coefficient was linearly proportional to the frequency of both the P and S waves. Interestingly, the attenuation coefficient of granite is more than 22% higher than that of siltstone, sandstone and shale for both P and S waves. The P and S wave velocities were calculated based on their recorded travel time, and these velocities were then used to calculate the dynamic mechanical properties including elastic modulus (E), bulk modulus (K), shear modulus (µ) and Poisson's ratio (ν). The P and S wave velocities for the selected rock types varied in the ranges of 2.43–4.61 km s−1 and 1.43–2.41 km h−1, respectively. Furthermore, it was observed that the P wave velocity was always greater than the S wave velocity, and this confirmed the first arrival of P waves to the sensor. According to the experimental results, the dynamic E value is generally higher than the static E value obtained by unconfined compressive strength tests.

Study on the Acoustic Emission Characteristics of Different Rock Types and Its Fracture Mechanism in Brazilian Splitting Test

“…These smaller scale fractures usually release AE signals with high frequency. Therefore, for granite, marble, and limestone, the signal peak frequencies larger than 300 kHz are in the majority, whereas for sandstone, the high frequency component of AE signal will be greatly attenuated due to high porosity [46][47][48][49], thus resulting in relatively larger proportion of low frequency signals being collected.…”

“…Rock AE signal frequency has a close relationship with its internal crack scale, by constructing the relevant earthquake model and simulating the earthquake process in the laboratory. It is concluded that the rupture scale (or magnitude) M 0 is proportional to the −3 power of the frequency f. The scaling relation of size-frequency can be used to estimate the scale of the source of different frequency signals [12,48,49], and the relationship between source scale and frequency can be as follows [50]:…”

On possible fluid detection in equivalent transversely isotropic media

“…Based on works of Castagna and Smith () and Wanniarachichi et al . (), the approximate ranges of these coefficients are and , where GPa are gigapascals. However, taking into consideration that the values of λ that are close to 0 might cause the issue within Backus averaging (Bos et al .…”

“…We examine only the ranges of Lamé coefficients that are relevant to sandstones (brine sands, gas sands and others). Based on works of Castagna and Smith (1994) and Wanniarachichi et al (2017), the approximate ranges of these coefficients are λ ∈ [−3 , 20] GPa and μ ∈ [1 , 30] GPa, where GPa are gigapascals. However, taking into consideration that the values of λ that are close to 0 might cause the issue within Backus averaging (Bos et al 2018;Kudela and Stanoev 2018), we set the ranges to be λ ∈ [3 , 20] GPa and μ ∈ [1 , 30] GPa.…”

The Impact and Causes of Subsidence in the Exmouth Sub-basin, Northern Carnarvon Basin

“…The cross-section has 6 sedimentary layers. We digitised the cross-section and converted the depth information from two-way time (seconds) to km using average speed of seismic waves in various lithologies multiplied by time, which gives layer thicknesses (Toksoz et al, 1976;Wanniarachchi et al, 2017). We then calculated the depth of the stratigraphic layer boundaries (Figure 1), followed by the collection of sedimentary layer features for example; lithology, absolute age, porosity, and many others as listed in Table 2, below.…”