Numeric Simulation of Frequency-dependent Seismic Response and Hydrocarbon Detection, a Turbidite Reservoir in JZ Area, the Bohai Sea, China

The numerous recent laboratory and field examples show the potential benefits of seismic low frequencies in hydrocarbon detection. To simulate the frequency-dependent response of turbidite reservoirs in JZ Area, the Bohai Sea, China, and then implement frequency-dependent detection of the hydrocarbon accumulation based on low frequencies, we expanded a diffusive and viscous wave equation (DVWE), which takes into account the diffusive and viscous attenuation, and velocity dispersion in fluid-bearing poroelastic media. In design of the reservoir equivalent geologic model, by applying a 90o-phase shift on the raw seismic section contained turbidites, we pick zero crossing on converted section to produce interface (reflectivity) section. We then assign the parameters such as density, velocity, diffusive, viscosity, and Q for each stratum to produce a physical parameters section. Such parameters originate from the well log and rock physical experiments. The DVWE-based simulation result not only shows the characteristic reflection and geometry of turbidites, that are consistent desirably with the reflection feature on the raw seismic section, but also delineates the phase delay, instantaneous dominant frequency decrease and magnitude attenuation related to the gas-bearing reservoir. By conducting instantaneous spectral decomposition on the simulation section, the common frequency sections indicate bright strong energy of the gas reservoir and low-frequency shadow lie immediately underneath the reservoir at 8Hz and 12Hz. At 20Hz and 28Hz, the gas reservoir is brighter than the shadow that becomes weaker but still persists. The shadow is almost gone at 36Hz. We carry out instantaneous spectral decomposition and calculate fluid mobility (permeability to viscosity ratio) on the seismic data. At 12Hz, the hydrocarbon accumulation in tubidites show clear bright spot on common frequency section and horizon slice at the reservoir. There show abnormally strong low-frequency shadow on the common section and the slice for a 40ms window immediately below the reservoir. At 20-36Hz, the reservoir remains bright, and the shadow has gradually disappeared. The fluid mobility calculated at 12Hz also clearly delineates the bright gas reservoir and its spatial distribution. In this case study presented above, the low-frequency effects are especially important for delineation of the fluid signature. It is necessary to develop more numerous deeper experiments and theoretical research in the near future. INTRODUCTION It has been found for many years that low-frequency seismic signals can be successfully AAPG Search and Discovery Article #90142 © 2012 AAPG Annual Convention and Exhibition, April 22-25, 2012, Long Beach, California used for accurate delineation of hydrocarbon reservoirs. This frequency dependence of seismic reflections from fluid-saturated porous media has been detected in different geologic environments. The numerous recent laboratory and field examples show the potential benefits of seismic low frequencies anomalies in hydrocarbon detection. Taner et al.(1979) once noted in complex seismic trace analysis that the lower apparent frequency occur from reflection immediately underneath the petroliferous zone. The low-frequency shadow (LFS) is often observed on common frequency sections decomposed by instantaneous spectral analysis (Castagna et al., 2003; Wang, 2007; Liu and Marfurt, 2007; Chen et al., 2009). However, although Ebrom (2004) summarized ten possible factors responsible for this phenomenon. There are still no definite explanations about its mechanism (Tai et al., 2009). It is necessary for understanding its mechanism to further theoretically study and numerically simulate (Liu, 2004). Goloshubin and Korneev (2000; 2002; 2004) studied the seismic response of fluid-saturated porous media by a series of rock physical model experiments. Their observations in laboratory indicate that reflections from a fluid-saturated layer have increased amplitude and delayed traveltime at low frequencies. They proposed a 1D scalar diffusive and viscous wave-propagation equation, but they did not study its numeric implementation. Another low-frequencies-dependent hydrocarbon indication is the fluid mobility (permeability to viscosity ratio), which is velocity dispersion dependent due to the hydrocarbon-saturated reservoirs. Silin et al. (2004) derived an asymptotic scaling of the frequency-dependent component of the reflection coefficient with respect to a dimensionless parameter depending on the reservoir fluid mobility. Goloshubin et al. (2008) used fluid mobility to estimate the permeability of reservoirs. In this paper, we present a 3D diffusive and viscous wave equation (DVWE), and simulate the frequency-dependent seismic response on the geological model, which is created by integrating real seismic data, rock physics and well log. By following the guidance of this simulation, we use both LFS and fluid mobility to indicate the hydrocarbon in a turbidite reservoir from JZ Area, the Bohai Sea, China. THEORY AND METHODS Seismic data based geological model building Figure 1 Schematic workflow of the model building In the equivalent geological model building, we firstly transform the raw seismic sections contained my target interval by applying a 90o-phase shift, which have superior geologic interpretive advantages for thin bed mapping that advocated by Zeng and Backus (2005). Next, Seismic sections 90o-phase shift Pick interface Physical parameters filling we automatically pick zero crossing on the converted sections to produce interface (reflectivity) sections. We then assign the physical parameters such as the density, velocity, diffusiveness, viscosity, and Q of the fluid and skeleton for each stratum, and produce physical parameters sections. Such parameters originate from the well log and rock physical experiments. The schematic workflow of the seismic data based geological model building is given in Figure 1. Diffusive and Viscous Wave Equation (DVWE) Considering the diffusiveness and viscosity of the fluid, Goloshubin and Korneev proposed a 1D scalar diffusive and viscous wave-propagation equation, which is given by (Goloshubin et al., 2000; Goloshubin et al., 2002; Korneev et al., 2004) 2 3 2 2 2 2 2 0 u u u u v t t z t z                (1) where u is the displacement,  is diffusive attenuation parameter in unit of Hz,  is viscous attenuation parameter in unit of m/s, and v is the wave-propagation velocity in unit m/s. We generalize equation (1) and obtain its 3D form, which can be expressed as