Source Pattern Analysis

Introduction:

Various seismic sources including pressure source, vertical source, radial source, transverse source and explosive source have been widely used in exploration seismology. Double couple (DC) source and sources defined by general moment tensor are commonly used for simulation of earthquake and microseismic waves. Processing of the recorded data helps reconstruct the subsurface image and locate the microseismic events for better understanding and interpreting the geologic structures and formations. Individual sources excited at different times will form simultaneous sources. Implementation of simultaneous sources provide flexibility in survey geometries and increase spatial sampling via increased source sampling and vector-offset sampling.

Here we describe the physical mechanisms of several typical seismic and microseismic sources and their impact on the synthetic wavefield. Microseismic modeling in viscoelastic media is discussed by using simultaneous sources with different source characteristics. Through the back propagation of the recorded data, the location of microseismic events can be imaged without picking the events.

Excitation of seismic sources:

Seismic exploration uses viscoelastic forward modeling to simulate wave motion in real media and extract subsurface information. The nature of the source event of viscoelastic forward modeling may be described as a system below. This system integrates all types of source mechanisms which are regularly performed for monitoring hydraulic fracture and validating seismic data for subsurface imaging and illumination. The excitation system of seismic sources involved in the modeling procedure can be characterized in the followings:

1. Pressure source: a stress-rate source applied to stress component which only generates P-wave.

2. Vertical source: a body force source applied to vertical component (z) of particle velocity which generates
both P-wave and S-wave.

3. Radial source: a body force source applied to radial component (x) of particle velocity.

4. Transverse source: a body force source applied to transverse component (y) of particle velocity. Radial and transverse source can generate both P-wave and Swaves.

5. Vector source: a body force source designed for simulating microseismic mechanism. Vector source is performed on all three components (x, y and z) of particle velocity with directional coefficients of each
component.

6. DC source: a shear dislocation source representing earthquake source. DC source specifies the source
radiation pattern and can be described as an equivalent distribution of body force source. Four fault parameters such as source magnitude, strike angle (φ), dip angle (δ) and rake angle (λ), are introduced to describe the DC source mechanism.

 

 

 

 

7. Moment tensor source: a generalized source that represents the response of a fault of any arbitrary
orientation.

The following figure shows the characteristic comparison of three different seismic moment tensor sources with Vp=2.0 km/s, Vs=1.0 km/s, maximum frequency is 40 Hz. All snapshots and shot gather traces are extracted from vertical particle velocity component. The shot gather trace is picked from offset at 0.4 km.

 

Viscoelastic modeling of simultaneous microseismic sources

Carcione (1993) investigated attenuation in viscoelastic media and developed the corresponding stress-velocity wave equations, where memory variables are introduced to model the relaxation mechanism. Attenuation effect is represented by quality factor Q defined as:

 

 

The relation between quality factor Qp (for P-wave), Qs (for S-wave) and the relaxation time in standard linear viscoelastic equations is described as (Blanch et al., 1995):

 

 

 

 

Following figure shows a snapshot of simultaneous source system in a constant velocity model. In this example P-wave velocity is 2.0 km/s, S-wave velocity is 1.0 km/s and density is 2.0 kg/m3. The quality factor for Qp and Qs are identical and equal to 20. Ricker wavelet is used with maximum frequency 30 Hz. Three moment tensor sources (located at 1.0 km below surface) are introduced with delayed excitation time 0.0 s, 0.2 s, 0.4 s respectively. By implementing multiple source system with viscoelastic modeling, it is possible to extract source parameters from the seismic events and understand the information such as fault dimension and seismic moment when comparing with field microseismic data.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Below describes the distribution of amplitude over frequency between different quality factors on viscoelastic modeling. Figure 4a shows that amplitude distribution at different frequencies with only one microseismic source (source index 1 in Figure 2). Figure 4b shows that the amplitude distribution with simultaneous sources (shown in Figure 3a). Introducing quality factor to wave simulation will absorb energy and lower the frequency band. This phenomenon not only appears on single microseismic source, but also takes effect on simultaneous microseismic sources. Several tests on point-source excitation also obtain same result.

 

Back propagation of the recorded microseismic data with simultaneous sources is used for locating the multiple microseismic events. Figure 6 shows an example of
imaging simultaneous microseismic events. The back propagated wavefield interferes constructively at the three source positions at different times. Over- or under-migrated events cause unfocused energy. Some artifacts are present on the image. Further study needs to be conducted to get a clean and high resolution image.

Conclusions:

Seismic source, which generates controlled seismic energy, is used to perform reflection, refraction and passive seismic surveys. In this abstract, we describe physical mechanisms of several popular seismic and microseismic sources. Different radiation patterns show different impacts on wavefield snapshot and seismogram. We discussed with a simultaneous source system which could use in arbitrary acquisition geometry for different purposes. Applying  simultaneous sources to viscoelastic modeling provides capability of understanding the source and fracture mechanisms, especially for microseismic data. Our test shows that the attenuation has a significant effect on seismograms. Wave amplitude will have greater loss rate when quality factor is less than 50 and frequency will shift to lower band. Test of back propagation with simultaneous sources shows that the recorded data could be used to locate microseismic events without picking for multiple source application. The imaged source locations and source mechanisms can help interpretation of the wave modes and behaviors of hydraulic fractures.

Reference: 2013_Viscoelastic modeling with simultaneous microseismic sources