American Oil and Gas Reporter - February 2017 - 44

SpecialReport: Shale Resource Science
Tomographic inversion is used to generate velocity models, making measurements of seismic waves passing through
a material. The character of these measurements is analyzed to make inferences
on the material the waves passed through
(velocity, density, etc.). The velocity of
compressional (p) and shear (s) waves
depends on the rheology of the material
they travel through (density and elasticity).
In short, variations in chemical compo-

sitions and thermal structure result in
changes in velocity.
The goal is to deliver a petrophsycially
meaningful microseismic 3-D velocity
model that honors the highest order of
anisotropy supported by the data, making
it possible to extract additional value
from higher-order anisotropy measurements in order to estimate the elastic
rock properties that are critical to developing shale assets.

FIGURE 1
Localizations of 56 Imaged Perforation Shots In Isotropic Velocity Model
3,200

8,430

8,720

2,560

9,010

1,920

9,300

Accurate Velocity Modeling

9,880

The polarization vector of particle
motion plays a fundamental role in processing and interpreting multicomponent
borehole microseismic surveys. We propose that a reliable and complete
hodogram analysis must consider the relationship between phase, group and polarization angles.
Anisotropic media present an added
level of complexity by deviating the polarization vector from its isotropic direction. The presence of anisotropy may
cause distortions to both the radiation
patterns and reflection coefficients.
Incomplete modeling or appreciation
of these interconnected parameters may
lead to under- or overcorrected azimuth
estimates, and overly simplified and
unrealistic velocity models. Further in
the workflow, this may bias or obscure
valid smaller-amplitude microseismic
events, and ultimately may lead to inaccurate and biased estimates of event
locations. The net result of these errors
can yield a suboptimal map of microseismic events.
Accurate effective velocity modeling
also is essential for travel-time computations used for a least-squares minimization
of localization. This study contrasts
isotropic, fine-layer, effective medium
modeling with one- and three-dimensional
vertical transverse isotropy (VTI)
anisotropic models. Forward modeling
is performed by two-point ray tracing
following Fomel velocity approximations.
For modeling, layers are constructed with
an arbitrary spacing, and the inversion is

9,590
10,170

1,280

10,460
10,750

640

11,040

0

-3,300

-2,640

-1,980

-1,320

-660

0

660

1,320

1,980

2,640

8,430

-640

8,720
9,010

-1,280

9,300
9,590

-1,920

9,880

10,170
10,460

-2,560

10,750
-2,280 -1,900 -1,520 -1,140

-760

-380

0

380

760

11,040

1,140

-3,300

-2,640

-1,980

-1,320

-660

660

0

1,320

1,980

2,640

FIGURE 2
Final 1-D (left) and 3-D (Right) VTI Velocity Profiles
12,456

13,148

Value
13,840

14,532

15,224

12,562
4,320

4,860

4,860

5,400

5,400

13,133

Value
13,704

14,275

TVDSS

TVDSS

4,320

5,940

5,940

6,480

6,480

Sonic Log (ft/s)

To test the premise and benefits associated with honoring an increasingly
more sophisticated level of physics when
building velocity models, varying levels
of anisotropy were tested through tomography using a microseismic dataset
for a horizontal well in the West Texas
Wolfcamp trend. The results demonstrate
that higher orders of anisotropy have increased perforation accuracy and aided
in developing a geologically relevant velocity volume.
Utilizing this anisotropic velocity volume, rock property estimates were extracted from perforation data, adding interpretative value.

Horizontal Velocity Proﬁle (ft/s)

44 THE AMERICAN OIL & GAS REPORTER

Vertical Velocity Proﬁle (ft/s)

14,846
American Oil and Gas Reporter - February 2017

Table of Contents for the Digital Edition of American Oil and Gas Reporter - February 2017
