American Oil and Gas Reporter - February 2017 - 46

FIGURE 3
Ray Paths and 3-D VTI Model Move-Out
Corrected Gathers of Perforation Shots from Stage 9
1 3 5 7 9

12 14

17

20

23 26

29

32 35

POD
38 41

44 47 50

53 56

59 62

65

68

71 74

77 80

300

400

500

600

700

constrained by inequalities that impose
smoothness between layers.
The most advanced case that one can
attempt to solve highlights the benefits
of moving to a fully elastic, arbitrarily
anisotropic forward ray-tracing and inversion routine. This technique operates
on the 21 independent elements of the
stiffness tensor, and consequently permits
not only a more accurate result, but sets
the table to deliver advanced interpretative
results that add significant value.
The general tomography workflow includes anisotropic log upscaling, split
shear-wave picking, inputting perforation
and events, ray tracing, sensitivity analysis,
parameter restrictions, and updating the
model/perforations locations.
Wolfcamp Case Study

FIGURE 4A
Perforation Localizations in Isotropic Velocity Model
11,280

6,560
6,630

10,750

6,700
6,770

10,220

6,840
6,910

9,690

6,980
7,050
7,120

9,160

7,190

-3,420

8,630

-3,250

-3,080

-2,910

7,220

7,790

8,360

-2,740

-2,570

-2,400

-2,230

-2,060

-1,890

-1,720

5,940

8,100

6,160
6,380

7,570

6,600
6,820

7,040

7,040
7,260

6,510

7,480
7,700
7,920

-6,640 -6,270 -5,900 -5,530 -5,160 -4,790 -4,420 -4,050 -3,680 -3,310 -2,940

8,930

9,500

10,070

10,640

11,210

11,780

12,350

FIGURE 4B
Perforation Localizations in Anisotropic 1-D VTI Velocity Model
12,360

2,225H

11,680

11,000

6,030
6,220
6,410
6,600
6,790
6,980

10,320

7,170
7,360

9,640

7,550
7,740

8,960

8,280

7,600

6,920

-5,040

-4,410

-3,780

-3,150

-2,520

-1,890

-1,260

-630

0

630

6,030
6,220
6,410
6,600
6,790
6,980
7,170

6,240

7,360
7,550

-6,280 -5,940 -5,600 -5,260 -4,920 -4,580 -4,240 -3,900 -3,560-3,220 -2,880

7,740

7,330

7,960

8,590

46 THE AMERICAN OIL & GAS REPORTER

9,220

9,850

10,480

11,110

11,740

12,370

13,000

The case for directly linking location
accuracy to velocity model efficacy was
demonstrated in the West Texas microseismic project. The lateral of the Wolfcamp "B" horizontal well is ≈6,000 feet
long and follows a general northwest trajectory. It has a nearly horizontal profile.
Lateral depth is ≈9,760 feet total vertical
depth. The well was completed with 24
hydraulic fracturing stages.
The workflow generated a single geologically meaningful 3-D VTI velocity
model that solved for all recorded perforation shots acquired from the well completion. The microseismic data were
recorded over two weeks using two spatially dense, 40-level, three-component
borehole geophone arrays in two offset
observation wells located on either side
(to the east and west) of the heel of the
treatment well.
Figure 1 shows the localizations of 56
imaged perforation shots within an isotropic velocity model. There is clear bias at
the toe in depth and azimuth. There also
is an off-wellbore bias noted in a lineation
from stages 9 through 24.
Since the perforation and microseismic
signals are processed in the same manner,
the average perforation location error is
a proxy for the average microseismic location uncertainties. The total perforation
errors ranged from a minimum of 22.4
feet to a maximum of 212.0 feet, with an
average error of 75.8 feet and a standard
deviation of 40.0 feet.
Originally, there was discussion about
whether azimuthal anisotropy was the
primary issue causing this problem. The
total error developed from the initial
American Oil and Gas Reporter - February 2017

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