SPE

Society of Petroleum Engineers

SPE 14399

Fracture Evaluation Using Borehole Sonic Velocity Measurements

by O.Y. Liu, Schlumberger Well Services

Copyright 1985, Society of Petroleum Engineers

This paper was prepared for presentation at the 60th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers held in Las
Vegas, NV September 22-25, 1985.

This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the
author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the
author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers
presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Permission to copy is
restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment of where
and by whom the paper is presented. Write Publications Manager, SPE, P.O. Box 833836, Richardson, TX 75083-3836. Telex, 730989 SPEDAL.

ABSTRACT resulting from reflection, refraction, and mode conversion

Using waveforms acquired by multitransmitter and/or are often, if not always, difficult to separate. These in-
multireceiver borehole sonic tools, acoustic velocity (or terference patterns can also be generated by bed boundaries
slowness) was computed for compressional, shear, and and borehole irregularities as well as by fractures.
Stoneley waves using direct phase determination (DPD), 1 The proposed method in this paper uses the differences
the slowness time coherence (STC) method,2 and multiple in measurements of receiver and transmitter l::lt of compres-
shot processing (MSP).3 It was found that values of flt (or sional, shear, and Stoneley waves. This method has been
slowness) computed using the receiver locations, referred shown to be very reliable, is little affected by bed boun-
to as the receiver flt, were different from those computed daries, and can distinguish fractures from borehole
using the transmitter locations, referred to as the transmit- irregularities.
ter l::lt, when the sonic tool was in the vicinity of fractures.
The amount of differences in compressional, shear, and . INTERFERENCES AT FRACTURES
Stoneley l::lt depends on the size and nature of the fractures.
The interferences observed in sonic waveforms can be
It is possible to determine the height, width, and dip angle
classified into three broad categories:
of the fracture when these differences are used in conjunc-
tion with Dual Dipmeter* tool data. 1. Interferences resulting from waves reflected off a discon-
tinuity in formation or borehole wall
INTRODUCTION 2. Mode conversion occurring when the incident wave Type
Acoustic waveforms, in addition to methods based on con- A is converted to wave Type B
ventional logs, have been used to detect and characterize 3. Crack couplingt
fractures intersecting the borehole. These log-based methods Because of the changing travel path of the wave as the
include differential response to porosity measurements, tool approaches or moves away from the reflecting surface,
magnitude of apparent power law exponent relating to for- interference Categories 1 and 2 exhibit patterns visible as
mation porosity, and various resistivity log curves resulting a W, V, or X shape on the variable density displays.l2,13
from different responses to fracture conductivity. 4,5 The Interference Categories 1 and 2 occur when the discontinuity
acoustic methods, whether seismic,6 sonic,7,8 or is either a formation bed boundary, borehole enlargement,
ultrasonic,9 are usually based on the scattering and absorp- or fracture. Interference Category 3 occurs only at open
tion of acoustic energy associated with the various wave fractures. ·
components contained in the full waveforms.
Fig. 1 shows waveforms from one receiver in the vicini-
Recent borehole studies have suggested that the reduc- ty of a fracture. Some minor changes of compressional wave
tion in Stoneley or tube wave energy can be used as a reliable shape become noticeable at 2,815 ft. Drastic changes can
indicator of fractures open to fluid flow.6,10 be seen at 2,810 ft where the compressional mode-converted
However, each method is subject to certain limitations Stoneley wave is visible. The first cycle of compressional
in its own application. Various components in the waveform arrival is distinctly different from that in the unfractured
* Mark of Schlumberger
t J. E. White, Colorado School of Mines, personal communication (1980-1985).
2 FRACTURE EVALUATION USING BOREHOLE SONIC VELOCITY MEASUREMENTS SPE 14399

Figure 1-Waveforms in the vicinity of a fracture.

zone. Fig. 2 shows the variable density display of the frac-

tured interval where waveforms were depth stacked to
remove the compressional and direct Stoneley arrivals. Two
mode-converted Stoneley arrivals from the compressional

Figure 3-Fracture-generated "crack coupling."

APPARENT TRANSMITTER AND

RECEIVER tlt ESTIMATION

Figure 2-Depth-stacked variable density log with compres- Two principal methods were used in estimating the values
sional arrivals time shifted. of tlt. The DPD method was used for the nonarray
waveforms obtained from the Long Spaced Sonic tool. In
order to get a 2-ft depth resolution and the transmitter dt
wave can be seen here. A Stoneley wave converted from for the waveforms from a research experimental sonic ar-
shear wave can also be seen. Upward- and downward- ray tool, the MSP method was used. The heart of the MSP
reflected Stoneley waves at the top and bottom borehole is a semblance processing called the STC method. This
fracture entries are visible here. method detects arrivals in the windowed waveforms by com-
The physics of the "crack coupling" is not yet fully puting the scalar semblance for a large number of possible
understood, especially for frequencies higher than seismic arrival times and slownesses. The local maxima of the two-
frequency. However, a simplified low-frequency approach dimensional semblance function are identified by a peak-
explanation can be given as follows. When compressed by finding algorithm. The corresponding slowness of compres-
an arriving compressional wave, weakly reflected and sional, shear, and Stoneley waves can then be determined.13
strongly transmitted, the fracture, or ~he crack, acts like The STt processing makes few a priori assumptions about
a source of blocked pressure Pc. Fig. 3 depicts the fracture the contents of waveforms. The transmitter and receiver
coupling and mode conversions when the transmitter and configurations for both receiver tlt and transmitter tlt are
receiver straddle a horizontal fracture. A tube wave, or low- shown in Fig. 4. The 2-ft resolution is achieved by using
frequency Stoneley wave, is generated with pressure Pr at waveforms from only five receivers. The semblance func-
the crack opening. The pressure at the receiver will be the tions overlap the depth interval being logged. For the
sum of the borehole coupling and the fracture-generated transmitter dt, the waveforms from different firings were
coupling. regrouped as shown in Fig. 4.
SPE 14399 0. YLIU 3

In the labeling process13 of identifying compressional,

OVERLAPPING SUBARRA YS shear, and Stoneley arrivals, strict time-to-slowness relation-
ships based on transmitter /receiver spacing are imposed to
- - - - .....--,------ SUBARRAY IN
screen out all peaks associated with the reflections and mode
~ - ___ ( ., 'TRANSMITTER" SECTION
conversions resulting from fractures, bed boundaries, and
SUBARRAY IN
borehole irregularities.
.,"""'. ""'0'~
FIELD EXAMPLES
/\ (1}----i
I o
[---
DEPTH INTERVAL
The logs of transmitter and receiver flt values for compres-
I
sional, shear, and Stoneley waves have been compared with
o
I
1 I
' BEING LOGGED
0 I
\) (1)------llll-----
core data and results from other fracture detection methods
for a number of wells. The most complete set of data is from
the Conoco 33-1 test well where the following examples were
X Transmitter of Sonic Array Tool chosen. These fractures are all in constant lithology and
•
lXI
Receiver of Sonic Array Tool
Transmitters of Long Spaced Sonic Tool
gauged boreholes unless otherwise mentioned. Values shown
(!) Receivers of Long Spaced Sonic Tool are MSP results of the sonic array waveforms. Table 1 sum-
marizes the fracture description, transmitter/receiver flt dif-
ferences, and lithology of all examples.

Figure 4-Transmitter and receiver configuration of sonic

tools.

Table 1
SUMMARY OF FRACTURE EXAMPLES

Fracture T/R .1t Difference (/Ls/ft)

Depth Fracture Description Comp. Shear Stoneley Lithology and Other Remarks
1849.5 Borehole enlargement, from 8.5 to 13 in. 2-20 0-40* 0-20 Concentration of shell fragments. 3 in. of
to mudstone. Very fissile.
1852.5
(Fig.S)
2161 Vertical, partially healed fractures 1-10 10-25 2-5 Limestone, very shaly; subhorizontal beddings.
to
2169
(Fig. 6)
2209 Vertical open fracture 1-5 5-20 1-3 No mineralization or slickensides. Fracture stops
to Very thin < < 0.5 mm abruptly at thin sandstone level and exits core.
2213
(Fig. 8)
2240 Open, large vertical fracture, variable width; 5-12 5-30* 5-25 Shale, heavily burrowed; sand filled some of the
to 0.5 mm to 1.5 em (on reconstruction) burrows. Slickensides apparent in upper section
2247 where fracture surface uneven.
(Fig. 10)
2214 Large, complex, vertical, partially filled 0-10 0-20 0-8 Massive sandstone, medium-grained. Fractures
to fractures. Variable width up to 5 mm. filled with dolomite, calcite crystals up to 2 em
2219 Completely filled in the middle. long. Complex fractures with many branches.
(Fig. 12)
2514 Heavily fractured zone with several distinct 0-8 2-20 5-20 Sandstone, medium-to very fine-grained.
to vertical fractures. Top section up to 2 mm Shale interbeds, minor ripples. Fracture
2527 wide, 70 to 90% filled with calcite or Si02 . exits core horizontally in 2-cm-thick shale
(Fig. 14) Open fractures with width 1 mm or larger bed. Large unfilled fracture terminated at
in the middle section. Bottom section is shale may be coring induced. Intense
very complex fracture system, up to 1.5 em vuggy porosity filled with well developed
wide, 50 to 90% filled. Mississippi Valley-type minerals.
2738 Irregular vertical fractures some open with 0-8 2-35 2-15 Limestone, grainstone/packstone. Abundant
to carbonate lining, stylolites, oil-stained. dissolution pits, trace bioclasts. Enlarged
2745 borehole at one place.
(Fig.17)
2812 Open vertical fracture, < 0.5 mm wide. No 0-6 5-10 5-10 Shale, silty. Trace oxidized, few burrows,
to mineralization or slickensides. Very uniform carbonate stringers.
2820 and straight.
(Fig.19)
2832 Long, very thin, < 0.1 mm straight fracture, 0-1 2-8 0-1 Same as above.
to = 100%. Filled with calcite cement and
2842 minor pyrite.
(Fig.20)

• Shear arrivals were missing from sonic array waveforms. Values are estimated from 8- to 1O-ft, 10- to 12-ft long-spaced waveforms by DPD method.
4 FRACTURE EVALUATION USING BOREHOLE SONIC VELOCITY MEASUREMENTS SPE 14399

Example 1
The transmitter/receiver !:J..t logs shown in Fig. 5 are recorded
from a borehole enlargement, not from a fracture. The pat-
tern for compressional wave is as one would expect. Note
that the transmitter !:J..t (T!:J..t) becomes faster as the tool ap-
proaches the enlarged section and then becomes slower as
the tool moves away from it. A skew-symmetric pattern is
found for the receiver Llt (R!:J..t). The pattern for Stoneley
wave is more complicated. No shear wave was observed by
the sonic array tool, possibly because of its narrower source
frequency band. An estimate from the long spaced
waveforms by the DPD method is given in Table 1.

Example 2
Figs. 6 and 7 show the results of thin vertical partially fill-
ed fractures in limestone. The size of the !:J..t differences is
apparently related to the degree of openness of these frac-
tures. Note that the pattern of !:J..t is reversed from that of
borehole enlargement.

1844

--------
1846- ====::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
=-=-=-=-==--=-=-=
1848- =~==~~======
-------
'./
Core 84 =-=-=-=-=-=-=-=
1852- ===============
-=-=-=-=-=-=-=-
-=-=-=-=-=-=-=-

-==--=-=-=-=-=-=-
==-=-=-=-=-=--==-=
1856- =:: : :=: : : =: : : =: : : =: : : =: : : =: : : =
--------
-==--=-=-=-=-=-=-
pb Tat
.75 glee 2.75 250. ~sift 50.
e.o____ -i,;.-- --Tifo
CALl
2-so~- ------ Rdl
--~9711 ______ ----5o.

Figure 7-Core photograph of vertical partially filled fractures.

Figure 5-Borehole enlargement.

Example 3
Fig. 8 shows the results of a single vertical open fracture
2160

2204

2182

2206

2164
(I
2208

2166

2210

2168

2212

2170 / .

2214

2216

Core 79

Figure 6-Vertical partially filled fractures. Figure 8-Very thin vertical open fracture.
SPE 14399 0. Y LIU 5

with a width much less than the 0.5-mm width depicted in 0ZZJII)

Fig. 9. The fracture stops abruptly at a thin sandstone level

and exits the core.

INTERVAL: 2210 TO 2212 FT

3 1/2-IN. DIAMETER

02Z48)

~ i<o~o------1--VERY THIN FRACTURE ,

<< 0 . 5 MM RECONSTRUCTED

e£2:10)
Figure 9-A vertical fracture stopped abruptly at a thin sand-
stone level.

Example 4
Large open vertical fractures can cause transmitter/receiver
fl.t values to be different by more than 10 ~-tslft for com-
Figure 11-Borehole televiewer log.
pressional and 20 ~-ts/ft for shear and Stoneley waves as
shown in Figs. 10 and 11. For the Stoneley wave, this is
an extremely large change in slowness because Stoneley Example 5
velocity does not vary much in fast and medium-fast The power of the method described in this paper can best
formations .14 be demonstrated in the examples shown in Figs. 12 and 13.
This is a large complex vertical fracture partially filled in
the top section and closed in the middle as indicated by the
small difference in Stoneley fl.t. The fracture is open again
· -
------- -
......... ~....._ . ......... . . . _ _ - . J ..... .
in the lower section.
=-=-=-=-=-=-=-::::
- - - - - - -· /1'1

=-=-=-I:-L;L=-=-=-:
-=-=-=-=-=-=~.r:

~ =-=-=-=-=-
- ------
-- =:
2418

-:r---=-= -=-=-=-: ~

2420

-=-=~~-=-=-=-=
2422 .... . .... . ....
:·.............
:·:· :· :· :· :· :· :· :::·:· :· · /1'1
. . ... .. . . ... .. .
2424 ·.·:::.·:::.·::.·::
.. .. . . . ... . ... .....-(1
.. . .
··· ·· ····· · ···
2426

....
. . . .... ... ...
. .. .....
·. ·..._· I I 2428
...

To t
250. p s /ft SO .

2so.- - - - - - - - pstTt- -
Ro t
- - - - - - so.

Figure 10-Large vertical open fracture. Figure 12-Large open-closed open fracture.
6 FRACTURE EVALUATION USING BOREHOLE SONIC VELOCITY MEASUREMENTS SPE 14399

to 90% filled where the transmitter /receiver l:::.t differences

are again small. Open fractures with a width of 1 mm or
larger are in the middle section. Fig. 16 is a detailed sketch
by a geologist of the lower section.

~2 4 l0)

Figure 13-Borehole televiewer log.

Example 6
Shown in Figs. 14 and 15 is a heavily fractured zone with
several distinct vertical fractures. The top section is 70ffJo Figure 15-Borehole televiewer log .

2514 INTERVAL: 2525 FT 5 IN. TO 2527 FT 2 IN .

2516

1\/1"\&-----:-+-+- 5-CM-THICK
FRACTURE ZONE
2518

2520
CM

2522 ....... .. . ... .... .

. . . ... .... 1'&;-t-----t- 3.5-CM-THICK
FRACTURE ZONE

2524 .... . .. .. .. ... ......,..,

INTERVAL : 2528 FT TO 2528 FT 4 IN .
.. . . . .. . . . ... .
··· · · ·· · · · ··· ·
2526

2528

2530
THICK BRECCIATED
FRACTURE ZONE
VUG FILLED WITH
DOLOMITIZED &
LARGE SADDLE
2532 ·-_....._---_. . . ....:..---_
.. ...... MINERALIZED
_a.....J_~·-
DOLOMITE CRYSTALS

Tt.t
~·Itt 50.
Rt.t ___________50.
- - - - ------;Siil
1.2 CM

Figure 14-Heavily fractured zone . Figure 16-A detailed sketch of lower section of Example 6.
SPE 14399 0. Y LIU 7

Example 7
More of the irregular vertical fractures, some open and some
partially filled, are shown in Fig. 17 and 18. Borehole
enlargement was found at 2,745 ft.

--------
2734
-=-=-=-=-=-=-=-:

2738

2740

2742

2744 ®
® ®
2746

:!!~~=·%
©
,;:::>
::::::::: : :::: A ©
2748
\
2750

2752

... . ........ .
-
250. ps lft 50.
Rat
2so.- ---- --- -liSitT ___ --------5o.

Figure 17-lrregular vertical fractures.

Example 8
A thin straight vertical fracture less than 0.5 mm wide, but
open, is shown in Fig. 19. This can be compared with a
similar one that is 100% filled, shown in Fig. 20. The totally
Figure 18-Core photograph of irregular vertical fracture .
filled fracture shows no effect on either compressional or
Stoneley fj.f but a noticeable difference is found in shear fj.t. ....... ........................ .............. ......:..._.
. _ ......_ .......
_ .........,..!.._-_ ......._ .......
__ ~ ·

2832
® I
,.I
l
Rat-e"-...
\)
2834

2814
' Ta t-e

2836

....-~-.~- ....
~_,_--;--L.~___....-.L....J-......::·

2816 .:....--=...-_......._-_:-~.............::::-"-=
\ 2838
Rat-c " ' i Rat ~/
·==-="7==-=~=-=-=-=-=·
~--4-----~~~.~:+--------+--~
· ....:.._:......!_.:....-:-~--;-....:...~.....:.-~-~·

2818 ..:....=-=-=-=-=-=--:::-=-=~7. .
r -. . - . ..,- ....... r -...... .
------ - · ...,....~

2840
. -~-.~-.__.- ......-;-......-.:.....-.......- ......-:: .
2820
r-.
- - - -r-;-.-,...,...,...,
--- -
.-.-r-.~

Ta l-c
"-.....
.-
~-.~- ....-;-_,_....-........--;-___....-.__-;-.._-:: . 2842
/
~-.....:...-_..........
_~---
....:..---~
Tdt-St "
2822 Rat-St

2844 ......_....-:.......-;-_._-'---.1.-.L....J-...L...--;-....L..--:'"__j
___:r-...:..:·
. _ ......_ ......._ ...........!..........._ ,....._.......

2824

Tat Tat
250. ps lft 50. 250. ps lft 50.
Rat __ --------R~ t _ _ _ - - - - - - -
2-50.------- --- -listlt--- - ---- - ---SO. 250. pslft 50.

Figure 19-0pen vertical fracture less than 0.5 mm. Figure 20-Long, very thin 100% filled fracture.
8 FRACTURE EVALUATION USING BOREHOLE SONIC VELOCITY MEASUREMENTS SPE 14399

CONCLUSIONS and Exposition, Los Angeles, Oct.

From the examples shown and from other cases not shown 2. Kimball, C. V. and Marzetta, T. L.: "Semblance Pro-
here, the following conclusions can be drawn: cessing of Borehole Acoustic Array Data,'' Geophysics
1. Large open fractures produce extremely large changes (Mar. 1984) 274-281.
and differences in compressional, shear, and Stoneley 3. Hsu, K. and Chang, S. K.: "Multiple Shot Processing
transmitter/receiver !l.t values. The extent of the frac- of Array Sonic Waveforms for High Resolution Sonic
ture height can be estimated from these differences. Logs," paper presented at the 1985 SEG Annual Inter-
2. Large vertical and subvertical open fractures result in national Meeting and Exhibition, Washington, D.C.,
very large differences in transmitter/receiver !l.t shear. Oct.
3. Filled fractures have very little or no effect on 4. Boyeldieu, C. and Winchester, A.: "Use of the Dual
measurements of compressional and Stoneley transmit- Laterolog for the Evaluation of Fracture Porosity in
ter/receiver !l.t, but produce noticeable differences in Hard Carbonate Formations,'' paper presented at the
transmitter /receiver !l.t shear. 1982 Offshore Southeast Asia Conference.
4. Transmitter /receiver !l.t differences caused by borehole 5. Suau, J. and Gardner, J. S.: "Fracture Detection from
enlargements appear to show a reverse pattern from Well Logs," The Log Analyst (March-April 1980).
those of fractures. 6. Huang, C. F. and Hunter, J. A.: "The Correlation of
5. The bed boundaries have a very small effect on the Tube Wave Events with Open Fractures in Fluid-Filled
transmitter/receiver !l.t differences. Depth shifts may be Boreholes," Current Research, Part A, Geological
required because of the difference in refraction angles. Survey of Canada, paper 81-1A (1981).
Changes in waveform amplitude and attenuation may 7. Norris, R. L., Grine, D. R., and Arkfeld, T. E.: "The
provide direct evidence for the existence of fractures, but Use of Compressional and Shear Acoustic Amplitude
changes in the apparent velocities provide important infor- for Location of Fractures," J. Pet. Tech. (June 1964)
mation on the type and size of the fractures. Precise height 623-635.
and dip angle of the fracture can be determined when these 8. Paillet, F. L.: "Acoustic Propagation in the Vicinity of
differences are used in conjunction with dipmeter or Fractures Which Intersect a Fluid-Filled Borehole,''
borehole televiewer data. The method described here is sen- Trans., 1980 SPWLA Annual Logging Symposium,
sitive to fracture-generated couplings, is reliable, is little af- DD1-DD33.
fected by bed boundaries, and can distinguish fractures from 9. Vogel, C. B.: "Method and Apparatus for Detecting
borehole enlargements. Fractures," U.S. Patent No. 3,775,739 (Nov. 1973).
On-going research both in theoretical investigation and
10. Hsu, K., Brie, A., and Plumb, R.: "New Method of
scale model work is being conducted to gain full under-
Fracture Identification Using Array-Sonic Tool," paper
standing of the effect of fractures on sonic and other log-
SPE 14397 presented at the 1985 SPE Annual Technical
ging measurements.
Conference and Exhibition, Las Vegas, Sept. 22-25.
11. Koerperich, E. A.: "Investigation of Acoustic Boundary
ACKNOWLEDGEMENTS Waves and Interference Patterns as Techniques for
The author wishes to thank J. E. White, Colorado School Detecting Fractures," J. Pet. Tech. (Aug. 1978)
of Mines, Golden, CO, for his in-depth discussions on crack 1197-1207.
coupling; and David King, Schlumberger Well Services, for 12. Cheung, P. S. Y.: "Fracture Detection Using the Sonic
his detailed descriptions and drawings of fractures. The Tool," 1984 International Formation Evaluation Sym-
author also thanks Conoco Inc. for their permission to posium, Communication No. 42, Oct.
publish results used in the examples.
13. Wu, P.: "Comparison of Digital and Analog Techni-
ques for Determining Sonic Velocity from Borehole
REFERENCES Acoustic Waveforms," paper SPE 13286 presented at
1. Ingram, J. D., Morris, C. F., MacKnight, E. E., and the 1984 SPE Annual Technical Conference and Exhibi-
Parks, T. W.: ''Direct Phase Determination of Shear tion, Houston, Sept. 16-19.
Velocities from Acoustic Waveform Logs," paper 14. Liu, 0. Y.: "Stoneley Wave-Derived !l.t Shear Log,"
presented at the 1981 SEG Annual International Meeting Trans., 1984 SPWLA Annual Logging Symposium.