Fracture Gradient Prediction and Its
Application in Oilfield Operations
Ben A. Eaton, SPE-AIME, Continental Oil Co.
Introduction
The subject of many discussions and technical papers
in the last 20 years has been the prediction of the well-
bore pressure gradients that are required to induce
or extend fractures in subsurface formations. The sub-
ject merits this attention because of the frequently
recurring problems that arise from an inability to
predict fracture pressure gradients.
Encountered in several common types of opera-
tions in the oil industry are problems associated with
the prediction of formation fracture pressure gradi-
ents. When wells are being drilled in both new and
old fields, lost circulation is often a very troublesome
and expensive problem. Complete loss of circulation
has been disastrous in some cases. Many times, such
disasters could have been avoided if techniques for
calculating fracture pressure gradient had been em-
ployed in the well plans, and if casing strings had been
set, and mud weight plans had been followed accord-
ingly. In areas of abnormally pressured formations,
the prediction of fracture gradients during the well-
planning stage is extremely important. In fact, it is as
important as the prediction of formation pressure
gradients, which has received a great deal of attention
in recent years.
There are several published methods used to deter-
mine fracture pressure gradients. However, none of
these methods appears to be general enough to be
used with much reliability in all areas. In 1957, Hub-
bert and Willis published a classical paper that in-
cluded the development of an equation used to predict
In arriving at a new method of predicting formation fracture gradients, it was found that
overburden load, Poisson's ratio for rocks, and pressure gradients vary with depth.
Although the method was developed specifically for the Gulf Coast, it should be highly
reliable for all areas, provided that the variables reflect the conditions in the specific
area being considered.
OCTOBER, 1969
Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023
the fracture extension pressure gradient in areas of
incipient normal faulting. 1 Overburden stress gradi-
ent, formation pore pressure gradient and Poisson's
ratio of rocks were the independent variables that
were shown to control fracture pressure gradient, the
dependent variable.
In 1967, Matthews and Kelly published another
fracture pressure gradient equation that is different
from that of Hubbert and Willis in that a variable
"matrix stress coefficient" concept was utilized. 3 Later
the same year, Costley wrote about a similar idea. 5
Goldsmith and Wilson used a least-squares curve-
fitting technique and field data from the Gulf Coast
area to correlate fracture pressure gradient with for-
mation pore pressure gradient and formation depth. 4
They noted that the fracture pressure gradient in-
creased with increasing depth while the pore pressure
gradient remained constant.
In each of these cases, the problem for which a so-
lution was sought was to determine the bottom-hole
pressure gradient required to initiate or extend a frac-
ture. Results of the previous work show that fracture
pressure gradient is a function primarily of overbur-
den stress gradient, pore pressure gradient, and the
ratio of horizontal to vertical stress. There is argu-
ment for a fourth variable in that in many cases break-
down fracture pressure gradient is greater than the
fracture extension pressure gradient. However, if the
fracturing fluid is able to penetrate the formation
through the pores or existing cracks, there is very little
1353
difference in the two fracturing pressure gradients. data of Figs. 1A and lB. Note that PwlD increases
Therefore, since field data are used here and since it only when plD increases. Similarly, the log data for
is not possible to know for certain whether or not a Well B were used with Fig. IB to determine the pres-
wellbore contains natural fractures or faults, this dif- sure gradient of Well B. The results make up Curve 1
ference will not be discussed further. of Fig. 5. Curve 2 of Fig. 5 shows the fracture gradi~
One independent variable that partially controls the ent computed by the Hubbert and Willis equation.
fracture pressure gradient is the formation pore pres- The same behavior is to be observed as in the first
sure gradient. Formation pressure gradients may be example. However, experience has shown that PwlD
calculated from logs or determined by bottom-hole increases with depth, regardless of the pressure be-
pressure bombs. Since formation pore pressure is not
the subject of this paper, it is assumed that accurate
formation pressure data are available to use in the o
- W-:-I I
fracture pressure gradient calculations.
Another important variable is the overburden stress
liES
CONDUCTIVITY 1
1/
INTERVAL._
TRANSIT
,TIME 1
!-
gradient. Overburden stress gr:'1dient is normally as- III
sumed to be 1.0 psi/ft. But this is a high average 2
I~ NORMAL ,#'

Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023

value for a nonconstant variable, and one should real- 1 LINE
3
ize that it may be seriously in error in some areas, I '1
such as on the Gulf Coast at shallow depths. ill
4
The amount of horizontal stress caused by the ver- (I
tical stress (net overburden) is a function of Poisson's 5 .1 !
ratio of the rocks in question and is another important 1.., el.
variable. The horizontal-to-vertical stress ratio has I ..,.7 1/
6
been considered to be constant with depth in many :: -..' II
0
cases, and equal to one-third. This corresponds to a 0 Ij
Q 7 J-'
Poisson's ratio equal to 0.25. However, this frequently I
:I:
used average value for Poisson's ratio for rocks may f-
a. a
well be in error. Values from laboratory experiments w
c
vary from well under to well over 0.25, but are never 9
greater than 0.50. 1
Several authors have questioned the foregoing as- 10 I-
....\ .•..
sumptions. Hence, it seems worthwhile to review and
compare the various techniques used to calculate II
:\. '
)
I .... ('"
fracture pressure gradients. To do so, the data of Figs. ! ,
1A and 1B were used to calculate the fracture pres- 12
i
sure gradient by several methods. 1 :""
,
13
Previous Methods of Predicting
Fracture Pressure Gradient 14
I .'
••.. ,
I "
i .(, I i
Hubbert and Willis showed the fracture pressure I I l
15
gradient to be a function of overburden stress gradi- 200 400 1000 2000 50 100 300
ent, formation pressure, and stress ratio. l This leads 300 600 3000 70 200
to Eq. 1: Fig. lA-Log data from Frio formation,
Nueces County, Tex."
Pw =
D
(~_l!-)(_v + L
D D I-v
) D'
(1) 0.4

When the assumptions are made that SID = 1.0 and

\.
v = 0.25, Eq. 1 reduces to 0.5

'\0 - 10.0

p", = ( 10 + 2") ._1 (2)

-
"'"
0.6
D . D 3'

'" '"
c:
12.0 ~
J;;
which is known as the Hubbert and WiIlis equation. 0.7 '0
....... ....z
Under these conditions, the fracture gradient is cal- o0 14.0 'cc":
culated to be a constant with increasing depth for all
normally pressured formations. This is known to be 0.8
0"'0 ~
:E
'"
(5
:<
....

0~ 0
l"--.... 16.0 ii
untrue for many cases in the Gulf Coast. Actually,
~
Eq. 2 predicts values that are usually too low com- 0.9 """- .0 -
r-
pared with values from field data. However, this does 18.0
not mean that Eq. 1 is invalid. It is more probable
that the assumptions used to obtain Eq. 2 are in error. 1.0
1.0 1.5 2.0 3.0 4.0 5.0
Eq. 2 was used for comparative purposes to calculate NORMAL Rtsh)/OBSERVEO R(shl

Curve 2 of Fig. 2, using the pressure gradient curve Fig. lB-Relationship between shale resistivity parameter
data_ (<:::l,lJ:'v.~_ C()(Fj£,_n.flS .detepnin~dfrom the log Rn('h)/R.b\o.h) and reservoir fluid pressure gradient.·
13_~4 JOURNAL OF PETROLEUM TECHNOLOGY
havior, until the abnormal pressure section is trav- load is 1.0 psi/ft. To calculate a fracture gradient by
ersed. Then Pw/D may decrease as shown by the this method one must use the following procedure.
other curves in Figs. 2 and 5. 1. Obtain the formation pore pressure.
Matthews and Kelly3 presented a fracture gradient 2. Determine the effective stress, (J' = 1.0 D - p.
equation similar to Eq. 1. However, they introduced 3. Determine the depth D i for which the matrix
the concept of a variable horizontal-to-vertical stress stress (J' would be the normal value:
ratio. Fig. 3 is a reproduction of their curves showing
the variable stress ratio as a function of depth for two
areas. Again one must assume that the overburden
4. Use the value of D i from the preceding step with
o Fig. 3 to determine K i •
II CURVE I. FORMATION PRESSURE 5. With the resulting data, calculate the fracture
2
I f~ \ GRADIENT
2. HUBBERT Af'olP WILLIS
gradient using the Matthews and Kelly fracture gra-
~ \ FRACTURE GRADIENT
WITH. =0.25
dient equation, which follows:
,\ 3. SAME AS 2. EXCEPT

~=K{;)+~ .
4 • = 0.32
(3)
l~ 4. MATTHEWS AND KELLY

Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023

FRACTURE GRADIENT
.......
6 1-/ 5. GOLDSMITH AND WILSON
FRACTURE GRADIENT (Note: Matthews and Kelly call this the breakdown
./ / 6. NEW METHOD WITH

8
. \ VARIABLE • gradient and say that it is higher than the fracture
extension gradient.)
t 1;-..... 2 1/3 \/4 6. Plot the fracture gradient as a function of depth.
o
o
o
.. 10 ~ ............... .
\r.5
\ In this manner, Curve 4 of Fig. 2 and Curve 3 of
Fig. 5 were generated. The effect of depth and forma-
t
'"
o ~ f\~\rs tion pressure is readily evident. However, there ap-
12
i 1\, pear to be two weaknesses in the approach, one of
.\ which is the assumption that the overburden stress is
14 Il" equal to 1.0 psi/ft of depth. The other weakness is
that the stress ratio used in calculating the fracture
/ 1/ , I
J gradient in abnormally pressured formations is that
,,,,
I
16
I
of the deepest normally pressured formation. The
I Matthews and Kelly approach represents a significant
8
I
I
I
I', advancement in fracture gradient technology, and the
! I
, I
I variable stress ratio concept is quite valid when com-
I
11
MUD WE\~HT
13 I 17 I
pared with field data analysis.
~i
20
o .2 .4 .6 1.0 .8 1.2 1.4
Recently, Goldsmith and Wilson 4 found that the
PRESSURE GRADI ENT - psllll presently existing techniques for calculating fracture
Fig. 2-Formation and fracture pressure gradients. gradients were inadequate. Using a great deal of data
on lost circulation and squeeze pressure, they devel-
oped empirical equations, using least-squares curve-
2 fitting, that appear to predict very well the fracture

~
gradients for a localized area. The equations are long

4
"~ ~ SOUTH TEXAS GULF COAST
and somewhat difficult to use, but this technique was
employed in calculating Curve 5 of Fig. 2. The vast
disagreement of these techniques is well illustrated.

''""
Costley5 recently published yet another technique
I'" ~
that is similar to the Matthews and Kelly method. The
same basic assumptions are involved; therefore, the
method will not be discussed further. Nevertheless,
8
o
o
Q
\ \
\
the data published by Costley were used with other
data to develop the method that follows.

\ \
I
:I:
l-
lL A Revised Approach
~ 12

1\ '~
LOUISIANA GULF COAST
Throughout the remainder of this work, it is postu-
lated that the assumotions leading to Eq. 1 are valid

\ \ and that all of the independent variables are functions

,
of depth. The problem is to determine the relationship

\, \
\6
of overburden stress, pore pressure, and Poisson's
ratio with depth. Since it is accepted here that abnor-
mal formation pressure gradients may be determined
from logs, that aspect of the problem is solved. The
20
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 next steps are to assume that· the overburden stress
MATRIX STRESS COEFFICIENT- Kj
gradient is 1.0 psi/ft, then to solve Eq. 1 for the stress
Fig. 3-Matrix stress coefficient: ratio group
OCTOBER, 1969

(4)
and to evaluate Eg. 4 with field data.
A great deal of data from the analysis of hydraulic
fracturing treatments in West Texas was published by
Crittendon. 6 These data were used to develop the left
curve of Fig. 4. It can be seen that for the producing
formations of the West Texas area, the assumptions
SjD = 1.0 and v = 0.25 are valid.
Data given by Costley were used to back-calculate
the middle Poisson ratio curve of Fig. 4. Note the
curvature of the trend of Poisson ratio vs depth for
the Gulf Coast area. This is caused by the sediments'
available. These logs were used to plot bulk density
vs depth, which is shown in Fig. 6. The values for
bulk density were read at the mid-point of each 1,000-
ft interval and averaged step by step downward to
20,000 ft of depth. In this manner, the overburden
stress curve of Fig. 7 was produced. The value of
overburden stress read from the curve at any depth
represents the real average overburden gradient at
that specific depth. Further averaging need not be
done.
The same procedure was used for similar data from
wells in the Santa Barbara Channel. Bulk densities
from logs and the resulting overburden stress gradient
curve are shown in Figs. 10 and 11, respectively, with
the results being similar to those given in Figs. 6 and
7. It was concluded that variable overburden stress

Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023

being younger and more compressible near the sur-
face, but less compressible and more plastic with
depth. For this reason the curve approaches 0.5 as an
upper limit. This limit is the Poisson ratio of an incom-
pressible material in the plastic failure environment.
Example results of fracture pressure gradient cal-
culations using the middle Poisson ratio curve of Fig.
4 and Eq. 1 are included for comparative purposes.
These are shown by Curve 6 of Fig. 2 and Curve 4
of Fig. 5.
From the preceding calculations it becomes evident
that the variation of overburden stress with well depth
must be determined where formations are compressi-
ble, such as in the Gulf Coast area. A composite
group of density logs from many Gulf Coast wells was
gradient curves, as determined from density logs of
good quality, are far superior to any assumed constant
number.
Based on Eq. 4, the same field data, and Fig. 7,
the Poisson ratio trend for the Gulf Coast area was
back-calculated and plotted on the right side of Fig.
4. Note that this curve approaches the middle curve
at greater depths where the overburden stress gradi-
ent does approach 1.0 psi/ft. It was concluded that
Figs. 4 and 7 could be used with formation pressure
data and Eq. 1 to predict accurately the fracture gra-
dients in the Gulf Coast. The same method will apply
in other areas, provided that the overburden stress
gradient and Poisson ratio curves are determined from
good data.

o
, GULFCOAs~.1
VARIABLE
2 I\. \. OVERBURDEN
o

4
'\\ 2
-LEGEND-
I. PRESSURE GRADIENT
FROM LOG

\
OVERBURDEN

6
EQUALS 1.0psi/lt
SHALES
11\ 2. FRACTURE GRADIENT
a
HUBBERT WILLIS
v -0.25

\\ 4
3. F'RACTURE GRADIENT

1\' .~-
I-
i-
;j
a:
w
6
\ a
MATTHEWS KELLY
4. FRACTURE GRADIENT
PRESENT METHOD

10

WEST TEXAS \ w
::E
w oo 8
i\
"-""," \
OVERBURDEN a:
e- EQUALS 1.0 psi 1ft
l- 9
12 ~- I

PR6D~_ J:
I-
FORMATIONS ::;10

'""" ~ ~
c
14

12

/ ~ !}I
16

14
I. 2. 3.4.
18

16
I
20
o 0.1 0.2 0.3 0.4 0.5 0.6 o 0.2 0.4 0.6 O.B 1.0 1.2 1.4
POISSON'S RATIO - v GRADIENT- psi/fl

Fig. 4-Variation of Poisson's ratio with depth. Fig. 5-Well B, East Cameron.
Fracture pressure gradient comparjson.

1356 JOURNAL OF PETROLEUM TEcHNOLOGY

 The preceding discussion illustrates a method
whereby all three variables that control the fracture
pressure gradient are determined from well data. A
nomograph for solving Eq. 1 to predict fracture pres-
sure gradients is shown, with an example calculation,
in Fig. 12.
How to Apply the New Technique
To demonstrate better how to apply this method of
predicting fracture gradients, an example follows,
showing how the technique should be employed in
drilling well plans.
1. Fig. 8 shows a plot of resistivities for a well in
the East Cameron area. Data from Fig. 8 are used
with Fig. IB to produce the formation pressure curve
of Fig. 9. Mud weights for this area should be deter-
mined in this manner during the well planning stage.
according to the curve of Fig. 7. However, all that
must be done to arrive at the correct fracture gradient
curve is to use Fig. 7 and interpolate between the
various gradients of Fig. 9. The interpolation is shown
as the correct fracture gradient curve on Fig. 9. Note
that near the surface, the curve is between the 0.8 and
0.9 curves, and at the 4,000 ft mark it crosses the 0.9
curve and approaches the 1.0 curve with depth. This
illustrates somewhat elaborately the effect of over-
burden stress gradient on fracture pressure gradient.
However, in actual practice only the true fracture
gradient should be used.
3. The casing points can be selected by making use
of the formation pressure gradient curve (minimum
mud weight curve) and the true fracture gradient
curve. In the example of Fig. 9, surface pipe was set
at 3,100 ft. If we assume that minimum mud weights

Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023

The mud weight scale at the bottom of Fig. 9 and the
formation pressure gradient curve dictate the mini-
mum mud weight program.
2. The next step is to determine and plot fracture
gradient vs depth. (This is illustrated in Fig. 9, which
shows five fracture gradient curves.) It is assumed
that the overburden gradient averages 0.8 psi/ft all
the way down. The graphic solution of Eq. 1 shown
by Fig. 12 is used with the variable Poisson's ratio
curve (right-hand curve of Fig. 4) to find the fracture
gradient vs depth. The process is repeated for assumed
overburden stress gradients of 0.9,1.0, and 1.1 psi/ft.
These four fracture gradient curves are not correct,
because the overburden gradient varies with depth
were used during drilling, a protective string of pipe
must be set at no deeper than 10,400 ft. A vertical
line from the pressure gradient curve at 10,400 ft
extends upward to a point slightly above that repre-
senting the surface pipe shoe. In this case, break-
down and lost circulation would not occur. However,
if drilling were to continue to 11,000 ft before a pro-
tective string is set, breakdown would occur. A verti-
cal line from the pressure gradient curve at the 11,000
ft mark extends up and intersects the fracture gradient
curve at the 5,000 ft mark. Breakdown would occur
anywhere between 5,000 and 3,100 ft of depth. This
is simply to say that casing points should be selected
by using both the formation pressure and the fracture

o o
"~
"-
'"~" \
"~ ""
4 0"
\~
4 \
6
\ l\.,,- 6 ~
I\~
UPPER LIMIT

8 \ \~
OFALLD~A POINTS

8
\
;;;
o
Q
~IO
ll.
....
o

12
--
LOWER LIMIT \
OF ALL DATA POINTS \ ~\\
~\\\\
\ 1\
\
\
\ \ .\
14
\
\
~ \ 14
\
\ \\ \
,
16 16

18
1\ \ \
18
\
20
1.9 2.0 2.1 2.2 2.3
\\\
2.4 2.5 2.6
20
0.70 0.75 0.80 0.85 0.90 0.95 1.00 1.05
8ULK DENSITY-GM/CC OVERBURDEN STRESS GRADIENT (psi/If)

Fig. 6--Composite bulk density curve from density log Fig. 7-Composite overburden stress gradient for all
data for the Gulf Coast. normally compacted Gulf Coast formations.

oCtOBER,- 1969 - - 1351

 0 o
~
2
~
.,
4 v 2
'\
6
!(\
"I\.
ti
3 \
0
0
2 8 l\ ....... 4
1\
\
I
~
~
0- ~ 0
0
.... 2,
0
10
~ \ 5

~
~
LJ
/' \ ~
lL
....
Ilt.,
0

Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023

12 \ 6

<;
\
\ 7 \
\
14
\ SUPER
t::> PRESSURff

16 I~ 8

.1 .2 .3.4.5.6.8 1.0 2.0 3.0 4.05.0

R.h
9
Fig. 8-Log data, Well C, East Cameron.

10
2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7
BULK DENSITY - GM/CC

Fig. lQ-Bulk density curve from density logs,

Santa Barbara Channel.

o
, 1\
o

1 2
\
\,
§ 0.8 0.9 1.0 1.1
2

4
FRACT:L ~ ~\t\ 3

\ \, \ \
PRESSURE

4
\

\
6

....... \ \\ \ t
5

\
o 8

i\\~1\ "
.....
\
8 o
, ~ 8
-;- 6
~

Ii:.... r--....
10

\
~

/--...... ~
o

12
FORMATION ___
PRESSURE N \ \\ \
lL
....
o
7

14
\ \
\
\
I
8
~
,
l\.- " \

9
6

EQUIVALENT MUD WEIGHT LBS /GAL

9 ID II 112 13 14 15 16 I17 I 18 19 10
0.7 0.75 0.8 0.85 0.9 0.95 1.0 1.05
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
GRADIENT- psi/II OVERBURDEN LOAD - PSI/FT

Fig. 9-Fracture gradient with variable overburden and Fig. ll-Overburden stress gradient,
Poisson's ratio included, Well C, East Cameron. Santa Barbara Channel.

1358 JOURNAL OF PETROLEUM TECHNOLOGY

gradient plots. All of the foregoing serves to illustrate .7

the application of fracture gradient prediction tech-

niques in drilling well plans. .6

Other Applications
A knowledge of fracture gradient prediction methods
is extremely useful in such everyday operations as ....
.~
.5
- -- - v- V- i
I
I
I
I
I

cementing, sand consolidating, matrix and fracture .!. .4

z
acidizing, and hydraulic fracturing.
Another important application is in secondary re- .
OJ
i5
~ .3
covery. In most injection operations, it is desirable to 0-
...'" ..... CORRESPONDS TO A PORE
stay below the fracturing pressures to prevent chan- PRESSURE OF 1455psi
.2
neling from injector to producer. Of course, in the
event that such vertical fractures all line up parallel I
with lines of injectors, good linear sweep patterns .1
o 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16
result. In such cases, the general direction of the frac- q - bbl/day x 10'2
tures should be determined and injection and pro- Fig. 13-Fracture gradient of a West Texas

Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023

ducing wells should be lined up to take advantage of water injection well.
induced fractures.
In many cases, injection is started in old wells that
have been producers for years. Here the formation
pressures are usually very low in the surrounding for-
mation. Low formation pressures cause low fracture
pressure, much the same as high formation pressures
cause high fracture pressures. Eq. 1 predicts this be-
havior. However, more vivid proof is given in Fig. 13,
in which the results of a pressure-rate test are shown.

'-D 1.00

0.95

SiD P/
D
1.0 0.' 0.90

0.8
;: O.8~

...
.....
...
iii ~~ II)

0.' ~ 0.7 0.80

OJ
z
OJ
0:
:::>
"-
"
0

~
II)

.
0: II)

...
-
:::> OJ
...
0:
0.8 o· ~O 0 0.75 ."
0:

\ '"'"
-------
OJ
.~
> z o· U>
0 0 II)
't-. c;
0.8 ~~ 0.8
0 ... 0
oJ>
1> 0.70 '"'"

-----
0: ~
Q. '!>~
...0 '0 ."
U>
"-
...
0.4 0.&5
,..

0.60

EXAMPLE: DETERMINE FRACTURE GRADIENT AT 12,000' WITH

FORMATION PRESSURE OF 0.67 PSI/FT. AT 12,000' 0.55

OVERBURDEN LOAD IS 0.96 AND POISSON'S RATIO

IS 0.46. FROM NOMOGRAPH FRACTURE GRADIENT
IS 0.91 PSI/FT 'OR 17.5 PPG. 0.80

0.45

0.40

Fig. 12-Fracture gradient nomograph.

OCTOBER, 1969 . 1359

Fracturing occurred when a gradient of 0.57 psi/ft
was reached, as shown by the sudden change in slope
at about 700 BID injection. New in-fill wells in the
same area are hydraulically fractured with gradients
of 0 68 to 0.70 psi/ft. The relatively low fracture
gradients of old producers are due to low formation
pressures. Many old producing wells are fractured
unintentionally when converted to injection wells.
The pressure in the producing formation is so low
that it cannot withstand even the gradient of fresh
water.
One more example of the effect of formation pres-
sure on fracture gradient is that three pressure-rate
tests were run on a California injector on Dec. 13,
1961; June 11, 1962; and Dec. 3, 1962. Fracture
pressure gradients were determined at sudden changes
in slope of pressure-rate plots. These gradients were
0.748, 0.864, and 0.993 psi/ft on their respective
dates. These data illustrate how the fracture pressure
gradient is affected by changes in formation pressure
around a wellbore.
How to Develop the Technique
for Other Areas
The data necessary to develop this method for other
tectonically relaxed areas of the earth are as follows:
1. Overburden stress gradient vs depth. Such data
can be derived from bulk densities taken from logs,
seismic data or shale density measurements. A plot
of bulk density vs depth can then be converted to a
plot of average overburden stress gradient vs depth.
2. Actual fracture pressure gradients for several
depths. These can be lost-circulation or squeeze data
or actual fracturing data.
3. Formation pressures that apply to the data in
Item 2. (In Items 2 and 3, the depths must corre-
spond.)
With these data and Eq. 4, the Poisson's ratio curve
for the area can be back-calculated and plotted vs
depth. The result will be a curve similar to those of
Fig. 4 and a curve similar to that of Fig. 7. With these
curves and Fig. 12, fracture gradients can be pre-
dicted quite easily and quickly. These values can be
plotted as a function of depth and the resulting curves
can be used in all the operations previously described.
Conclusions
In summary, several conclusions have been drawn.
1. In drilling well plans, well stimulation plans, and
secondary recovery plans, fracture pressure gradient
should be considered.
2. Poisson's ratio for rocks increases with depth in
the Gulf Coast area.
3. The Poisson's ratio trend, which can be back-
calculated using field data, will not be exactly the same
for data from different areas.
4. In the Gulf Coast area, the average overburden
stress gradient does not equal 1.0 psi/ft, but instead
is about 0.85 psi/ft near the surface and increases
1360
smoothly to 1.0 psi/ft at about 20,000 ft of depth. A
similar trend has been established for the Santa Bar-
bara Channel area.
5. The present method is a modification of the Hub-
bert and Willis approach, using a variable Poisson
ratio and a variable overburden stress gradient with
depth. The method will predict the fracture gradient
for the Gulf Coast area with as much accuracy as the
log-derived pressure gradient.
6. Using this approach and field data, the same
method can be developed for other areas.
7. The method is very simple, and Fig. 12 may be
used as a working chart for any area.
Nomenclature
D = depth, ft
Downloaded from http://onepetro.org/JPT/article-pdf/21/10/1353/2221855/spe-2163-pa.pdf/1 by guest on 31 December 2023
D i = equivalent depth of lowermost normally
pressured formation, ft
K i = matrix stress coefficient
Pw = wellbore pressure, psi
P = formation pressure, psi
S = overburden stress, psi
0' = net effective overburden stress, 0' = S - P,
psi
v = Poisson's ratio
Acknowledgments
I should like to express appreciation to the manage-
ment of Continental Oil Co. for permission to publish
this paper. Special thanks are given to C. H. Hefner
for his preparation of the nomograph, Fig. 12.
References
1. Hubbert, M. King and Willis, D. G.: "Mechanics of Hy-
draulic Fracturing", Trans., AIME (1957) 210, 153-166.
2. Howard, G. C. and Scott, P. P.: "An Analysis and the
Control of Lost Circulation", Trans., AIME (1951) 192,
171-182.
3. Matthews, W. R. and Kelly, John: "How to Predict Forma-
tion Pressure and Fracture Gradient", Oil and Gas J.
(Feb. 20, 1967).
4. Goldsmith, R. and Wilson, G.: staff engineers, Production
Engineering Services, Continental Oil Co., Houston, Tex.,
private communication.
5. Costley, R. D.: "Hazards and Costs Cut by Planned Drill-
ing Programs", World Oil (Oct., 1967) 92.
6. Crittendon, B. C.: "The Mechanics of Design and Inter-
pretation of Hydraulic Fracture Treatments", J. Pet. Tech.
(Oct., 1959) 21-29.
7. Wuerker, R. G.: "Annotated Tables of Strength and Prop-
erties of Rocks", Drilling, SPE Petroleum Transactions
Reprint Series No.6 (1963) 23-45.
8. Hottman, C. E. and Johnson, R. K.: "Estimation of For-
mation Pressures from Log-Derived Shale Properties",
J. Pet. Tech. (June, 1965) 717-722. JPT
Original manuscript received in Society of Petroleum Engineers
office July 2, 1968. Revised manuscript received June 26. 19~9.
Paper (SPE 2163) was presented at SPE 43rd Annual Fall Meeting
held in Houston, Tex., Sept. 29·0ct. 2, 1968; and at SPE Fifth Annual
Eastern Regional Meeting held in Charleston, W. Va., Nov. 7·8,
1968. © Copyright 1969 American Institute of Mining, Metallurgi·
cal, and Petroleum Engineers, Inc.
This paper will be printed in Transactions volume 246, which
will cover 1969.
JOURNAL OF PETROLEUM~CHNOWGY