SURFACE DEFORMATION ASSOCIATED WITH OIL AND GAS
EIELD OPERATIONS IN THE UNITED STATES
R.F. YERKES and R.O. CASTLE
U S . Geological Survey, Menlo Park, California 94025
ABSTRACT
Surface deformation associated with oil and gas field operations in the United States
consists of (i) differential subsidence centering on the fields, (2) centripetally directed
horizontal displacements, and (3) faulting. The number of documented examples and
maximum measured movement for each effect are:
Number of fields
Effect
California Texas
Maximum recorded movement
Amount
(meters)
Period
Field
~
Differentia] subsidence
Horizontal displacement
Faulting
22
3
5
-47
> 8.8
3.66
0.74
1928-1966 Wilmington,Calif.
1937-1966 Wilmington, Calif.
1932-1967 Buena Vista, Calif.
Faulting is commonly high-angle, normal, and peripheral to the subsidence bowl
(Goose Creek and Mykawa, Texas; Inglewood and Kern Front, California), but may be
low-angle,reverse,and central (Buena Vista, California).
Strain patterns developed over these fields are similar to those measured bv D.V.Deere
over a Texas sait dome where all three surface effects were observed during Frasch
extraction of sulfur. In this case, a central compressional zone extended out to beyond
maximum subsidence) and a surrounding tensional zone extended
the 0.35 S isobase (S
to the periphery of the bowl, wherc high-anglenormal faulting occurred.
=
RESUME
L a déformation superficielle subséquente à l’exploitation de gisements pétrolifères
aux Etats-Unisse compose (1) d’un affaissement dont 12 maximum est centré au-dessusdes
gisements,(2) de déplacements horizontaux centripètes,et (3) de faillage. L e nombre de cas
suivis et le déplacement maximum associé aux divers types de déformation sont repris
ci-dessous:
Déplacement maximum observé
Nombre de champs
Type de déformation
Californie Texas
Subsidence
Déplacement horizontal
Faillage
22
3
5
-47
Total
(mktres)
> 8,8
3,66
0,74
Période
Champ
1928-1966 Wilmington (Cai.)
1937-1966 Wilmington (Cai.)
1932-1967 Buena Vista (Cal.)
Les failles sont en général de pendage élevé, normales et en bordure du bassin de
subsidence (Goose Creek et Mykawa (Texas); Inglewood et Kern Front (Calif.)), mais
il en cst de faible pcndage, inverse et occupant le centre du bassin (Buena Vista,
California).
Les types dc déformation observés sont semblables à ceux qui ont été mesurés par
D.V. Deere au sommet d’un dôme de sel (Texas). Ici, les trois effets de surface ont
été obsirvés lors de l’extraction du soufre par IC prxédé Frasch. Dans ce cas, une zone
de compression centrale s’étendait jusqu’à l’isobase 0,35 S (S
subsidence maximum)
et &tait ceinturée par une zone de tension périphérique où s’est produit un faillage normal
avec pendagcs élevés.
=
55
R.F. Yerkes and R.O. Castle
INTRODUCTION
More than 40 known examples of differential subsidence, horizontal displacement,or
surface faulting have been associated in time and space with operation of 27 California
and Texas oil and gas fields (table 1). Altough the magnitudes of these movements rarely
attain those developed over the Wilmington oil field,each effect has led to costly damage
or destruction of surface structures.The necessity of avoiding such effects in urban areas
emphasizes the importanceof detecting and monitoring them;this necessity may require
volume-for-volumereplacement of withdrawn liquids.
DIFFERENTIAL SUBSIDENCE
Differential subsidence is the most common and widespread ofthe effects,but it is easily
detected only in shoreline areas. Where level surveys-havedetermined the size and shape
of the subsidence bowl, it centers over and extends well beyond the producing area. Production in the listed fields comes from unconsolidated to poorly lithified and poorly
sorted sands,generally Miocene or younger in age. Median depths of production range
from about 360 to 3 900 m and exceed 1 800m in only four cases.
Differential subsidence associated with oil field operations was first recognized in the
Goose Creek oil field,on the Texas Gulf Coast.The producing strataconsist of essentially
unconsolidated sands and intercalatedclay stringers;median depth ofproduction is about
600 m.Production began in 1917; by 1925 subsidence centered over the producing area
had exceeded 1 m and involved an area of about 1 1 km2,and normal faults as long as
0.7 km,having displacements downwareon the oilfield side,as great as 0.4m,had formed
along of the long margins of the eliptical subsidence bowl (fig. 1). The subsidence was
accompanied by submergence of considerable near-sealevel areas within the producing
limits.Litigation was then initiated by the state over title to the newly-submerged“tidelands”and the value of the extracted minerals.Title was awarded to the oil field operators
who contended that the subsidence was caused by removal of large volumes of materials
and was therefore an act of man.
The most spectacularand costly case of differential subsidence is that of the Wilmington oil field,near Long Beach,California.By 1966,after 30 years ofproduction and 8 of
repressuring by injection,an elliptical area of more than 75 km2 had subsided more than
8.8m (see Poland and Davis,1969, for a comprehensive,up-to-datereview). Production
is from poorly sorted,unconsolidated to semi-consolidatedsand containing thin interbeds
of silty shale and the median depth of production is about 1 O00 m (fig.2a). Successive
collar surveys of well casings in the subsiding area at Wilmington for the first time have
quantitatively related surface subsidence to subsurface compaction.Although the producing section extends to depths greater than 2 300m,subsurface compaction was concentrated in the 650to 1 200-meterinterval(fig.2b). Cumulativecompaction ofnearly 3 m
had occurred in the Upper Terminalzone (960-1125 m)by 1960;a similar amount occurred in the Tar and Ranger zones above,which aggregate about the samethickness (180m)
as the Upper Terminal.The cumulative compaction in the several zones agrees roughly
with the surface subsidence at the well during the same time interval (Allen,1968, p. 25).
The surficialdeposits above 650 m were in tension and well casingswere stretchedat least
0.6m before full scale injection began in 1957;during the succeeding 2% years of repressuring the casings were shortened as much as 0.75 m as the underlying producing zones
expanded.
Large subsidence bowls have also been delineated over the Huntington Beach oil field,
at the shoreline just southeast of Wilmington; over the Long Beach field immediately
northeast ofWilmington;and over the Inglewoodfield,an inland field about 30 km northwest of Wilmington. Although differential subsidence over these fieldshas been minor
56
1
0
2 2
r-m
00
2
CI
A
m
m
CI
O
P.
2
O 0
r-E
i n *
O
f3
CI
*
?
O
2
o o o v l o o v l o o
m o t - m m o b ~ w V l N m o ~ o m o
~
~
m
~
r
n
~
~~ l
t o -v lwc om m m ~ r n
i m
3
3
m - 3
m
vlooooooo
Surface deformation associated with oil and gas field operations in the United States
E
.z-
57
m
8
A
m
r.
A A
gm vz
0
m o
O
0
'O
0
O
o o m
2 8 2
A A
w m o
\o991
O 0 0
0
R.F. Yerkes and R.O.Castle
58
3
3
m m
"o\?
A A A
- 8 0 0
0 0 0 0 0 0 0
m
b o o o o e
m N
m w m e
m m -w
Surface deformation associated with oil and gas field operations in the United States
59
R.F. Yerkes and R. O. Castle
(less than 1.75 m) in comparison with that at Wilmington,that at Inglewood was accompanied by horizontal displacement and surface faulting that led to costly destruction of
surface structures near and beyond the periphery of the bow! (see California Department
of Water Resources, 1964).
Fipure 2-0
Fiqute 2-b
FIGURE2. Upper producing zones and compaction at the Wilmington oil field,southern California
2a, approximate thickness and age of upper producing zones near the center of subsidence: 2b,
cumulative compaction in the upper three zones during 1945-1965 as based on casing joint measurements in a well ( W-2)about 600 m southeast of the center of subsidence. Adaptedfrom Allen (1968,
figures 2 and 5)
HORIZONTAL DISPLACEMENT
Centripetally directed horizontal displacements commonly accompany differential subsidence,but these can be documented only by triangulation and hence have been determined
in only three United States oil fields. In the Inglewood and Wilmington fields the maxim u m horizontal displacementexceeded 35 percent of the maximum differentialsubsidence;
it was located about halfway up the flanks of the subsidence bowl, from which location
it decreased progressively to zero at both the center and periphery of the bowl. The maxim u m known horizontal displacement occurred within the Wilmington oil field (fig.3),
where 3.66 m was measured between 1937 and 1966.This displacement resulted in horizontal strains greater than 1.2 percent. Between 1957 when repressurization began and
1967, several survey stations along the east part of the subsidence bowl (an aera where
vertical rebound of about 17 percent of the total differential subsidence has occurred)
recovered as much as 80 percent of their measured horizontal displacement.
SURFACE F A U L T I N G
In addition to being the most prominent and easily documented effect, surface faulting
m a y develop suddenly and it is therefore potentially more damaging to surface structures.
Surface deformation associated with oil and gas field operations in the United States
Subsidence-associatedsurface faulting is most commonly high-angleand normal,peripheral to the subsidence bowl and downthrown on the oil field side; it commonly trends
subparallelto the isobase contours. Exemples of this type are over the Goose Creek and
Mykawa fields in Texas,and over the Inglewood and Kern Front fieldsin California.
FIGURE3. Surface deformation over the Wilmington oil field. 3a, horizontal strain; 3b, horizontal
displacement during 1937-1962 and uertical subsidence during 1928-1962,Profile is drawn along
the minar axis of the subsidence b o d . Figure 36 from an unupblished study by Jan L a w and D .R.
Allen, Deparfment of Oil Properties, Long Beach, California
The surfacefault at Kern Front is,at 5 km,the longest known;it followsa pre-existing
subsurface fault of even greater extent that forms the east boundary of the Kern Front
field. Displacementon the fault has not been great,but movement has been continuous
for more than 20 years. The displacement has been described in the context of tectonic
faulting unaccompanied by seismic activity (Hill,
1954,p. li). However,the fact that it
(1) is associated with measured differential subsidence,(2) continues with continuing
extraction,(3) occurs along a pre-existing fault that dips toward the differential subsidence
over the Kern Front field,and (4) has not been accompanied by seismicity,suggests that
it is attributable to effects of subsurface compaction.
A second type of faulting,low-angle,reverse,and central to the subsidence bowl,has
been recognized in only one field. A gently north-dippingreverse or thrust fault about
2.6 km long in the center of the Buena Vista oil field,California (fig. 4
), has been known
for more than 35 years. Dip slip of at least 0.74m accumulated during the interval
1932-1967(Nason and others,1968,fig.2), giving a average,fairly constant rate of about
2 cm/year-a rate that continues.Faulting between 1942 and 1964 was accompanied by
differential subsidence exceeding 2.3 m,and by horizontal displacementof0.39rn between
1932 and 1959. Even though the fault cannot be traced to depth in the numerous wells,
it has long been cited as an example of active tectonism (Koch, 1933;W
ilt,1958); it has
more recently been related to withdrawal of oil (Whitten, 1961, p. 319).
61
R.F. Yerkes and R.O.Castle
1
.:
4
.
3
u
:
..
..
.. ..
.. ..
. . ..
'
:
..
..
.. ...
T
e
<u
a
8
'
$
,..
. ..
...
.
..
:
i
Y
*T
:
:
:
;
:
._......
O
1
I
0
."..o...
-. o
:e
. 2 i
L
: o
.
. c
u
O
3
m
U
L
pi
.-
I
L
FIGURE4. M a p of Buena Vista oilfield,Kern County, California,showing (I) approximate boundaries of the Buena Vista Front and Buena Vista Hills areas of the Buena Vistafield and two asfiacent
oilfields; (2) trace of active thrustfault along the south Bank of the Buena Vista H
ills area (Wilt
1958,pp. 170, 172); (3) adjusted elevation changes for the period 1957-1964at selected bench
marks (US Coast and Geodetic Survey, 1966,pp.5-6,18) and (4)horizontal displacementfor the
Whitten, 1961,pp. 318-319)
period 1932-1959relative to an undefinednetwork (
62
Surface deformation associated with oil and gas field operations in the United Stares
MODELS
The published literature contains few models that relate subsidence and other surface
effects to extraction of subsurface materials. Probably,the most carefully documented
“controlled experiment” is the case history of surface effects measured during Fraschprocess extractionof sulfur described by Deere (1961). Deere’sresults verify the analytical
“tension center” model described by Stanford Research Institute(1949), and also resemble the results of analysis of cumultative surface deformation over the Wilmington oil
field.
FRASCH-PROCESS CASE HISTORY
Deere (1961)has described differential subsidence,simultaneoushorizontal displacements,
and surface faulting associated with Frasch-processextraction of sulfur from a depth of
397-488m within the cap rock of a Texas Gulf Coast salt dome;the cap is overlain by unconsolidated sands,gravels,clays,and clay-shales.During thefirst31 months ofoperation,
differential subsidence as great as 1.75 m developed over an elliptical area exceeding 5 km2
that centered directly over the narrow linearproducing zone.A normal fault about 650 m
long,downdropped as much as 0,l m on the mining side and peripheral to the subsidence
bowl,developed suddenly during the fifth month of production. By the 31st month,it had
FIGURE5. Surface deformation during Frasch-process extraction of sulfurfrom cap rock of
Texas
Gulf Coast salt dome. 5a, horizontal strain in percent deueloped between 9 and 31 months’production
(after faulting had begun) ;56, vertical differentialsubsidence during the same period, and approxi
mate total differential subsidence after 31 months production. Based on Deere (1961,jîgs. I, 4,
and 5)
63
R.F. Yerkes and R. O. Castle
increased to about 800 m in length and 0.3 m in displacement. The fault dipped about
40” inward directly toward the top of the producing zone, and it formed at a point of
maximum surface tension as determined by surveys (fig. 5).
QUALITATIVE
MODELS
The “tension center” model was developed during a private study of the Wilmington oil
field subsidence (Stanford Research Institute, 1949). The model assumes (1) a homogeneous, isotropic earth; (2)that all stresses remain within the elastic limit;(3) a spherical
compacting region or “tension center”;and (4)a negligible weight for the removed material. This model predicts a central zone of compression surrounded by an annular zone
of tension that extends out to the periphery of the bowl’, a strain pattern similar to that
developed in the Frasch-mining case.
The quantitative results of Deere’s investigation also support the qualitative models of
Grant (1954), Rellensmann (1957), and Lee and Shen (1969). Grant (1954,p. 21) compared
the subsiding prism of deposits at Wilmington to a bending beam, in which the greatest
horizontal displacement of points on the surface occurs over the point of inflection (or
point of greatest slope of the subsidence profile). Rellensmann (1957, fig. 2)presented
dimensionless profiles of subsidence, horizontal displacement (“shift ”), and horizontal
strain as developed at the surface over relatively shallow mining excavations. Very similar
I
FIGURE6. Qualitative comparison of horizontal strains developed during subsidence over (I) the
Wilmington oil field (relatively deep extraction of fluids over a broad area), (2) Frasch-process
extraction of sulfur (from a narrow zone at intermediate depths), and (3) shallow mining of solid
ores (Rellensrnan,1957,fig.2). Relatiue limits of producing areas are also indicated
i. The Stanford Research Institute report (1949, pp. 76-68) presents an expression relating
horizontal displacement(u) at a radius (r) from the center of subsidence to the differential
subsidence(w)at that radius and the depth (h) beneath the original surface to the center of
compaction: u = rw/h. By this expression, horizontal displacement varies from zero at the
center of subsidence(where r = O), through a maximum, and back to zero at the periphery
(where w = O).
64
Surface deformation)zssociated with oil and gas field operations in the United States
dimensionless curves are presented by Lee and Shen (1969,fig. i), who investigated horizontal displacement during subsidence due chiefly to surface loading;they supported the
curves with both model studies and theoretical finite element analyses. A profile of the
horizontal strain developed during subsidence over the Wilmington oil fieId (extraction
offluids over a broad area from a median depth of about 1 O00 m) is compared in figure 6
with that developed during Frasch mining of sulfur (extraction from a narrow zone at
a median depth of about 440m),and with that developed over relatively shallow mining
or tunnelling.
The following conclusionsmay be drawn from this brief review:(1) Differential subsidence commonly accompanies extraction of fluid or solid materiais from poorly consolidated sequences; (2)subsidenceis commonly accompanied by centripetally directed horizontal displacement,which varies from zero at the center of subsidence,through a maxim u m over the point of steepest slope of the subsidence profile, and back to zero at the
periphery (as described by the tension center model and by Grant and Rellensmann);
(3) the accompanying horizontal strains are most intense at or near the center of subsidence,where greatest compression is attained and thrust or reverse faulting may occur.
In all cases analyzed, the greatest tension, and tensional faulting,develop in the outer
parts of the subsidence bowl,peripheral toithe annulus at which the steepest slope of the
subsidenceprofile develops.
ACKNOWLEDGEMENTS
Partial support for this study was provided by the Division of Reactor Development and
Technology,U.S. Atomic Energy Commission.
REFER ENCES
ALLEN,D.R. (1968): Physical changes of reservoir properties caused by subsidence and repressuring operations, four. Petroleum Technology,vil. 20,no. 1, pp. 23-29.
California Department of Water ResoÙrces (1964): Investigationof failure Baldwin Hills reservoir,
California Dept. Wafer Resources, 64 p.
DEERED.U.(1961): Subsidence due to mining-a case history from the Gulf Coast region of
Texas, in Mining engineering series -Proceedings of the 4th symposium on rock mechanics,
1961,Pennsylvania State Uniu.Mineral Industries Expt. Sta. Bull.,76 p, 59-64.
GRANT, U.S.(1954): Subsidence of the Wilmington oil field, California, (Pt.) 3 in Chap. 10 of
Jahns, R.H.,ed., Geology of southern California, California Diu.Mines Bull. 170, pp. 19-24.
HILL,M.L.(1954):Tectonics of faulting in southern California,(Pt.) 1 in Chap. 4 of Jahns,R.H.
ed.,Geology of southern California: California Diu.Mines Bull.,170,pp. 5-13.
KOCH, T.W.(1933): Analysis and effects of current movement on an active thrust fault in Buena
Vista Hills oil field, Kern County, California: A m Assoc. Petroleum Geologists Bull.,vol. 17,
no. 6, pp. 694-712.
LEE, K.L. and SHEN, C.K. (1969): Horizontal movements related to subsidence: Jour. Soil
Mechanics and Found. Diu.,vol. 95, S M 1, pp. 139-166.
NASON,R.D., COOPER,
A.K.and TOCHER, Don (1968): Slippage on the Buena Vista thrust fault,
in Geology and oilfields,west side southern San Joaquin Valley, 1968 Guidebook,Am. Assoc.
Petroleum Geologists-Soc. Exlor. Geophysicists-Soc. Econ. Paleontologists and Mineralogists, Pacific Sctions,pp. 100-101.
POLAND, J.F. and DAVIS,G.H. (1969): Land subsidence due to withdrawal of fluids, in Varnes,
D.J., and KIERSCH,
George, eds., Reviews in Engineering Geology, Volume II: Boulder,
Colorado, The Geological Society of America, p. 187-269.
PRATT, W.E.and JOHNSON,D.W . (1926): Local subsidence of the Goose Creek oil field (Texas):
four. Geology, vol. 34, no. 7, pp. 577-590.
RELLENSMANN,
Otto (1957): Rock mechanics in regard to static loading caused by mining excavation: Colorado School Mines Quart., voi. 52, no. 3, pp. 35-49.
65
M.N.M a y u g a and D.R. Allen
Stanford Research Institute (1949): Report on subsidence in the Long Beach-San Pedro area
Rept. to the Long Beach Harbor Subsidence Comm. (Unpub.), July, 1949, 295 p.
U.S.Coast and Geodetic Survey (1966): Vertical control data, Quad. 351 192:Washington,D.C.,
U.S.Coast and Geod. Survey, 113 p.
WHITTEN,
C.A. (1961): Measurement of small movements in the earth’s crust:Acad. Sci. Fennicae
Annules, ser. A, vol. 3, Geologica-Geographica,Suomatainen Tiedeakatimia,No. 61,pp. 31 5320.
WILT,
J. W .(1958): Measuredmovementalong thesurfacetraceof an active thrustfault in theBuena
Vista Hills,Kern County, California: Seismol. Soc. America Bull., vol. 48,no. 2, pp. 169-176
DISCUSSION
Intervention of Dr. Sakuro
MURAYA
(Japan)
Questiob:
This morning, Dr. Miyabe mentioned acute subsidence due to the earthquake in Tokyo.
I think there were some earthquakes in Long Beach. D o you have any experience about
influence of earthquake on the movement of surface?
Answer of Mr. YERKES:
There has been earthquakes during oil production in Long Beach but the relations are
complex and I do not clearly understand.
Intervention of Dr. Jose G.MENDEZ
(Venezuela)
Question:
I think there have to be room for certain legal aspects.W h o is respensible for the damage?
Answer of Mr.YERKES:
I a m not in the position to answer that.
SUBSIDENCE IN THE WILMINGTON OIL FIELD,
LONG BEACH, CALIFORNIA, U.S.A.
M.N.MAYUGAI and D.R. ALLEN2
ABSTRACT
The subsidence area is in the shape of an elliptical bowl superimposed on top of
California’s largest oil giant, the Wilmington Oil Field. The center of the bowl has subsided
over 9 meters (29 feet) since 1926.Horizontal and vertical movements have caused extensive
damage to wharves, pipelines, buildings, streets, bridges and oil wells necessitating costly
repairs and remedial work, including the raising of land surface areas to prevent inundation by the sea. Remedial costs have already exceeded US$lOO million. Most investigators
(i) Assistant Director and
(2) Subsidence Control Engineer, Department of Oil Properties, City of Long Beach,
California, U.S.A.
66