Geofluids (2002) 2 , 147–161
A . B O E H M
1
A N D J . C . M O O R E
2
1 IIS – ETH Zu¨rich, Zu¨rich, Switzerland; 2 Earth Sciences UC Santa Cruz, Santa Cruz, CA, USA
The late Miocene sandstone intrusions of northern Santa Cruz County, California, are the largest subaerial exposures of clastic intrusions on earth. The intrusions are sourced from a sandstone, underlying mudstone, accumulated in an outer shelf to upper slope environment. Dikes are the most frequent intrusion type, reach the greatest thickness and tend to strike north-east and dip steeply. One giant dike is more than 150 m wide. Sills are least frequent, locally > 8 m thick and have no clear preferred geographical distribution. Clustered intrusions are commonly < 10 cm thick and mostly composed of dikes of various attitudes. The majority of the intrusions probably were injected shallowly as some extrude onto the seafloor. The local seafloor extrusion also indicates injection during the deposition of the Santa Cruz
Mudstone (7–9 Ma). The intrusions are concentrated at the basin margin. Fluid pressure at the centre of the basin and perhaps hydrocarbons were communicated to the basin margin through the then sand, causing fluid overpressures that contributed to the fluidization and intrusion into the overlying mudstone.
Primarily north-east-striking, steeply dipping dikes and secondarily, shallowly dipping sills are most significant in terms of regional connectivity of intrusions and physical dilation of the formation. The orientation of the dikes and sills indicates a regional stress field with a horizontal NE–SW maximum and a NW–SE minimum compressive direction.
The simultaneous development of dikes and sills suggests similar magnitudes of the minimum and intermediate principal stresses. Preferential weakness along bedding contributed to the development of sills. Palaeomagnetic data indicate no significant block rotation around a vertical axis. The maximum principal stress direction indicated by the intrusions is about 55 8 to the San Gregorio Fault and about 70 8 to the San Andreas Fault during the late Miocene.
This stress field is similar to the modern stress field and suggests moderate fault weakness.
Key-words: contour plots, high-permeability conduit for the transmission of fluid pressure, paleostress field from preferred intrusions orientation, regional strain, sandstone fluidization and intrusion
Received 23 June 2001; accepted 20 December 2001
Corresponding author: Dr Anja Boehm, IIS – ETH Zu¨rich, Gloriastr. 35, 8092 Zu¨rich, Switzerland.
E-mail: anjaboehm@gmx.net. Tel: þ 41-1-6324664. Fax: þ 41-1-6321194.
Geofluids (2002) 2 , 147–161
Recently, intrusive sandstone bodies have been recognized as integral parts of commercial hydrocarbon reservoirs in the
North Sea (Dixon et al . 1995; Lawrence et al . 1999) and offshore West Africa. Sandstone intrusions are significant as they increase connectivity between existing stratigraphically controlled reservoirs. Moreover, clastic sediment intruding up from a reservoir provides an ever-shallower level in which hydrocarbons can be sequestered. Finally, intrusions form cross-stratal migration paths that could even connect to the surface. Molyneux (1999) has identified the clastic intrusions of the Santa Cruz Mountains (Eldridge 1901;
Newsom 1903; Phillips 1990; Thompson et al . 1999) as the largest subaerial intrusions in the world and the best glo-
# 2002 Blackwell Science Ltd bal analogues for the North Sea examples. We have studied the geometry, extent and thickness distribution of the Santa
Cruz intrusions to provide a quantitative analysis of their form for use in hydrocarbon exploration and to shed light on the intrusive conditions and stress field during their emplacement.
The orientations of the intrusions provide an opportunity to define the palaeostress field during their intrusion. Injecting dikes open fractures that propagate as a tensile crack in a plane normal to the least compressive stress direction (Delaney et al . 1986). Previous studies of the intrusions have emphasized the orientations of the dikes, which are the dominant intrusion type (Phillips 1990; Thompson et al . 1999).
Yet, intrusions in other orientations, including sills, occur, raising the question: what are the orientation patterns of all

148 A. BOEHM & J. C. MOORE intrusions and how are they related to the geology? In addition to the control of the stress field, the geometry of intrusions can be influenced by the anisotropy of the host rock or pre-existing fractures (Jolly & Sanderson 1995; Pollard
1973).
In order to estimate the overall geometry of the sandstone intrusions in northern Santa Cruz County, we have analysed the exposures along Highway 1 outcrops and all accessible beach portions between the beach north of Needle Rock
Point, north of Santa Cruz, and the city of Davenport
(Fig. 1). We have completed a nearly continuously exposed one-dimensional transect of this intrusive system in which we have analysed 328 intrusions.
The coastline west of Santa Cruz, where the sandstone intrusions occur, is part of the Salinian Block that here consists of
Palaeocene to Pliocene marine and nonmarine deposits which rest disconformably on pre-Tertiary granitic rocks and metamorphic basement (Figs 2 and 3) (Greene 1990;
Phillips 1990). In the study area, the basement is commonly
Fig. 1.
Regional geology north of Santa Cruz
County. Note location of cross-section (Fig. 3) on both location and geological map.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 149
Fig. 2.
Regional stratigraphy of western Santa
Cruz County.
Fig. 3.
Cross-section paralleling coastline from Needle Rock Point to 3 km north-west of Scott Creek (see Fig. 1 for location). Note significant thinning of Santa
Cruz Mudstone from north-west to south-east and concentrations of intrusions at south-west end of section. Wells P1 and S1 confidently determine the depth to
Santa Margarita Sandstone at the ends of the section. Depth of Santa Margarita Sandstone in wells D1 and D3 projected down regional 10 8 dip, that is about normal to section. Geology along section from Clark (1981).
overlain by the transgressive, middle Miocene Lompico
Sandstone and the overlying Monterey Formation. The upper Miocene Santa Margarita Sandstone is a transgressive, shallow marine unit (Phillips 1990; Stanley 1990), which
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161 consists mainly of sandstones and conglomerates, and is typically unconformable overlying the Monterey Formation. The
Santa Margarita Sandstone, especially its coarser grained facies, may have served as a migration path or reservoir for

150 A. BOEHM & J. C. MOORE hydrocarbons sourced in the underlying Monterey Formation (Phillips 1990). Geochemical studies by Lillis & Stanley
(1999) indicate that the Santa Cruz Mudstone may also be a possible petroleum source. The Santa Margarita Sandstone hosts the largest accumulation of petroleum in the Santa
Cruz Mountains (Phillips 1990), and this lithology is believed to be the source of the sandstone intrusions studied here.
The sandstone intrusions occur in the Santa Cruz Mudstone that overlies the Santa Margarita Sandstone (Fig. 3).
The upper Miocene Santa Cruz Mudstone is an organic-rich, shelfal depositional unit of mainly thickly bedded, siliceous mudstone and thin porcelanite layers. Dolomite and calcite concretions are locally abundant (Stanley 1990). Diatom biostratigraphy suggests that the Santa Cruz Mudstone accumulated approximately 9.0–7.0 Ma ago (Barron 1986). Silica diagenesis studies by El-Sabbagh & Garrison (1990) suggest a minimum burial depth of 1000 m for exposed sections of the Santa Cruz Mudstone. Rapid thickening of the mudstone north-west of Santa Cruz implies that extensive erosion preceded the deposition of the younger Purisima Formation (El-
Sabbagh & Garrison 1990). Clark et al . (1984) report about
200–500 m of vertical uplift in the western Santa Cruz
Mountains area since deposition of the Santa Cruz Mudstone.
The overlying Purisima Formation is separated from the
Santa Cruz Mudstone by a hiatus that occurred between
6.0 and 7.0 Ma ago (Barron 1986). This formation is composed chiefly of sandstone, mudstone, conglomerate and tuff that were deposited in marine environments (Stanley 1990).
The area of sandstone intrusions is bounded by the San
Gregorio Fault to the west and the Ben Lomond Fault to the east (e.g. Stanley 1990) (Fig. 1). The San Gregorio Fault is a major branch of the right-lateral San Andreas Fault
(Greene 1990; Weber 1990). Right-lateral strike–slip offset is about 150 km since 12 Ma (Clark et al . 1984; Graham &
Dickinson 1978). These faults are components of the strike–slip regime that developed along the Pacific-North
America transform margin at about 30 Ma. According to
Atwater & Stock (1998), the Pacific-North America plate motion shifted from WNW to NNW in northern California about 8 Ma ago. Thus the Pacific-North America plate motion changed from slightly extensional to essentially parallel to the trend of the continental margin in northern California at this time.
The Santa Cruz Mudstone and underling deposits thicken to the north-west offshore of the Santa Cruz Mountains.
TheNeogene section dips gently into this basin (Erickson et al . 1994). The centre of the study area of sandstone intrusions is characterized by a gently south-west-plunging open anticline in the vicinity of Majors Creek. This structure is interpreted as a compaction anticline that served as a structural trap for the upward migrating petroleum (Phillips
1990).
The clastic intrusions occur as tabular dikes and sills (Figs 3, 4 and 6). Locally, surface extrusions of sand occur (Fig. 4). The sand is very well sorted and commonly of medium grain size.
Its colour varies from dark brown to grey; locally dark grey or black colours are due to fine-distributed bitumen. Hydrocarbon staining is homogeneous where it occurs. Only the two big intrusive bodies at Yellow Bank Creek show a well-developed primary flow structure.
At a few outcrops, dikes branching into sills exist. Crosscutting intrusions, however, are rare. Where recognized, dikes cut sills. Mixing of intrusive material with the host sediment could be observed only on a small scale and very locally.
More commonly, angular inclusions or rafts of the host mudstone occur within the intrusions. Inclusions are only present in thicker intrusions and are more frequent in sills than in dikes. Generally, intrusive contacts show sharp angular changes in orientation over short distances with sharp nearly perpendicular offsets of contacts on a 1–10 cm scale; smooth bends of intrusion contacts are rare.
On the basis of field observations, we can classify the intrusions into clustered intrusions, which have various attitudes, single dikes and sills. Statistical analyses lead to a narrower distinction later in this article where dikes are referred to as intrusions with an azimuth between 97 8 147 8 or 277 8 327 8 and a dip of 60 8 (Fig. 5); sills are referred to as intrusions dipping 25 8 and oblique dikes are all other directions.
Dikes
Clustered intrusions
Although small intrusions show a great variety of attitudes, a preferred orientation is commonly obvious. The majority of dikes dipping less than 65 8 (Fig. 5) belong to this intrusion type. The margins of the clustered intrusions are sharp and occur in subhorizontal as well as subvertical orientations.
The margins are more often irregular than those of single dikes. They show multiple branching into thinner intrusion parts, which takes the form of fingering because no cross-cutting relationships could be recognized (Fig. 6). Locally intrusive branches link, bifurcate and show small dilational jogs
(terminology following Huang 1988; Lonergan et al .
2000). Individual intrusive branches of a cluster are thin, rarely exceeding several centimetres. However, the clustered intrusions rapidly change in thickness, commonly when their margins change in direction. The clustered network-forming intrusions tend to extend only a few metres across outcrops, in contrast to rarer thicker intrusions that may transect the height of the sea cliff exposure. The complicated structure of the clustered intrusions indicates good local three-dimensional interconnectivity, but over a limited extent.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 151
Fig. 4.
(a) Intrusive body south of Yellow Bank
Creek beach. Lack of intrusive or fault contact and sedimentary homogeneity between the sill and dike portions suggest syngenesis of sill and dike parts. (b) Extrusive sand bed. The clearly bioturbated, irregular base, transitional top and the downward connection to a dike indicate that this sand bed at Majors Creek beach is a sand extrusion fed by this dike. The burrowing in the sand makes it unlikely that the sands were oil saturated when they erupted.
Single dikes
Single dikes are mostly vertical with straight or just slightly irregular margins. Parallel steps of the intrusion margins at bedding borders may be present but are not pronounced.
Single dikes are generally more than 10 cm thick with a consistent thickness. They extend up cliffs to 25 m in height. We rarely view along-strike exposures due to the nature of the cliff outcrops. However, we infer from the vertical exposures, which show only very minor changes in intrusion thickness, that changes in thickness along strike are very probably minor as well. Branching and rejoining of single dikes only occurs rarely. At a few sites, a steeply dipping dike is associated with a sill (Fig. 4A). Cross-sections between Yellow Bank Creek and Laguna Creek (Thompson 1995) show an actual thickness of the Santa Cruz Mudstone at the coast between 40
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161 and 75 m. Thus, the clastic intrusions may extend at least this distance from their source at the Santa Margarita Sandstone to the surface of current exposure. Single dikes are preferably injected into the limbs of the broad anticline centred at Major
Creek (Fig. 7).
Sills
Sills associated with the above-described clustered intrusion complexes are mostly thin and are locally branches of dikes.
The margins and their geometry are comparable to those of the dikes of the clustered complexes. Singular sills may reach a considerable thickness of up to 90 cm (with the exception of the giant sills in the Yellow Bank Creek area). The margins of sills are often wavy and irregular, mostly due to changes in

152 A. BOEHM & J. C. MOORE
Fig. 5.
Statistical distribution of intrusions (excluding both big intrusive complexes in the Yellow Bank Creek area). Although dikes of various orientation (oblique dikes) are statistically most frequent, thickest intrusions are either sills or north-east-striking dikes, and by far the greatest cumulative thickness is reached by northeast-trending dikes.
Fig. 6.
Sketch of a larger intrusion network of intrusions with a great variety of attitudes at the beach near Yellow Bank Creek. This structure is referred to as clustered intrusions by the authors.
Branching and rejoining of intrusions are common, whereas no cross-cutting could be recognized. This structure implies a good local three-dimensional interconnectivity. Bedding and fractures in host rock inclusions within intrusions show the same orientation as the surrounding mudstone, suggesting that the jointing is later than the intrusion event.
thickness. Branching and rejoining are common in thinner sills.
Intrusive bodies at Yellow Bank Creek
Intrusions in the Yellow Bank Creek beach area (Fig. 3) and just to the south-east (Fig. 4A) are discussed separately here because they are significantly larger than any other intrusions in the study area, and in fact the largest subaerially exposed intrusions known anywhere (Molyneux 1999). The partly hydrocarbon-saturated giant intrusive sandstone body at Yellow Bank Creek (Thompson et al . 1999) consists of a funnelshaped body and a north-east-striking dike. The dike part is more than 150 m thick and can be mapped inland away from the beach for at least 600 m; the beach outcrop shows a thickness of 13 m for the funnel-shaped body, which also extends away from the beach for several hundred metres. According to Thompson et al . (1999), intrusion apparently occurred in two stages: after intrusion of initially water-saturated sands, hydrocarbons migrated in the sand body by fluid replacement mechanisms and perhaps locally by intrusion of hydrocarbonbearing sand. Another dike and sill complex of considerable size is located some 300 m south-east of the previously described intrusion. It comprises a 10 m thick dike connected with a sill body whose exposed thickness is 10 m (Fig. 4A).
The dike strikes N35 8 E and dips 78 8 W. Where it is least
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 153
Fig. 7.
The statistical distribution by locality (excluding both big intrusive complexes in the Yellow Bank Creek area) shows a systematic pattern with respect to the regional anticline (at site 4: Majors Creek). Oblique dikes are most common near the anticline where the top of the Santa Cruz Mudstone is shallowest. However, north-east-striking dikes, which are most significant in terms of total and individual thickness, occur mostly among the fold limbs.
weathered, at the base of the cliff, the intrusion is hydrocarbon saturated.
Surface extrusions of sand
Previously, workers interpreted all bedding-parallel intrusions as sills. New evidence suggests that some may be extrusions. The beach outcrops at Majors Creek (site 4, Fig. 1) are characterized by bedding-parallel bituminous sand layers up to 3 m thick, that include lamination and cross-bedding.
They are internally bioturbated, and have had the sand bioturbated down from and up into the mudstone (Fig. 4B).
Thus, we believe that they are surface deposits. The burrowing in the sand beds makes it unlikely that the sands were oil saturated when they erupted. The extrusive sand layer extends several hundred metres north-west and south-east of the mouth of Majors Creek. The extrusive sand layer is generally slightly coarser grained than the intrusive sand; subjacent dikes are both truncated by and feed into the bottom of the extrusive sand beds (Fig. 4). Apparently, the intrusions fed sand extrusions that were worked or reworked to develop the sedimentary and biological structures. At Majors Creek, the extrusive sand bed is stratigraphically higher than any of the intrusions, suggesting that it represents the final phase of intrusive/extrusive activity here.
In addition to the extrusive sand layers described above, we have observed several extrusive bituminous sands in quarries
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161 in the Santa Cruz Mudstone about 5 km north-east of the mouth of Majors Creek. These layers are about 2 m thick, parallel to enclosing strata and show sand-filled burrows at the basal contact. Intrusive sills also occur in these quarries
(Page & Holmes 1945). Sand layers are otherwise unknown in the Santa Cruz Mudstone (Clark 1981).
Phillips (1990) and Thompson (1995) demonstrated that the sandstone dikes studied here strike predominantly northeast; previous workers have not measured the orientation of either dikes or sills and have not provided any information on the thickness of the intrusions. Quantification of the orientation and thicknesses of the intrusions provides clues to the stress field at the time of intrusion as well as the magnitude of dilation. This information also provides a basis for judging whether the intrusions could be independent reservoirs or provide unanticipated connectivity between reservoirs.
Overall distribution and geometry
We compiled histograms of the thicknesses of 328 intrusions, specifically excluding the large intrusions in the Yellow Bank
Creek area. Intrusions were grouped into three categories

154 A. BOEHM & J. C. MOORE depending on where their poles plotted on a stereographic projection: north-east-trending dikes, sills and oblique dikes
(all other directions). Statistical analyses indicate that very thin intrusions are most common (Fig. 5A). Intrusions of various orientations (oblique dikes) are dominant in terms of their frequency in most thickness ranges. North-east-trending dikes are more frequent than sills. Intrusions greater than
40 cm, however, are nearly exclusively dikes or sills
(Fig. 5A,B). It should be noted that, when comparing the frequency of these three intrusion types, the area limits in the projection defining sills and north-east-striking dikes are less than that of the oblique dikes. The cumulative thickness of each thickness range (Fig. 5B) shows that, despite their smaller number, north-east-striking dikes and sills are more significant than oblique dikes in terms of total thickness and total dilation by orientation.
Local distribution and geometry
Examination of the intrusion frequency and thickness data by different locations shows a systematic pattern in respect of the regional anticline at Majors Creek (Figs 3 and 7). Oblique dikes are very common around the fold axis, but their cumulative thickness is generally less than the north-east-striking dikes and/or sills. This pattern may indicate a preferred occurrence of clustered intrusions and oblique dikes close to the lithological boundary of the Santa Margarita Sandstone to the Santa Cruz Mudstone, as the top of the Santa
Margarita Sandstone (most likely source) is shallowest near
Majors Creek.
Sills do not seem to have a preferred distribution in respect to both frequency and cumulative thickness. Most of the thickest sill intrusions occur around Majors Creek. Because the thicker sills are not numerous, this distinction may not be significant.
The orientation of all clastic intrusions is shown in
Fig. 8(A). The attitude of the intrusions varies greatly: dikes and sills of all possible strike and dip directions are present. A weak girdle distribution is recognizable which shows a slight predominance of intrusions with an orientation varying around a NE–SW-orientated axis. Two density maxima indicate a preferred occurrence of mainly sills and NE–SW-orientated dikes. Figure 8(B) is a contour plot that shows all intrusions weighted for their thickness. This plot covers all studied intrusions, except the very large intrusive bodies at the beaches in the Yellow Bank Creek area. In contrast to
Fig. 8(A), which shows a widespread point distribution, the contour plot in Fig. 8(B) has a very pronounced bipolar distribution. North-east-trending, steep dipping dikes show the greatest cumulative thickness, which is 12.4 m. Roughly horizontal intrusions form the other maximum with a cumulative thickness of approximately 9 m. The cumulative thickness of north-east-trending dikes and horizontal intrusions is calculated by summing the true maximum thickness of the single intrusions. The maxima of the contour lines in Fig. 8(B) represent all intrusions; only intrusions following the definition of Fig. 5 for north-east-trending dikes and sills are used to calculate the overall thickness (north-east-trending dikes have a dip direction of 97 8 147 8 or 277 8 327 8 and a dip
60 8 ; sills have a dip 25 8 ). The pronounced maxima of
Fig. 8.
Spherical distribution of intrusions. (a) Orientation of all intrusions. Plot of points to pole of all intrusions shows a widespread distribution with an only weak girdle distribution, indicating a preferred occurrence of mainly sills and NE–SW-orientated dikes. Point to pole; N ¼ 328; contour interval: 1, 3, 5%; rhomb: eigenvector (e1, eigenvector with maximum eigenvalue; e3, eigenvector with minimum eigenvalue). (b) Weighting the intrusions for their thickness results in a plot with a very pronounced bipolar distribution emphasizing the maxima of the previous plot. Contour lines represent the distribution density of points to poles. This plot shows more clearly the previous observation from Fig. 5 that sills and north-east-trending dikes reach considerable cumulative thickness, whereas all other directions are subordinate and do not show any preferred orientation. These two orientations agree with the trends of the very large intrusions around Yellow Bank
Creek which are excluded in this plot.
N ¼ 297; no smoothing; contour interval: 2, 4, 8, 12, 16%.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 155
Fig. 9.
Contour plots of thickness-weighted orientation by locality. These contour plots show that either horizontal NW–SE or vertical dilation dominate also on a local scale and that s 2 and s 3 have regionally similar values. Although the e3 axis of site 1 (beach north of Needle Rock Point) is vertical, nearly all intrusions of this outcrop are north-east-striking, steeply dipping dikes; the minimum eigenvector/eigenvalue is extremely low and close to the intermediate eigenvector/eigenvalue.
For this reason, this site is compatible with the trend shown by the other plots. The plot of site 4 (beach at Majors Creek) is characterized by a NNW-orientated minimum eigenvector/eigenvalue. This beach shows the most scattered orientation of dikes. However, it also shows a clear dominance of sills when weighted for the intrusion thickness, indicating mainly vertical extension. Rhomb: eigenvectors e1, e2 and e3.
dikes and sills in Fig. 8(B) provided the basis for subdividing the intrusions into dikes, sills and oblique dikes as shown in the histograms (Fig. 5).
The subdivision of contour plots per locality provides a sense of the regional variability of intrusion orientation
(Fig. 9). All plots show either a vertical or a NW–SE-orientated maximum eigenvector/eigenvalue, and a majority show a continuous NE–SW-orientated, horizontal minimum eigenvector/eigenvalue. Six out of eight sites show an
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161 approximately horizontal, NW–SE-orientated maximum eigenvector/eigenvalue. The beaches at Majors Creek and
Yellow Bank Creek and the roadcut at Highway 1 are characterized by a pronounced vertical dilation.
The inclination of the maximum eigenvector/eigenvalue is related to the regional open anticline: sites 5–8, located north-west of the open anticline, generally have a maximum eigenvector/eigenvalue located in the north-west quadrant of the plot; sites 1–3, south-east of this anticline, show the

156 A. BOEHM & J. C. MOORE maximum eigenvector/eigenvalue located in the south-east quadrant. This implies that the orientation of the dikes is controlled by this larger scale fold structure. It may reflect the impact of stress distribution on the convergent, fanning-like, preferred opening orientation of fractures within a fold structure (Thompson et al . 1999) or postintrusion anticlinal flexure. The local predominance of either vertical or horizontal dilation (as indicated by the maximum and intermediate eigenvectors/eigenvalues) indicates regionally similar values of s 2 and s 3.
Summary
Clearly, the NE–SW-striking dikes and shallowly dipping sills are the dominant intrusive types in the Santa Cruz Mudstone.
The oblique dikes, which tend to form the thinner clustered intrusions, are less important in terms of cumulative thickness. It is noteworthy that these major trends defined by the small intrusions agree with the trends of the very large intrusions around Yellow Bank Creek. The slightly to the north-east-tilted vertical extension defined by the sills reflects the mean dip of the Santa Cruz Mudstone (10 8 to the southwest) in the study area. This confirms that the bedding planes are the main opening direction for the sills.
The intrusions studied are only locally affected by deformation. Brittle deformation features evident in the intrusions include faults and deformation bands (Aydin 1978) locally mineralized with calcite. Deformation bands enriched in carbonate cement generally do not exceed a thickness of 2 mm.
Faults that cut intrusions and continue into the host mudstone are invariably normal faults and typically strike to the north-west.
In addition to the faulting described above, some intrusions are folded or offset by faults that do not continue into the adjacent mudstone. In both examples, the faults that cut the intrusions cannot be traced into the adjacent mudstone.
Figure 10 shows an example of a faulted intrusion.
Most parts of the Santa Cruz Mudstone have vertical extensional fractures. The vertical fractures form a perpendicular joint set striking north-east and north-west (Fig. 11).
Hence, one direction appears to be subparallel to the main striking trend of the dikes. Field observations show that, despite having a similar orientation, the dikes mostly do not intrude the adjacent joints. In addition, fractures in host rock inclusions within intrusions show the same orientation as the surrounding mudstone, suggesting that the jointing occurred later (Fig. 6). Locally joints cut the intrusions.
Overall, it appears that the jointing occurred later than the intrusions and the lack of similar jointing intensity between the host mudstone and intrusions is due to a difference in material properties.
Fig. 10.
Normal fault cutting the intrusion but not extending into adjacent mudstone indicates early intrusion into the mudstone. The difference in material properties between host mudstone and sand intrusion caused normal faulting of the intrusion during consolidation of the mudstone.
Fig. 11.
Rose diagram of predominant joint set (shaded) and north-eaststriking steep dipping dikes (dotted). One direction of the joint set seems to be subparallel to the main trend of the dikes. However, field observations show that the joints typically terminate into the intrusions, which indicates that at least the majority of the joints formed after sandstone intrusion. For details, see text. Joints, N ¼ 197; dikes, N ¼ 108; radius of circle, 25, where 25 is the number of dikes/joints.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 157
Several lines of evidence indicate that the sandstone intrusions in the Santa Cruz Mudstone occurred early in its history. The most compelling evidence is that some intrusions extruded onto the seafloor during the time of Santa Cruz
Mudstone deposition (Fig. 4B). Secondly, the faulting of intrusions by structures that do not continue into the adjacent mudstone (Fig. 10) suggests that these periods of shortening occurred during consolidation of the Santa Cruz
Mudstone. Obviously, intrusion would have to have occurred earlier. Joints cutting the mudstone around the folded and faulted intrusions (Fig. 10) are not deformed, indicating that the jointing occurred later than the sandstone intrusion.
Finally, we observed intrusions that are enclosed by the later growth of concretions (Fig. 12). These concretions are considered to have formed early in the history of the Santa Cruz
Mudstone (R. Garrison, personal communication, 2000).
The margins of intrusions are sharp and generally planar, suggesting intrusion into cohesive sediment. However, because mud can become cohesive at a very shallow depth, this need not obviate a shallow depth of intrusion of the sandstones.
Because dikes and sills are continuous, these features formed simultaneously.
Although rare cross-cutting relationships, banding and internal layering in intrusions indicate different pulses of intrusion, we could not distinguish the unique lithological characteristics of different intrusion generations. The uniform sedimentary composition and texture of the intrusions imply that they are from the same source. Similarly, the huge intrusive bodies around Yellow Bank Creek could not be lithologically or stylistically distinguished from the other smaller intrusions.
The clustered intrusions consist of networks of variably orientated thin intrusions. We believe that these intrusions are similar to small cracks forming in a rock under extension and prior to propagation of a through-going crack (van der
Pluijm & Marshak 1997). We interpret the clustered intrusions as representing the initial state of intrusion propagation
(Fig. 13) (Jolly et al . 1998).
Fig. 12.
Sketch showing sill enclosed by concretion at roadcut along Highway
1 at Majors Creek. As these concretions are considered to have formed early in the history of the Santa Cruz Mudstone and at very shallow sediment depth, the intrusion predates the inclusion.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161
Fig. 13.
Clustered intrusions, characterized by often branching and rejoining, are common around the anticline centred at Majors Creek. North-easttrending dikes tend to be most common on the flanks of the anticline. They may have a minimum length of > 40 m. Sills are the least common intrusion type. Sand extrusions were identified only at the beach at Majors Creek. The greatest regional dilation during intrusion occurred along NW–SE and vertical.
The amount of NW–SE dilation is 192 m which accounts for 3.6% of the studied coast length. The maximum principal stress was orientated approximately horizontal along N29 8 E at the time of intrusion. Thus, late Miocene stress orientation is similar to the orientation of the present principal compressive stress.
In summary, the similar style, evidence for continuity between dikes and sills and lack of compositional differences suggest that all intrusions formed in one generation of intrusive activity, albeit with pulses of emplacement in individual large intrusions. Because some of the intrusions connect with seafloor extrusions, we know that intrusive activity occurred from 9 to 7 Ma during the deposition of the Santa Cruz
Mudstone.
The overall geometry of the sedimentary basin (Fig. 3) provides clues to the driving forces of the clastic intrusions.
The intrusions are clustered at the thinned margin of the basin as shown in the cross-section (Fig. 3). A compilation of seismic reflection data indicates that the Tertiary section thickens offshore perpendicular to the cross-section in
Fig. 3. The erosion of the Santa Cruz Mudstone is estimated to be only 200–500 m with no significant variation along the section in Fig. 3 (El-Sabbagh & Garrison 1990). Therefore the major thickness differences shown in Fig. 3 were present during intrusion, both along the line of section and offshore.
The thickness differences in the Santa Cruz Mudstone basin drive fluids to the basin margin and spawn the clastic intrusion process (Fig. 14). The Santa Margarita Sandstone, just below the Santa Cruz Mudstone and the source of the intrusions, provides a high-permeability conduit for the transmission of fluid pressure from the deeper portions of the basin to the margin. The difference in overburden at the time of intrusion was at least a factor of five between the deeper portions of the basin and the margin. Thus, even

158 A. BOEHM & J. C. MOORE
Fig. 14.
Schematic diagram showing how moderately overpressured fluid conditions in the basin centre can be communicated up a permeable sand layer resulting in supra-lithostatic conditions at the basin margin, and dike and sill intrusion. Fluid overpressures in the basin centre could be due to compaction, hydrocarbon generation or tectonics.
a moderately abnormally high pressure gradient in the deeper basin would significantly exceed the lithostatic pressure and induce hydrofracture when communicated to the edge via the Santa Margarita Sandstone (Fig. 14). The drive of the intruding sand was probably facilitated by the presence of low-density hydrocarbons, now residing in the intrusions, in addition to basinal overpressure due to compaction or tectonics.
The sandstone intrusions caused a measurable positive dilation of the Santa Cruz Mudstone. A comparison of the thickness of the intrusions to the overall length of the studied area measured parallel to the maximum dilation direction gives an approximate quantification of the maximum regional strain related to the intrusion process.
The amount of the actual regional dilation is the sum of the thicknesses of intrusions in a particular direction. We analysed the overall dilational strain of the intrusive event by determining the maximum, minimum and intermediate directions of cumulative thickness. These directions are represented by eigenvalues and eigenvectors for the thickness-weighted distribution of all intrusions, excepting the huge bodies of the
Yellow Bank Creek area (Fig. 8B).
The three main principal eigenvectors and eigenvalues calculated by the Bingham method (Fisher et al . 1987) are shown in Fig. 8(B). These eigenvectors and respective eigenvalues are an orthogonal set of axes best approximating the maximum, intermediate and minimum concentration of points. The eigenvectors and eigenvalues describe the symmetry of dilational strain associated with the intrusions. They define the strain ellipsoid that is a visual representation of the volume change of a body. The orientation of the maximum, minimum and intermediate eigenvalues/eigenvectors can be interpreted in terms of the orientations and relative magnitudes of the principal stress axes.
The largest eigenvector/eigenvalue in Fig. 8(B) is orientated NW–SE, nearly horizontal, indicating maximum dilation perpendicular to the north-east-orientated dikes. The intermediate eigenvector/eigenvalue corresponds to the orientation of the sills and suggests that dilation perpendicular to bedding was about 80% of that in the NW–SE direction. The minimum eigenvector/eigenvalue indicates the least dilational strain in a north-east direction. The intermediate and maximum eigenvectors/eigenvalues are slightly displaced in relation to the density contour in the area of the two highest maxima. This is because the eigenvectors are averages based on all data points from a skewed distribution and would not correspond exactly to the maxima of the contoured data. The contour diagrams in Figs 8 and 9 are a representation of point to pole densities in a stereographic projection.
Our estimates of dilational strain associated with the sandstone intrusions represent a one-dimensional coastal transect of the intrusive system and may be biased by location. The large intrusions of the Yellow Bank Creek area have both prominent dikes and sills, with the former thicker than the latter.
Compaction would preferentially reduce the overall thickness of sills after their emplacement.
The absolute NW–SE horizontal regional extension due to the north-east-striking dikes is small in comparison to the coastal length studied. We were able to access and study a total length of 4768 m along a line parallel to the maximum dilation axis. The maximum total dilation is 192 m. The single big intrusion at the beach at Yellow Bank Creek accounts for 170 m, or 3.2% of the length of the coast studied. The large dike just south of Yellow Bank Creek (Fig. 4A) accounts for another 10 m, or 0.22%. All other north-east-trending dikes (excluding intrusions at roadcuts along Highway 1) total 11.6 m, or another 0.25% of the pre-intrusive coast length. Our estimate of dilation is slightly exaggerated because it represents a simple sum of the thicknesses of any dike orientations in the area of north-east-striking dikes
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 159
(Fig. 5), and not all dikes are parallel to the maximum eigenvector/eigenvalue direction.
The regional stress field controls the geometry of clastic intrusions. For intrusion to occur, the fluid pressure in the sand has to exceed the minimum principal stress plus the tensile strength of the host rock (Price & Cosgrove 1990).
Intrusions into a homogeneous medium open as tensile cracks parallel to the s 1 direction and perpendicular to the s 3 direction (Anderson 1951). Injections into pre-existing fractures in the host sediment must have a pressure exceeding the normal stress ( s n) which acts across the fracture walls
(Delaney et al . 1986). Within the study area, only 6% (19 of 328) intrusions appear to have injected pre-existing fractures. Thus, the majority of the intrusions should reflect the orientation of the palaeostress field at the time of injection (about 8 Ma).
The north-east-striking dikes suggest a minimum principal stress shallowly plunging to the north-west, and the simultaneously intruded sills indicate a minimum principal stress with a subvertical orientation. This apparent inconsistency may be explained by the dikes intruding from the subjacent source rock perpendicular to the minimum principal stress and the sills forming simultaneously due to the weaker sediment cohesion parallel to bedding. Thus, the north-westorientated stress controlling dike formation ( s 3) and the overburden stress ( s 2) were relatively similar in magnitude, different only by the anisotropy in tensile strength or elastic moduli of the Santa Cruz Mudstone (Pollard 1973). This geometry of intrusion constrains the maximum principal stress to be along the minimum eigenvector/eigenvalue
(209/05) or NE–SW.
Because our study area is located between the San Andreas and San Gregorio strike–slip faults, the rocks and any palaeostress indicators may have been rotated around a vertical axis
(Horns & Verosub 1995). Late Neogene clockwise rotations of 35 8 60 8 occur within a kilometre of the San Gregorio
Fault zone (Horns & Verosub 1995). However, magnetostratigraphic studies of 6–7 Ma rocks directly overlying the
Santa Cruz Mudstone around the city of Santa Cruz indicate no significant rotation (Madrid et al . 1986). On the basis of these studies, we believe that there has been little or no vertical axis rotation since the deposition of the Santa Cruz
Mudstone for the coastal area of intrusion between Santa
Cruz and Davenport. Thus, we use the current orientation of the minimum eigenvector/eigenvalue as a direct palaeostress orientation indicator.
Our studies of the sandstone dikes indicate that the principal compressive direction during intrusion (7–9 Ma) was
NNE (N29 8 E), which is at a high angle to the mean strike of the San Andreas and San Gregorio Faults. This stress is sig-
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161 nificantly more oblique than the most favourable orientation for slip on the fault and at least suggests that the argument for a ‘weak’ San Andreas Fault (Zoback et al . 1987) could extend back into the Miocene.
Although Anderson’s theory of faulting for strike–slip systems predicts an angle of about 30 8 between the principal strike–slip faults and the s 1 direction (Anderson 1951), this angle for the San Andreas Fault System in the Pacific plate is much greater (Zoback et al . 1987). This contradiction can be explained by the weak fault model (e.g. Hardebeck 1999).
Elevated pore fluid pressure and inhomogeneous stress along the fault, in combination with the presence of weak clay minerals, are believed to reduce a fault’s shear strength.
We measured 197 vertical joints along the whole length of the dike intrusion area to evaluate the relationship between joints and sandstone intrusions. Sites with well-developed joints near intrusions were chosen to ensure that representative data were taken in host rock with very similar rheological properties to the sections in which intrusions injected.
Systematic joint sets in the Santa Cruz Mudstone form a steeply dipping, orthogonal set striking north-east and north-west. The north-west joint set is the most prominent with the north-east joints typically terminating into it. Thus, the north-west joint set probably formed first. Although the north-east-trending joint set and the dikes overlap partly, the maxima of these two planar structures nevertheless have different orientations. The majority of these joints strike between 30 8 and 70 8 , whereas the majority of the dikes strike between 20 8 and 40 8 .
The measured joints are mode 1 or extensional fractures, include ribs or arrest lines and are interpreted to have formed perpendicular to the minimum principal stress (van der
Pluijm & Marshak 1997). Thus, they have the same dynamic significance as the sandstone intrusions. The slight difference in the frequency maxima of Fig. 11 implies a different time of origin for the intrusions and the north-east joint system. It also emphasizes the field observation that dikes often have a very similar orientation as their adjacent joints, but do not intrude the joints. As mentioned earlier, field observations indicate that at least the majority of the joints formed after sandstone intrusion.
The sandstone intrusions in the late Miocene Santa Cruz
Mudstone of northern Santa Cruz County include dikes and sills as single bodies, and clustered intrusions. Locally, the intrusions connect to extrusive sand bodies. Intrusions,
1 cm to about 1 m thick, are widespread. Several much larger intrusions (10–150 m thick) occur in the north-west part of

160 A. BOEHM & J. C. MOORE the study area. The clustered intrusions are commonly a few centimetres thick and often characterized by branching and rejoining. They are interpreted as the termination regions of dikes or sills (Fig. 13). Thicker north-east-striking, steeply dipping dikes occur as singular bodies, not associated with clustered intrusions. North-east-trending dikes tend to be most common on the flanks of the anticline centred at Majors
Creek. Sills are the least common intrusive type and may be transitional to dikes. Extrusions occur at only one locality along the coast.
Smooth, mostly planar contacts and the lack of differential offsets along intrusive contacts suggest formation by simple dilation. Intrusion at shallow depths during the 7–9 Ma deposition of the Santa Cruz Mudstone is indicated by surface extrusions of sand fed from dikes. Joints postdate intrusions. Intrusion and extrusion of the sand through the poorly consolidated muds must have been driven by copious water flow or natural gas. Local oil saturation occurred later. The homogeneous texture and composition of the intrusions suggest a common source in the underlying Santa Margarita
Sandstone. Stratigraphic arguments suggest that singular dikes may have a minimum length of > 40 m. Fluid pressures at the centre of the basin and perhaps maturing hydrocarbons were communicated to the basin margins through the Santa
Margarita Sandstone. These fluid overpressures at the basin margins contributed to the fluidization and intrusion of this then sand into the overlying Santa Cruz Mudstone.
The magnitude of dilation due to intrusion is relatively small. The greatest regional dilation of 192 m during intrusion occurred along a NW–SE horizontal axis. North-westorientated horizontal dilation due to all dikes accounts for
3.6% of the studied coast length.
Palaeomagnetic studies show no significant block rotations, suggesting that the orientation of the intrusions has remained unchanged since their genesis. North-east-striking dikes and simultaneously intruded sills indicate that the maximum principal stress was orientated approximately horizontal along N29 8 E at the time of intrusion. This late Miocene stress orientation is similar to the orientation of the present principal compressive stress. The late Miocene principal stress orientation implies weakness along the San Andreas and San
Gregorio Faults at that time.
Our results have been obtained from the numerous small intrusions. Interestingly, the big intrusive bodies mirror the trends shown by the smaller intrusions in terms of geometry and sequence of fluid migration. Therefore, the big intrusions are exceptional only in size.
Acknowledgement is made to the Petroleum Research Fund of the American Chemical Society for support of this research through grant ACS/PRF 35329-AC2. Support was also provided by Exxon-Mobil Upstream Research Co. We thank
Lidia Lonergan, Robert Garrison, Rick Stanley and Peter
Vrolijk for valuable discussions and suggestions. Gerald
Weber provided the well data used in Fig. 3. The manuscript was improved by reviews of Davide Durante and Andrew
McCaig.
Aiello IW, Stakes DS, Kastner M, Garrison RE (1999) Carbonate vent structures in the upper Miocene Santa Cruz Mudstone at
Santa Cruz, California. In: Late Cenozoic fluid seeps and tectonics along the San Gregorio fault zone in the Monterey Bay Region,
California (eds. Garrison, RE, Aiello, I, Moore, JC), pp. 35.
American Association of Petroleum Geologists, Pacific Section,
Bakersfield, CA.
Anderson EM (1951) The Dynamics of Faulting and Dyke Formation with Application to Britain . Oliver & Boyd, London.
Atwater T, Stock J (1998) Pacific-North America plate tectonics of the Neogene southwestern Unites States — an update. In:
International Geological Review, Integrated Earth and Environmental Evolution of the Southwestern United States; the Clarence A.
Hall, Jr. Volume (ed. Ernst, WG), pp. 393–420. Bellwether
Publishing, Columbia, MD.
Aydin A (1978) Small faults as deformation bands in sandstone.
Pageoph., Birkhauser-Verlag, Basel , 116 , 913–30.
Barron JA (1986) Update biostratigraphy for the Monterey
Formation in California. In: Siliceous Microfossil and Microplankton of the Monterey Formation and Modern Analogs, Book 45 (eds
Casey RE, Barron JA), pp. 105–20. Pacific Section Society for
Sedimentary Geology (SEPM), Bakersfield, CA.
Clark JC (1981) Stratigraphy, paleontology, and geology of the central Santa Cruz Mountains, California Coast Ranges.
US
Geological Survey Professional Paper , 1168 , 61.
Clark JC, Brabb EE, Greene HG, Ross DC (1984) Geology of
Point Reyes Peninsula and implications for San Gregorio fault history. In: Tectonics and Sedimentation along the California
Margin, Society of Economic Paleontologists and Mineralogists,
Pacific Section, 38 (eds Crouch JK, Bachman, SB), pp. 67–86.
Society of Economic Paleontologists and Mineralogists., Los
Angeles.
Delaney PT, Pollard DD, Ziony JI, McKee EH (1986) Field relations between dikes and joints: emplacement processes and paleostress analysis.
Journal of Geological Research , 91 , 4920–38.
Dixon RJ, Schofield K, Anderton R, Reynolds AD, Alexander RWS,
Williams MC, Davies KG (1995) Sandstone diapirism and clastic intrusion in the Tertiary submarine fans of the Bruce-Beryl
Embayment, Quadrant 9, UKCS. In: Characterization of Deep
Marine Clastic Systems, Geological Society Special Publication, 94
(eds Hartley AJ, Prosser DJ), pp. 77–94. Geological Society,
London.
Eldridge GH (1901) The asphalt and bituminous rock deposits of the United States.
US Geological Survey 22nd Annual Report , 1 ,
381–407.
El-Sabbagh D, Garrison RE (1990) Silica diagenesis in the Santa
Cruz Mudstone (Upper Miocene), La Honda Basin, California. In:
Geology and Tectonics of the Central California Coast Region (eds
Green HG, Weber GE, Wright TL), pp. 123–32. American
Association of Pacific Geologists, Pacific Section, Bakersfield, CA.
Erickson CL, Moore JC, Garrison RE (1994) The influence of changing tectonic styles on petroleum migration and accumulation in a small Pacific Rim basin: the Majors tar sand deposits, Santa
Cruz County, California.
Research Report.
University of California
Energy Institute, California.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161

Fluidized sandstone intrusions 161
Fisher NI, Lewis T, Embleton BJJ (1987) Statistical Analysis of
Spherical Data . Cambridge University Press, Cambridge.
Graham SA, Dickinson WR (1978) Evidence for 115 km of right slip on the San Gregorio–Hosgri Fault Trend.
Science , 199 , 179–81.
Greene HG (1990) Regional tectonics and structural evolution of the Monterey Bay Region, Central California. In: Geology and
Tectonics of the Central California Coast Region (eds Green HG,
Weber GE, Wright TL), pp. 1–30. American Association of Pacific
Geologists, Pacific Section, Bakersfield, CA.
Hardebeck JL, Hauksson E (1999) Role of fluid in faulting inferred from stress field signatures.
Science , 9 , 236–39.
Horns DM, Verosub KL (1995) Paleomagnetic investigation of late Neogene vertical axis rotation and remagnetization in central coastal California.
Journal of Geophysical Research , 100 ,
3873–84.
Huang Q (1988) Geometry and tectonic significance of Albian sedimentary dykes in the Sisteron area, SE France.
Journal of
Structural Geology , 10 , 453–62.
Jolly RJH, Cosgrove JW, Dewhurst DN (1998) Thickness and spatial distribution of clastic dykes, northwest Sacramento Valley,
California.
Journal of Structural Geology , 20 , 1663–72.
Jolly RJH, Sanderson DJ (1995) Variation in the form and distribution of dykes in the Mull swarm, Scotland.
Journal of
Structural Geology , 17 , 1543–57.
Lawrence DA, Sancar B, Molyneux SJM (1999) Large-scale sandstone clastic intrusion in the Tertiory of block 24/9,
Norwegian Sea; origin, timing and implications for reservior continuity. In: Americal Association of Petroleum Geologists
International Conference and Exhibition; Abstracts, AAPG
Bulletin , 83 , p. 1324.
Lonergan L, Lee N, Johnson HD, Cartwright JA, Jolly RJH (2000)
Remobilization and injection in deepwater depositional systems:
Implications for reservoir architecture and prediction.
GCSSEPM
Foundation 20th Annual Research Conference, Deep-Water
Reservoirs of the World, December 2000 , (eds Weimer P et al.) pp. 512–32. Published as a CD ROM. Houston, Texas.
Madrid VM, Stuart RM, Verosub KL (1986) Magnetostratigraphy of the late Neogene Purisima Formation, Santa Cruz County,
California.
Earth and Planetary Science Letters , 79 , 431–40.
Molyneux S (1999) Giant clastic dykes and sills of Santa Cruz,
Coastal California.
Petroleum Exploration Society of Great Britain,
January, 1999 Newsletter , 118–125.
Newsom JF (1903) Clastic dikes.
Geological Society of America
Bulletin , 14 , 227–68.
Page BM, Holmes CN (1945) Geology of the bituminous sandstone deposits near Santa Cruz County, California.
US Geological Survey
Oil and Gas Investigations Preliminary Map , 27 .
Phillips RL (1990) Depositional and structural controls on the distribution of tar sands in the Santa Cruz Mountains, California.
In: Geology and Tectonics of the Central California Coastal Region,
San Francisco to Monterey, Volume and Guidebook, Pacific Section,
American Association of Petroleum Geologists, Book GB67 (eds
Garrison R, Greene HG, Hicks KR, Weber GE, Wright TL), pp.
105–21. American Association of Petroleum Geologists, Pacific
Section, Bakersfield, CA.
van der Pluijm BA, Marshak S (1997) Earth Structure: an
Introduction to Structural Geology and Tectonics . McGraw-Hill,
New York.
Pollard DD (1973) Derivation and evaluation of a mechanical model for sheet intrusions.
Tectonophysics , 19 , 233–69.
Price JW, Cosgrove JW (1990) Analysis of Geological Structures.
Cambridge University Press, Cambridge. pp. 502.
Stanley RG (1990) Evolution of the Tertiary La Honda Basin,
Central California. In: Geology and Tectonics of the Central
California Coast Region (eds Green HG, Weber GE, Wright
TL), pp. 31–56. American Association of Petroleum Geologists,
Pacific Section, Bakersfield, CA.
Stanley RG, Lillis PG (2000) Oil-bearing rocks of the Devenport and point Reyes area and their implications for offset along the San
Gregorio and Northern San Andreas Faults. In: Proceedings of the
3rd Conference on tectonic problems of the San Andreas Fault
System, September 6–8, 2000, Stanford University, Stanford (ed
Kovach R), Stanford University Press, Stanford, CA.
Thompson B (1995) Geometry and fluid flow mechanisms of the bituminous sandstone intrusion at Yellow Bank Creek, western
Santa Cruz County, California. Unpublished MS Thesis. UC Santa
Cruz, Santa Cruz.
Thompson BJ, Garrison RE, Moore JC (1999) A late Cenozoic sandstone intrusion west of Santa Cruz, California: Fluidized flow of water- and hydrocarbon-saturated sediments. In: Late Cenozoic
Fluid Seeps and Tectonics along the San Gregorio Fault Zone in the
Monterey Bay Region, California, GB-76 (eds Garrison RE, Aiello
I, Moore JC), pp. 53–74. American Association of Petroleum
Geologists, Pacific Section, Bakersfield CA.
Weber GE (1990) Late Pleistocene slip rates on the San Gregorio
Fault Zone at Point Ano Nuevo, San Mateo County, California.
In: Geology and Tectonics of the Central California Coast Region
(eds Green HG, Weber GE, Wright TL), pp. 193–202. American
Association of Petroleum Geologists, Pacific Section, Bakersfield,
CA.
Zoback MD, Zoback ML, Mount VS, Suppe J, Eaton JP, Healy JH,
Oppenheimer D, Reasenberg P, Jones L, Raleigh BL, Wong IG,
Scotti O, Wentworth C (1987) New evidence on the state of stress of the San Andreas fault system.
Science , 238 , 1105–11.
# 2002 Blackwell Science Ltd, Geofluids , 2 , 147–161