Contrasting Settings of Serpentinite Bodies, San Francisco ... · cut sandstones of the Great Valley Group (GVG), forearc basi n deposits that loca lly overlie the Coast Range ophiolite

International Geology Review, Vol. 46, 2004, p. 1103–1118.Copyright © 2004 by V. H. Winston & Son, Inc. All rights reserved.

Contrasting Settings of Serpentinite Bodies, San FranciscoBay Area, California: Derivation from the Subducting Plate

vs. Mantle Hanging Wall?JOHN WAKABAYASHI1

Geologic Consultant, 1329 Sheridan Lane, Hayward, California 94544

Abstract

Although the best-known outcrops of serpentinite in the California Coast Ranges are part ofthe Coast Range ophiolite (CRO), significant serpentinite bodies crop out within the Franciscansubduction complex, and others have ambiguous affinity. These serpentinites provide evidence ofsubduction-accretion processes.

The Hunters Point shear zone (HPSZ), an intra-Franciscan structural horizon that consists ofboth a regionally extensive serpentinite body and shale matrix mélange, extends tens of km alongstrike, and has a structural thickness of 1 to 1.5 km. The HPSZ exhibits little mixing of serpentiniteand shale. The HPSZ consists of a structurally high shale matrix mélange with mostly sandstoneblocks, an intermediate zone composed of a serpentinite sheet (or sheets) containing metagabbrolenses, and a structurally low shale matrix mélange including a variety of different block types, butlacking serpentinite and gabbro. These field relations suggest that the serpentinite in the HPSZbecame part of the subduction complex by offscraping of the remnants of a mantle core complex fromthe downgoing plate, rather than having originated from the upper mantle hanging wall of the sub-duction zone.

In contrast, serpentinite exposed within a km east of the Hayward fault in the southern HaywardHills is intermixed with shale on scales of tens of meters to centimeters. This serpentinite is presentin shear zones, ranging from 50 m to several centimeters in structural thickness; these shear zonescut sandstones of the Great Valley Group (GVG), forearc basin deposits that locally overlie the CoastRange ophiolite in depositional contact. The shear zones also include blocks of of basalt, blueschist,amphibolite, and gabbro. Serpentinite bodies in the southern Hayward Hills may have been derivedfrom the mantle hanging wall of the subduction zone relatively early in the history of the subductionzone before much tectonic underplating occurred; after substantial underplating, the mantle hangingwall would have been largely blocked from contributing material to the subduction channel. Theserpentinite and shale matrix were exhumed to approximately the same crustal level as the CRO andGVG, before a final stage of faulting mixed pieces of both into the mélange and emplaced themélange into shear zones cutting GVG rocks.

Introduction

SERPENTINITE OCCURS in a variety of tectonic set-tings in California, as summarized by Coleman(2000). Of these tectonic settings, perhaps the bestknown are those associated with major ophioliteunits such as the Coast Range ophiolite (CRO),Josephine ophiolite, Trinity ophiolite, and others(Coleman, 2000). Although described in detail byColeman, serpentinites that occur within subductioncomplexes have not received as much attention asthose associated with major ophiolite sheets. In thispaper, I describe reconnaissance-level field rela-

tionships of two serpentinite occurrences, one fromwithin the Franciscan subduction complex andanother of somewhat ambiguous affinity. Field rela-tionships of these serpentinites may illuminate somedetails about the relationship between subductionprocesses and the incorporation of serpentinites intosubduction complexes or related tectonic settings.Before describing the field occurrences, I summa-rize the tectonic setting of the Franciscan Complexand Coast Range ophiolite in California.

The Franciscan Complex, an assemblage ofvariably deformed rocks with widespread high-pres-sure/low-temperature metamorphism (includingblueschist-facies metamorphism), makes up the1Email: [email protected]

11030020-6814/04/774/1103-16 $25.00

1104 JOHN WAKABAYASHI

largest outcrop area of any rock unit in the Califor-nia Coast Ranges (Fig. 1). The Franciscan formed asa subduction complex associated with east-dippingsubduction at the western North American platemargin from the Late Jurassic through the Miocene,an interval of over 140 million years (e.g., Hamilton,1969; Ernst, 1970; Page, 1981). Franciscan rocksconsist predominantly of shales and sandstones with

subordinate basaltic volcanic rocks, chert, serpen-tinite, and minor limestone. Most Franciscan rockunits were tectonically scraped off the downgoingplate, originating either as trench sediments (sand-stones and shales), with a component of olistostromeblocks of varied lithologies from the upper plate(MacPherson et al., 1990), or as the upper part of thedowngoing oceanic crust (pelagic and volcanic

FIG. 1. Distribution of Franciscan Complex, Coast Range ophiolite/Great Valley Group and other basement rocks ofcentral and northern California. Modified from Wakabayashi (1999a).

CONTRASTING SETTINGS OF SERPENTINITE 1105

rocks; e.g., Hamilton, 1969; Dickinson, 1970). TheCoast Range ophiolite structurally overlies theFranciscan, and is in turn depositionally overlain bywell-bedded sandstones and shales of the GreatValley Group (GVG) that are coeval with the Fran-ciscan (e.g., Dickinson, 1970). The CRO and GVGlack high P/T metamorphism and the degree ofdeformation exhibited by the Franciscan. The CROcrops out in mostly sheetlike remnants or in a nearlycontinuous belt along the eastern margin of the

northern Coast Ranges (Fig. 1). CRO lithologiesinclude serpentinite, gabbro, quartz diorite,diabase, basalt, and silicic volcanic rocks; remnantsrange in structural thickness from a few hundredmeters to 5 km (Hopson et al., 1981).

Hunters Point Shear Zone

The Franciscan Complex of the San FranciscoBay area comprises a stack of coherent or nearly

FIG. 2. Map showing Franciscan Complex and Coast Range ophiolite exposures in the central San Francisco Bayarea. Modified from Wakabayashi (1999b).


coherent thrust nappes separated by mélange zones(Blake et al., 1984; Wakabayashi, 1992; Fig. 2).Mélange zones separating coherent nappes arepredominantly shale matrix mélanges, althoughvariable amounts of serpentinite are present in most(Blake et al., 1984; Wakabayashi, 1992).

One Franciscan structural horizon, the HuntersPoint shear zone (HPSZ) in San Francisco, containsespecially abundant serpentinite (Schlocker, 1974;Wahrhaftig, 1984; Figs. 2 and 3). The HPSZ forms aNW-trending belt 12 km long and up to 3 km wide,with a structural thickness of 1 to 1.5 km. The ser-pentinite of the HPSZ has a prominent magneticanomaly associated with it, which suggests that theHPSZ extends an additional 20 km to the southeastof Hunters Point beneath San Francisco Bay(Jachens and Roberts, 1993; Fig. 3). The serpen-tinite consists primarily of serpentinized harzburgite(Fig. 4). Replacement of primary harzburgite miner-

als by lizardite and chrysotile ranges from 100% toabout 60%. Serpentinite outcrop textures rangesfrom sheared, foliated serpentinite to massive,blocky serpentinite, with local sheared zones (Figs.5 and 6).

Small (centimeters in size) lenses of nearly freshclinopyroxenite occur locally, as well as lenses of met-agabbro that range up to a few tens of meters in size.The metagabbro has undergone amphibolite-faciesmetamorphism; most consists of brownish green horn-blende, and plagioclase has been altered to fine-grained mats of secondary or retrograde metamorphicminerals, including pumpellyite (Fig. 7). The horn-blende has a preferred orientation that defines a prom-inent foliation within each metagabbro lens. Pyroxeneis rare in the metagabbro. No high-pressure/low-tem-perature metamorphic minerals—such as sodicamphibole, sodic pyroxene, or lawsonite—are presentin thin sections of the metagabbro.

FIG. 3. Map showing distribution of serpentinite in the Hunters Point shear zone. Note that “hu” marks uncertainlithology and includes serpentinite. Adapted from Schlocker (1974), Bonilla (1971), and Wahrhaftig (1984).


FIG. 4. Photomicrograph of serpentinized harzburgite from the Potrero Hill area of the Hunters Point shear zone.Some unserpentinized orthopyroxene (opx) and olivine (ol) are present. Mesh-textured serpentine is evident. Field ofview is about 1.5 mm. Cross-polarized light.

FIG. 5. A. Massive, blocky, serpentinized harzburgite, Hunters Point shear zone along Innes Avenue, Hunters Point,San Francisco. View to southeast. The location of this outcrop is shown on Figure. 3. B. Close-up view of massive serpen-tinized harzburgite in the right-hand part of photo A.


Whether or not the serpentinite associated withthe HPSZ forms a continuous sheet or comprises anumber of separate blocks is unclear because of thelack of exposure along much of its strike length (Fig.3). The largest exposures of serpentinite, in thesoutheast part of the exposed HPSZ (Fig. 3) appear

to be part of a continuous body, because there aremany extensive exposures (Figs. 5 and 6 show partsof such exposures) in this area with no other rocktypes. Collectively these outcrops are part of aserpentinite body that extends at least 6 km in strikelength by 2 km in map width, with a structural thick-

FIG. 6. View of outcrop of sheared serpentinite (sp) with lenses of metagabbro (gb), southern base of Potrero Hill,Hunters Point shear zone, north of Cesar Chavez Street west of Connecticut Street. View to the north. The foliation in theserpentinite dips gently to the right (northeast). The location of this outcrop is shown on Figure 3.

FIG. 7. Photomicrograph of metagabbro from a lens in serpentinite in the Potrero Hill area of the Hunters Point shearzone. This rock is composed of greenish-brown hornblende (hb) and what appears to have formerly been plagioclase (pl)that was largely replaced by albite and pumpellyite. Field of view is about 3 mm. Plane light.


ness of about 1 km. Sea cliff exposures at Fort Pointat the northwestern end of the HPSZ also form acontinuous outcrop of serpentinite, rather than anumber of blocks separated by other lithologies. Inboth Fort Point and the Hunters Point–Potrero Hillexposures, foliation in the serpentinite strikes north-west and dips at shallow angles (generally 15–30°)to the northeast (Fig. 6). Exposures are scarce in thecentral part of the HPSZ. Whether or not the FortPoint serpentinite exposures form the western (onland) end of an unbroken slab that includes theexposures at Potrero Hill and Hunters Point is notclear, inasmuch as much of the intervening rock hasbeen mapped as undifferentiated serpentinite,shale, and other lithologies (Schlocker, 1974) andfew of these urban outcrops are visible or accessibletoday. These serpentinite outcrops do not appear tobe part of a serpentinite-matrix mélange. The onlynon-serpentinite rocks found within serpentinitebodies are the rare clinopyroxenites and metagabbrolenses. “Exotic” lithologies such as sandstones,shales, cherts, or volcanic rocks do not occur asblocks within the serpentinite.

Shale matrix mélange crops out on the margins ofthe HPSZ. The best exposures of the shale matrixmélange are at Fort Point/Baker Beach and at Hunt-

ers Point, where the mélange forms the southernmargin of the HPSZ. At those localities, the shalematrix mélange is structurally beneath the serpen-tinite and consists of sheared shale containingblocks of sandstone, chert, and basalt (Fig. 8). Atboth localities, the shale matrix mélange foliation issubparallel to the foliation in the serpentinite, strik-ing NE and dipping NE. At Fort Point/Baker Beach,the shale matrix mélange crops out along the base ofthe sea cliff, where much of it is obscured by land-slides, whereas intact serpentinite crops out at thetop of the cliff (Wahrhaftig, 1984). In the HuntersPoint–Potrero Hill area, the shale matrix mélangecrops out along the southwestern base of the hillsthat are composed of serpentinite (Bonilla, 1971;Schlocker, 1974).

Serpentinite blocks are rare or absent in most ofthe shale matrix mélange exposures, although somefault-bounded slivers of serpentinite occur locallywithin the shale matrix mélange (Fig. 3). Mostblocks in the shale matrix mélange are prehnite-pumpellyite grade, with the exception of severalamphibolite and blueschist blocks found on thebeach south of Fort Point, for which the structuralaffinity is more complex. The metamorphic blocksappear to have slid to the beach from a horizon at or

FIG. 8. Within the shale matrix mélange, structurally lowest part of the Hunters Point shear zone at Baker Beach, SanFrancisco. Some of the shale matrix is shown here between two metabasalt blocks. The location of this outcrop is shownon Figure 3.


near the contact between serpentinite and shalemélange. One amphibolite block exhibits a piece ofserpentinite attached to its southern margin (Fig.9). The border zone, up to about 80 cm wide,between this attached piece of serpentinite and theamphibolite block is highly sheared and recrystal-lized with metamafic, meta-ultramafic, andmetashale materials mixed at mm to cm scales(Figs. 10 and 11). Fine-grained actinolite/tremoliteis common in this border zone, and occurs in felted,folded, and brecciated mats. The attached piece ofserpentinite consists of serpentinized harzburgiteand is composed primarily of lizardite. The amphib-olite block has been variably overprinted by blue-schist-facies assemblages, similar to othermetamorphic blocks studied at the beach (Waka-bayashi, 1990).

Sandstone and shale of the Alcatraz terranestructurally overlies the HPSZ and bounds the shearzone on its northeastern margin. Owing to small-

scale folding and erosion, some klippe of theAlcatraz terrane overlie the northeastern part of theHPSZ, and shale matrix mélange appears to bedeveloped structurally above the serpentinite andbelow the Alcatraz terrane (Schlocker, 1974; Fig. 3).The most common blocks in the mélange above theserpentinite are sandstone, and exotic blocks suchas chert and basalt are less common than themélange structurally below the serpentinite.

In summary, the HPSZ comprises three struc-tural subhorizons (Fig. 3): a thin structurally lowshale matrix mélange including common exoticblocks; a thick structurally intermediate horizonconsisting of a large sheet or several large sheets ofserpentinite, and an intermediate thickness struc-turally high shale matrix mélange or broken forma-tion that contains fewer exotic blocks than thestructurally low mélange. The HPSZ strikes NW anddips NE, and is bounded above by the Alcatrazterrane and below by the Marin Headlands terrane,

FIG. 9. Border between amphibolite border and serpentinite, Hunters Point shear zone, Baker Beach. The borderzone consisting of mixed layers of metashale, metamafic and meta-ultramafic material narrows from nearly a meter wideat the base of the block to about 30 cm wide near its top. The location of this outcrop is shown on Figure 3.


both of which are metamorphosed to prehnite-pumpellyite grade.

Serpentinite in the Southern Hayward Hills

In contrast to the serpentinite of the HPSZ, ser-pentinite exposed in the southern Hayward Hills(Figs. 2 and 12) appears to form both blocks andmatrix in mélange zones. Whether this serpentiniteis associated with the Franciscan, the CRO, or theGVG is problematic. Serpentinite crops out withinan extensive belt of CRO serpentinite, gabbro, andsilicic volcanic rocks that parallels the Haywardfault (Fig. 2). Within this belt, Franciscan rocks areexposed only in small, isolated outcrops south ofOakland, whereas GVG rocks form extensive expo-sures in depositional contact above CRO rocks, orare faulted against them.

At the study locality, serpentinite is present inshear zones, ranging from 50 m to several centime-ters in structural thickness, which cut sandstones ofthe GVG (Figs. 12 and 13). A geotechnical investi-gation for a proposed housing development provided

an opportunity to examine details of the site geologythat would not be possible from surface exposuresalone. The investigation involved the excavation ofnumerous test pits and drilling of many exploratoryborings, all of which contributed to the geologic datashown in Figure 12, and afforded excellent opportu-nities to study the matrix of the mélange/shearzones. The shear/mélange zones large enough toshow on Figure 12 are up to at least 50 m thick, andhave a serpentinite and shale matrix in which thetwo rock types are interleaved and mixed at scalesdown to centimeters (Fig. 13). One shear zone, onlycentimeters thick, was identified in a test pit expo-sure (location on southern part of Fig. 12); it cutGreat Valley Group sandstones, and this shear zoneconsists of both sheared shale and serpentinite. Thelargest piece of serpentinite without other litho-logies mixed in it is probably on the order of 10meters in maximum dimension; serpentinite ispresent as both blocks and granular layers withinthe mélange. Much of the serpentinite appears to bederived from harzburgite, and it displays higher-temperature metamorphic minerals than are typical

FIG. 10. Close-up view of part of the border zone between the amphibolite block and serpentinite shown in Figure 9.Orientation of figure corresponds with that of Figure 9.


for most CRO or Franciscan serpentinite bodies.The serpentinite includes metamorphic mineralssuch as tremolite, talc, and antigorite (Fig. 14),although lizardite and chrysotile are also present inmany samples.

A number of block types occur in the mélangezones, including chert, basalt, fine- and coarse-grained blueschist, amphibolite, actinolite schist,gabbro, and sandstone. Gabbro comprises the larg-est blocks within the mélange; the largest gabbroblock at the study site is up to 20 m in long dimen-sion. Unlike the metagabbro of the HPSZ, thisgabbro has well-preserved pyroxenes and lacksmetamorphic hornblende. Apparent igneous plagio-clase is replaced by very fine grained secondaryminerals, and rare actinolite occurs mostly as over-growths on pyroxene. The metamorphic blocks andthe chert blocks in the mélanges are Franciscan inorigin. In contrast, the gabbro appears to be of CROaffinity, based on the lack of high-pressure meta-morphism and rarity of gabbro in the Franciscan.The basalt lacks high-pressure metamorphic miner-als, but without geochemical data or clear-cut fieldrelations it is difficult to determine whether theserocks are low-grade Franciscan or CRO basalts.Some sandstone blocks within the mélanges areGVG sandstones similar to those that bound theshear zones (lithic-rich sandstones lacking meta-morphic minerals), whereas blocks of unusualvolcanic-clast-rich sandstone (consisting of about80% lithic clasts, most of which are volcanic) found

in the eastern shear zone shown on Figure 12 areFranciscan, based on a well-developed cleavage andmetamorphic pumpellyite. In summary, the blockpopulation of the mélange appears to have a mixedFranciscan, CRO, and GVG source.

The serpentinite and shale matrix mélanges cutGVG rocks, strike NW, and dip NE (Fig. 12). Thisorientation appears to be subparallel to the beddingin the Great Valley Group sandstones that bound theshear zones.

In addition to the locality described here insouthern Hayward, I have also observed similaroccurrences (not shown on published maps) of Fran-ciscan metamorphic rocks in shear zones (matrix notwell exposed, but containing serpentinite in at leastone locality) cutting GVG and/or CRO rocks at twoother localities about 4 and 15 km northwest alongthe Hayward fault, respectively.

Discussion: Speculation on the Originof the Serpentinite Bodies

The southern Hayward Hills and HPSZ serpen-tinites clearly have drastically different local andregional field relationships. In order to speculate onthe origins of the serpentinite at the two localities, Ifirst review some of the proposed sources for serpen-tinites within an accretionary complex and forearcsetting.

A variety of tectonic origins have been proposedfor serpentinites in the California Coast Ranges

FIG. 11. Photomicrograph of part of the border zone between the serpentinite and amphibolite block shown in Figures9 and 10. Fine-grained, folded mats of actinolite (am) and quartz-rich zones (q) are shown along with a clast of clinopy-roxenite with chromian spinel (Cr). Field of view is about 2 mm. Plane light.


(Fig. 15). Perhaps the most commonly proposed tec-tonic setting from which serpentinites in the CoastRanges are derived is the basal ultramafic part ofthe Coast Range ophiolite (e.g., Bailey et al., 1970;Hopson et al. 1981; Page, 1981). Such serpentinitebodies are present both along the eastern margins ofthe Coast Ranges and as outliers that have been dis-placed to positions farther west in the Coast Rangesby dip-slip (thrust or normal) faulting and/or strike-slip faulting (e.g., McLaughlin et al., 1988). Serpen-tinites derived from the CRO structurally overlie

rocks of the Franciscan Complex, in contrast tothose serpentinites structurally interleaved with it.

Other serpentinites associated with the upperplate of the forearc system appear to occur as sedi-mentary deposits that depositionally overlie theCRO (Phipps, 1984). Such deposits locally make uppart of the basal GVG, and Fryer et al. (2000)suggested that they represent deposits from forearcserpentinite mud volcanoes. The recognition ofsedimentary serpentinite deposits in the GVGalso raises the possibility that some Franciscan

FIG. 12. Geologic map of serpentinite and shale matrix mélange zones in part of the southern Hayward Hills. “West-ern sheet,” “central sheet,” and “eastern sheet” denote the three slabs of Great Valley Group sandstone that are sepa-rated by mélange zones and are the source of samples for the corresponding photomicrographs shown in Figure 16.


serpentinites also may have a sedimentary or olis-tostromal origin (Phipps, 1984).

Coleman (2000) suggested that some serpen-tinites within the Franciscan Complex, particularlythe larger bodies, originated from the downgoing

plate, probably as the offscraped and underplatedremnants of mantle core complexes formed nearspreading ridge-transform intersections (e.g.,Tucholke et al., 1998; Karson, 1999). Other intra-Franciscan serpentinites may have been derived by

FIG. 13. Core from one of the serpentinite and shale matrix mélange zones in the Hayward study area. The serpen-tinite and shale are mixed at scales down to centimeters (inch and centimeter metal scale). A few examples of represen-tative clasts and layers are labeled.

FIG. 14. Photomicrograph of serpentinite from one of the shear zones in the Hayward Hills. The minerals in this vieware intergrown talc, tremolite, and antigorite. Field of view is about 1.5 mm. Cross-polarized light.


plucking from the mantle of the upper plate of thesubduction zone (e.g., Cloos, 1984).

Hunters Point shear zone serpentinitesThe HPSZ is a structural horizon within the

Franciscan, and it is not associated with the CRO orthe basal GVG. In addition, the serpentinite iseither massive or occurs as a sheared but not disag-gregated rock mass, in contrast to the textures ofsedimentary serpentinite, so did not originate assedimentary debris shed into a trench. It is difficultto envision the tectonic plucking of serpentinitebodies as large as those in the HPSZ from the deepmantle hanging wall of a subduction zone. Moreover,such a mechanism would suggest a deep source forthe serpentinites, because many Franciscan units,including blueschist-facies units still awaitingexhumation, already existed structurally above the

HPSZ (between the subduction channel and themantle hanging wall) at the time of the accretion ofthe HPSZ at approximately 100 Ma (Wakabayashi,1992, 1999a). If the serpentinite was derived fromsuch a deep source, metagabbro pods within the ser-pentinite should have experienced high-P/low-Tmetamorphism, but they lack such metamorphism.The amphibolite metamorphism of the metagabbropods may have occurred near a spreading ridge.Accordingly, I suggest that the HPSZ serpentiniteswere scraped off the downgoing plate, having origi-nated as mantle core complexes, as suggested byColeman (2000) for Franciscan serpentinites hetermed Franciscan peridotite wedges. The mantlematerial appears to have been offscraped as discreteblocks or slabs, rather than becoming tectonicallymixed with the shale matrix mélange, becauseexotic blocks do not occur within the serpentinite,

FIG. 15. Diagrams showing proposed tectonic origins of serpentinite in the California Coast Ranges. Possible sourcesof serpentinite in trench-forearc regions are indicated in bold type, with Franciscan Complex as an example.


with the possible exception of those at the borderwith the shale matrix mélange. Thus the HPSZconsists of structurally distinct shale matrixmélange and serpentinite horizons, rather thanbeing a serpentinite and shale matrix mélange.

The serpentinite attached to the amphiboliteblock at Baker Beach appears to have undergone adifferent history than the larger pieces of serpen-tinite within the HPSZ. Although the serpentiniteattached to the block no longer contains higher tem-perature meta-ultramafic minerals, the contact zonebetween the serpentinite and the amphibolite blockhas shale that is recrystallized to a degree muchgreater than the rest of the exposed mélange matrix.It is likely that the attached serpentinite once hadhigher-temperature meta-ultramafic minerals thatwere erased by retrogression to lizardite, and thatthe serpentinite was attached to the amphiboliteblock during at least part of its metamorphic evolu-tion from amphibolite- to blueschist-facies meta-morphic conditions. Speculation on how the blockbecame incorporated into the mélange is beyond thescope of this paper.

Southern Hayward Hills serpentinite

Unlike serpentinites of the HPSZ, serpentinitesfrom the southern Hayward Hills locality defy a sim-ple structural categorization. They occur in shearzones that cut Great Valley Group sandstones, partof the “upper plate” structurally above the Fran-ciscan Complex. Yet the shear zones contain blocksof Franciscan rocks with high-P/low-T metamor-phism in addition to blocks of apparent CRO andGVG affinity. Serpentinite is intermixed with shaledown to centimeter scales, in contrast to the largesheets and blocks of serpentinite in the HPSZ. Ser-pentinite at the southern Hayward Hills locality alsoexhibits higher-temperature metamorphic mineralsthan are commonly present in serpentinites in theCalifornia Coast Ranges. Such higher-temperaturemetamorphic minerals, and the occurrence of Fran-ciscan metamorphic rocks (including amphiboliteswith blueschist overprints) suggest that the serpen-tinite in these shear zones may have been derivedfrom the deep mantle hanging wall of the subductionzone, with considerable mixing between serpen-tinite and the shale of the subduction channel (e.g.,Cloos, 1984). The mechanism of eventual emplace-ment of the shale and serpentinite matrix mélangeinto shear zones cutting the GVG is difficult to envi-sion. Metamorphism of the Franciscan blocks in theshear zone, as well as the metamorphic mineralogy

of the serpentinites, suggests considerable tectonictransport (tens of km relative vertical movement)relative to the GVG on either side of the shear zones.Yet each shear zone does not have significant verti-cal separation across it because the GVG sand-stones on the hanging and footwalls of the shearzones are unmetamorphosed, and all of the tectonicslices of GVG rocks are petrographically identical(Fig. 16). A simple explanation for such field rela-tions might be diapiric emplacement of serpentiniteand shale mixtures upward along shear zones orfaults in the GVG (e.g., Dickinson, 1966; Lockwood,1972), but Phipps (1984) has shown that the densitycontrast between such serpentinite and the rocks ofthe GVG is too small (if it existed at all) to drivesuch diapirism. An alternative explanation may bethat the serpentinite and shale originated as sedi-mentary flows, similar to the GVG basal serpen-tinites, but the shale and serpentinite matrix of theshear zones at the southern Hayward locality ismuch more coherent in appearance than sedimen-tary serpentinites of the Coast Ranges. In addition,the higher-temperature meta-ultramafic mineralsfound in the southern Hayward Hills locality seemincompatible with a sedimentary serpentinite origin.

I propose the following speculative scenario forthe origin of these enigmatic serpentinites in thesouthern Hayward Hills. They were derived from thedeep mantle hanging wall above the subductionzone, with the serpentinite having been incorporatedand mixed at small scales into the shale matrix ofthe mélange subduction channel. The subductionchannel may have been particularly rich in serpen-tinite during the early history of subduction becausemost of the mantle hanging wall had not beencovered yet (blocked from providing material to thechannel) by tectonically underplated material.Exhumation by return flow in the channel broughtthe mantle material and associated high-P metamor-phic rocks to relatively shallow levels in the crust,equivalent to those of the Coast Range ophiolite andGreat Valley Group. Following this exhumation, asecond stage of deformation imbricated the GVGand associated CRO and resulted in the juxtaposi-tion of the shale and serpentinite mélanges withGVG and CRO rocks, as well as the inclusion ofCRO and GVG blocks in the mélange. Such fault-ing, whether originally dip-slip, strike-slip, oroblique-slip, lacked large-scale (several km ormore) offset because there is no lithologic contrastbetween the two sides of each shear zone.


Acknowledgments

I thank Berlogar Geotechnical Consultants, Inc.(BGC), for permission to examine and photographtheir drill core. My field study of the southern Hay-ward Hills serpentinite locality was part of consult-

ing projects funded by Northgate EnvironmentalManagement, Inc. and BGC. This paper wasimproved by helpful reviews by R. Erickson, W. G.Ernst, and T. Kato.

REFERENCES

Bailey, E. H., Blake, M. C., Jr., and Jones, D. L., 1970, On-land Mesozoic ocean crust in California Coast Ranges:United States Geological Survey Professional Paper700-C, p. 70–81.

Blake, M. C., Jr., Howell, D. G., and Jayko, A. S., 1984,Tectonostratigraphic terranes of the San Francisco BayRegion in Blake, M. C., Jr., ed., Franciscan geology ofNorthern California: Pacific Section Society of Eco-nomic Paleontologists and Mineralogists, v. 43, p. 5–22.

Bonilla, M. G., 1971, Preliminary geologic map of the SanFrancisco South quadrangle and part of the HuntersPoint quadrangle: U. S. Geological Survey Miscella-neous Field Studies Map MF-574.

Cloos, M., 1984, Flow mélanges and the structural evolu-tion of accretionary wedges, in Raymond, L., ed.,Mélanges: Their nature, origin, and significance: Geo-logical Society of America Special Paper 198, p. 71–80.

Coleman, R. G., 2000, Prospecting for ophiolites along theCalifornia continental margin, in Dilek, Y. D., Moores,E. M., Elthon, D., and Nicolas, A., eds., Ophiolites andoceanic crust: New insights from field studies and theOcean Drilling Program: Geological Society of Amer-ica Special Paper 349, p. 351–364.

Dickinson, W. R., 1966, Table Mountain serpentiniteextrusion in California Coast Ranges: Geological Soci-ety of America Bulletin, v. 77, p. 451–472.

Dickinson, W. R., 1970, Relations of andesites, granites,and derivative sandstones to arc-trench tectonics:Reviews of Geophysics and Space Physics, v. 8, p.813–860.

Ernst, W. G., 1970, Tectonic contact between the Fran-ciscan mélange and the Great Valley Sequence,crustal expression of a Late Mesozoic Benioff Zone:Journal of Geophysical Research, v. 75, p. 886–902.

Fryer, P., Lockwood, J. P., Becker, N., Phipps, S., andTodd, C. S., 2000, Significance of serpentine mud vol-canism in convergent margins, in Dilek, Y., Moores,E. M., Elthon, D., and Nicolas, A., eds., Ophiolitesand oceanic crust: New insights from field studies andthe Ocean Drilling Program: Geological Society ofAmerica Special Paper 349, p. 35–51.

Hamilton, W. B., 1969, Mesozoic California and underflowof the Pacific mantle: Geological Society of AmericaBulletin, v. 80, p. 2409–2430.

Hopson, C. A., Mattinson, J. M., and Pessagno, E. A., Jr.,1981, The Coast Range ophiolite, western California,in Ernst, W. G., ed. The geotectonic development of

FIG. 16. Photomicrographs of Great Valley Group sand-stone from the south Hayward Hills locality. These three sam-ples are taken from different fault-bounded sheets ofsandstone shown on Figure 12. A. From western sheet. B.From central sheet. C. From eastern sheet. These sandstoneshave not been metamorphosed and are nearly identical petro-graphically, indicating little displacement across the shearzones that separate them.


California: Englewood Cliffs, NJ, Prentice-Hall,Rubey Volume 1, p. 419–510.

Jachens, R. C., and Roberts, C. W., 1993, Aeromagneticmap of the San Francisco Bay Area: U.S. GeologicalSurvey, Geophysical Investigations Map GP-1007,scale 1: 286,500.

Karson, J. A., 1999, Geological investigation of a lineatedmassif at the Kane transformation fault: Implicationsfor oceanic core complexes, in White, R. S., Hardman,R. F. P., Watts, A. B., and Whitmarsh, R. B., eds.,Response of the Earth’s lithosphere to extension:Philosophical Transactions, Royal Society of London:Mathematical, Physical, and Engineering Sciences, v.357, p. 713–740.

Lockwood, J. P., 1972, Possible mechanisms for theemplacement of alpine peridotites: Geological Societyof America Memoir 132, p. 273–287.

Macpherson, G. J., Phipps, S. P., and Grossman, J. N.,1990, Diverse sources for igneous blocks in Fran-ciscan mélanges, California Coast Ranges: Journal ofGeology, v. 98, p. 845–862.

McLaughlin, R. J., Blake, M. C., Jr., Griscom, A., Blome,C. D., and Murchey, B., 1988, Tectonics of formation,translation and dispersal of the Coast Range ophioliteof California: Tectonics, v. 7, p. 1033–1056.

Page, B. M., 1981, The southern Coast Ranges, in Ernst,W. G., ed., The geotectonic development of California:Englewood Cliffs, NJ, Prentice-Hall, Rubey Volume 1,p. 329–417.

Phipps, S. P., 1984, Ophiolitic olistostromes in the basalGreat Valley sequence, Napa County, northern Califor-nia Coast Ranges, in Raymond, L., ed., Mélanges:Their nature, origin, and significance: Geological Soci-ety of America Special Paper 198, p. 103–125.

Schlocker, J., 1974, Geology of the San Francisco Northquadrangle, California: U. S. Geological Survey Pro-fessional Paper 782, 109 p.

Tucholke, B. E., Lin, J., and Kleinrock, M. C., 1998,Megamullions and mullion structure defining oceanicmetamorphic core complexes on the Mid-AtlanticRidge: Journal of Geophysical Research, v. 103, p.9857–9866.

Wagner, D. L., Bortugno, E. J., and McJunkin, R. D., 1990,Geologic map of the San Francisco–San Jose quadran-gle: California Division of Mines and GeologyRegional Geologic Map Series, Map 5A, scale:1:250,000.

Wahrhaftig, C., 1984, A streetcar to subduction: Washing-ton, DC, American Geophysical Union, 76 p.

Wakabayashi, J., 1990, Counterclockwise P-T-t paths fromamphibolites, Franciscan Complex, California: Relicsfrom the early stages of subduction zone metamor-phism: Journal of Geology, v. 98, p. 657–680.

Wakabayashi, J., 1992, Nappes, tectonics of oblique plateconvergence, and metamorphic evolution related to140 million years of continuous subduction, Fran-ciscan Complex, California: Journal of Geology, v. 100,p. 19–40.

Wakabayashi, J., 1999a, Subduction and the rock record:Concepts developed in the Franciscan Complex, Cali-fornia, in Sloan, D., Moores, E. M., and Stout, D., eds.,Classic Cordilleran concepts: A view from California:Geological Society of America Special Paper 338, p.123–133.

Wakabayashi, J., 1999b, Distribution of displacement on,and evolution of, a young transform fault system: Thenorthern San Andreas fault system, California: Tecton-ics, v. 18, p. 1245–1274.

Documents

Contrasting Settings of Serpentinite Bodies, San Francisco ... · cut sandstones of the Great Valley Group (GVG), forearc basi n deposits that loca lly overlie the Coast Range ophiolite