9
10 MELANGES FELSBAU 22 (2004) NR. 5 Geological Characterization of Melanges for Practitioners By John Wakabayashi and Edmund W. Medley Geologische Charakterisierung von Melangen für den Fachmann Unter Melangen versteht man ungeordnete Einheiten von Fels, der aus einer Mischung aus Felsmasse mit niedriger Fes- tigkeit und harten Gesteinsblöcken besteht. Obwohl Melangen weltweit anzutreffen sind und Gefügegeologen seit Jahrzehn- ten damit vertraut sind, wissen geotechnische und geologi- sche Fachleute nicht Bescheid über die neuesten geologischen Konzepte bezüglich Melangen und deren ingenieurtechni- schen Stellenwert; diese Unwissenheit resultiert in kostspieli- gen Planungsfehlern und unwillkommenen Überraschungen während der Baudurchführung. Basierend auf Erfahrungen in Franciscan Melange werden im Folgenden Identifizie- rungsmerkmale für Melangen und Zuordnungsmerkmale für externe und interne Details innerhalb der Melangeeinheiten M any geotechnical engineers and engineer- ing geologists (practitioners) believe that simply drawing contact lines or other features on a geologic map or cross section produces repre- sentative characterizations of the subsurface. But the results of most investigations are often grossly incorrect when working with melanges (from French: mélange, or mixture). Melanges are mappable but discontinuous, often chaotic rock units, composed of mixtures of often perva- sively sheared, weak matrix enclosing a variety vorgestellt und auch Richtlinien angeboten, die bei der Erstel- lung einer systematischen ingenieurtechnischen Charakteri- sierung von Melangen als Hilfestellung dienen. Melanges are chaotic bedrock units consisting of mixtures of weak matrix and stronger blocks. Although melanges are globally common and have been familiar to structural geolo- gists for decades, many geotechnical and geological practi- tioners are unaware of recent geological concepts of me- langes and their engineering significance: such ignorance results in costly design errors and unwelcome surprises dur- ing construction. Based on experience with Franciscan com- plex melanges, criteria are provided for identifying me- langes and mapping external and internal details within me- lange units, and guidelines offered for developing orderly engineering characterizations in melanges. Fig. 1 Principal mappable engineering geology characteris- tics of a melange. Bild 1 Ingenieur- geologische Grund- satzmerkmale für die Kartierung von Melangen. of stronger blocks of different lithologies and size (Figure 1). Melanges can form as submarine landslides (olistostromes), by tectonic processes as fault rocks, or by a combination of the two processes (1, 2, 3). The origins of melanges inter- est research geologists to the point of producing several thousand papers, but from an engineer- ing viewpoint, the processes all produce mix- tures of weak matrix and stronger blocks. Despite more than 40 years of geological un- derstanding of melanges and their origins, me-

By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

  • Upload
    vokhue

  • View
    219

  • Download
    0

Embed Size (px)

Citation preview

Page 1: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

10

ME

LA

NG

ES

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5

Geological Characterizationof Melanges for Practitioners

By John Wakabayashi and Edmund W. Medley

Geologische Charakterisierung vonMelangen für den Fachmann

Unter Melangen versteht man ungeordnete Einheiten vonFels, der aus einer Mischung aus Felsmasse mit niedriger Fes-tigkeit und harten Gesteinsblöcken besteht. Obwohl Melangenweltweit anzutreffen sind und Gefügegeologen seit Jahrzehn-ten damit vertraut sind, wissen geotechnische und geologi-sche Fachleute nicht Bescheid über die neuesten geologischenKonzepte bezüglich Melangen und deren ingenieurtechni-schen Stellenwert; diese Unwissenheit resultiert in kostspieli-gen Planungsfehlern und unwillkommenen Überraschungenwährend der Baudurchführung. Basierend auf Erfahrungenin Franciscan Melange werden im Folgenden Identifizie-rungsmerkmale für Melangen und Zuordnungsmerkmale fürexterne und interne Details innerhalb der Melangeeinheiten

Many geotechnical engineers and engineer-ing geologists (practitioners) believe that

simply drawing contact lines or other features ona geologic map or cross section produces repre-sentative characterizations of the subsurface.But the results of most investigations are oftengrossly incorrect when working with melanges(from French: mélange, or mixture). Melangesare mappable but discontinuous, often chaoticrock units, composed of mixtures of often perva-sively sheared, weak matrix enclosing a variety

vorgestellt und auch Richtlinien angeboten, die bei der Erstel-lung einer systematischen ingenieurtechnischen Charakteri-sierung von Melangen als Hilfestellung dienen.

Melanges are chaotic bedrock units consisting of mixtures ofweak matrix and stronger blocks. Although melanges areglobally common and have been familiar to structural geolo-gists for decades, many geotechnical and geological practi-tioners are unaware of recent geological concepts of me-langes and their engineering significance: such ignoranceresults in costly design errors and unwelcome surprises dur-ing construction. Based on experience with Franciscan com-plex melanges, criteria are provided for identifying me-langes and mapping external and internal details within me-lange units, and guidelines offered for developing orderlyengineering characterizations in melanges.

Fig. 1 Principalmappable engineering

geology characteris-tics of a melange.Bild 1 Ingenieur-

geologische Grund-satzmerkmale für

die Kartierung vonMelangen.

of stronger blocks of different lithologies and size(Figure 1). Melanges can form as submarinelandslides (olistostromes), by tectonic processesas fault rocks, or by a combination of the twoprocesses (1, 2, 3). The origins of melanges inter-est research geologists to the point of producingseveral thousand papers, but from an engineer-ing viewpoint, the processes all produce mix-tures of weak matrix and stronger blocks.

Despite more than 40 years of geological un-derstanding of melanges and their origins, me-

Page 2: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5 11

ME

LA

NG

ES

langes are unknown or misinterpreted by manypractitioners. Costly and imprudent consequen-ces derive from practitioners’ errors in the mis-characterization of melange structures as “layercake” strata, or incorrectly describing melangesas “soil containing boulders”, or “miscellaneoussoils”, for example. To confuse matters, the word“melange” is also used by some practitioners tomean any mixture of rock and soil materials,which is inappropriate given the long-used geo-logical meaning. Furthermore, some practitio-ners declare melanges as impossible to charac-terize and recommend geotechnical design bebased on the properties of the weak matrix. Suchsimplification can lead to too-conservative andinappropriate designs and costly surprises andunsafe ground failures during construction.

Researchers have recently developed ap-proaches to the engineering characterization ofmelanges and other bimrocks (block-in-matrixrocks) (4, 5, 6, 7, 8). Medley (9) defined bimrocksas geological mixtures of geotechnically signifi-cant blocks of rock within weaker, bonded rockmatrices. Geotechnical significance means thatthere is sufficient mechanical contrast betweenthe blocks and the matrix to force failure sur-faces to negotiate around the blocks in tortuousfashion; and that there is a sufficient size andnumbers of blocks to affect the overall mechani-cal properties of the geological mixture.

The authors of this paper, a Structural Geolo-gist (Wakabayashi) and a Geological Engineer(Medley), consider it necessary to apply bothfirst-order geologic field observations and quan-titative engineering methods to the characteriza-tion of melange once it is identified, and to thatend guidance for the identification, mapping,and characterization of melanges by practicinggeologists and engineers is provided.

Melanges – geologic conceptsand misconcepts

A brief history of styles of mapping of melangesof the Franciscan Complex (“the Franciscan”) ofcoastal California provides examples of how geo-

Fig. 2 Hypothetical geologic maps showing how prevailing geologic theories influencehow contacts are drawn on maps. Map A: outcrops. Map B: geology interpreted asstratabound layers. Map C: entire area interpreted as melange. Map D: area composedof both melange and coherent thrust sheets.Bild 2 Theoretische geologische Karten, die zeigen, wie vorherrschende Theorienin der Geologie die Darstellung von Kontaktflächen beeinflussen. Abbildung A: Auf-schlüsse. Abbildung B: Geologie interpretiert als stratigraphische Schichten. Abbil-dung C: Gesamtfläche interpretiert als Melange. Abbildung D: Fläche besteht ausMelangen und zusammenhängenden Überschiebungsdecken.

logic knowledge influence how contacts aredrawn on geologic maps and cross sections (Fig-ures 2 and 3). The Franciscan hosts some of theworld’s most famous melanges (10,11), as wellas engineering projects that have suffered prob-lems because of their chaotic conditions.

Page 3: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

12

ME

LA

NG

ES

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5

(Figure 2C). Some geologists even classified theentire Franciscan as one large melange bodyand neglected the internal details, a simplifica-tion that can still be encountered in some aca-demic research papers. By the mid-1970s, geo-logists such as Maxwell (14), began to discrimi-nate the Franciscan into “coherent units”, fault-bounded sheets of intact non-melange Francis-can geologic rock units; and discrete “melangeunits”. This concept was expanded in the 1980sas the “terrane” concept explained the complextectonic jigsaw of the North American Cordillera,with terranes being the individual puzzle pieces(15, 16). The Franciscan was then called the“Franciscan Assemblage”. Although the terraneconcept led to improved categorization of coher-ent units, identification of melanges regressed,as all Franciscan melange bodies were thencollected into one “Central Terrane”, based onan interpretation that all Franciscan melangesformed at the same time.

Wakabayashi (17, 18, 31) expanded on Max-well’s (14) concepts, by delimiting separate Fran-ciscan melanges and coherent units, and thencorrelating melange units and coherent units todiscrete structural levels within stacks of thrustnappes. Accordingly, an up-to-date structuralgeologist mapping Franciscan outcrops todaymight find and map both coherent and melangeunits, as shown schematically in Figure 2D. Thismodern approach reflects the appropriate cur-rent “Complex” suffix to “Franciscan Complex”.

During a century of geologic mapping in theFranciscan Complex, the rocks have not chang-ed, but the geologic maps have changed dra-matically. Although geologists long ago recog-nized melanges and how to map them, manypractitioners still treat melange bedrock as bed-ded geologic units. Others, also incorrectly, con-sider entire regions to be melange. Both groupsthus fail to secure the geologic information thatcan be collected and used for engineering pur-poses.

When mapping, geologists most commonlyencounter the erosion-resistant blocks of a me-lange (Figure 4), rather than the weak matrix,which easily erodes and seldom forms observ-able outcrops except in bare natural slopes, orartificial cut slopes. Hence, prior to the 1960’smost geologists mapped areas with scatteredoutcrops of sandstone, chert, basalt, or otherrock types (Figure 2A) and then interpreted themelanges into the layer-cake continuous strati-graphic framework of the Franciscan “Forma-tion” (12) (Figure 2B). If ignorant of melanges,many practitioners still map this way.

Greenly (13) first christened chaotic units inNorth Wales as “Autoclastic Mélange” but wide-spread recognition of melange structures did notfollow until Hsü (10) formalized the melangeconcept. Melanges were then recognized as glo-bally common, particularly in ancient orogenicbelts associated with old subduction zones (2,11). Following the acceptance of Hsü’s (10) me-lange concepts, geologists mapping in the Fran-ciscan and similar geologic confusion mappedoutcrops as blocks in the usually unseen matrix

Fig. 4 A view of landscape underlain by serpentinite matrix and shale matrix melange;Tiburon Peninsula, San Francisco area, California.Bild 4 Landschaftsansicht mit darunterliegender Serpentinit-Matrix und SchiefersteinMatrixmelange; Tiburon Halbinsel, Bezirk San Francisco, Kalifornien.

Fig. 3 Cross section-al diagrams showingthe difference betweenassuming stratigraphiccontinuity and assum-ing melange structurewhen interpretingborehole data.A: borehole observa-tions. B: Cross-sectionbased on interpreta-tion of strataboundgeology (layers). C:Cross-section basedon melange model.Bild 3 Querschnitts-diagramme zeigen denUnterschied zwischenangenommenerSchichtenfolge undvermuteter Melange-struktur bei der Inter-pretation von Bohr-kerndaten. A: Kernboh-rungsbeobachtungen.B: Querschnitt basie-rend auf Auswertungvon stratigraphischenSchichten. C: Quer-schnitt basierend aufeinem Melangemodell.

Page 4: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5 13

ME

LA

NG

ES

Mapping melanges –guidelines and cautions

In the Franciscan a gradation exists between co-herent units and melanges, with an intermediatelevel of stratal disruption, commonly referred toas a “broken formation” (2), that renders identi-fication of melange bimrocks for engineeringpurposes more difficult. The origins of melangesdictate the nature of the bounding contacts of amelange body. A purely sedimentary (or olisto-stromal) melange has sedimentary boundingcontacts unless modified by later faulting,whereas the contacts of a tectonic melange are,by definition, faults. Furthermore, in melanges,tectonic signatures may include pronouncedanisotropic rock mass fabrics that control matrixshears and block orientation (7, 8).

Despite the complexity of melanges, a knowl-edgeable and alert geologist can identify andmap much useful information, as shown in Fig-ure 1. Assuming that the melange has been cor-rectly recognized, the overall boundary contactsof melange bodies will require standard “exter-nal” mapping of faults or depositional contacts,depending on the origin of the melange. “Inter-nal” mapping of melanges requires detailed ob-servations. When working with coherent geology(intact geologic units), a geologist commonly lo-cates a few points along a contact and interpo-lates between them while “contact mapping”.However, internal mapping of a melange is bestaccomplished by “saturation” mapping of everyavailable outcrop. Detailed mapping will definethe external contacts of the melange body, delin-eation of the boundaries of larger blocks, provideinformation to estimate the proportion of theblocks in the melange, and information on thevariety of block lithologies. Several guidelinesand common errors are summarized below.

Recognizing melanges andgeomorphologic indicators

A melange must be recognized early in an inves-tigation. One of the most common errors by prac-titioners in this regard is: not consulting a geolo-gist nor reading a geological map. Even whenavailable geology maps identify melanges, manygeotechnical engineers (in particular) seem un-able to conceive of the possibility that a “claysoil” may actually be pervasively sheared shalebedrock; that “bedrock” is discontinuous blocks,and that “boulders” are blocks that may be hun-dreds of meters in dimension. Such ignoranceleads to mischaracterizations that could beavoided by consulting with a knowledgeable geo-logist.

Most units termed by structural geologists“melanges” have matrices with metamorphicgrades less than greenschist facies and so willconform to the engineering definition of abimrock. However, in some mappable melangebodies there may be areas that are bimrocks in

Page 5: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

14

ME

LA

NG

ES

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5

one place but not in another. For example, in thenorthern Sierra Nevada of California, there aremelange units of sub-greenschist metamorphicgrade (19, 20) that are bimrocks because littlerecrystallization has occurred to strengthen thematrix relative to the blocks. However, furthersouth, in the central Sierra Nevada, these samegeologic units occur with upper greenschist andhigher metamorphic grades (20) and the me-lange matrix is mechanically competent quartz-mica schist, and the melange is not a bimrock.

Melanges occur at all scales, from shear zonesthat are several km in outcrop dimension andstructural thickness, to fault zones of meter orsmaller scales. Franciscan Complex melangesare scale independent, meaning that melangehave block and matrix structure at any scale ofobservation (9, 21, 22). Perhaps the only (andquite rare) exception to this scale independenceare basalt matrix shear zones. The authors haveobserved such volcanic matrix limited to scalesbetween from microscopic (millimeters) up toabout a meter or so of structural thickness.

The most common field indicator of melangesis their geomorphologic expression. Becausemelange matrix is commonly weak, it is subjectto slope movement and easily eroded. As a con-sequence it tends to form rolling topographywith outcrops of larger blocks standing out incontrast, a geomorphology commonly referredto in California as “melting ice-cream topogra-

phy” (see Figure 4). However, this characteristicgeomorphic signature is not foolproof, for somecoherent chert and basalt units will form some-what similar topography with chert making upmost of the blocky outcrops (Figure 5).

In some areas, scattered exposures of chert orbasalt or limestone, in an area otherwise exhibit-ing only outcrops of sandstone and shale oftenindicates the presence of a melange, as does thepresence of rocks such as sandstone, shale,chert, or basalt in an area that is otherwiseserpentinite. Scattered metamorphic rocks thatare of different metamorphic grade than sur-rounding rocks are also useful field indicators ofa melange.

Serpentinite by itself is not necessarily an in-dicator of melange but it is commonly associatedwith melanges. Serpentinite in an area that isotherwise mostly sandstone and shale indicatesthe likelihood of the underlying rock unit being amelange. In serpentinite matrix melanges, thematrix is of sheared or disaggregated serpenti-nite and the most common blocks are usuallymassive serpentinites and less serpentinized ul-tramafic rocks, various mafic igneous rocks(gabbro, diabase, basalt), pelagic sedimentaryrocks (chert, limestone), and metamorphic rocks(23, 1). But many serpentinite bodies are notmelanges in a geologic sense, contrary to somemisconceptions. Such bodies of rock usually oc-cur as fault-bounded sheets or blocks and theserpentinite comprising them can range frommassive and strong to sheared. Hence, a sheetcomposed entirely of serpentinite may exhibitblock and matrix fabric, and thus be a bimrock,but not be a geologic melange.

Weathered melange exposures can be diffi-cult to distinguish from colluvial soils, particu-larly if the colluvium itself has a melange source.Melanges interpreted as colluvium may lead toincorrect conclusions as to the subsurface geo-metry, since a colluvium deposit will have a baseand a melange body may not. In a good exposure(such as the wall of a trench or test pit), somedifferences between melange-derived colluviumand weathered melange bedrock can be ob-served. Melange-derived colluvium will seldomhave well-developed matrix foliation that is con-tinuous over a square meter or so of exposure,whereas such foliation is commonly observableeven in weathered bedrock. Melange-derivedcolluvium may have apparent foliation orienta-tions that are fairly consistent and they will com-monly be sub parallel to the slope, but the areasover which this foliation is visible will be patchy,for they will consist of individual pieces of matrixthat have been incorporated into the soil. In me-lange-derived colluvium there may also be bits offormer melange matrix that are rotated so thatthere are abrupt discontinuities in foliation ori-entation, in contrast to folding of the foliation orwarping of foliation around blocks that charac-terize melange bedrock. The distinction between

Fig. 5 Photos show-ing how geomorpho-logy can be misleadingin identifying melange.A: melange-like topo-graphy with a chertoutcrop; B (taken afew hundred metersaway from A) showsthat the underlyingbedrock is not me-lange.Bild 5 Fotoauf-nahmen zeigen, wieirreführend Geo-morphologie bei derBestimmung vonMelange sein kann. A: Melange-ähnlicheTopographie mitKieselerdeaufschluss.B: (aufgenommen auseinigen 100 m Ent-fernung von A) zeigt,dass das darunter-liegende Felsgesteinnicht Melange ist.

Page 6: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5 15

ME

LA

NG

ES

weathered melange bedrock and melange-de-rived colluvium may be difficult to ascertain inborehole samples because a larger area of ob-servation is generally needed to apply the crite-ria noted above.

Mapping matrix and foliationsThe most common melange matrix types areshale/mudstone, sandstone, and serpentinite.Basalt or volcanic matrix (or mixed volcanic/shale matrix) is rare. Some melanges have amixed serpentinite and shale matrix in whichserpentinite can be interleaved as small as centi-meters, although it is more common to findserpentinite as blocks in shale matrix melange(24).

Mapping melange foliation it is no differentthen mapping foliation in a metamorphic rockunit. Melange matrix foliation locally wrapsaround blocks and will have variable orienta-tions, but over the extent of the mappable unitwill commonly have a comparatively consistentfoliation. When possible, the foliation orienta-tions should be mapped to provide clues aboutthe general orientation of the melange fabricwhich likely influences anisotropy in thestrength of the melange, as described by Medleyand Sanz (25). Shears may be so pervasive thatthe matrix is soil-like. Matrix sheared into scalyclay, in which the matrix is pervasively shearedand breaks into brittle chips of shale (ArgilleScagliose of Northern Italy), may also be foundand is diagnostic of melange.

Blocks – size distributions, lithologies,proportions, and orientations

Melange blocks vary greatly in character andsize. To be considered a block there must bemechanical contrast between the block and thesurrounding matrix, which can often be decidedon the basis of striking both with a rock pick andobserving the penetration or sound (4, 9). Theblock size distributions of observed Franciscanmelanges are scale-independent or fractal (9,22), and blocks will be found at all scales of engi-neering interest. In outcrops blocks are found assmall as sand, whereas in regional-scale me-lange (several km in structural thickness), blockscan exceed a km in maximum dimension. The“size” of a “block” is thus dependent on the scaleof observation and various criteria have beendeveloped for determining critical scales (4, 9,22). However, only rarely is the observed “size”of a block the same as the “diameter” of a block,for reasons explained by Medley (26, 27) andHaneberg (28). Once a “characteristic engineer-ing dimension” or scaling dimension is selectedthat represents the scale of engineering interestof the bimrock (e.g. slope height, footing width,diameter of triaxial specimen), blocks are de-fined as being within about 5 to 70 % of that di-mension, at least until the scale of interestchanges (4). Since scales will change from recon-

naissance-level site mapping to the scale of theproposed facility (e.g. cut slope, tunnel, founda-tion) it is best to decide early in the investigationwhat range of block sizes to examine and mea-sure.

The lithologies of blocks vary from melange tomelange and locally within any single melangeunit. In shale matrix melanges, the most com-mon block lithology is generally greywacke, withmuch smaller proportions of basalt, chert, lime-stone, plutonic and metamorphic rocks (9, 11).Identification of block lithologies and block dis-continuity fabric is important for engineeringpurposes because certain block lithologies maypose greater excavation challenges than others,owing to their mechanical and discontinuityproperties. For example an unexpected block ofintact, fresh greenstone with an unconfinedcompressive strength of 200 MPa (30 000 psi)can seriously frustrate tunneling that has beendesigned to accommodate more tractable frac-tured greywackes. Also, fractured, weak blocksmay offer little mechanical contrast with matrixand should thus prudently be assigned to matrixwhen considering overall geomechanical prop-erties of the bimrock.

The volumetric proportion of blocks in a me-lange is an important engineering geology pa-rameter because studies have shown that me-

Page 7: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

16

ME

LA

NG

ES

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5

lange strength is related to the volumetric pro-portion of blocks (5, 22, 29, 30). However, as de-scribed above, there are significant uncertain-ties to estimates of volumetric block proportionsbased on field observations (26, 27, 28).

Blocks in a melange will commonly have pre-ferred shapes and orientation, much like imbri-cated pebbles in a gravel deposit. For blocks thatare commonly disk shaped in three dimensions(“phacoids”) the disk plane is generally parallelto sub parallel to the melange foliation. In addi-tion, the long dimension of blocks in a melangemay also have a preferred orientation. Similar tothe matrix foliation, block shape orientation mayalso influence anisotropy in the overall strengthof the melange so this is field geologic informa-tion that should be recorded as recommendedalso by Haneberg (28).

Internally, the block arrays of many melangesdo not appear to exhibit any order, but somemelanges have mappable sub zones within them.These sub zones can be distinguished by differ-ences in block lithologies, block abundance, oreven matrix type. For example a melange mayconsistently have a structurally lower zone thathas common chert and basalt blocks, but have astructurally higher zone that lacks chert or ba-salt blocks. Different sub zones within a melangemay actually correspond to spatially distinct(and thus mappable) subunits of different blockproportions or block lithologies. This also applies

to some melanges that have gradational con-tacts: mapping from the outside of the unit to-ward the middle one might observe a gradationfrom intact sandstone and shale to broken for-mation (block-in-matrix structure but no blocktypes other then shale and sandstone) to a fullmelange with exotic blocks. This gradation cor-responds to a difference in block proportions,and such a gradation is commonly mappable.

Interpretations fromborehole observations

Interpretation of melanges from borehole datapresents considerable additional challenges asindicated in Figure 3. Whereas surface float orgeomorphic clues allows interpolation betweenoutcrops, interpolation of block boundaries fromboreholes is impossible unless the block isknown to extend between the boreholes. Be-cause of the potential for interpretation errors,backhoe pits or excavator trenches may yieldmore useful and economical subsurface infor-mation such as fabric orientations. Alternatively,as commonly performed in California, large di-ameter auger borings can be drilled to allow ac-cess by a geologist protected by a cage.

As noted previously, external contacts of me-lange can be interpolated between boreholes asfor any geologic contact or fault. If internal subzones are mappable, including gradations near

Page 8: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5 17

ME

LA

NG

ES

the external contacts, it may be possible toproject these sub zone boundaries betweenboreholes. A cautionary note: contacts, particu-larly external contacts of a melange body, mustbe recognized. For example if a borehole at thedipping external boundary of a melange bodypenetrates the melange and terminates within acoherent unit, the coherent unit may inadvert-ently be classified as a block, leading to a too-high linear block proportion.

The lengths of the intercepts between the coreand blocks (chords) can be totaled for severalboreholes and divided by the total length of theboreholes to yield a cumulative linear block pro-portion, that subject to adjustments for uncer-tainty (26), yields an estimate of the volumetricblock proportion of the melange explored. Withconsiderably greater potential errors, the chordsmay also crudely indicate blocks size distribu-tions subject to several cautions (27). Melangefoliation and block preferred shape may also berecorded in a borehole with oriented core.

A common error when logging core in me-lange is to describe the alternating matrix andblock intersections as “inter-layered” or inter-bedded” shale and sandstone. But such descrip-tions incorrectly imply stratal continuity and ifused to describe melanges in geological reportscan lead to misunderstandings when drawingcross sections, or to differing site conditions claimsfrom earthwork and tunneling contractors.

It is common practice in Northern Californiato extend exploration boreholes in Franciscanmelanges through soil and terminate the drilling1 to 2 m into bedrock. A common error whenexploring melanges to characterize them as “soilabove bedrock”, “miscellaneous soils” or “soilwith boulders”. The use of these inappropriateterms for Franciscan melange has been a factorin earthwork construction disputes. For examplecontractor have been known to excavate deeplyin attempts to locate the “basal failure surface”in a pervasively sheared “clay soil”, and to jack-hammer unexpected “boulders” in excess of 5 msize. Such problems are avoided if practitionersdo not draw straight lines between the “rock/soilcontacts” they identify in exploration borings.

Conclusions

Melanges and similar bimrocks are commonthroughout the world and many engineeringprojects are constructed in these chaotic rockbut the engineering geologic understanding ap-plied to many of these projects has been obsoletefor decades. The methods presented in this pa-per should help geologists and engineers learnhow to identify and characterize melange, sothat engineering assessment of melanges andother bimrocks can be performed. Admittedly,melanges are more difficult to characterize than“coherent” geologic units, but practitioners must

Page 9: By John Wakabayashi and Edmund W. Medley · WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS FELSBAU 22 (2004) NR. 5 ... externe und interne Details

18

ME

LA

NG

ES

WAKABAYASHI AND MEDLEY: GEOLOGICAL CHARACTERIZATION OF MELANGES FOR PRACTITIONERS

FELSBAU 22 (2004) NR. 5

learn to characterize geological chaos in an or-derly fashion, or else continue to perform costlyand imprudent mischaracterizations.

References1. Phipps, S.P.: Ophiolitic olistostromes in the basal GreatValley sequence, Napa County, northern California CoastRanges. In: Raymond, L.A. (ed.): Melanges: Their nature,origin, and significance. Geological Society America, SpecialPaper 198, pp. 103-125, 1984.2. Raymond, L.A.: Classification of melanges. In: RaymondL.A. (ed.): Melanges: Their nature, origin and significance.Geological Society of America, Boulder, Special Publication228, pp.7-20, 1984.3. Cowan, D.S.: Structural styles in Mesozoic and Cenozoicmelanges in the Western Cordillera of North America. Geo-logical Society of America Bulletin, Vol. 96 (1985), pp. 451-462.4. Medley, E.W.: Orderly characterization of chaotic Francis-can melanges. Felsbau, Vol. 19 (2001), No. 4, pp. 20-33.5. Lindquist, E.S. ; Goodman, R.E.: The strength and defor-mation properties of a physical model melange. In: Nelson,P.P.; Laubach, S.E. (eds): Proceedings of the 1st NorthAmerican Rock Mechanics Conference (NARMS), Austin,Texas, pp. 843-850. Rotterdam: Balkema, 1994.6. Goodman, R.E. ; C.S. Ahlgren: Evaluating safety of con-crete gravity dam on weak rock: Scott Dam. J. of Geotech-nical and Geoenvironmental Engineering, Vol. 126 (2000),pp. 429-442.7. Riedmüller, G. ; Brosch, F.J. ; Klima, K. ; Medley, E.W.:Engineering geological characterization of brittle faults andclassification of fault rocks. Feldsbau, Vol. 19 (2001), No. 4,pp. 13-19.8. Button, E.A. ; Schubert, W. ; Riedmüller, G. ; Klima, K. ;Medley, E.W.: Tunnelling in tectonic melanges – accommo-dating the impacts of geomechanical complexities and an-isotropic rock mass fabrics. Bulletin of Engineering Geologyand the Environment, (in press).9. Medley, E.W.: The engineering characterization of me-langes and similar block-in-matrix (bimrocks). Ph.D. disserta-tion, Dept. of Civil Engineering, University of California, Ber-keley, California, USA,1994.10. Hsü, K.J.: The principles of melanges and their bearingon the Franciscan-Knoxville paradox. Geological Society ofAmerica Bulletin, Vol. 79 (1968), pp. 1063-1074.11. Cloos, M.: Flow melanges and the structural evolution ofaccretionary wedges. Geological Society America, SpecialPaper 198, pp. 71-80, 1984.12. Taliaferro, N.L.: Franciscan-Knoxville problem. AmericanAssociation of Petroleum Geologists Bulletin, Vol. 27 (1943),pp. 109-219.13. Greenly, E.: The geology of Anglesey. Geological Surveyof Great Britain. London, 1919.14. Maxwell, J.C.: Anatomy of an orogen. Geological Societyof America Bulletin, Vol. 85 (1974), pp. 1195-1204.

15. Blake, M.C. Jr. ; Howell, D.G. ; Jayko, A.S.: Tectonostrati-graphic terranes of the San Francisco Bay Region. In: Blake;M.C. Jr. (ed.): Franciscan Geology of Northern California.Pacific Section Society of Economic Paleontologists and Mi-neralogists, Vol. 43, pp. 5-22, 1984.16. Blake, M.C. Jr. ; Howell, D.G. ; Jones, D.L.: Preliminarytectonostratigraphic terrane map of California. United StatesGeological Survey Open File Report 82-593, 1982.17. Wakabayashi, J.: Nappes, tectonics of oblique plate con-vergence, and metamorphic evolution related to 140 millionyears of continuous subduction, Franciscan Complex, Cali-fornia. Journal of Geology, Vol. 100 (1992), pp. 19-40.18. Wakabayashi, J.: Distribution of displacement on, andevolution of, a young transform fault system: the northernSan Andreas fault system, California. Tectonics, Vol. 18(1999), pp. 1245-1274.19. Day, H.W. ; Schiffman, P. ; Moores, E.M.: Metamorphismand tectonics of the northern Sierra Nevada. In: Ernst, W.G.(ed.): Metamorphism and crustal evolution of the westernUnited States. Rubey Volume VII, pp. 737-763. EnglewoodCliffs, New Jersey: Prentice-Hall, 1988.20. Sharp, W.D.: Pre-Cretaceous crustal evolution in the Si-erra Nevada region, California. In: Ernst, W.G. (ed.): Meta-morphism and crustal evolution of the western United States.Rubey Volume VII, pp. 823-865. Englewood Cliffs, New Jer-sey: Prentice-Hall, 1988.21. Medley, E.W.: Using stereologic methods to estimate thevolumetric block proportion in melanges and similar block-in-matrix rocks (bimrocks). Proceedings of the 7th Congressof the International Association of Engineering Geologists,Lisbon, pp.1031-1040. Rotterdam: Balkema, 1994.22. Medley, E.W. ; Lindquist, E.S.: The engineering signifi-cance of the scale-independence of some Franciscan me-langes in California, USA. In: Daemen, J.K. ; Schultz, R.A.,(eds.): Proceedings of the 25th US Rock Mechanics Sympo-sium, pp. 907-914. Rotterdam: Balkema, 1995.23. Hopson, C.A. ; Mattinson, J.M. ; Pessagno, E.A. Jr.: TheCoast Range ophiolite, western California. In Ernst, W. G.,(ed.): The Geotectonic Development of California, pp. 419-510. Englewood Cliffs, New Jersey: Prentice-Hall, 1981.24. Wakabayashi, J.: Contrasting settings of serpentinitebodies, San Francisco Bay area, California: Derivation fromthe subducting plate vs. mantle hanging wall. InternationalGeology Review, 2004.25. Medley, E.W. ; Sanz. P.R.: Characterization of Bimrocks(Rock/Soil Mixtures) With Application to Slope StabilityProblems. In: Schubert, W. (ed.): Proc. Eurock 2004 and 53rd

Geomechanics Colloquium, Salzburg, Austria. Essen: VerlagGlückauf GmbH, 2004.26. Medley, E.W.: Uncertainty in estimates of block volumet-ric proportion in melange bimrocks. In: Marinos, P.G. ;Kpukis, G. ; Tsiambous, G. ; Stournaras, G. (eds.): Proc Int.Symp. On Engineering Geology and the Environment, pp.267-272. Rotterdam: Balkema, 1997.27. Medley, E.W.: Estimating block size distributions of me-langes and similar block-in-matrix rocks (bimrocks). Proc. 5th

North American Rock Mechanics Symposium, Toronto,Canada, 2002.28. Haneberg, W.C.: Simulation of 3-d block populations tocharacterize outcrop sampling bias in block-in-matrix rocks(bimrocks). Felsbau, Vol. 22, No. 5, pp. 19-26 (this issue).29. Lindquist, E.S.: The strength and deformation propertiesof melange. Ph.D. dissertation, Dept. of Civil Engineering,University of California, Berkeley, CA, USA, 1994.30. Lindquist, E.S.: The mechanical properties of a physicalmodel melange. Proceedings of the 7th Congress of the Inter-national Association of Engineering Geologists, Lisbon,pp.819-826. Rotterdam: Balkema, 1994.31. Wakabayashi, J.: Subduction and the rock record: Con-cepts developed in the Franciscan Complex, California. In:Sloan, D. ; Moores, E.M. ; Stout, D. (eds.): Classic CordilleranConcepts: A View From California. Geological Society ofAmerica, Special Paper 338, pp. 123-133, 1999.

AuthorsJohn Wakabayashi, 1329 Sheridan Lane, Hayward, CA94544, USA, E-Mail [email protected]; Edmund W. Medley,Medley Geoconsultants, 1554 Winding Way, Belmont, CA94002, USA, E-Mail [email protected]