ARMA 08-357

FRANCISCAN COMPLEX, CALIFORNIA: PROBLEMS IN RECOGNITION OF MELANGES, AND THE GAP BETWEEN RESEARCH KNOWLEDGE AND PROFESSIONAL PRACTICE.

John Wakabayashi, Department of Earth and Environmental Sciences, California State University, Fresno, CA 93740, USA

Copyright 2008, ARMA, American Rock Mechanics Association

This paper was prepared for presentation at San Francisco 2008, the 42nd US Rock Mechanics Symposium and 2nd U.S.-Canada Rock Mechanics Symposium, held in San Francisco, June 29-July 2, 2008. This paper was selected for presentation by an ARMA Technical Program Committee following review of information contained in an abstract submitted earlier by the author(s). Contents of the paper, as presented, have not been reviewed by ARMA and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of ARMA, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of ARMA is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgement of where and by whom the paper was presented.

ABSTRACT

The Franciscan Complex comprises the largest bedrock unit of the California Coast Ranges, so it impacts many engineered projects in coastal California, but the ignorance of basics of Franciscan geology plagues many projects. The stratigraphic name given by engineers or geologists in many projects, 'Franciscan Formation', a term abandoned by research geologists nearly four decades ago, illustrates some of this misunderstanding. The "layercake" stratigraphy inferred by 'Formation' differs markedly from reality, for faults separate most Franciscan rock units, and chaotic block-in-matrix units, known as melanges, are widespread. While the layercake view results from a pre-plate tectonic understanding, others mistakenly believe that the entire Franciscan is a melange. Research shows that the Franciscan consists of both intact units ("coherent") and melanges. Franciscan geology knowledge continues to progress, but researchers have virtually no understanding of the engineering significance of these rocks, while the engineering community remains largely unaware of the current research knowledge. There is substantial potential benefit from better communication between the engineering and geologic research communities. Education is also critical, for as undergraduate field studies continue to diminish in geology programs, most existing field classes tend to focus on areas of layercake stratigraphy and excellent exposure, providing dismally poor preparation for professionals who might work in areas such as coastal California.

INTRODUCTION Melange units, composed of a weak matrix melange is present or not [12]. It is here where a huge enclosing stronger blocks, make up a significant gap exists between the state-of-knowledge of the proportion of the world's bedrock in areas of former research geologic community, in which melanges subduction zones [1,2]. Recognition of melanges is have been common knowledge for at least 40 years, important for engineering geology and related and the applied geoscience community (the term I will engineering because their physical properties differ use to define engineering and environmental geology markedly from simple bedded or isotropic rock and related engineering fields) where it is still [3,4,5,6). Because of their discontinuous, rather than common to think of all geologic units as either being continuous structure and/or stratigraphy, particular homogeneous and isotropic (as an idealized granite) or care must be taken in terms of interpretation of field a layered sedimentary stack (herein termed "layercake data, be it the interpolation between or extrapolation stratigraphy" as one might expect in the Grand from boring or surface geologic information, or the Canyon). In this paper I will discuss misconceptions scaling between outcrop and boring levels of about melanges (or failure to recognize them) within observations to the scale of engineering interest the context of the evolution of geologic thought on the [6,7,8,9,10,11]. These studies [3-11] provide a good Franciscan Complex of California, one of the world's framework for engineering characterization of most famous localities for melanges and also a melanges, but a non-trivial step is to recognize that a geologic unit upon which a large population lives and builds many engineered structures. I will then offer about the state of field studies in geologic curricula some tips for identifying melanges in the field and and its impact on melange characterization, with some show how melanges have a number of characteristics recommendation for better communication between that can be recorded that may be useful in engineering the academic ("research") and applied geosciences. characterization. I will conclude with some concerns

the concept became widely accepted throughout the geologic community [2], or more accurately, the MELANGES AND THE FRANCISCAN research geologic community. Front and center in COMPLEX: THE OUTCROPS DID NOT Hsü's work was the Franciscan Complex of coastal CHANGE, BUT OUR VIEWS DID California (Fig. 1). Today we know the Franciscan Complex as the world's type subduction complex, because the pioneers of the Plate Tectonic Revolution Melanges were first recognized by Greenly [13] first used the Franciscan to illustrate the connection who coined the term based on exposures in Wales, between subduction and high-pressure/low- although chaotic, non-stratiform (ie., not layercake) temperature metamorphism [15], and the tectonic rocks had been noted decades earlier by geologists shuffling of oceanic and continentally derived working in the Alpine region of mainland Europe [2]. materials [16]. The melange concept, while not In spite of Greenly's work, and research of others originally envisaged in a subduction context by (whose observations and interpretations were parallel Greenly or Hsü, now became part of the plate tectonic and independent of those of Greenly) conducted in paradigm [16], so the Franciscan became triply other parts of the world in the decades following recognized for three critical components in Greenly's work, nearly five decades elapsed before development plate tectonic theory. Hsü [14] published his historic paper on melanges and

We now view the Franciscan as a product of Investigations in the late 60's and early 70's showed subduction that began about 165 million years ago and that the Franciscan consists of both melange terranes continued for over 140 million years [17]. Faults and so-called "coherent" (non melange) terranes and bound major units, and the age of incorporation into that the proportion of melange to coherent terranes the Franciscan youngs structurally downward (across varied spatially within the Franciscan [21]. Workers major faults), as a consequence of progressive such as Blake et al.[22,23] and Wakabayashi [17,18] offscraping of material from the downgoing plate extended the studies of Maxwell [21] and his students, [17,18]. This contrasts markedly with a so that the dual melange and coherent components sedimentary/stratigraphic view held prior to the were mapped throughout the Franciscan. acceptance of plate tectonic theory and the melange As pointed by Wakabayashi and Medley [12], concept. Although the spatial distribution of ages outcrops have not changed over the past 40 years, but made little sense from a stratigraphic standpoint (ie a prevailing theories have, so that the same set of sedimentary sequence that youngs progressively outcrops have been or are interpreted three ways: (1) upward), researchers resolutely tried to force fit the stratigraphic (pre-1968), (2) all melange (the next Franciscan into a layercake stratigraphic model [19]. step, post-1968), (3) mixed coherent and melange The term "Franciscan Formation", a formal (Fig.2). This points out an often unappreciated aspect stratigraphic term that most researchers discarded in of field geology: Field studies do not represent a 1968, represents the pre-melange/pre-plate tectonic purely objective, non-analytical "stamp collecting" view of the Franciscan Complex as a sedimentary exercise as wrongly believed (and even ridiculed) by sequence. many field-phobic academicians. The latter see little Following the recognition of melanges, some value in geologic mapping, because the geologic maps workers viewed the entire Franciscan as a melange exist for every part of the Earth (at least above sea [20], and indeed the informal name of the entire level). Rather, field geology has a highly interpretive complex as the "Franciscan melange" still finds aspect to it, with our drawing of contacts tied to common usage. Such a view of the Franciscan can prevailing geologic thinking. Why? We never have a have as disastrous consequences in engineering fully 3-dimensional set of field data without gaps. investigations as the mistaken view of layercake Interpolation and extrapolation comprise a critical part stratigraphy, because some default the Franciscan to of our field mapping and the prevailing theories an intractably chaotic unit from which one cannot dictate how we "connect the dots". extract any meaningful spatial field data.

This summary of the evolution of thought on (consistent with the "Franciscan Formation" view), melange and the Franciscan Complex illustrates how continue occur at alarming rates. Such blunders much prevailing thoughts and knowledge have commonly have costly consequences. Although this changed over the past four decades in the research summary criticizes the ignorance shown by parts of community. Sadly, a large fraction of the applied the applied geoscience community, I hold research geoscience community still holds pre-1968 views of geologists equally culpable. Academic geologists the Franciscan. Fig.3 represents a cross section from a typically have little or no awareness of "real world" project somewhere in the Franciscan done within the (ie engineering or environmental) issues connected last 10 years (location not named for confidentiality, with the rocks they so enjoy studying. For example, in and the cross section has been redrafted from the spite of the interest that studies in melanges have original) wherein a layercake interpretation is given to generated in the academic community over the last geology that is almost certainly melange in part. few decades, I have yet to encounter a fellow Huge errors, such as mistaking the melanges for a academic who has even a remote understanding of the surficial unit ("soil with boulders"; imagine how far engineering issues associated with melanges. In a later deeply one might have to excavate to find "bedrock"), section, I will return to this theme to discuss role of or the usual layercake stratigraphic interpretation undergraduate geologic education in improving the awareness of melanges in the applied community.

MELANGES: DIFFERENCES BETWEEN GEOLOGIC ISSUES AND ENGINEERING ISSUES

The applied geoscience community commonly resistant quartz-mica schist [24,25] (Fig.1). In the has difficulty recognizing melange terranes, but Franciscan Complex, the vast majority of melanges academic geologists do not realize that not all are "engineering" melanges with significant block- melanges they map comprise melanges from the matrix strength contrast. A reason for this is the low standpoint of the engineer. As a result, applied metamorphic temperature (less than 300°C in most geoscientists reading the research literature may be led cases) that the matrix of Franciscan melanges were astray. An "engineering melange" has notably subjected during their sojourn at depth within the stronger blocks than matrix, whereas some "geologic" Earth [15]. melanges lose this contrast as a result of In the research community considerable debate metamorphism that results in the recrystalization of still rages over the comparative importance of tectonic the matrix to form a hard and strong schist. In (shearing, faulting) and sedimentary (submarine metamorphic grade, this commonly corresponds to landsliding) processes in the generation of melange. greenschist facies or higher (rock experienced These research issues have comparatively little metamorphic temperatures exceeding 300°C). For bearing on engineering issues, because origin of a example, large tracts of melange in the central and melange carries far less importance than the physical southern Sierra Nevada are not "engineering" melange and spatial properties of the unit. because former shale matrix that has recrystalized to a

WHERE IS THE MELANGE? USEFUL CRITERIA TO NOTE IN THE FIELD

The first challenge an applied geologist faces in that commonly represent the only exposed bedrock the Franciscan or similar terrane faces is whether or (the plums). This looks like classic landslide not the study area includes melange. After reviewing topography as well it should, because melanges are existing geologic mapping, geomorphic expression is notoriously prone to slope failure. Recognizing intact first level of analysis. Because of its lack of erosional melange from slope-mobilized melange presents a resistance, melange matrix seldom forms good challenge even to the most hardened outcrops. Melanges form lumpy topography "Franciscanologist", but at minimum a geologist commonly known as "plum pudding topography" should recognize that they have encountered either (Fig. 4) with irregular slopes and big resistant blocks melange or landslides within one.

Geologists should avoid the Franciscan melange this looks superficially similar to melange topography default mode, wherein they record "melange" simply the slopes attained on these rounded hills are steeper because geologic maps show their bedrock unit as than one typically encounters in melange terranes. Franciscan. Within coherent (non melange) Indeed, in poorly exposed areas the slope angle is a Franciscan, there exist erosional resistance contrast good indicator of coherent (or large block) versus that may falsely lead a geologist (especially with a melange. Another clue to local coherence in the preconceived all-Franciscan-as-melange view) to map Franciscan is the presence of strike ridges of various melange. In Fig. 5 the resistant outcrops consist of types of bedrock that one can walk out for some chert whereas the rocks not showing obvious outcrops distance (Fig. 6). Such a long aspect ratio of a block consist of coherent basalt and sandstone. Although is unlikely in a melange.

Geomorphology provides the best clue to the bedrock that is either coherent, or a large block in presence of melange, in part because the slope has a melange versus the plum pudding topography just direct relationship to the strength of the material. across the stream that has developed on melange However, melange zones that are fairly thin (in bedrock. Some of the steep slope in Fig.7 is laced with structural thickness) versus the slope they cross may very thin melange zones, although they make up a not result in diagnostic geomorphic chacteristics. Fig. small fraction of the slope compared to hard bedrock. 7 shows an example of a steep slope formed on

Where classic melange geomorphology is lacking, (probably does not crop out well, if at all) and some rock type distributions may indicate the sandstone (will commonly form decent outcrops), but presence of melange in the Franciscan or similar finds a few outcrops of rocks such as chert and basalt, subduction-related rock units. For example, sandstone a melange may be suspected owing to the fact that the (sometimes colloquially called graywacke) and shale chert and basalt are not expected in a trench-fill comprise the largest proportion of Franciscan bedrock. depositional sequence (of sandstone and shale). This If one maps in an area that comprises largely shale does not mean that if one encounters only sandstone outcrops that one is not in a melange, for "broken" "saturation" mapping in which every last outcrop is formation, derived from a bedded sandstone-shale recorded on the geologic map, in contrast to the sequence by deformation, may exhibit full block-in- traditionally taught "contact" mapping in which a few matrix character without exotic blocks (ie rock types points on contacts may suffice to characterize the that did not belong in an original sedimentary spatial distribution of units. Much has been written sequence). Some rock types almost certainly signal about the determination of appropriate block sizes for melange in an otherwise sandstone-shale terrane. If a given scale of interest and various techniques to one has outcrops of sandstone, but sees a few small determine block sizes for engineering purposes (the scraps of serpentinite it certainly points to the reader is referred to the references above), so I will presence of melange. The makers of the original cross not review them here; the remainder of this section section modified for Fig. 3 should have been aware of will outline other field parameters that are not as well this, for the presence of serpentinite there nearly publicized, but are still potentially important. guarantees the presence of melange. Blocks of higher Such field parameters include the external metamorphic grade than the surrounding rocks can contacts of a melange, provided they occur within the also be a good indicator melange presence, such as the study area. Although internally chaotic, the external occurrence of blueschist, amphibolite, or eclogite contacts of a melange present nothing extraordinary in blocks where surrounded by outcrops of sandstone terms of geologic mapping. These external contacts that is prehnite-pumpellyite facies. The latter criteria may be "contact" mapped as one would a normal will probably not prove useful for the vast majority of faulted or stratigraphic contact, although a gradational applied geologists who lack sufficient training to contact (many melanges have them) may represent an readily tell metamorphic grade contrasts but it additional challenge. They are no more likely to underscores the potential value of metamorphic change dramatically in strike and dip along their petrology, commonly thought of as purely an length as any other sort of contact. academic exercise by applied geologists. The preferred orientation (foliation) of a melange Note that serpentinite by itself is not necessarily matrix is commonly mappable and may contribute an indicator of melange as believed by many. Some asymmetry to physical properties [28], so the serpentinite bodies consist of primarily massive rock, geologist should record these orientations, if possible. whereas others are both geologic and engineering Blocks commonly exhibit a preferred flattening plane melanges because they consist of exotic blocks (non parallel to this orientation and they may exhibit a serpentinite) in a sheared serpentinite matrix, and collective lineation as well. Such block preferred some are engineering but not geologic melanges orientation may also influence the physical anisotropy because they may consist entirely of serpentinite (not of the melange [27]. geologic melange) but have an internal fabric of shear The lithologies of melange blocks should be serpentinite enclosing stronger blocks of massive recorded given that they can vary dramatically from serpentinite [12,26]. melange to melange and locally within a given melange. Different block types may have vastly ONCE A MELANGE IS IDENTIFIED: USEFUL different strengths and hardnesses that may have a FIELD DATA TO OBTAIN significant impact on how easily they may be excavated. In addition, although some melanges Because empirical and experimental studies have appear to exhibit a more or less random mixing of determined a positive correlation between block blocks, others appear to show systematic spatial proportions and melange strength, block proportions distribution of block lithologies and block proportions, are a particularly important parameter to record and this may be particularly the case along gradational [3,4,5,6]. Challenges come in selecting the contacts of melanges [12]. For example, a unit appropriate scaling for determination of what bordering a melange may be coherent sandstone and constitutes a block versus matrix, as well as shale that grades increasingly disrupted toward the understanding the uncertainties inherent in melange, to a broken formation (block-in-matrix interpolated between or extrapolating from surface or fabric with only sandstone blocks), into the melange boring data [6,7,8,9,10,11,27]. For field mapping the that will exhibit block-in-matrix fabric with exotic non-predictive nature of internal contacts demands blocks. The field criteria discussed in this section are diagrammatically shown in Fig. 8.

IMPROVING FIELD INSTRUCTION IN "saturation" mapping techniques necessary to surface UNDERGRADUATE GEOLOGY PROGRAMS map in melange terranes, for it was far easier (for the AS ONE PART IN RAISING MELANGE students and instructors) for field instruction to be AWARENESS conducted in areas of layercake stratigraphy and excellent exposure (this generally means somewhere in the desert in this part of the world). Unfortunately, The value of field work has not diminished at all the professional geologist in a place such as California in professional practice, yet the amount of field work commonly encounters poor exposure as well as the required of students in undergraduate geology discontinuous structure and stratigraphy. While a programs has decreased alarmingly over the three student made to map in melange (even for a short decades that I have observed various academic time) with poor exposure will be able to map programs. The cutback in field studies has many layercake stratigraphy with superb exposure with ease, origins, including the comparatively high cost of field the converse is not true. As much of a problem as this instruction and perceived liability risk, as well as the was a few decades ago, the problem has worsened as diminished role of detailed field mapping in many field time shrank in undergraduate programs. At sectors of the academic geosciences. Even three Fresno State, I have tried to overcome some of the decades ago, when students received far more field above problems by giving my advanced field methods instruction than today, few students were taught the students a mapping exercise that includes mapping academic community. In turn, academicians have a melange (Fig. 9), but I fear this may be unique among responsibility to learn more about what professional all the field mapping courses in this country. For the geologists do and what skills are important in the future I believe that there needs to be more dialogue world of "real geology", so that they may better between the applied geosciences and academia when prepare their students, the vast majority of whom will it comes to the issues faced by professionals having to become applied geoscientists rather than characterize melange. Professionals have a academicians. Finally, both academic and research responsibility to update themselves with current communities need to establish better communication understanding in geology rather than holding onto in order to bridge large gaps in understanding and views abandoned more than 40 years ago by the knowledge.

REFERENCES and the evolution of geologic thought. Boulder, Colorado, Geological Society of America Special 1. Raymond, L.A., 1984, Classification of melanges. in: Paper 373, p. 385-445 Raymond L.A., ed.: Melanges: Their nature, origin and 3. Lindquist, E.S., 1994, The strength and deformation significance. Boulder, Colorado, Geological Society of properties of melange. Ph.D. dissertation, Dept. of America Special Publication 198, p.7-20. Civil Engineering, University of California, Berkeley, 2. Sengor, A.M.C., 2004, The repeated discovery of California1994. melanges and its implications for the possibility nd the 4. Lindquist, E.S., 1994, The mechanical properties of a role of objective evidence in the scientific enterprise: in physical model melange. Proceedings of the 7th Dilek, Y., and Newcomb, S., eds. Ophiolite concept Congress of the International Association of Engineering Geologists, Lisbon, Balkema, Rotterdam, Franciscan Complex, California: Journal of Geology, v. p.819-826. 100, p. 19-40 5. Lindquist, E.S.; Goodman, R.E., 1994, The strength and 18. Wakabayashi, J., 1999, Subduction and the rock record: deformation properties of a physical model melange: Concepts developed in the Franciscan Complex, in: Nelson, P.P. and Laubach, S.E., (eds): Proceedings California: in: Sloan, D., Moores, E.M.; Stout, D., eds.: of the 1st North American Rock Mechanics Conference Classic Cordilleran Concepts: A View From California. (NARMS), Austin, Texas. Balkema, Rotterdam, p. Boulder, Colorado, Geological Society of America 843-850. Special Paper 338, p. 123-133, 1999 6. Medley, E.W, 1994, The engineering characterization of 19. Taliaferro, N.L., 1943, Franciscan-Knoxville problem: melanges and similar block-in-matrix (bimrocks): American Association of Petroleum Geologists Ph.D. dissertation, Dept. of Civil Engineering, Bulletin, v. 27 , p. 109-219. University of California, Berkeley, California. 20. Cloos, M., 1984, Flow melanges and the structural 7. Medley, E.W., 1994, Using stereologic methods to evolution of accretionary wedges. in Raymond, L.A., estimate the volumetric block proportion in melanges ed., Melanges and their significance, Boulder, and similar block-in-matrix rocks (bimrocks): Colorado, Geological Society of America Special Proceedings of the 7th Congress of the International Paper 198, p. 71-80. Association of Engineering Geologists, Lisbon. 21. Maxwell, J.C., 1974, Anatomy of an orogen: Geological Balkema, Rotterdam, p.1031-1040. Society of America Bulletin, v. 85, p. 1195-1204. 8. Medley, E.W., 1997, Uncertainty in estimates of block 22. Blake, M.C. Jr.; Howell, D. G.; Jayko, A. S., 1984, volumetric proportion in melange bimrocks: in: Tectonostratigraphic terranes of the San Francisco Bay Marinos, P.G.; Kpukis, G.; Tsiambous, G.;Stournaras, Region: in: Blake; M. C., Jr. ed., Franciscan Geology G., eds: Proc Int. Symp. On Engineering Geology and of Northern California: Pacific Section Society of the Environment. Balkema,Rotterdam, p. 267-272. Economic Paleontologists and Mineralogists, v. 43, p. 9. Medley, E.W., 2001, Orderly characterization of chaotic 5-22. Franciscan melanges: Felsbau, v. 19, p. 20-33. 23. Blake, M.C., Jr.; Howell, D. G.; Jones, D. L., 1982 10. Medley, E.W., 2002, Estimating block size distributions Preliminary tectonostratigraphic terrane map of of melanges and similar block-in-matrix rocks California: United States Geological Survey Open File (bimrocks): Proceedings of the 5th North American Report 82-593, 1982. Rock Mechanics Symposium, Toronto, 24. Day, H.W.; Schiffman, P.: Moores, E.M., 1988, 11. Medley, E.W.; Lindquist, E.S., 1995, The engineering Metamorphism and tectonics of the northern Sierra significance of the scale-independence of some Nevada. In Ernst, W.G., ed., Metamorphism and Franciscan melanges in California, USA: in: Daemen, crustal evolution of the western United States. Rubey J.K.; Schultz, R.A., eds: Proceedings of the 25th US Volume VII, Prentice-Hall, Englewood Cliffs, New Rock Mechanics Symposium, Balkema, Rotterdam, p. Jersey, p. 737-763. 907-914. 25. Sharp, W.D.,1988, Pre-Cretaceous crustal evolution in 12. Wakabayashi, J., and Medley, E.W., 2004, Geological the Sierra Nevada region, California: in: Ernst, W.G., characterization of melanges for practitioners: Felsbau ed., Metamorphism and crustal evolution of the v. 22, no. 5, p. 10-18. western United States. Rubey Volume VII, Prentice- 13. Greenly, E., 1919, The geology of Anglesey: Hall, Englewood Cliffs, New Jersey, p. 823-865. London,Geological Survey of Great Britain, 980 p 26. Wakabayashi, J., 2004, Contrasting settings of 14. Hsü, K.J., 1968, The principles of melanges and their serpentinite bodies, San Francisco Bay area, California: bearing on the Franciscan-Knoxville paradox: Derivation from the subducting plate vs. mantle Geological Society of America Bulletin, v. 79, p. 1063- hanging wall: International Geology Review, v. 46, p. 1074. 1103-1118. 15. Ernst, W.G., 1970, Tectonic contact between the 27. Haneberg, W.C., 2004, Simulation of 3-d block Franciscan melange and the Great Valley Sequence, populations to characterize outcrop sampling bias in crustal expression of a Late Mesozoic Benioff Zone: block-in-matrix rocks (bimrocks): Feldsbau, v. 22 Journal of Geophysical Research, v. 75, p. 886-902 (September). 16. Hamilton, W. B., 1969, Mesozoic California and 28. Medley, E.W.; Sanz. P.R., 2004, Characterization of underflow of the Pacific mantle: Geological Society of Bimrocks (Rock/Soil Mixtures) With Application to America Bulletin, v. 80, p. 2409-2430. Slope Stability Problems: in: Schubert, W. ed: 17. Wakabayashi, J., 1992, Nappes, tectonics of oblique Proceedings of the. Eurock 2004 and 53rd plate convergence, and metamorphic evolution related Geomechanics Colloquium, Salzburg, Austria. to 140 million years of continuous subduction,