Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
COCORP and Fluids in the Crust 8 INTRODUCTION During a little more than a decade, COCORP (Consor- tium for Continental Reflection Profiling) has collected deep seismic reflection data in the United States for pro- files totaling over 8000 km in length (Figure 8.1~. These data, though trivial in quantity compared to the huge vol- ume of industrial seismic data for the sedimentary basins, 128 JACK E. OLIVER Cornell University ABSTRACT Using the seismic reflection profiling technique, COCORP has probed the conti- nental basement along lines traversing a wide variety of major geological fea- tures. The results are correspondingly diverse, and they are informative on various aspects of crustal geology, including the geology of fluids, the particular aspect emphasized in this chapter. There are at least three ways in which COCORP data may relate to the understanding of fluids in the basement. First, some anomalously strong reflectors, typically near-horizontal and at depths of about 20 km, may correspond to pockets of magma or other fluids trapped within the basement. Second, as suggested by others, some or even all basement reflectors may correspond to zones of fluid-fi~led porous rocks. Third, new insights into structures suggest that tectonic models include inferences about the role of fluids in the crust. For example, fluids expelled tectonically from sedi- ments buried in erogenic belts may play a role in migration of hydrocarbons, transport of minerals, diagenesis, crustal rheology, chemical Demagnetization, and a variety of other phenomena. constitute the largest and most comprehensive set of land- based seismic reflection data on the deep crust of the continents anywhere. BIRPS (British Institutions Reflec- tion Profiling Syndicate) has collected a comparable quan- tity of marine data on the deep crust beneath the continen- tal shelf surrounding Great Britain, and at the time of this writing some 20 countries, including the United States and the United Kingdom, have surveyed a total of 20,000 to
COCORP AND FLUIDS IN THE CRUST I NW _ / CORDILLERA - / _ ' OREGON 11 ( N. CA. ~ _ NEVADA ,~LARAMIE RANGE UT H ; KANSAS HOA COALINGA I \ ~ D.VAL. PAl~ttFIELD id, AVE '- - ~AZ- NH . IND RIVER V -\___ ~ INNESOTA' o PA ~ ENGLAND> K~.lFt :~1~: ~: OKLAHOMA 80CORRo HARDEMAN (ARKANSAS SOUTHERN it, APPALA~IANSr ~ ~RLE5T~ GEORGIA ~ t of · Current Operation -Completed Prof lies 25,000 km of seismic reflection profiles of the deep conti- nental crust throughout the world. This total is exclusive of any proprietary industrial data on the deep crust. This quantity of data corresponds to exploration of only a small fraction of the total volume of the continents. Nevertheless, these data represent early exploration of a new frontier, and as such they have already provided sur- prising and important information on many aspects of continental crustal geology. This chapter, however, passes over most of these aspects and, because of the nature of this volume, focuses solely on just one aspect of crustal geology fluids in the crust. Furthermore, the chapter is based almost exclusively on COCORP data, all taken in the 48 contiguous states. For a review of COCORP stud- ies in general, see Brown et al. (1986~. COCORP observations relate to the study of fluids in the crust in at least three different ways and even more if various indirect lines of reasoning not emphasized here are pursued. First, the observations occasionally reveal, at mid- crustal depths, exceptionally strong, near-horizontal re- flectors that are probably associated with magmas and/or other fluids. This information is some of the very best evidence in existence on spatial locations of fluids in deep basement rocks at the present time. Second, the COCORP observations reveal a variety of weaker reflectors through- out the continental basement, some, or even all according to one hypothesis, of which may be related to fluids in some manner. Third, COCORP data provide novel infor- mation on the structure and tectonics of the crust and thereby stimulate thinking about crustal tectonic models that have implications about fluid behavior in the crust. This chapter discusses these topics in order and with some 129 FIGURE 8.1 Map showing deep seismic profiles surveyed by COCORP from the inception of the project through December 1986. emphasis on the third point because of the nature of this volume. STRONG REFLECTORS IN THE MID-CRUST In 1975 an early COCORP survey sensed a strong re- flector at a depth of about 20 km in the crust beneath the Rio Grande Rift north of Socorro, New Mexico. This location is just where Allan Sanford (Sanford et al., 1973, 1977), basing his ideas on earthquake and other geophysi- cal and geological data, had earlier postulated that a magma body exists. The COCORP reflection survey supported Sanford's hypothesis, delineated certain features of the body, and, if Sanford's well-supported hypothesis is in- deed correct, provided a type example of reflection data for a magma body in the mid-crust. Later, while profiling in Death Valley, COCORP de- tected a strong reflection that in many ways resembles that type example from beneath the Rio Grande Rift. The Death Valley reflector is at about the same depth, is nearly horizontal, and is in a similar extensional tectonic environ- ment. An obvious possible, though not unrefutable, inter- pretation is that another magma body exists beneath that part of Death Valley (de Voogd et al., 1986~. In fact, the data also reveal a dipping reflector that rises from the vicinity of the proposed magma body and extends to near the surface at a place where recent volcanic activity has occurred. It has thus been proposed (de Voogd et al., 1986; Serpa et al., 1987) that this dipping reflector corre- sponds to a fault zone that acted as a conduit for rising magma. Whether the fault zone continues to hold magma, in its lower if not upper reaches, is not resolved.
130 Since the Death Valley survey, COCORP has found at least two other comparable reflectors, one in Nevada and one in Arizona. Both locations are in the Basin and Range Province. These two reflectors are somewhat less promi- nent features, but they are of the same general character and at about the same depth as the Rio Grande Rift and Death Valley reflectors. They may also correspond to magma at some stage of cooling. These strong mid-crustal reflectors, coupled with addi- tional observations by COCORP and by others on various aspects of the buried continental crust, are prompting new speculation about the formation and evolution of the deep crust and its role in the evolution of modern surface geol- ogy. The subject of the structure, composition, evolution, and tectonics of the deep continental crust and uppermost mantle is a particularly important and exciting one at pres- ent. The speculation takes many forms and produces diverse hypothetical models. Figure 8.2 shows one example that corresponds to an old idea that magmas generated in the mantle intrude the mid-crust near the top of a zone of earlier intrusions. Subsidence and phase change continu- ally convert the rocks at the base of the crust to denser and higher-velocity phases that become uppermost mantle or "anomalous" mantle. In a similar model (not shown) intrusion occurs at the base of the crust and successive intrusions build up from below, adding to anomalous mantle. Other ideas to explain the nature of the deep crust without major intrusion are also under consideration; the two above are mentioned here because of their bearing on fluid penetration of the crust and their partial base in COCORP data. The examples cited above of strong mid-crustal reflec- tors all occur in the Basin and Range Province, a region characterized by extension in the most recent tectonic ~ , ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ ' · ' · ' · ' · ' · ' · ' · ' ~ /\'l\~/\~/\~/~/~/~/~'l\'l\'l\~/\'l\~/\~/\~/\~/\~/\~/\~/~/\~/ ,', i',, A,, ;',', <', ,U\P ,P\E OR\, CRUST,',-`',', `',',n',',-\' _ _ i_ ~ ~ _~__~` ~ __ _ x-~` ~-~-~-~\-~\-~-~ V~\ - ~ - ~\ - ~\ - _ \ _ ~ _ \ _ \ _ \ _ \ _ \ _ \ _ \ _ \ \ _ \ _ \ \ _ \ _ \ _ \ _ \ \ _ \ _ \ ' \ ~ _ _~ FIGI~E 8.2 An example of one hypothesis proposed to explain the structure of the lower crust in some areas. Magma from mantle intrudes crust near top of zone of previous intrusion. Deeper portion of zone of intrusions is converted to mantle-like material (anomalous mantle) at base of crust. This hypothesis is rid --a designed to explain structure and dynamics of lower crust and uppermost mantle. but one example of many JACK E. OLIVER episode. COCORP has acquired data elsewhere on still another strong mid-crustal reflector, this one of somewhat different seismic character and somewhat different mod- ern tectonic character (Brown, 1987~. The reflector lies beneath the coastal plain of Georgia (i.e., a province dis- tinctly different tectonically from the Basin and Range Province). This reflector is called the Surrency bright spot "Surrency" because that is the name of a small town nearby and "bright spot" because the reflection resembles in many ways a bright spot of the type familiar to the petroleum industry. Oil industry bright spots typically occur in shallow sedimentary rocks. They are commonly drilled and commonly produce hydrocarbons. The typical bright spot is a consequence of occurrence of fluids that affect the acoustic impedance of porous rocks and of grav- ity which tends to make boundaries between fluids flat and horizontal. In spite of the striking similarity, the Surrency bright spot differs dramatically from a typical oil industry bright spot in that it occurs at a reflection time of about 6 s (i.e., a depth of about 15 to 18 km) and hence must be well within the basement that lies beneath the sediments of the coastal plain of Georgia. The Surrency bright spot, which is described in more detail elsewhere (Brown, 1987), is located in the Paleozoic suture zone between former African and North American crust. How to interpret this unusual and very robust obser- vation is not completely clear. Because of lack of any evidence for young volcanism or tectonic movements within the coastal plain, it seems unlikely that there is magma in the crust at this location. A lithologic boundary may be considered as an explanation, but a strikingly flat and horizontal boundary between two lithologies of such great impedance contrast as would be necessary to produce the observed reflection also seems unlikely in this highly deformed zone. Perhaps the most reasonable suggestion is that fluids caught up and trapped in the collision process, or trapped later in the deformed rocks of the collision zone, account for the observation. The fluids might be water, He, CO2, hydrocarbons, or something else-just what is uncertain at present. But it seems obvious that information on such prominent, almost certainly fluid- related, features of the crust must be important to anyone attempting to understand the geology of fluids in the base- ment. Unfortunately, with only a small fraction of the total crustal volume so far explored, it is not possible to say whether this occurrence is unique or one of tens or hundreds or thousands. Nor is it known whether, if there are many such features, all are in suture zones or some are distributed through other parts of the crust as well. Detec- tion of such features is so easy and straightforward techni- cally, however, that a great opportunity for further study is readily apparent from this first striking observation.
COCORP AND FLUIDS IN THE CRUST THE REFLECTIVE CRUST In addition to the rare, near-horizontal, strong reflectors of the mid-crust just discussed, in most areas the entire continental crust is generally reflective to some degree. Flat and curved reflectors, dipping reflectors, laminated reflectors and diffractors are common in the basement, although generally they are weaker than those in sedimen- tary rocks. Reflections within Phanerozoic, and in some cases Proterozoic, sedimentary basins are studied inten- sively by the petroleum industry. Such reflectors may correspond to lithologic changes, time-stratigraphic bounda- ries, faults, fluids, etc., and, because of the great industrial effort, understanding of them is relatively advanced. For reflectors within the crystalline basement, most not drilled and most so deep as to be undrillable, identification and interpretation are normally less certain. Sometimes a particular reflector can be identified by tracing it to the surface. Sometimes the spatial configuration of a reflector gives it a place in an interpretation of geologic structure that in turn gives confidence to speculation on the nature of the reflector. Commonly, however, it is not known for certain just what physical properties produce a particular reflector at a particular place. Furthermore, it is likely that a combination of factors is involved. One possible explanation is that some, or in the extreme even all, such basement reflectors correspond to fluid- filled fractures in the crust. From drilling (Koslovski, 1984; Clarke et al., 1986) it is known that, at least for one location, fluid-filled fractures prevail to depths of at least 11 km. That such fractures are reflective was demon- strated by Mair and Green (1981) and Green and Mair (1983) for a case of very shallow reflectors in a granitic intrusion. Matthews and Cheadle (1986) suggested that all reflectors throughout the basement are associated with fluids, an extreme but nevertheless provocative sugges- tion. Among other things, this suggestion implies that the reflection Moho, which is commonly taken as the base of the reflective zone, marks the base of fluid-filled fractures, and hence fluids, in the crust. An explanation in terms of a mobile deep fluid boundary is thereby provided for the vertical mobility of the Moho that is now well demon- strated by seismic reflection studies. This hypothesis is not without its difficulties. For example, mere draining of fluids from a part of the lower crust would likely not elevate velocities to sub-Moho levels, and hence the hy- pothesis would not account for the observed coincidence in reflection and refraction Mohos at most places. Clearly the reflective crust is not yet fully understood and remains in a speculative stage. It would be risky, for example, to base hydrologic or geochemical studies of the crust on the concept that all basement reflectors necessar 131 fly indicate fluid-filled fractures of the crust, but it would also be risky to ignore the possibility that many of them do. A SYNTHESIS OF INFORMATION ON CRUSTAL STRUCTURE, TECTONICS, AND FLUIDS IN THE CRUST The COCORP data provide information on the struc- ture and evolution of the continental crust. Many other sources contribute important, different, and highly diverse information on this subject. All such information must ultimately be fit into a common story. What follows is an attempt to synthesize observations from a wide variety of disciplines in earth science, all related in some way to the story of fluids in the crust. The role of COCORP data in this synthesis arises primarily from its contribution to the tectonic model and not so much from direct detection of fluids in the crust as described earlier. The reader should be aware that (a) much of what follows is speculative. The hypothesis to be presented may be incorrect; some parts are surely controversial at present. Although a substantial amount of evidence in support of the hypothesis is presented, there remains much opportunity for further testing. Should the hypothesis be more or less correct, however, it will be of considerable importance because it links many geological observations of the continents to the great global geodynamical pro- cesses through the mechanism of plate tectonics. (b) Parts of what follow will be familiar to some, for the hypothesis and synthesis incorporate some concepts that have been proposed and discussed, though not necessarily agreed upon, previously. This chapter should not be mistaken as a claim for originality with regard to those concepts. Instead the chapter is an attempt at synthesis on an unusually broad scale. Appropriate references to concepts proposed earlier are provided insofar as possible. (c) The author is not a specialist in most of the wide variety of disciplines from which observations are drawn and would appreciate notification from specialists of any case in which data from a specialty have been misinterpreted, particularly if that misinterpretation affects the testing of the hypothesis. The synthesis of this chapter uses as a framework a hy- pothesis that may be described in general terms as follows (Oliver, 1986~. At the surface of the Earth, continents (and smaller land masses) drift about within the large sea that occupies most of the surface. The continents have limited freeboard, and their margins are generally awash in the sea. The conti- nents are porous and permeable, at least in their upper parts, and are for the most part saturated with fluids. When two landmasses converge and collide, the process is an
132 asymmetric one. One landmass is partially subducted beneath the other. Put another way, the margin of one landmass is commonly buried beneath the accretionary wedge on the leading edge of the other. According to the hypothesis, fluids from the sediments of the buried conti- nental margin are expelled and travel toward the interior of the partially subducted continent carrying heat, miner- als, and organic compounds (Figure 8.3~. These fluids may leach or deposit materials as they travel; hence, they redistribute materials and leave a lasting record of their passage. As an illustration of the hypothesis, the Appala- chian orogen can serve as the type example. The Appalachian orogen is a consequence of conver- gence and closing of the proto-Atlantic Ocean during Paleozoic time. The process included several progenies and culminated in the collision of the North American and African continents. Subduction during this closing was generally to the east, so a large accretionary wedge or thrust sheet was forced onto and over the sediments of the North American margin. Some sediments were buried, some were carried along with the thrust sheet, and some were reworked to become younger strata. The load of the thrust sheet caused a large depression near the sheet and a more distant forebulge, both changing dynamically as the process evolved. The depression, or foreland basin, filled with sediments from the thrust sheet. Some of these sedi- ments were also buried by the advancing sheet. The sedi- ments involved were generally porous and full of pore fluids, and they included hydrous minerals. Perhaps one- third to one-half of the total volume was made up of water in the pores and in the minerals. As the sediments were buried, water was expelled, some from shallow depths at modest temperatures and some from greater depths at higher temperatures. According to this hypothesis, some of these fluids, or brines, were expelled into rocks of adjacent parts of the continent. Some of the fluids may be an important part of phe- nomena of the metamorphic core of the orogen, perhaps producing veins, dikes, and intrusions and facilitating thrust- ing, but they are not the phenomena of interest here. This chapter is about fluids that are expelled into sediments of FIGURE 8.3 Block diagram of orogen as accretionary wedge or thrust sheet over- ndes preexisting continental margin. Sub- duction is to the east. Heavy arrows sche- matically illustrate flow of tectonic brines expelled from buried sediments. Gas and anthracite deposits are closer to orogen than oil and bituminous coal, respectively. Continental crust is -35 km thick; hori- zontal dimension of diagram is ~500 km (from Oliver, 1986~. JACK E. OLIVER the platform and foreland basin, perhaps eventually reach- ing the more distant continental interior. These fluids carry heat, minerals, and organic compounds, some from the marginal sediments and perhaps some leached from sediments along the way. Expelled by the tectonic load, and perhaps by heating that produces pressure of volatiles, perhaps by other volume changes, they leave the orogen and travel long distances in the style of long-distance hydrologic flow (Figure 8.3~. Hydrologic flow for which meteoric waters are the source is thought to carry fluids for many hundreds of kilometers through parts of the continent, although the details of that flow are not yet well known (Sun, 1986~. Fluids ejected tectonically into the continent may propagate regionally in similar manner. How much mixing occurs between water from meteoric sources and tectonic fluids is not specified here, but it seems that such mixing will be dependent on various local parameters and so may vary greatly from one region to another. Nor does this hypothesis specify such matters as the chemistry of the migrating brines or the particular place and time when organics mature into hydrocarbons. Such maturation presumably may occur at many places- in the sedimentary trough of the margin prior to subduc- tion, during the subduction process, or in the foreland basin sediments almost independently of the tectonics. It remains to be seen whether data on the details of migration for a particular hydrocarbon province can be fit into this hypothesis, but, as shown later, more general information on hydrocarbon migration seems compatible with it. Fluids injected into the hydrologic system must perturb the thermal regime. Precisely what form such perturba- lions take cannot easily be specified in the absence of information on parameters such as channeling, depths, rates of subduction, etc. Nevertheless, as the temperature of tectonic fluids from depth is likely to be high, the vertical temperature gradient is likely to be increased, thereby enhancing certain kinds of metamorphism. Not all features or processes that might be associated with this model are closely specified here. At this stage of development it seems prudent to maintain flexibility in the hypothesis, or model, so as to preserve the opportunity to / 91TUMiN~S ANTHRACITE '`-; \ THRUST \~ SHEET '' .',,.: ·:--'', v ': it_ CONT I N ENTAL ~RIJ9T ~ /' Forebulge Subsidence due to load | ~ TECTONIC BRINES SCHEMATIC: NOT TO SCALE
COCORP AND FLUIDS IN THE CRUST refine the model as observations dictate. Let us now consider some of the data that provide support for the model. All observations reported here are taken from the literature. The data are many; only a summary is pre- sented here. For a more complete discussion of the obser- vations, see Oliver ( 1990~. Coal It is consistently observed that in foreland basins coal is generally metamorphosed to higher grade near the oro- genic belt; the grade decreases more or less gradually toward the continent. Thus, in Pennsylvania anthracite is found to the east and successively lower grades to the west. This point has been demonstrated not only for the Appalachian orogen but also for the Ouachita and Cordil- leran orogens by Thom (1934~. This pattern is conven- tionally explained by deeper burial, and hence the higher temperatures needed to metamorphose the coal, near the orogen. In such explanations the deep burial is followed by sometimes rapid uncovering to bring the coal near the surface where it is mined (Levine, 1986~. Some coal geologists are not in accord with the burial hypothesis, however, and have suggested lesser depths of burial and anomalous sources of heat such as intrusions or hydrother- mal fluids (Damberger, 1974; Haquebard and Donaldson, 1974~. An alternative explanation is that a pulse of heat carried by tectonic fluids caused the coal to be metamor- phosed at depths shallower than those usually called upon under the burial depth hypothesis. Brines There is a great deal of evidence that suggests or is compatible with the idea that a pulse of hot mineral-laden brines propagated through permeable sediments of the continent at the time of orogeny in a nearby erogenic belt. The evidence comes from study of Mississippi Valley- type (MVT) lead-zinc ores, fluid inclusions, certain occur- rences of dolomite, paleoremagnetization, and some other sources. After some years of controversy and discussion, eco- nomic geologists seem to be settling on the origin of the strata-bound sphalerite and galena deposits commonly called MVT Pb-Zn ores as, quoting Guilbert and Park ( 1986), "deposition from connate basinal water that moved updip in response to compaction or other loading pres- sure." Some geologists call upon expulsion of fluids from compacting basins to provide such brines. The view of this chapter is that basins subjected to tectonic deforma- tion and burial are more likely to expel fluids in greater volumes than are simple compacting ones, although there is no reason to conclude that compaction might not be a ]33 source in some cases, and of course some basins may be affected by both self-compaction and tectonic activity. The map pattern of MVT lead-zinc occurrence, shown in Figure 8.4, illustrates that such occurrences bear a spa- tial relation to erogenic belts similar to that for hydrocar- bons (see later section) except that, for reasons unknown, Pb-Zn deposits tend to occur between basins rather than within them. MVT deposits tend to occur in carbonates with a strong bias toward dolomites (Anderson and Mac- Queen, 19824. This observation suggests a common ori- gin for MVT Pb-Zn and some dolomites. The subject is not free of controversy, but in recent years at least sedimentologists have come to believe that some dolomites are formed at depth as a consequence of passage of basinal brines through limestones (see Zenger and Dunham, 1980, for a review). This view is in contrast to the hypothesis that all dolomites are formed at or near the surface. Gregg (1985) proposed that circulation of basinal brines through an underlying sandstone caused a 6-m layer of dolomite at the base of a limestone and shale horizon in southeastern Missouri. He associated this dolo- mitization with the emplacement of nearby MVT Pb-Zn ores and also with the mineralizing waters that Leach et al. (1984), on the basis of fluid inclusion data, had postulated as flow from the Arkoma Basin. In other words, dolomi- tization is apparently related to MVT ores in some cases because of related origin of both as a consequence of mineralizing migrating brines that may have originated in erogenic belts. Thus, brines producing the MVT deposits of Missouri likely originated during the Ouachita orogeny to the south, although the possibility of some effects from the Appalachian orogen to the east must also be kept in mind. The dating of emplacement of MVT Pb-Zn deposits is not easy, but some information is available. Sphalerite (ZnS) twins in deformation. Taylor et al. (1983) inter- preted sphalerite twinning in intermediate layers of multi- layered sphalerite in eastern Tennessee to indicate deposi- tion during the Alleghenian orogeny and attributed the mineralization to brines expelled from the shale basin to the east. Beales et al. (1980) used paleomagnetic data to determine an Upper Pennsylvanian age for MVT ores and speculated on the role of tectonic processes in ore forma- tion and hydrocarbon migration. Such dates, as available, for MVT ore deposition support the tectonic fluids hy- pothesis. Fluid inclusions are an important source of information on the history of fluids in the crust. Roedder (1984) prepared an excellent summary of this subject, and his book includes many points relevant to the subject matter of this chapter. For example, Roedder noted that evidence from fluid inclusions in Ohio, New York, Iowa, Missouri, and Tennessee are similar and hence suggest that scattered
134 FIGURE &.4 Map, modified from Cathles and Smith (1983), showing Mississippi Valley-type lead-zinc occurrences. This map is an approximation. Occurrences in small quantities may be more widespread. occurrences of sphalerite, etc., formed from the same kinds of fluids as MVT ores. This point implies a possible pulse, or pulses, of mineralizing brines that covered large re- gions. Roedder also noted that organic material is com- mon in the inclusions of both ores and scattered minerals. Leach (1973) attributed fluids from sphalerite in a coal mine and MVT PB-Zn to a single episode of fluid flow. Leach et al. (1984), Rowan et al. (1984), and Leach and Rowan (1986) studied fluid inclusions from Missouri, Kansas, Arkansas, and Oklahoma and found lateral gradi- ents in temperature increasing to the south. They suggest that fluids expelled tectonically from the Ouachita Arkoma basin at 200° to 300°C migrated northerly to deposit MVT Pb-Zn. Much evidence from fluid inclusion studies sup- ports the concept of tectonic expulsion of fluids from erogenic belts. Certain aspects of diagenesis, a term used here in its general sense to include epigenesis, of sediments may be a consequence of migrating tectonic fluids. Morton (1985) dated, by the Rb-Sr method, diagenetic illite from the Upper Devonian black shales of Texas and found a date equivalent to the time of the Ouachita orogeny. He sug- gested that brines from the Ouachita tectonic zone were JACK E. OLIVER . , , 1 1 1 1 _. ~ ma_ . _ _ ~J-i `__} J.~\ ,- -\ ~"`';'`~ \ / ~ LEAD-ZINC OCCURRENCES ~ ~ c, x~ All rot ~\J \\ ~~'~ _~ J ~ ~ , . ~` ~ ~!' . . . . ~ - ~ ' ~,; ::.: TECTONIC 8ELtS 0 200 ItlLo - 7EnSk ~ ~ ~ BAS I NS the cause and, following Dickinson (1974), that hydrocar- bons may also have migrated to West Texas at that time. Authigenesis, the enlargement by overgrowth, of potas- sium feldspars in Cambro-Ordovician rocks of western Maryland was dated by Hearn and Sutter (1985) using the 40Ar/39Ar method. They found ages corresponding to the Alleghenian orogeny and suggested that brines from the erogenic zone produced the authigenesis and affected hydrocarbon migration and ore deposition. Hearn et al. (1985) obtained results that were consistent with and that extended these conclusions for feldspars in Pennsylvania, Virginia, and Tennessee. Authigenesis by migrating brines is playing a role of growing importance in the subject of paleoremagnetiza- tion, a major new dimension in the field of paleomagnet- ism. In recent years paleomagneticists have demonstrated widespread demagnetization of sedimentary rocks for large parts of North America (Van der Voo, 1986~. One pos- sible explanation is based on thermally activated viscous magnetization. The other, and apparently the preferred one, is chemical demagnetization through authigenesis. The demagnetization is dated as the time of the Alleghenian orogeny in some areas. Van der Voo and French (1977)
COCORP AND FLUIDS IN THE CRUST found Alleghenian dates in Virginia, West Virginia, and Pennsylvania. Scotese et al. (1982) found similar dates from a study of Silurian and Devonian rocks in New York. McCabe et al. (1983) studied demagnetization of folded Silurian-Devonian rocks in New York using the fold test and found that demagnetization occurred during folding (i.e., during the Alleghenian orogeny). McCabe et al. (1984) found similar dates in Quebec, Ontario, and New York. Kent and Opdyke (1985) revised an earlier study con- trasting paleolatitudes of the New England-Canadian Maritime region and the North American craton and showed an error as a result of paleoremagnetization that negated earlier and enigmatic conclusions about major relative tectonic movement between these two provinces. To summarize, there is widespread and growing evidence from studies of paleomagnetism that supports the concept of authigenesis of magnetic minerals in sediments of the continental interior as a consequence of brines migrating at the time of orogeny. Hydrocarbons In the preceding portion of this chapter evidence is presented in support of the concept of long-distance migration of brines expelled from erogenic belts. This section considers the possible effects on hydrocarbons, particularly the spatial distribution of hydrocarbons as a consequence of that brine migration. Petroleum geologists are divided over the question of long-distance migration of hydrocarbons. Some believe little or no such migration occurs, others call for substan- tial migration, and still others take intermediate positions. This chapter takes the position that for some hydrocar- bon provinces those unaffected by tectonic fluid flow- lateral migration is typically modest at most. For those provinces affected by tectonic fluids, however, lateral migration up to distances of many hundreds of kilometers may occur. The spatial pattern of hydrocarbons through- out the world may be examined in this framework. A second factor influencing the occurrence and abun- dance of hydrocarbons in the case of tectonically influ- enced deposits is the nature of material on the down-going slab of the subduction zone. If the sediments are volumi- nous and organic-rich, as in the case of a large delta, the development of hydrocarbons is enhanced. Of course, many other factors that are part of the rich literature of petroleum geology but not discussed here, such as seals, traps, and the nature of organic materials in source rocks, influence the spatial pattern of hydrocarbon deposits; however, the emphasis here is on active tectonism and the factors discussed above. As an example consider the Gulf Coast province, which 135 holds one of the Earth's largest hydrocarbon deposits. Drainage from a large part of the North American conti- nent has resulted in a large and complex delta with organ- ics of terrestrial and marine origin. Organics are buried to maturation conditions and converted to hydrocarbons. The hydrocarbons migrate vertically and, at most, to modest distances (perhaps tens of kilometers) horizontally and are found in traps relatively near to where the hydrocarbons were formed. Now imagine what would happen if the Gulf of Mexico were to close, so that the North American continent was partially subducted to the south beneath an accretionary wedge on the leading edge of a landmass impinging from the south. Fluids from the Gulf Coast sediments would be expelled and, according to the hypothesis, travel toward the interior of the continent (i.e., to the north). The fluids would be brines carrying minerals and both organics and hydrocarbons. Some organics may have already matured; some might encounter conditions for maturation along the way. The foreland basin resulting from the tectonic load- ing would fill with sediments. These sediments might produce some of their own hydrocarbons that might or might not be affected by the flow of tectonically generated fluids. Nevertheless, the general pattern of the hydrocar- bon occurrences that resulted would reflect the effect of the migrating fluids. This experiment need not be done solely in the mind. It may represent about what happened at the time of the Ouachita orogeny. Thus, the spatial pattern of the hydro- carbon province extending from West Texas through Oklahoma and Kansas (Figure 8.5) may be at least partly a consequence of the Ouachita orogeny and the expulsion of fluids from the Ouachita erogenic belt. The pattern is sharply defined by a smooth boundary along the south and east sides. Along the north and west it has the feather edge that might be expected of the squeegeeing effect of Ou- achita overthrusting. This view resembles that of Salis- bury (1968), who suggested years ago that such a phe- nomenon occurred in the case of the Marathon erogenic belt and associated hydrocarbons to the north of that belt. Gas in the Ouachita hydrocarbon province tends to occur near the orogen, oil farther away. The hypothesis suggests that some hydrocarbons of this province have not migrated much, but others have traveled as organics or hydrocar- bons long distances from where the former continental margin lies buried beneath the Ouachita orogen. To the east and north the similarity of the pattern of the boundary of hydrocarbon occurrences in Kentucky and Illinois with that of the Ouachita orogen raises the specu- lation that some flushing of these hydrocarbons to the north may have occurred. In like manner one can interpret other hydrocarbon provinces of the map of Figure 8.5. Williston Basin, Michi
~- ~ art 1 `,~ Oil fields or scattered occurrences `~ Gas fields or scattered occurrences Tectonic belts JACK E. OLIVER N_._ __ ~ V~ ~ By, - ~.~.~. .~ By." ' . ' =\ I _ ,, ~ .,, . , . . . . . . ..... ~ . , FIGURE 8.5 Map, generalized from PennWell map (Wilkerson, 1982), showing regions of oil and gas fields in the United States and adjoining parts of Canada and Mexico. Also shown are erogenic belts. The spatial distribution of hydrocarbons in the gan Basin, and California hydrocarbons are found essen- tially where they were formed, with only modest migra- tion within the particular basin. Appalachian oil has migrated to the west as described by Woodward (1958) as a consequence of burial of an organic-rich continental margin by convergence and collision during Paleozoic time. The oil of the Findlay arch in Ohio and Indiana was part of the westward migration in the early Paleozoic and concentrated in the arch when it was formed as a forebulge resulting from the loading of the crust by later Appala- chian tectonics. Albertan hydrocarbons migrated to the east, the conse- quence of burial of an organic-rich margin beneath over- thrust slices from the west. Demaison (1977) proposed such an explanation for the Alberta tar sands (and also for large tar sands elsewhere). He called upon burial of a foreland basin delta, with source to the east, beneath thrust Appalachians, Alberta, the Cordilleran province, the West Texas- Oklahoma-Kansas province, and possibly the Illinois basin are in part a consequence of fluid migration from nearby erogenic belts according to the hypothesis discussed here. ing from the west that drove the hydrocarbons updip through the permeable beds of the delta. The pattern of Cordillera hydrocarbons, though grossly similar to that of Alberta, is not so simple. This region has been disrupted by more recent tectonics in the foreland basin that must have af- fected the spatial distribution of hydrocarbons. Further- more, the Cordillera hydrocarbons may not be associated with a subducted continental margin as rich in organics as that of Alberta. Nevertheless, the pattern is not in dis- agreement with the hypothesis. Information on directions of migration of hydrocarbons in the 48 states is summarized in a map by Momper (1978) reproduced in Figure 8.6. Note that all of the directions on Momper's map are compatible with those predicted by the tectonics-based hypothesis discussed here, even to the extent of asymmetrical migration from the Illinois basin as op- posed to symmetrical migration from other interior basins
COCORP AND FLUIDS IN THE CRUST (i.e., Williston and Michigan). The surprisingly good fit of an independently collected set of data with a hypothesis based initially on entirely different considerations must be considered strong support for the hypothesis. Similar reasoning about the spatial distribution of hydro- carbons may be applied to other parts of the world. For example, the petroliferous deltas of Africa and the rifts of China are places unaffected by major compressional orog- eny and also where hydrocarbons have not migrated far from the place of maturation. The hydrocarbons of the Middle East, as was pointed out by Dickinson (1974) in one of the first and most important early attempts to relate hydrocarbon accumula- tions to plate tectonics, likely migrated updip to the west from the Zagros erogenic belt as a consequence of subduc- tion of the leading edge of the African continent, which then included the Arabian peninsula, to the east beneath the overriding Asian plate. Dickinson focused on hydro- carbon migration and not the associated brines. However, in the context of this chapter, one might speculate that, as a consequence of the paleodrainage of the African conti- nent, a large organic-r~ch delta existed on the leading edge of the African plate that was subducted in the Zagros, with the consequence that fluids from the delta were expelled into parts of Iran and the Arabian peninsula. The Red Sea, a later feature, was not part of the proposed paleodrainage pattern. An interesting point here concerns the Persian Gulf, which has been in existence through much of the period of subduction and hence through the period of proposed hydrocarbon migration. Consequently, in this area there was not topographic relief with elevations de LONG - DISTANCE Old MIGRATION NINE GENERAll2ED EXAMPlES IN U.S.A. :~ ~: ~.: SOURCE PROVINCE '-_ L ~its ~ PH0ttA. AlSEtrA. loll mopl llllSTON. O \ WlillSION, M ~lillNoISe M ANADAlKO, ~ MICHIGAN. I`l ~ i ~ (it) ~41~AND, ~APPALACHIA {Mull;) "C\ 137 creasing from the erogenic belt into the foreland basin (i.e., towards Arabia). Thus, conditions were not favorable for hydrologic flow of meteoric waters from the erogenic belt toward the foreland basin in a manner comparable to that described by Garven and Freeze (1984) for Alberta. Consequently, such hydrologic flow is not likely the cause of hydrocarbon migration in the Middle East, and the migration is more likely a consequence of tectonic expul- sion and updip transport of already formed hydrocarbon fluids and perhaps immature organic materials. The locations of the hydrocarbons of the North Slope of Alaska may be at least partly a consequence of expulsion of fluids from the overthrust zone or subduction zone of the Brooks Range to the south. Difficulties of associating source rock and oil (Magoon and Claypool, 1985) in this region may be a consequence of failure to make full use of the tectonic framework as a basis for sample selection and interpretation of data. Within the context of this hypothesis one might ask why greater accumulations of hydrocarbons are not found in India in the foredeep of the Himalayas, for example. A possible reason is that a large delta never existed on the leading edge of India, the edge that was subducted beneath the Himalayas. As India drifted across the Tethys, the paleodrainage was limited to the Indian subcontinent. The area of this subcontinent is not very large, and the major drainage may have terminated at other parts of the margin, a hypothesis that could be tested by offshore seismic stud- ies. Prior to the time of drift the portion of India that made up the margin of Gondwanaland may not have been the site of a major delta. FIGURE 8.6 Map of Momper (1978) showing summary of information on mi- gration direction of hydrocarbons in the United States. Note general agreement between the migration directions of this map and those suggested by the hypothe- sis on fluid expulsion from erogenic belts as described in the text. S - SOURC E AREA - MIGRATION pArMs
138 The eastern margin of the United States seaward of the Appalachian hydrocarbon province is not very productive of hydrocarbons, whereas the Gulf Coast that occupies a similar position relative to the Ouachita hydrocarbon prov- ince is. According to the hypothesis, this situation occurs because the Gulf Coast is and has been the deposition area for drainage of a large part of the North American conti- nent and the East Coast has not. Of course, other factors such as climate, conditions of deposition, and marine sources must be involved. To summarize this section, it appears that a large amount of evidence from a wide diversity of specialties favors the hypothesis that fluids expelled from erogenic belts play an important role in many geological phenomena such as the spatial distribution of certain hydrocarbon and mineral deposits, authigenesis, diagenesis, coal metamorphism, dolomitization, and paleoremagnetism. The case is not closed, however, for there remains opportunity for further testing, and the attempt at such broad-based synthesis is so new that many aspects require further attention. Should the hypothesis be more or less correct, however, a means for relating many types of geological observations of the continent to plate tectonics, and hence the great geody- namical processes of the Earth, may be in sight. ACKNOWLEDGMENTS This chapter, as are all papers based on COCORP data, is highly dependent upon the efforts of many scientists and others who are part of the COCORP project. The CO- CORP project is funded by National Science Foundation grant EAR-8418157 and by the Cornell Program for the Study of the Continents. The story of fluids expelled from erogenic belts is not strictly a part of the COCORP project but is a speculative synthesis stimulated by COCORP and other data. This is contribution no. 79, INSTOC, Cornell University. REFERENCES Anderson, G. M., and R. W. Macqueen (1982~. Ore deposit models. 6. Mississippi Valley-type lead-zinc deposits, Geos- cience Canada 9, 108-117. Beales, F. W., K. C. Jackson, E. C. Jowett, G. W. Pearce, and Y. Wu (1980~. Paleomagnetism applied to the study of timing in stratigraphy with special emphasis to ore and petroleum prob- lems, in The Continental Crust and its Mineral Deposits, D. W. Strangway, ea., Geological Association of Canada Special Paper 20, 789-804. Brown, L. D. (1987~. Lower continental crust: Variations mapped by COCORP deep seismic profiling, Annales Geophysicae SB(4), 325-330. JACK E. OLIVER Brown, L. D., M. Barazangi, S. Kaufman, and J. Oliver (1986~. The first decade of COCORP, 1974-1984, in Reflection Profil- ing: A Global Perspective, M. Barazangi and L. Brown, eds., Geodynamics Series, vol. 13, American Geophysical Union, Washington, D.C., pp. 107-120. Cathles, L. M., and A. T. Smith (1983~. Thermal constraints on the formation of Mississippi Valley-type lead-zinc deposits and their implications for episodic basin dewatering and de- posit genesis. Economic Geology 78, 983-1002. Clarke, J. W., R. C. McDowell, J. R. Matzko, P. P. Hearn, D. J. Milton, D. J. Percious, D. B. Vitaliano, and G. Ulmishek (1986~. The Kola superdeep drill hole [detail summary of Kol'skaya sverkhglubekaya by Ye. A. Kozlovskiy (1984~], U.S. Geological Survey Open File Report 86-517, 249 pp. Damberger, H. H. (1974~. Coalification patterns of Pennsylva- nian coal basins of the eastern United States, in Carbonaceous Materials as Indicators of Metamorphism, R. E. Dutcher, ea., Special Paper 153, Geological Society of America, Boulder, Colo., pp. 53-74. Demaison, G. J. (1977~. Tar sands and supergiant oil fields, American Association of Petroleum Geologists Bulletin 61, 1950-1961. de Voogd, V., L. Serpa, L. Brown, E. Hauser, S. Kaufman, J. Oliver, B. W. Troxel, J. Willemin, and L. A. Wright (1986~. Death Valley bright spot: A midcrustal magma body in the southern Great Basin, Geology 14, 64-67. Dickinson, W. R. (1974~. Subduction and oil migration, Geology 2, 421-424. Garven, G., and R. A. Freeze (1984~. Theoretical analysis of the role of groundwater flow in the genesis of stratabound ore deposits. 1. Mathematical and numerical model, American Journal of Science 284, 1085-1124. Green, A. G., and J. A. Mair (1983~. Subhorizontal fractures in a granitic pluton: Their detection and implications for radioac- tive waste disposal, Geophysics 48, 1428-1449. Gregg, J. M. (1985~. Regional epigenetic dolomitization in the Benneterre dolomite (Cambrian), southeastern Missouri, Ge- ology 13, 503-506. Guilbert, J. M., and C. F. Park (1986~.- The Geology of Ore Deposits, W. H. Freeman, New York, 985 pp. Haquebard, P. A., and J. R. Donaldson (1974~. Rank studies of coal in the Rocky Mountains and inner foothills belt, Canada, in Carbonaceous Materials as Indicators of Metamorphism, R. E. Dutcher, ea., Special Paper 153, Geological Society of America, Boulder, Colo., pp. 75-93. Hearn, P. P., and J. F. Sutter (1985~. Authigenic potassium feldspar in Cambrian carbonates: Evidence of Alleghenian brine migration, Science 228, 1529-1531. Hearn, P. P., J. F. Sutter, M. J. Munk, and H. D. BeLkin (1985~. Evidence for Alleghenian brine migration in the central and southern Appalachians: Implications for Mississippi Valley- type sulfide mineralization, Geological Society of America Abstracts with Programs 17, 606. Kent, D. V., and N. D. Opdyke (1985~. Multicomponent mag- netizations from the Mississippian Mauch Church follllation of the central Appalachians and their tectonic implications, Journal of Geophysical Research 90, 5371-5383.
COCORP AND FLUIDS IN THE CRUST Koslovski, Ye. A. (1984~. Kol'skaya sverkhglubokaya, Moscow, Nedra, 490 pp. Leach, D. L. (1973~. Possible relationship of Pb-Zn mineraliza- tion in the Ozarks to the Ouachita orogeny, Geological Society of America Abstracts with Programs 5, 269. Leach, D. L., and E. L. Rowan (1986~. Genetic link between Ouachita foldbelt tectonism and the Mississippi Valley-type lead-zinc deposits of the Ozarks, Geology 14, 931-935. Leach, D. L., J. G. Viets, and L. Rowan (1984~. Appalachian- Ouachita orogeny and Mississippi Valley-type lead-zinc deposits, Geological Society of America Abstracts with Pro- gramsl6,572. Levine, J. R. (1986~. Deep burial of coal-bearing strata, anthra- cite region, Pennsylvania: Sedimentation or tectonics? Geol- ogy 14, 577-580. Magoon, L. B., and G. E. Claypool, eds. (1985~. Alaska North Slope OillRock Correlation Study, Studies in Geology 20, American Association of Petroleum Geologists, Tulsa, Okla., 682 pp. Mair, J. A., and A. G. Green (1981~. High-resolution seismic reflection profiles reveal fracture zones within a homogeneous granite batholith, Nature 294, 439-442. Matthews, D. H., and M. J. Cheadle (1986~. Deep reflections from the Caledonides and Variscides west of Britain and Comparison with the Himalayas, in Reflection Seismology: A Global Perspective, M. Barazangi and L. Brown, eds., Geody- namics Series, vol. 13, American Geophysical Union, Wash- ington, D.C., pp. 5-19. McCabe, C., R. Van der Voo, D. Pracor, C. R. Scotese, and R. Freeman (1983~. Diagenetic magnetite carries ancient yet secondary remanence in some Paleozoic sedimentary carbonates, Geology 11, 221-223. McCabe, C., R. Van der Voo, and M. M. Ballard (1984~. Late Paleozoic demagnetization of the Trenton limestone, Geophysi- cal Research Letters 11, 979-982. Momper, J. A. (1978~. Oil migration limitations suggested by geological and geochemical considerations, in AAPG Continu- ing Education Course Note Series 8, American Association of Petroleum Geologists, Tulsa, Okla., B1-B60. Morton, J. P. (1985~. Rb-Sr dating of diagenesis and source age of clays in Upper Devonian black shales of Texas, Geological Society of America Bulletin 96, 1043-1049. Oliver, J. (1986~. Fluids expelled tectonically from orogenic belts: Their role in hydrocarbon migration and other geologic phenomena, Geology 14, 99-102. Oliver, J. (1990~. The spots and stains of plate tectonics, in preparation. Roedder, E. (1984~. Fluid Inclusions, Reviews in Mineralogy, Vol. 12, Mineralogical Society of America, Washington, D.C., 644 pp. Rowan, L., D. L. Leach, and J. G. Viets (1984~. Evidence for a late Pennsylvanian-Early Permian regional thermal event in Missouri, Kansas, Arkansas, and Oklahoma, Geological Soci- ety of America Abstracts with Programs 16, 640. Salisbury, G. (1968~. Natural gas in Devonian and Silurian rocks 139 of Permian basin, in Natural Gases of North America; A Sym- posium, B. W. Beebe, ea., Memoir 9 (2), American Associa- tion of Petroleum Geologists, Tulsa, Okla., pp. 1433-1445. Sanford, A. R., O. S. Alptekin, and T. R. Toppozada (1973~. Use of reflection phases on microearthquake seismographs to map an unusual discontinuity beneath the Rio Grande rift, Seismo- logical Society of America Bulletin 63, 2021-2034. Sanford, A. R., R. P. Mott, Jr., P. J. Shuleski, E. J. Rinehart, F. J. Caravella, R. M. Ward, and T. C. Wallace (1977~. Geo- physical evidence for a magma body in the crust in the vicinity of Socorro, New Mexico, in The Earth's Crust, J. G. Heacock, ea., Monograph 20, American Geophysical Union, Washing- ton, D.C., pp. 385-403. Scotese, C. R., R. Van der Voo, and C. McCabe (1982~. Paleo- magnetism of the Upper Silurian and Lower Devonian carbon- ates of New York State: Evidence for secondary magnetiza- tions residing in magnetite, Physics of Earth and Planetary Interiors 30, 385-395. Serpa, L., B. de Voogd, L. Wright, J. Willemin, J. Oliver, E. Hauser, and El. Troxel (1987~. Structure of the central Death Valley pull-apart basin from COCORP profiles in the southern Great Basin, Geological Society of America Bulletin 100, 1437- 1450. Sun, R. J., ed. (1986~. Regional aquifer-system analysis program of the U.S. Geological Survey, Summary of projects, 1974-84, U.S. Geological Survey Circular 1002, 264 pp. Taylor, M., W. C. Kelly, S. E. Kesler, J. E. McCormick, F. D. Resnick, and W. V. Mellon (1983~. Relationship of zinc mineralization in east Tennessee to Appalachian orogenic events, in International Conference on Mississippi Valley Type Lead-Zinc Deposits, Proceedings Volume, G. Kisvarsanyi, S. Grant, W. Pratt, and J. Koenig, eds., University of Missouri- Rolla, pp. 271-288. Thom, W. T., Jr. (19341. Present status of the carbon-ratio theory, in Problems of Petroleum Geology, W. E. Wrather and F. H. Lahee, eds., American Association of Petroleum Geolo- gists, Tulsa, Okla., pp. 69-95. Van der Voo, R. (1986~. Late Paleozoic remagnetization in the Appalachians and their foreland: Is it related to fluid migra- tion? Oral presentation at 14th Meeting, COPSTOC, Cornell University, April 15-16, 1986. Van Der Voo, R., and R. B. French (1977~. Paleomagnetization of the Late Ordovician Juniata follllation and the remagnetiza- tion hypothesis, Journal of Geophysical Research 82, 5796- 5802. Wilkerson, R. M., ed. (1982~. Oil and Gas Fields of the U.S., PennWell Publishing Co., Tulsa, Okla., 1-p. map. Woodward, H. P. (1958~. Emplacement of oil and gas in Appa- lachian basin, in Habitat of Oil, A Symposium, L. G. Weeks, ea., American Association of Petroleum Geologists, Tulsa, Okla., pp. 494-510. Zenger, D. H., and J. B. Dunham (1980~. Concepts and models of dolomitization: An introduction, in Special Publication No. 28, Society of Economic Paleontologists and Mineralogists, Tulsa, Okla., pp. 1-9.
