Abstract This chapter describes an integrated approach to reservoir characterization and three-dimensional (3-D) geologic modeling of the San Andres Formation at Vacuum field, New Mexico, United States. We present techniques to identify significant heterogeneities within a carbonate reservoir using stratigraphic, petrophysical, and 3-D multicomponent seismic data. This integrated approach provides a detailed static description of reservoir heterogeneity and improved delineation of the reservoir framework in terms of flow units. We use a petrophysics-based method to identify hydraulic flow units within a sequence-stratigraphic framework. Flow units are characterized within high-frequency carbonate sequences through analysis of the vertical variation of flow (kh) and storage capacity (ϕh) and pore-throat radius (R35) associated with successions of subtidal, intertidal, and supratidal rocks. Pore-throat radii from cored wells are used to modify the empirically derived Winland equation to estimate values of pore-throat radius in non-cored wells. Flow profiles, constructed from log porosities and neural-network permeabilities, are correlated and used to build a 3-D geologic-model framework. Characterization of both matrix and fracture properties within a reservoir is possible using 3-D multicomponent seismic data and wire-line logs. Compressional- and shear-wave amplitude attributes together provide more accurate porosity estimates than those determined from compressional-wave data alone. Shear-wave anisotropy measurements provide information about inferred fracture density and orientation that can be used to modify permeability models to account for regions with open fractures. Because of this study, reservoir-simulation models that incorporate modified permeability distributions more accurately account for unexpected early CO 2 -breakthrough times observed in the field. In addition, flow-simulation results indicate that the need to upscale the geologic model was significantly reduced or eliminated by describing flow units using the combined sequence-stratigraphic- and petrophysics-based method.

Abstract Muspac and Catedral are two of the most important gas- and condensate-producing fields in southern Mexico. They produce from Cretaceous fractured carbonates. The objective of this integrated study is to define the stratigraphic and structural controls that caused early water production in those fields. Open-hole log correlation of 45 wells served to define eight reservoir zones, based on petrophysical characteristics. Petrophysical properties were mapped using a volumetric parameter to analyze the anisotropy of the gas-storage capacity in the fields. Dipmeter and borehole image logs were interpreted in 14 wells using cumulative dip and vector plot techniques to define unconformities, flooding surfaces, and faults. Borehole images from five wells were extremely useful in detecting evidence of sedimentologic and structural features. Fracture density depends on petrophysical properties of the reservoir rocks. In-situ stress directions were determined in 18 wells using borehole breakouts to define the predominant northwest-southeast orientation of the open fractures. Two dominant fracture sets, determined from seismic attributes and borehole images, are parallel to seismically determined faults. According to this study, early water production is caused by coning through fractures, faults, and karstic zones. Some water-producing intervals depend on the location of perforations, especially when these are located in highly fractured rocks and close to the gas-water contact. To minimize early water production, the operator must avoid wells in fault zones, wells on the flanks close to the gas-water contact, and deviated wells drilled perpendicular to the direction of open fractures.

ABSTRACT Eolian reservoirs exhibit significant compartmentalization and directional permeability caused by the processes taking place during accumulation of sediments within an eolian system. The contrast in grain packing across erosional bounding surfaces is one of the primary controls of fluid-flow patterns within eolian reservoirs. Better prediction of the geometry of flow units bounded by erosional surfaces can be made by reconstructing the type of bedform that formed the accumulation. Subsurface study of the occurrence and the frequency of erosional bounding surfaces has been limited by the availability and quality of core data. However, using borehole images, specifically FMI and FMS logs, the orientation of stratification can be resolved, and the cross-cutting relationships produced by erosional bounding surfaces can be identified. Comparison of the stratification orientation above and below an erosional bounding surface makes it possible to classify the erosional bounding surface within a process-oriented hierarchy. Using the foreset and bounding surface orientations gathered from the FMI and FMS log data, and using computer simulation methods for bedform reconstruction, a bedform that reflects the observed variations in stratification can be constructed. An integrated study of FMS logs, FMI logs, and cores from the Tensleep Sandstone in the Oregon Basin Field, Bighorn Basin, Wyoming indicates that erosional bounding surfaces can be identified and classified. The FMI and FMS logs also allow delineation of eolian facies such as interdune accumulations.

ABSTRACT Modified Fischer plots can be used in subsurface reservoirs for parasequence-scale correlations. Although the technique is applied to deep-marine sediments, it is not restricted to such sediments. The approach is based on high-resolution bedboundary detection using borehole-imaging logs such as FMS (Formation MicroScanner, Trademark of Schlumberger, Inc.), FMI (Formation Microlmager, Trademark of Schlumberger, Inc.), or any of the other electrical or acoustic dipmeter devices. The operator interactively selects dips at the top of every clay-sand and clay-chalk interbed while viewing images on a computer workstation. A spreadsheet is prepared which lists the depths to the top of each selected bed boundary. From all contiguous cycles, one calculates a mean-cycle thickness and the departure of each cycle thickness from the mean. Dimensionless cycle thickness is then calculated by dividing mean-cycle thickness into departure from mean-cycle thickness. Interpretations are made on a cross plot of cycle number (a function of depth) vs. cumulative dimensionless departure from mean cycle thickness. The curve that results has a distinctive shape which is based on changes in cycle thickness. This curve can be used like a well-log trace to correlate cycles between wells, look for missing section, and hypothesize about changes in sea level, channel migration, and lateral continuity of facies.

Abstract Porous dolomites are present below a distinctive stratigraphic marker within the lower Quintuco Formation (Lower Cretaceous, Berriasian-lower Valanginian) in the eastern Neuquén basin, Argentina. Dolomitized pack-stones and wackestones with moldic and sucrosic porosity provide the main reservoir facies in Rio Neuquén field and perhaps other oil fields in the area. Lower Quintuco carbonates are comprised of: (1) oolitic grainstones, (2) burrowed, dolomitized oolite-skeletal-peloid packstones/wackestones, (3) dolomudstones and bedded anhydrites, and (4) very fine-grained, superficially coated oolite grainstones. These sediments are commonly packaged into shoaling- and coarsening-upward parasequences. Reservoir-quality porosity and permeability exist almost exclusively in burrowed, dolomitized packs tones and wackestones. These strata are interpreted as off-bar facies deposited on the landward side of bar complexes, similar to modern facies analogs known in the Joulters Cay area of the Bahamas. In the lower Quintuco Formation, dolomite preferentially replaced carbonate mud. Below an inferred widespread paleo-exposure surface, ooid-skeletal-peloid grains were then dissolved to leave an open pore network with abundant moldic and intercrystalline porosity.

Abstract A mineralogical, geochemical, and rock-magnetic investigation of sediments deposited during the last 300 years in Lake Greifen, a hard-water lake with moderate sulfate concentrations (<250 μmol/L) and seasonal anoxia, shows that both authigenic single-domain biogenic magnetite and multidomain detrital titanomagnetite were preserved within the bioturbated marls deposited prior to the onset of anthropogenically induced eutrophication. Subsequently, in response to a gradual change from oxic to dysaerobic to anoxic bottom-waters, the deposition of organic carbon-rich varved sediments occurred and the degree of magnetite preservation decreased as altered diagenetic conditions resulted in the rapid dissolution and sulfidization of the biogenic and detrital magnetite. The occurrence of both biogenically produced magnetite and detrital titano-magnetite within the upper 4 cm of sediment indicates that (1) biogenic magnetite may be produced within the near surface organic carbon-rich sediments, probably on an annual basis when the overlying waters are oxygenated, and (2) detrital magnetite is continuously deposited. Changes in magnetic properties below this zone of surface magnetite production and microscopic examination of corroded fine-grained biogenic magnetite extracted from this interval indicate the rapid destruction of the most recently produced (or deposited) magnetite. Our findings demonstrate that (1) lacustrine sedimentary magnetic properties may reflect redox conditions, which are in the case of Lake Greifen determined by productivity, and (2) rapid destruction and sulfidization of fine-grained and coarse-grained magnetite can occur in lacustrine systems that are characterized by high productivity, low available lake-water sulfate, low concentrations of dissolved sulfide, and rapid sediment accumulation rates. These findings differ from marine studies in which magnetite dissolution and sulfidization is postulated to occur in systems characterized by high productivity, high concentrations of dissolved sulfide, and low sediment accumulation rates. Based on our observations, we propose that microbially mediated processes are contributing, either directly or indirectly, not only to the authigenesis of magnetite in the Lake Greifen sediments but also to its destruction.

Abstract Glacio-lacustrine varved clay of late Wisconsinan age in western New York has stable remanent magnetization and anisotropic magnetic susceptibility (AMS). Remanence is carried by interacting single-domain grains of magnetite, but coarse multidomain grains of magnetite are also present. Remanent inclination is anomalously shallow, given the latitude of the area of deposition and the existence of a geocentric dipole field at the time of deposition. The AMS consists of a foliation that is gently inclined to bedding and a weaker lineation in the plane of foliation. Independence of magnetic fabric and direction of remanence is demonstrated by comparison of remanence and AMS at closely spaced sites within individual clay laminae. Magnetic fabric resulted from alignment of multidomain grains by transient density currents at the time of deposition; alignment of single-domain grains by the geomagnetic field occurred later in a dilute slurry at rest on the surface of deposition to produce a post-depositional detrital remanent magnetization (pDRM). Rapid deposition and compaction resulted in anomalously low remanent inclination. Remanent declination and magnetic lineation were unaffected by compaction, but magnetic foliation may have been increased.

Abstract We have been studying the effects of diagenesis on the paleomagnetic signal of lacustrine sediments by examining tephra layers found in unaltered and altered states in the same diagenetic environment. In this paper, we report rock magnetic data from seven such layers in six diagenetic environments ranging from saline-alkaline to mildly alkaline. The effects of diagenesis can be complicated and varied but certain patterns are evident. These patterns indicate that both the physical state of the magnetic grains and the geochemistry of the porewaters are important in determining the effects of diagenesis. In some cases, diagenesis simply reduces the intensity of the original magnetization, but in others it produces a new magnetization that completely overwhelms the original one. Measurements of various rock magnetic parameters have allowed us to relate the changes in intensity to changes in the particle-size distribution of the magnetic carriers. In particular, decreases in intensity appear to be associated with selective dissolution of the fine-grained magnetic carriers or with a general reduction in the quantity of magnetic carriers of all grain sizes; increases in intensity seem to result from a reduction in the numbers of coarse-grained magnetic carriers. Plots of the demagnetization behavior of natural and saturation isothermal remanent magnetizations may be useful in discriminating between unaltered and altered material, especially when used in conjunction with other rock magnetic information.

Abstract Eight short stratigraphic sections within the Moenkopi Formation of northeastern Arizona verify a network of magnetic polarity stratigraphy previously observed in that area. The magnetostratigraphic signature was used to test the time relationships of numerous vertebrate faunal occurrences and to test the synchroneity of a major change in facies over this part of the depositional basin. The magnetic polarity data show that the vertebrate occurrences are not all of the same age; two or three different ages of fauna are indicated. Parallelism of the changes of magnetic polarity and lithology indicate a relatively rapid spread of sand-laden streams across the depositional basin. The near time-synchroneity of this lithologic change over much of the area indicates that the revised definition of the boundary between the Moqui and Holbrook members of the Moenkopi Formation as the first occurrence of a persistent, ledge-forming sandstone (Purucker and others, 1980) is a well-founded definition. A paleopole position was calculated from those samples that exhibited demagnetization behavior that is univectorial to the origin of orthogonal axes plots; this early Middle Triassic (early Anisian) paleopole is located at 94.8°E, 58.5°N (alpha-95 = 3.4°). Global correlation of the magnetostratigraphic pattern of the Moenkopi Formation to the patterns of two marine sequences indicates that Moenkopi deposition began in the Early Triassic mid-Griesbachian Stage and continued until it probably was interrupted by a hiatus that represents much of the Smithian. Deposition resumed in late Smithian and continued until late Spathian. The latest Spathian is represented by a hiatus. Deposition resumed again in the early Middle Triassic (early Anisian), thus the Moenkopi Formation provides a heretofore unknown record of geomagnetic field polarity during this time interval. A widespread Smithian hiatus, i.e., a lowstand, is suggested by comparisons of observations in south China, the Moenkopi Formation, and the Chugwater Group to those of the Arctic stratotypes.

Abstract Magnetostratigraphic dating of sedimentary strata is often the most precise technique available for temporally constraining the evolution of and controls upon sedimentary basins over I Ma in age. Uncertainties in the absolute dates derived by this technique are often difficult to assess quantitatively, despite the desirability of specifying their precision. An explicit discrimination should be made between correlations of the local magneto-polarity stratigraphy (MPS) to the global geomagnetic polarity time scale (GPTS) based on independent biostratigraphic or radiometric time control and those based on the smoothest derived sediment-accumulation rates. Situations in which there is a single, compelling correlation and those in which the correlation is the most reasonable of several possibilities should also be explicitly distinguished. In the latter case, alternative feasible correlations should be illustrated in order to permit a qualitative assessment of the uncertainties involved. Two classes of uncertainties are associated with the temporal calibration of magnetostratigraphic sections: those related to the creation of the local MPS and those related to the GPTS. Imprecision in measured stratal thicknesses and in the position of magnetozone boundaries can produce significant (up to 50 percent) uncertainties both in magnetozone patterns and in derived rates of sediment accumulation. Uncertainties in the GPTS result from uncertainties in the radiometric calibration of magnetic anomaly patterns. Comparison of available GPTS’s indicates uncertainties of (1) as much as 100 percent for sediment accumulation rate calculations involving intervals of less than 1–2 my and (2) up to 3 my in absolute ages. An example drawn from the Late Cretaceous to Eocene Axhandle thrust-top Basin of central Utah illustrates these uncertainties.

Abstract A high-resolution paleomagnetic record of a magnetic polarity episode has been recovered from an 8-m section of diato-maceous lacustrine sediments from Pringle Falls near La Pine, Oregon (43.7°N, 238.6°E). A total of 79 samples (172 specimens) was collected at about 2- to 5-cm intervals from the 20-m section. Biostratigraphic dating from diatoms gives an approximate age around 150 ka for this record. The characteristic magnetization of the samples was determined by AF demagnetization to 15 mT, after progressive demagnetization experiments from 2.5 to 60 mT. The average normal polarity stable declination and inclination of the non-reversing polarity episode sampled section (61°; α 95 = 2.6°) is statistically equivalent to that expected for this site for a geocentric axial dipole (62°). Rock magnetic experiments such as isothermal remanent magnetization (IRM) indicate that titanomagnetite is the dominant ferrimagnetic mineral. Interpretation of saturation IRM, susceptibility (X), and cleaned intensity (J) over anhysteretic remanent magnetization (J/ARM) experiments indicate that the behavior of the intensity of magnetization of these sediments is a reflection of the decrease of the intensity of the Earth’s magnetic field during the Blake polarity episode. The record provides clear documentation of three conspicuous inclination features, labelled A, B, and C. A well-defined virtual geomagnetic pole (VGP) path also has been obtained and shows two preferred bands of longitude over the Americas and antipodal to them. Correlation between a previously published record obtained from a locality about 1.5 km from the site of the record reported here indicates that there is a repeated sequence of inclination features (labelled A, B, and C) between the two records. These inclination features from the Pringle Falls sites can be correlated with a record from the Mediterranean (oceanic record) and from Gioia Tauro, Italy. Also, the same inclination features are repeated and correlated with the only igneous record for the Blake episode, obtained from the island of Reunion in the South Indian Ocean. The Pringle Falls sections, as well as the oceanic cores from the Mediterranean, produced a series of intermediate directions that coincide with the more continuous sedimentary record of the Blake polarity episode. The correlation within (local) and between (global) sites of lake sediments (Pringle Falls), oceanic sediments (Mediterranean cores), sedimentary rocks (Gioia Tauro, Italy), and ultimately lava flows (Réunion record) is important because the VGP paths correlate with other geophysical phenomena which indicate that episodes and reversals yield key data about the dynamo process and the core-mantle boundary. It appears that the paths show channels of north-south flow in the outer core, which are related to the nature of the base of the mantle, and that the geometry of the transitional field and the fluid motion in the Earth’s core are to a large extent determined by the thermal structure of the lower mantle. Thus, paleomagnetic records of the geomagnetic field on the surface of the Earth during polarity transitions may lead to a better understanding of the dynamics of the Earth’s deep interior. In summary, it is valid to assume that if the record of the Blake episode presented here and its correlation to other records of the same episode are valid, the Blake geomagnetic polarity episode can and should be used as a stratigraphic marker because of its world-wide presence as a geomagnetic feature.

Abstract A preliminary reversal stratigraphy has been resolved for three Lower to Middle Ordovician units in northern Arkansas. At least three magnetic components are present in these units. The first component removed (component A) with both thermal and AF demagnetization is steep and northerly. It reflects a modern viscous remanent magnetization (VRM), with possible minor contribution by recent goethite. The second component (component B), which yields the reversal stratigraphy, is shallow and easterly or westerly. The third component (component C) is shallow and south-southeasterly. Isothermal remanent magnetization (IRM) acquisition and thermal demagnetization reveals magnetite as the dominant magnetic phase in all samples. Many component B samples and some component C samples contain hematite as well. Component B directions are similar to interpreted Ordovician directions from previous studies. A conglomerate test on mineralized breccias from the Rush MVT district demonstrates that component B was acquired prior to brecciation. Resolvable component B directions are much more common in sandy dolomite and sandstone of the Everton Formation and St. Peter Sandstone than in samples with no terrigenous sand. This suggests a link between the influx of cratonic material (including detrital magnetite) into the carbonate depositional basin and acquisition of component B magnetization. Component C yields a paleopole at 119.2°E and 49.5°N, which falls very near the middle portion of the Permian segment of the apparent polar wander path for stable North America. Component C is likely a chemical remanent magnetization (CRM) acquired during Permian time. Polarity changes in component B can be correlated stratigraphically between sampled sections within the study area. Most of these correlations are consistent with correlations in lithology and magnetic susceptibility. The presence of a correlatable magnetostratigraphy is further evidence that component B was acquired during or soon after deposition. The preliminary polarity reversal pattern from this study crudely matches that of Piper (1987) but contains considerably more reversals than a compilation by Torsvik and Trench (1991).

Abstract Magnetic polarity stratigraphy (magnetostratigraphy) offers a powerful stratigraphic tool of great promise for high-resolution age-dating and correlation in the middle Miocene Monterey Formation of California in which precise age-dating and long-range correlation have not been possible due mainly to the absence or inadequate preservation of age-diagnostic siliceous microfossils (diatoms). For this reason, we have conducted a detailed magnetostratigraphic study of a 290-m thick Monterey section at Shell Beach in Pismo basin, central California. The results of this study, which is the first that provides high-resolution numerical age data for an entire section of the Monterey, clearly illustrate the potential that magnetostratigraphy holds for establishing a much-needed temporal framework for the Monterey. The lack of precise chronologic control in the Monterey has posed a serious obstacle to understanding the true origin of this economically and scientifically important unit of California. Detailed stepwise demagnetization analysis of some 1,021 closely spaced oriented samples of the Monterey rocks from the Shell Beach section, the age of which is constrained by limited diatom and calcareous nannofossil biochronologic control, shows that the Monterey lithologies, especially its dolomite, have faithfully preserved the geomagnetic reversal record that these rocks acquired at or near the time of deposition during the middle Miocene. Our data, which pass both the fold and reversal tests at greater than 95 percent confidence level, result in the recognition of 17 well-defined, lithologically independent and stratigraphically controlled magnetozones. Nine of these zones are of normal polarity (N1-N9) and eight are of reversed polarity (R1-R8). We correlate these magnetozones with the interval from the lower part of magnetic polarity Chron 5B to the lower part of Chron 5r of the standard geomagnetic polarity time scale of Harland and others (1982). From the ages of polarity zone boundaries and extrapolation of sedimentation rates to the base and top of the section, we conclude that the Shell Beach section of the Monterey was deposited between approximately 15.15 Ma and 11.0 Ma at an average post-compaction sediment-accumulation rate of 94 meters per million years. Our data indicate that the facies boundary between the lower calcareous-phosphatic facies and upper siliceous facies is marked by a hiatus/disconformity that lasted about one million years, from approximately 14.3 Ma to 13.25 Ma. We speculate that this hiatus may have been caused by a eustatic fall in sea level that began around 14.3 Ma due to global climatic cooling. The onset of predominantly siliceous sedimentation may signal the intensification of coastal upwelling caused by the increased pole-to-equator temperature gradients accompanying this same climatic cooling, an event clearly recorded by the Monterey rocks at around 13.25 Ma. Sandstone of the overlying Pismo Formation truncates the Monterey section at Shell Beach. We suggest that this erosional contact may have been caused by a latest middle Miocene sea-level fall at around 11.0 Ma. We note that a large eustatic fall in sea level is proposed by Haq and others (1987) at around 10.8 to 11.0 Ma.

Abstract Paleomagnetic, petrographic, and geochemical results, as well as field relationships, are used to relate Late Paleozoic chemical remanent magnetizations (CRMs) to the migration of basinal fluids in Ordovician carbonates in the Arbuckle Mountains, southern Oklahoma. The Viola Limestone contains a pervasive Pennsylvanian synfolding CRM residing in magnetite and a localized Permian CRM which resides in hematite and occurs in alteration zones around veins mineralized with calcite and Mississippi-Valley-type oxides and sulfides. The vein mineralization precipitated from basinal fluids that were warm, saline, and radiogenic. Radiogenic 87 Sr/ 86 Sr ratios of the limestones in the alteration zones and the fact that there is more significant alteration closer to the veins suggest that the basinal fluids were also responsible for alteration in the limestones. The coincidence of the geochemical and remagnetization trends suggest that the Permian CRM dates the migration of the basinal fluids in the veins. Geochemical results from the Viola with pervasive CRM indicate that it is relatively unaltered, with no evidence for radiogenic basinal fluids. This suggests that a mechanism that does not necessarily require exotic externally-derived fluids is needed to explain the origin of the pervasive CRM. Liesegang-banded carbonate around calcite-filled fractures in the Kindblade Formation also contains a Permian CRM residing in hematite. The fluid that precipitated the hematite liesegang bands emanated from the fractures and, based on geochemical results, was basinal in origin. The results of this study suggest that basinal fluids migrated through the carbonates in the Arbuckle Mountains during the Permian, although perhaps in several episodes. The flow of basinal fluids was focused in veins and only locally altered the host carbonates.

Abstract Paleomagnetic data have been obtained from heterogeneous, shallow-water, miogeoclinal carbonate rocks of the Pogonip Group (Early Ordovician) in the Desert Range of southern Nevada, the Egan Range of east-central Nevada, and the southern House Range of western Utah. These rocks locally contain abundant replacive chert that preserves relict textures from the host limestones as well as clearly detrital grains (e.g., blue-luminescing feldspars). Stylolites are abundant and are interpreted as late diagenetic features, as they cut late cements and truncate bedding lamination. Differential compaction along stylolites wrapping around the chert masses has resulted in macroscopic deformation, as evidenced by tilting of bedding of over 25° about chert masses in some cases. We have used the differential compaction fabrics in these rocks to test for the age of acquisition of a generally well-grouped and well-defined characteristic magnetization. All three carbonate sections give a low-inclination, southerly to southeasterly magnetization residing in magnetite (e.g., Decl. = 152°, Incl. = -21°, α 95 = 3°, kl = -61, k2 = -21, N = 48 independent samples, site 12; Pogonip Group, Sawmill Canyon, Egan Range). The magnetization is interpreted to be secondary and acquired after local compaction because directions of magnetizations from different samples are not dispersed by the compaction deformation. The uniform reversed polarity in addition to the direction of the magnetization, moreover, is interpreted to suggest a late Paleozoic age of remagnetization. In the Desert Range, the remagnetization had been previously attributed to a viscous partial thermoremanent magnetization (VPTRM) from deep burial. Based on several observations, we now argue for a chemical origin from late diagenetic magnetite, such as is now well-documented in the Appalachians and mid-continent. The cherts are almost nonmagnetic, as would be expected from their impermeability if the magnetite were precipitated from late fluids. Abundant authigenic alkali feldspar in the Desert Range is also consistent with late metasomatism. Finally, in the Egan Range, the remagnetization extends through a section exceeding 3 km in thickness, into rocks as young as Mississippian, which were never buried as deeply and thus not heated to the same degree as lower Paleozoic strata. These results underscore the utility of integrating observations based on paleomagnetic data with carbonate textures. “Micro”-field tests can constrain both the timing of magnetization acquisition and of diagenetic events. The micro-fold tests discussed apply to features that are not formed by tectonic deformation. The availability of field tests from early formed textures in sedimentary rocks is especially important given the recent recognition of widespread remagnetization in ancient rocks.

Abstract Studies of the paleomagnetism and rock magnetism of speleothems have shown that their natural remanent magnetizations (NRMs) are either detrital remanent magnetizations (DRMs), chemical remanent magnetizations (CRMs), or a mixture of both. The DRM arises from magnetic particles deposited by cave floods, and the evidence is unequivocal. The CRM results from organic-chelated iron or bacteria, and the evidence is circumstantial. The frequent presence of organic materials in speleothems is easily demonstrated, and they may be soluble, insoluble, or both. Organics account for the color of many speleothems. Whether the remanence is DRM or CRM, the signal is contemporaneous with the precipitation of the host calcite matrix and is measurably free of depositional inclination error effects. It is suggested that the studies summarized here have relevance for the timing of remanence in diagenetic limestone calcite, faithfulness of this remanence in its alignment along the ambient magnetic field at the site, and to hydrocarbon-magnetite associations in carbonate rocks.

Abstract Some salt dome cap rocks contain stratiform laminae of sulfides, which accreted in an orderly sequence as the cap rock formed by underplating at the salt/cap-rock interface. The occurrence of the magnetic mineral pyrrhotite makes it possible to use the remanent magnetization to date the timing of mineralization and its relationship with tectonic, sedimentary, and fluid evolution events within the local basin. Detailed paleomagnetic analyses of the cap rock of the Winnfield salt dome in northern Louisiana delineate a magnetic reversal pattern that can be correlated with the geomagnetic time scale, thus providing the first direct determination of the age of cap-rock formation. The sampled section, which represents about two-thirds of the total anhydrite thickness, formed between 157 and 145 Ma (latest Jurassic). Anhydrite accumulation rates calculated from these data decrease from 5.5 m/my to 2.6 m/my for the younger strata. Because the mother salt at Winnfield contains about 3 percent anhydrite, the growth rate of the salt diapir must have been at least 30 times faster, yielding values similar to other geologic estimates. The best documented cap rock-hosted sulfide concentrations occur at the Hockley salt dome in south-central Texas. The Hockley cap rock is believed to have developed within the last 45 Ma. Several magnetic reversals have been observed, but the data set is too incomplete at this time to further constrain the time of mineralization. The major metal concentrations are found within a 20-m zone within the central cap-rock stratigraphy. If this zone formed at rates comparable to the Winnfield cap rock, then the main pulse of mineralization at Hockley lasted only a few million years.

Abstract Post-depositional iron-sulfide (Fe-S) minerals that are related to hydrocarbon seepage have changed the original magnetizations at Cement oil field (Anadarko basin, Oklahoma), at Simpson oil field (North Slope basin, Alaska), and above deep Cretaceous oil and gas reservoirs, south Texas coastal plain. At Cement, ferrimagnetic pyrrhotite (Fe 7 S 8 ) formed with pyrite and marcasite in Permian red beds. The Fe-S minerals contain sulfur from two sources: (1) abiogenic sulfide, which has positive δ 34 S values, derived from thermochemical reduction of sulfate in deep reservoirs; and (2) biogenic sulfide, which has negative δ 34 S values, produced by reactions mediated by sulfate-reducing bacteria fed by leaking hydrocarbons. At Simpson, ferrimagnetic greigite (Fe 3 S 4 ) dominates magnetizations in nonmarine Upper Cretaceous clastic beds that contain epigenetic sulfide (δ 34 S > +20 per mil) and seeping biodegraded oil. In this setting, the authigenic magnetic sulfide mineral apparently incorporated sulfide produced by bacterial sulfate reduction under limited sulfate conditions. An inferred hydrocarbon food source for the sulfate-reducing bacteria links the hydrocarbon seepage to the greigite. The greigite is perhaps forming today. In middle Tertiary sandstones of southeast Texas, pyrite and marcasite formed when abiogenic H 2 S (enriched in 34 S) migrated upward from deep reservoirs, or when H 2 S (depleted in 34 S) was produced at shallow depths by bacteria that used organic material dissolved in migrating water from depth. The pyrite and marcasite replaced detrital magnetic iron-titanium oxide minerals. The degree of such replacement appears to increase toward faults that connect deep petroleum reservoirs to shallow sandstone. Our results show that abiologic and biologic mechanisms can generate different magnetic sulfide minerals in some sulfidic zones of hydrocarbon seepage. More commonly the magnetizations in such zones would be diminished as a result of the replacement of detrital magnetic minerals by the common nonmagnetic sulfide minerals, or would remain unchanged if such detrital minerals were originally absent.

Abstract Paleomagnetic, rock magnetic, petrographic, and geochemical studies indicate that hydrocarbons can cause either an increase or decrease in the magnetization in sedimentary rocks. For example, hydrocarbon-impregnated Permian calcite speleothems in southwestern Oklahoma contain a Permian chemical remanent magnetization (CRM) that resides in magnetite. A positive relationship between extractable organic matter and the natural remanent magnetization (NRM) suggests that the chemical conditions created by the hydrocarbons caused precipitation of the magnetite and acquisition of the associated CRM. There is no correlation, however, between percent asphaltenes and NRM in the speleothems. In addition, bitumen speleothems with high NRMs are, in general, less extensively biode-graded. These results suggest that a chemical process, and not biodegradation, is the mechanism for magnetite authigenesis in speleothems. The results from the speleothems suggest that hydrocarbons can cause acquisition of magnetization that can be dated using paleomagnetic analysis. Development of this dating approach, however, requires more work to better understand the mechanism(s) of magnetite precipitation. Studies of red bed and hydrocarbon-impregnated samples from cores of the Lyons Sandstone in the Denver Basin and from outcrops of the Maroon Formation (Schoolhouse Member) in northwest Colorado indicate that while hydrocarbons can cause precipitation of magnetite, they can also reduce the NRM by dissolution of hematite. This has implications for the various types of magnetic prospecting techniques that have been proposed.