Paleoredox conditions and the origin of bedded barites along the Late Devonian North American continental margin

The most important controls of the redox state of natural waters are the flux of organic matter from the photic zone, the degree of physical mixing, and oxygen concentrations of any connecting water masses. In coastal upwelling zones, these factors can be described with a simple box model constrained by data from modern field settings. Onshore-offshore mixing is largely controlled by the intensity of coastal currents that generally are found at depths of <400 m. As a result, onshore-offshore chemical gradients are greater in deeper waters. Application of the mass balance models to Late Devonian paleogeographic reconstructions of western North America illustrates how bedded barites in this setting may have formed in conjunction with a sulfate-reducing coastal upwelling zone. Low-oxygen waters from an extensive equatorial undercurrent entered the upwelling zone during regressive periods of the Late Devonian. The steep bathymetry of emergent bank carbonates to the east meant that the most intense portion of the coastal upwelling system occurred over deep water. Sinking organic matter was remineralized in these deep waters rather than being incorporated into sediments of the continental shelf and shelf break. The influence of coastal currents, and therefore onshore-offshore mixing, was minimal, thereby leading to the depletion of available oxygen and nitrate. Recent geochemical studies of particulate and dissolved barium in the modern ocean in conjunction with the mass balance model illustrate how waters with only minor sulfate-reduction and elevated dissolved barium concentrations may have led to extensive barite flux to the Late Devonian deep-water sediments off the North American continent.

Paleoredox Conditions and the Origin of Bedded Barites along the Late Devonian North American Continental Margin1 Paul W. Jewell Department of Geology and Geophysics, University of Utah, Salt Lake City, UT 84112 ABSTRACT The most important controls of the redox state of natural waters are the flux of organic matter from the photic zone, the degree of physical mixing, and oxygen concentrations of any connecting water masses. In coastal up welling zones, these factors can be described with a simple box model constrained by data from modern field settings. Onshore-offshore mixing is largely controlled by the intensity of coastal currents that generally are found at depths of <400 m. As a result, onshore-offshore chemical gradients are greater in deeper waters. Application of the mass balance models to Late Devonian paleogeographic reconstructions of western North America illustrates how bedded barites in this setting may have formed in conjunction with a sulfate-reducing coastal upwelling zone. Low-oxygen waters from an extensive equatorial undercurrent entered the upwelling zone during regressive periods of the Late Devonian. The steep bathymetry of emergent bank carbonates to the east meant that the most intense portion of the coastal upwelling system occurred over deep water. Sinking organic matter was remineralized in these deep waters rather than being incorporated into sediments of the continental shelf and shelf break. The influence of coastal currents, and therefore onshore-offshore mixing, was minimal, thereby leading to the depletion of available oxygen and nitrate. Recent geochemical studies of particulate and dissolved barium in the modern ocean in conjunc­tion with the mass balance model illustrate how waters with only minor sulfate-reduction and elevated dissolved barium concentrations may have led to extensive barite flux to the Late Devonian deep-water sediments off the North American continent. Introduction The study of oxygen and nutrients in natural wa­ters is critical to understanding a variety of geologi­cal and geochemical phenomenon. The cycling and preservation of many elements in water and sedi­ments is dependent on the redox state of the water column (Stumm and Morgan 1981). Barite (BaS04) is particularly interesting in this regard because its solubility product is low (Church 1979), dissolved barium behaves much like a nutrient (Turekian and Johnson 1966), and sulfate concentrations are dependent on the redox state of the water. Barite concentrations as high as several weight percent on a carbonate-free basis in oceanic sedi­ments beneath productive areas of the ocean have long been known (Goldberg and Arrhenius 1958). In recent years, considerable attention has been given to the paleoceanographic significance of bar­ite and its relationship to biological productivity 1 Manuscript received July 20, 1993; accepted November 15, 1993. in the ocean. Several studies have shown that bar­ite and barium in carbonates is a reliable proxy for surface productivity (Schmitz 1987; Shimmield et al. 1988; Lea et al. 1989; Dymond et al. 1992). Specific relationships of export productivity, dis­solved barium concentrations, and barite flux have also been developed (Dymond et al. 1992). Coastal upwelling zones are characterized by some of the greatest surface productivities found in the modern ocean (Berger 1989) and as such, constitute one of the primary mechanisms by which carbon and other nutrients are removed from the upper ocean and ultimately sequestered in marine sediments. The relationship between barium flux and coastal upwelling has only re­cently been considered in detail (e.g., von Brey- mann et al. 1990). This is in contrast to phospho­rites (e.g., Burnett et al. 1980; Sheldon 1980; Coles and Snyder 1985; Coles and Varga 1988) and or­ganic carbon-rich shales (e.g., Parrish 1982; Parrish and Curtis 1982; Schopf 1983), which have been [The Journal of Geology, 1994, volume 102, p. 151-164] © 1994 by The University of Chicago. All rights reserved. 0022-1376/94/10202-003$!.00 151 152 PAU L W. JEWELL extensively studied within the context of coastal upwelling throughout the geologic record. A significant difference between barite depos­ited in the modern ocean and that found in ancient sedimentary rocks is the form and concentration of the barite. Massive, bedded barite deposition is a common feature of some periods in the geologic record, particularly the early to middle Paleozoic (e.g., Papke 1984; Maynard and Okita 1992). In many of these settings, barite bed thicknesses ex­ceed 1 m and constitute 50% or more of the vol­ume of a given section of rock. Similar massive accumulations of barite are not found in the mod­ern ocean. Any model of barite formation in Paleo­zoic rocks must therefore account for these abnor­mally high concentrations of barite. The fact that barite is a redox-sensitive mineral and intimately associated with the biological cy­cles of the ocean suggests that the formation of massive bedded barite may be understood by con­sidering the redox controls of highly productive settings such as coastal upwelling. The work pre­sented here consists of a simple mass balance cal­culations that allow key upwelling processes to be highlighted and permit examination of how redox processes may have been different along the mar­gins of ancient oceans. A specific application of the model is then made to paleogeographic features and the formation of bedded barite of the Late De­vonian continental margin of North America. The Late Devonian of North America has long been rec­ognized for its widespread black shale and phos­phorite deposition (Conant and Swanson 1961; Ettensohn 1985; Coles and Varga 1988), the oc­currence of nodular and bedded barites (Holden and Carlson 1979; Pepper et al. 1985; Dube 1988; Poole 1988; Graber and Chafetz 1990; Jewell and Stallard 1991), and the development of extensive carbonate platforms (Heckel and Witzke 1979; Sandberg et al. 1988). The model presented in this paper is used to explain how these features may have been genet­ically related. Oxygen and Phosphorous Mass Balance Models The physical and chemical controls of dissolved oxygen and nutrients in natural waters are varied and complex. Physical exchange, either with the atmosphere or other water bodies, is one important factor. The flux of organic matter as a result of surface productivity is also very influential. Redox conditions in most bottom waters are the result of some combination of these factors. The goal of the mass balance models presented here is to deter­mine the relative importance of these variables. Particular attention is given to factors that might cause complete oxygen depletion and sulfate- reduction in ancient oceans. Results of the paleore- dox model are then used to examine the behavior of dissolved barium and barite flux in a sulfate- reducing environment. A brief summary of the most important features of coastal upwelling is given as a prelude to the discussion of mass balance models of oxygen, nu­trients, and barium. Particular attention is given to the Peru upwelling system, because of its rel­atively low subsurface oxygen concentrations, extensive data bases, and steep continental mar­gin which allows examination of deep water (>1000 m) geochemistry. Coastal Upwelling Overview. Most coastal up­welling zones exhibit a two-layer system of cur­rents (Smith 1981, 1992). The surface Ekman layer and associated offshore transport generally occurs in the upper 10-50 m of the ocean. Onshore trans­port below the surface Ekman layer (typically found between 100-300 m depth) is often diver­gent, i.e., a certain percentage of the onshore trans­port becomes entrained in longshore currents (Smith 1981). Poleward undercurrents constitute much of the subsurface, longshore transport in coastal upwelling zones. The dynamic origin of these undercurrents is not well understood, al­though they are typically strongest near the shelf break of coastal upwelling zones (Johnson and Rockcliff 1986). Poleward undercurrents are re­sponsible for most of the kinetic energy found at depths of <400 m in coastal upwelling zones (e.g., Brockman et al. 1980; Hagan 1981). Currents and kinetic energy below 400 m depth are considerably less than that found at more shallow depths. Upwelling water in coastal settings has widely varying nutrient and oxygen concentrations. In general, these concentrations reflect the character of intermediate-depth water which feeds into the coastal upwelling zone. For instance, the Peru- Chile Undercurrent is the source of upwelled wa­ter off the coast of Peru between 5°S and 15°S. The Peru-Chile undercurrent has its origins in the Pa­cific Equatorial Undercurrent, which traverses much of the Pacific equatorial region between 2°N and 2°S (Pickard and Emery 1983). Divergence of trade winds in the equatorial Pacific causes upwell­ing across much of the area. Sinking organic matter is remineralized and advected eastward by the un­derlying equatorial undercurrent. Equatorial wa­ters thus become progressively more nutrient-rich and oxygen-poor to the east. As the undercurrent encounters the South American continental mar­ Journal of Geology O RIGIN OF BEDDE D BAR ITES 153 gin and flows southward, coastal upwelling causes additional oxygen depletion and nutrient enrich­ment in the underlying water. As a result, nitrate reduction is common in deep coastal waters off Peru (e.g., Codispotti 1983). Under certain condi­tions, sulfate reduction has been observed (Dug- dale et al. 1977). Surface biological productivity of coastal up- welling zones is most intense where intermediate depth water comes to the surface (generally within a few tens of kilometers of the coast) and drops off dramatically in the offshore direction. Cross-shelf gradients in surface productivity (figure 1) tend to produce subsurface, cross-shelf gradients of oxygen and nutrients as a result of organic matter that sinks out of the photic zone. Off the coast of Peru, cross-shelf phosphate and oxygen gradients show (1) (1) (2) 'O =o o o <Q <r> 3 150 100 50 DISTANCE (km) Ekman transpo P Pa deep deep ocean -► coastal . (tel (d<?) dc Figure 1. Average primary productivities and a simple onshore-offshore box model of the Peru upwelling zone. 1 = the near shore productivities reported by Packard et al. (1983). 2 = offshore productivities from Suess et al. (1987). Ekman transport represents onshore move­ment of nutrient-rich water and offshore movement of nutrient-depleted water. A certain fraction of biogenic particles produced at the surface (adc • P) is remineralized in the deep coastal (dc) box. The deep coastal box re­ceives a higher amount of biogenic particle flux than does the deep ocean (do) box due to higher near-shore surface productivities. Exchange between the two deep- water boxes is represented by the mass flux Kd (m3/s per unit length of shoreline). considerable variability, but generally increase with depth (figure 2, table 1). This feature is proba­bly the result of decreasing onshore/offshore veloc­ities with depth. Although mean velocities are close to zero, velocity variances are clearly smaller in deeper water (table 2). Measurements of kinetic energy also show marked decreases with depth off of Peru (Hagen 1981). As a result, phosphate which is remineralized by sinking organic matter adja­cent to the coastline is less likely to be advected offshore in these deeper waters. Model Formulation. Many of the features of modern and ancient coastal upwelling chemistry can be explained by simple relationships that de­scribe the vertical and cross-shelf flux of oxygen and nutrients. In this section, surface productivity, organic carbon flux, and onshore-offshore oxygen and nutrient data from the Peru upwelling system and a simple box model are used to examine the fate of organic matter that settles into deep coastal water in coastal upwelling zones. It is important to emphasize that a box model such as this is fun­damentally different from advection-dispersion models that calculate continuous variations of properties with differential equations. Among the most simple and effective advection-dispersion models have been 1-dimensional, vertical simula­tions of oxygen and nutrients in the open ocean (Wyrtki 1962; Southam and Peterson 1985; Wilde 1987). The 2- and 3-dimensional nature of coastal upwelling requires considerably more complex ad­vection- dispersion models (e.g., Walsh 1975). The box model presented here represents a simple yet effective method of examining the deep-water, on­shore- offshore biogeochemistry of these settings. Deep-water (>300 m) coastal upwelling features are modeled with onshore and offshore boxes cou­pled by onshore-offshore mixing processes (figure 1). Surface water transport is represented by a con­tinuous Ekman transport loop, which moves on­shore in the subsurface and offshore in the surface Ekman layer. Ekman transport moves nutrient- rich water into the productive photic zone, which in turn produces biogenic particles, P. A small per­centage of the organic matter, adc • P, is remineral­ized in the deep coastal zone. The level of nutrients or oxygen in the deep coastal zone (dc box in figure 1) is a function of adc • P, the degree of cross-shelf mixing (Kd), and the chemistry of the water enter­ing the deep coastal zone (i.e., the do box). These relationships can be expressed with the mass bal­ance relationship: P04>dc = PO 4, do (1)154 PAU L W. JEWELL Figure 2. Cross-shelf phosphate concentrations at 15°S off the Peru coast. The left-hand panel is from Stations 411-416 and the right-hand panel is from Stations 367-373 of the JOINT-II data set (Hafferty et al. 1978). Note the increase in onshore-offshore phosphorous gradients with depth. In equation (1), Kd has the units of m3/s per meter of longshore distance, i.e., 2-dimensional mass flux (m2/s). The depth of the deep coastal (dc) zone is arbitrary. In other words, the deep coastal zone can be represented as a series of boxes rather than one single box. The analysis presented here considers Table 1. Summary of Onshore-offshore Phosphate and Oxygen Gradients at 15°S Stations Transect length (km) Depth (m) APO4 (ixmol/L) APO4 (ixmol/L) 81-89 55 355 -.16 405 -.03 455 .09 10.1 560 .15 9.6 810 .06 -16.3 367-373 44 330 .53 410 .39 510 .24 15.6 610 .68 0.3 765 .54 7.3 1020 .86 -13.2 382-387 51 410 .07 510 .01 6.0 610 -.11 10.8 765 .10 1.3 960 -.14 -3.1 411-416 50 510 .14 -7.4 610 .21 2.1 770 .20 -10.5 1025 .20 -8.3 "slices" of the water column that are 100 m thick and are not in contact with bottom sediments. Comparison of model output is made with chemi­cal data from similar 100-m thick slices of the open ocean portion in the Peru upwelling system which likewise are not in contact with the bottom sedi­ments. An expression which is similar to equation (1) can be written for oxygen: 2,do P'«A Kd (2) Source. Hafferty et al. (1978). In this case r represents the P04:02 ratio of approx­imately 170 observed in the deep ocean (Takahashi et al. 1985). Equations (1) and (2) are solved by using modern oceanographic data for all variables except Kd. This variable cannot be measured and must be con­strained indirectly. In modern oceanographic settings, the amount of organic matter fluxed through the water column shows considerable spatial and temporal (seasonal to daily) variability. A few long-term (>2 month) summaries of surface productivity in upwelling zones have been published. Among these data sets, the JOINT-II survey at 15°S off the coast of Peru is noteworthy for its nearly continuous occupation of a series of cross-shelf stations over a 3-month period. Average surface productivity within 40 km of the shore at 15°S is reported to be 1400 gC m2/ yr. Within 40-80 km of the shore, productivity isJournal of Geology ORIGI N OF BEDDED BARITES 155 Table 2. Summary of Cross-Shelf Velocities in the Peru Upwelling System Depth (m) Latitude Longitude Cross-shelf velocity (cm/s) 197 5°or 81031' 1.5 ± 4.0 560 do. do. -.3 ± 3.6 860 do. do. -.3 ± 1.9 195 5°00' 81°44' .0 ± 6.2 235 do. do. .0 ± 4.2 415 do. do. -2.1 ± 4.0 183 15°ir 75°34' -.3 ± 3.4 283 do. do. -.2 ± 2.1 115 15°10' 75°36' -1.1 ± 4.7 214 do. do. -.3 ± 3.1 512 do. do. .9 ± 1.6 Source. Brockmann et al. (1980). reported to be 630 gC m2/yr (Packard et al. 1983). In the box model formulation, the width of the do and dc boxes would thus be 40 km. Box model surface productivity, P, would be the observed dif­ference between the 0-40 and 40-80 km widths, i.e., 770 gC m_2/yr. Over the past three decades, sediment-trap data from a variety of oceanographic stations has been used to construct relationships that relate surface productivity, water depth, and subsurface carbon flux. In general, carbon flux decreases exponen­tially with depth. A summary of these relation­ships can be found in Berger et al. (1989). In inter- mediate-depth water (500-1500 m), approximately 1-3% of sinking organic matter is remineralized while sinking through 100 m of water. For the pur­pose of solving equations 1 and 2, a value of 2% is assumed (i.e., adc = .02). The value of adc • P is therefore 15 gC m_2/yr. Estimates of Kd can then be obtained by substituting cross-shelf phosphate concentrations from the Peru upwelling system (table 1) into equation (1). Kd values of approxi­mately 0.05 m2/s are obtained for deep water (>500 m), while values of >0.30 m2/s are obtained for shallow water (<500 m). For oxygen data (table 1), Kd is approximately 0.15 m2/s in deep water. Model Application. Plots of equations (1) and (2) illustrate the most important controls of nutrient enrichment and oxidant reduction in upwelling zones. In natural waters, certain oxidants which remineralize organic matter are energetically fa­vored over others. In seawater the sequence of oxidation is 02 -» NO3 -> Mn+4 -» Fe+3 -> SO4- (Stumm and Morgan 1981). Since the amount of dissolved Mn and Fe in seawater is virtually nonex­istent, the primary oxidants are generally 02, NO3 , SOI", in that order. N03" concentrations in deep seawater are considerably smaller (20-40 |xmoles/L) than either 02 (0-300 |xmoles/L) or SOl~ (approximately 28 millimoles/L) concentra­tions. In marine basins with restricted circulation (e.g., the Black Sea and Cariaco Trench), oxygen- depletion and sulfate-reduction are closely related spatially. In highly productive, open marine set­tings such as the Peru upwelling system, nitrate- reduction is common in intermediate-depth water. In equation (2), "02" values of 0 to -30 ixmoles/ L are therefore taken to represent nitrate-reduc- tion, while values <-30 |xmoles/L represent sulfate-reduction. This is a simplification of ac­tual geochemical processes. For instance, nitrate- reduction commences when 02 concentrations are low, yet greater than zero. For the sake of the pres­ent discussion, the simplification is useful for il­lustrating processes which occur in the deep coastal waters of upwelling zones. The degree of phosphate enrichment and oxygen depletion is relatively minor at high values of Kd (figure 3a). At low values of Kd (approximately .05 m2/s), phosphate enrichments up to 0.5 |xmoles/L in the dc box are possible (figure 3a). Nitrate- and sulfate-reduction can be brought on by increasing the amount of oxidants feeding into the do box (i.e., 02/do) or by decreasing the amount of onshore- offshore mixing, Kd (figure 3b). It is interesting to note that nitrate- and sulfate-reduction is possible at low onshore-offshore mixing rates, regardless of the deep ocean oxygen concentrations which feed into the deep coastal zone. Late Devonian Paleogeography of Western North America Bedded barites, bedded cherts, phosphorites, and black shales are common constituents of the Late Devonian rocks of western North America. A brief overview of these rocks is given as a prelude to consideration of the possible paleoceanographic conditions which led to the formation of the mas­sive bedded barites. Late Devonian sedimentary rocks of the west­ward- facing continental margin of the North American continent in the present-day Great Basin of the United States are characterized by a deep- water siliceous facies, a shallow-water carbonate facies, and a poorly-exposed, transitional facies be­tween the shallow and deep water rocks. These fa­cies have been structurally juxtaposed against each other by the Roberts Mountains thrust fault. The thrusting has typically been considered to be cor­relative with the Antler orogeny of Late Devonian-156 PAUL W. JEWELL A B Figure 3. Calculated concentrations of [a) phosphate (|xmoles/L) and (b) oxygen (|xmoles/L) in the deep coastal [dc) zone as a function of cross shelf mixing (Kd) and the phosphate and oxygen concentrations which enter from the deep offshore (do) layer of the box model in figure 1. Early Mississippian age (Roberts et al. 1958), al­though a Mesozoic age of the fault is advocated by some workers (e.g., Ketner and Smith 1982). The allochthonous, western assemblage rocks are known as the Roberts Mountains Allochthon, a term retained here. The deep-water siliceous rocks of the Roberts Mountains allochthon are composed primarily of Ordovician through Devonian shales, cherts, and siltstones with lesser amounts of greenstone, lime­stone, bedded barite, and phosphorite. A major, barite-bearing Devonian unit within the Roberts Mountain allochthon is the Slaven Chert, which is exposed extensively in central Nevada. The Slaven Chert spans most of the Devonian (Stewart 1980), although the phosphorite- and barite-bearing por­tions appear to be late Late Devonian (Famennian) (Sandberg et al. 1988). Additional phosphorite- and barite-bearing sequences in the Roberts Mountain allochthon include the Late Devonian-Early Mis­sissippian Pinecone sequence in the Toquima Range (Coles 1991) and the Late Devonian Beacon and lower to middle Devonian South Creek units in the Tuscarora Mountains (Dube 1988). The de­positional environment of these rocks appears to be that of siliciclastic-starved, pelagic, and hemi- pelagic sedimentation beneath a zone of high sur­face productivity (Coles 1991). Several studies suggest that the Late Devonian deep-water marine facies in central Nevada were deposited under a coastal upwelling system off the North American continent. Evidence includes the abundance of black shales with very high organic carbon contents (up to 12 wt %) (Poole and Clay- pool 1984), phosphorite lenses and nodules in shales and chert (Coles and Snyder 1985; Coles and Varga 1988; Graber and Chafetz 1990; Jewell and Stallard 1991), and paleogeographic reconstruc­tions which demonstrate an appropriate west- facing, low-latitude position of these rocks during the Late Devonian (Scotese and McKerrow 1990). The presence of calc-alkalic lamprophyres of appar­ent continental or transitional affinity that cross­cut the deep water facies (Dube 1988), siliciclastic facies gradational into shallow water carbonates to the east (Stewart and Poole 1974; Smith and Ketner 1975), and minor detrital carbonate (Coles 1991) suggest that the deep-water facies were formed ad­jacent to a continent. An extremely thick sequence of lower- to mid­dle- Paleozoic carbonate rocks is found throughout central Nevada and western Utah. These carbon­ates represent more or less continuous deposition along a stable cratonic margin from Late Cambrian through Late Devonian (Stewart 1980). Carbonate units dated as Late Devonian in central Nevada were deposited in environments that ranged from deep subtidal to supratidal (Sandberg et al. 1988).Journal of Geology OR IGIN OF BEDDE D BAR ITES 157 Detailed Late Devonian chronostratigraphic corre­lations of carbonate rocks in Nevada and Europe suggests a series of eustatic transgressions and re­gressions (Johnson et al. 1985; Sandberg et al. 1988). The overall trend through the Famennian appears to be regressive (figure 4). One persistent feature of Devonian rocks in western North America is a seaward zone of bank carbonates from Alberta to Nevada as late as the marginifera conodont zone (middle Famennian) (figure 5). These bank carbonates form the upper portion of the Guilmette Formation, which is ex­tensively exposed throughout central Nevada. A less extensive, subtidal bank is believed to have been deposited during the expansa conodont zone (figure 5) (late Famennian). Following this time, carbonate deposition along the continental margin of North America ceased (Sandberg et al. 1988). Proposed Climatic and Bathymetric Setting. De­position of carbonates generally takes place in warm water (>20°C) at latitudes within 30° of the equator (Lees 1975). Bank carbonate deposition along the North American margin appears to have occurred during relatively high sea level stands of a warm climate (Heckel and Witzke 1979; Sand­berg et al. 1988). The regressive portions of the late Late Devonian of North America (figure 4) may have been the result of glaciation on the Gondwana continent (Caputo and Crowell 1985), changes in mid-ocean spreading center rates (John­son et al. 1985), or localized tectonic uplift as a result of the Antler orogeny in the late Late De­vonian (Coles 1991). Coastal upwelling would have brought rela­tively cool water into the continental shelf areas of North America, and for this reason it is unlikely that bank carbonate deposition and upwelling oc­curred concurrently. Although evidence for exten­sive continental glaciation during the Late Devo­nian is sketchy (Caputo and Crowell 1985), it is very likely that warm-cool, Milankovitch-type cli­mate cycles were operative during the 10 m.y. time span of the Fammenian. If so, then the close spatial association of coastal upwelling zone and bank car­bonates suggested by field evidence from the Late Devonian record in North America is not necessar­ily contradictory. Modern coastal upwelling zones are located along continental margins dominated by siliciclas- tic deposition. These margins often have more gen­tle slopes than margins dominated by carbonate deposition (Schlanger and Camber 1986). For in­stance, in the Bahamas banks the steepest slopes (up to 20-30°) are located on the seaward-facing, north slope of the Little Bahamas (Mullins and Neuman 1979). The area is bordered by a 10 km- Figure 4. Conodont zones, dated portions of the upper Slaven Chert, and general sea level curve of the late Frasnian and Famennian stages. Adapted from Johnson et al. (1985) and Sandberg et al. (1988). Symbols Ilb-IIf refer to specific transgressive-regressive cycles of the Late Devonian158 PAUL W. JEWELL Figure 5. General paleogeographic reconstruction showing the position of the proposed phosphogenic province and bank carbonates in the western United States during the Late Devonian. Lower marginifera conodont zone and lower expansa conodont zone are adapted from Sandberg et al. (1988) and Coles (1991). Abbreviated locations (TUS = Tuscarora Mountains, NSR = Northern Shoshone Range, TOQ = Toquima Range) refer to dated bedded barite deposits shown in figure 4 and described by Dube (1988). Dashed pattern represents the Late Devonian phosphogenic province described by Coles (1991). The positions of the TUS, NSR, and TOQ locations relative to the carbonate banks are only approximate and are not palinispastically reconstructed. Deposition of the carbonate banks and phosphatic sediments is not necessarily contemporaneous (see text). wide zone of carbonate talus accumulation. Simi­lar talus zones have not been observed in Late De­vonian rocks of central Nevada, although Coles (1991) does report detrital limestone and rare car­bonate slide blocks in some of the deep-water fa­cies rocks. If the bathymetry of the shallow-water Nevada carbonates was similar to that of the modern-day Bahamas, then the locus of upwelling would occur over relatively deep water. This is in contrast to the high sea-level stands of the modern interglacial ocean in which the most intense productivity of upwelling systems is located over the outer conti­nental shelf (e.g., northwest Africa) or over the shelf break and slope (e.g., Peru). In the Late De­vonian, upwelling adjacent to an inactive carbon­ate bank would cause significant amounts of or­ganic matter to be fluxed into the deep ocean rather than being sequestered on the relatively shallow continental shelf. Origin of the Bedded Barite As shown above, oxygen-depletion and nitrate- reduction is an observed feature of the subsurface waters beneath coastal upwelling zones. In modern settings, complete oxygen- and nitrate-depletion and subsequent sulfate-reduction is only rarely ob­served (Dugdale 1977). In the Late Devonian ocean off North America, factors not present in the mod­ern ocean may have caused sulfate-reduction to be more common, however. Formation of Sulfate-Reducing Water. As men­tioned above, water that feeds into the modern Peru upwelling system has undergone significant nutrient-enrichment and oxygen-depletion as a re­sult of the "nutrient-trapping" effect of equatorial wind divergence and eastward advection by the Pa­cific Equatorial Undercurrent. The longer the resi­dence time of water in this undercurrent, the greater the nutrient-trapping effect (figure 6). In the Late Devonian, the equatorial ocean circled much of the entire globe (Scotese and McKerrow 1990). As such, the equatorial nutrient-trapping mecha­nism observed in the modern Pacific Equatorial Undercurrent would have been very pronounced prior to encountering the west-facing, Laurasian land mass, of which modern North America was a part. Water entering the Late Devonian coastal upwelling system would logically have undergone greater oxygen- and/or nitrate-reduction than that which enters the modern Peru coastal upwelling system. A second factor which may have enhanced ni-Journal of Geology ORIGI N OF BEDDED BARITES 159 Maximum Productivitv Figure 6. Surface nitrate concentrations in the equato­rial Pacific (from Toggweiler et al. 1991). Surface produc­tivity in the equatorial Pacific (from Berger et al. 1989 and Najjar 1992). trate- and/or sulfate-reduction was the bathymetry of the inactive carbonate bank along the North American continental margin. The zone of greatest surface productivity would overlie relatively deep water, rather than shallow water of the continental shelf (figure 7). Relatively high amounts of organic matter would sink into deep water (>400 m) and thus not be under the influence of coastal under­currents (table 2). Cross-shelf mixing at these depths would be minimal, and the water would have been more likely to be nitrate- or sulfate- reducing (figure 3b). As shown previously, nitrate- and/or sulfate-reduction and phosphorous enrich­ment is a natural consequence of low cross-shelf mixing, even when the oxygen concentrations which feed into it are relatively high (figure 3b). It is important to re-emphasize that while ni- trate-reduction is widespread in the Peru upwell­ing system, the total amount of nitrate available as an oxidant in seawater is quite small (<40 fxmoles/L). Even though sulfate reduction has only rarely been observed in open ocean settings, factors peculiar to certain paleoceanographic configura­tions (a globe-encircling equatorial undercurrent and steep carbonate bank bathymetry) could easily consume nitrate and bring about sulfate reduction. Maximum j productivity C>2 minimum zone / /\Sulfate / / \Reduction / / n. Zone zone: of x. COASTAL CURRENT\ INFLUENCE \ Figure 7. Idealized cross-shelf bathymetry for a modern siliciclastic margin and a steep slope carbonate margin which may have existed off North America during the Late Devonian. Barium-Productivity Relationships. Bishop (1988) has shown that particulate barite in sinking or­ganic matter in the water column is the result of sulfate-rich microenvironments in the sinking or­ganic matter. Although most areas of the ocean are undersaturated with respect to pure barite (Church 1979), decaying organic matter provides the neces­sary sulfate to achieve barite supersaturation on a microscopic scale. Uptake of barium into sinking organic matter does not alter the nutrient-like, dis­solved barium profiles observed in the deep ocean (e.g., Chan et al. 1976). Dymond et al. (1992) show that the Corg/Ba ratio of three open ocean localities decreases with re­spect to depth. As sinking organic matter is remin­eralized, additional sulfate is released, thereby fix­ing additional barite. Dymond et al.'s observations160 PAUL W. JEWELL are therefore consistent with Bishop's (1988) the­ory of how particulate barite forms in the water column. A relationship that relates new productiv­ity (i.e., the flux of organic matter that leaves the photic zone under steady state conditions) to bar­ium flux, water-column barium concentrations, and depth has been proposed by Dymond et al. (1992). PBa = 2056{Pn \0.665~ - 0.476 + .004785a r ^ 0.171 Ba111& (3) FSa is particulate barium flux (ixgrams cm 2/yr), Pnew is export productivity, and Ba is dissolved bar­ium in nanomoles/L. Recently, Falkner et al. (1993) found that the barite-forming mechanism of Bishop (1988) was re­versed (i.e., dissolved barium increases) when sink­ing organic matter encountered deep anoxic seawa­ter in the Cariaco Trench, Black Sea, and an anoxic fjord in Norway. The increase in dissolved barium is pronounced at low H2S concentrations (figure 8). Implications of the Dymond et al. (1992) and Falk­ner et al. (1993) studies are directly applicable to the formation of bedded barites in the Late De­vonian ocean. Formation of Bedded Barites. Many previous re­searchers consider the barites in deep water facies of the Roberts Mountain allochthon to have a hy­drothermal exhalative origin (e.g., Papke 1984; Poole 1988; Dube 1988; Maynard and Okita 1992). Other than small amounts of diagenetic pyrite, base metal sulfides are very rare in these sedi­ments, however. Furthermore, stable isotope data suggest that 34S and possibly 87Sr/86Sr in the bedded barite was equivalent to that of contemporaneous Devonian seawater (Rye et al. 1978). Jewell and Stallard (1991) recently elaborated on an idea origi­nally proposed by Shawe et al. (1969) that bedded barite in the Shoshone and Toquima Ranges of cen­tral Nevada was formed by biogenic processes in a highly productive marine setting. Here, a deposi­tional model for the bedded barite is outlined within the context of a sulfate-reducing upwelling system outlined above and geochemical studies of barium from the modern ocean. Organic particles formed by export productivity (i.e., the net loss of organic carbon from the photic zone) would scavenge dissolved barium from the oxygenated, upper portions of the water column. As the waters were advected in the longshore di­rection and encountered sulfate-reducing waters at depth, barium would be released from the organic particles in a manner similar to that observed in modern anoxic settings (figure 8). The sulfate- reduction zone of the upwelling system would cause dissolved barium concentrations in water exiting the sulfate-reduction zone to increase dra­matically (figure 9). Organic particles settling through oxygenated, barium-rich water outside the sulfate-reduction zone would result in very high barium flux to the sediments (equation 3, figure 10). The net effect of the upwelling zone would therefore be an "upgrade" of barium flux to sedi­ments downstream from the upwelling zone. Two points of this genetic model for bedded bar­ites merit special emphasis. First, increased dis­solved Ba in the sulfate-reduction zone could occur at very low (tens of |xmoles/L) H2S concentrations (figure 8). In other words, because (02 + NO^) con­centrations are low in some modern upwelling zones, only minimal additional oxidant (02 + NO3 + Son reduction (the order of tens of (xmoles/L) would be necessary to bring about sig­nificantly higher barite flux. Second, modest in­creases in dissolved barium exiting a sulfate- reduction zone would dramatically increase barite flux to the sediments (figure 10). For instance, if Figure 8. Plots of dissolved H2S (closed circles) and dissolved Ba (open triangles) for three modern anoxic settings (data from Falkner et al. 1993 and Millero 1991). Cariaco Trench H2S(mM) Framveran Fjord H2S(mM) Black Sea H2S(/x M)Journal of Geology ORIGI N OF BEDDED BAR ITES 161 O2 minimum zone normal Ba flux low Ba flux high Ba flux Figure 9. Diagrammatic representation of the long­shore dimension of a sulfate-reducing coastal upwelling zone. Normal Ba flux to the sediments would be ex­pected below an 02 minimum zone similar to that found in the modem ocean. If the upwelling resulted in H2S production, barite particles produced by export produc­tivity [Pnew] would dissolve, resulting in low Ba flux to the underlying sediments. Longshore currents would ad- vect the Ba-rich water out of the sulfate-reducing zone, where sinking organic matter would produce elevated Ba flux to the sediments. Figure 10. Barite flux (ixgrams cm_2/yr) as a function of new productivity (Pnew) (gC m_27yr) and dissolved bar­ium concentrations (nanomoles/L) in the water column. New productivity is assumed to be one-half of nearshore productivity reported by Packard et al. (1983) for the Peru upwelling system. Arrow shows the expected dissolved barium increases which would result from modest sul­fate reduction (figure 8). the dissolved barium concentrations in the mod­ern Pacific Ocean were doubled, barite mass flux rates computed with equation 3 would be compara­ble to published rates of Mesozoic bedded chert deposition (table 3). This is consistent with field observations of economic barite deposits in the Roberts Mountains allochthon, which often show roughly equal barite-to-(chert +shale) volumes (Papke 1984). Discussion Many mid-Paleozoic rocks of North America other than the Roberts Mountains allochthon are charac­terized by abundant black shale, barite, phospho­rite, and chert deposition. Examples of Middle Devonian to Middle Mississippian barite occur­rences with no associated metal sulfides include the Jefferson Formation of Montana (Berg 1988), the Lower Earn Group of the Yukon Territory (Ly- don et al. 1985), the Hanover Shale of upstate New York (Pepper et al. 1985), the Cleveland and Bed­ford Shales of Ohio (Holden and Carlson 1979), the Stanley Formation of Arkansas (Howard and Ha- nor 1987), and the Needmore Shale, Marcellus- Millboro Shale, and the Tully Limestone of the central Appalachians (Nuelle and Shelton 1986; Clark and Mosier 1989). Many of these occurrences are thought to have been formed by seafloor hydro­thermal processes (e.g., Lydon et al. 1985; Nuelle and Shelton 1986), while others are believed to be the result of cycling through anoxic zones in a manner similar to that described here (e.g., Clark and Mosier 1989). Most of the barite occurrences mentioned above are associated with phosphatic black shales and cherts. Circumstantial evidence of highly productive, nutrient-rich surface waters therefore exists in these rock sequences, although most are not adjacent to bank carbonates. It is interesting to note that all barite occur­rences mentioned above occur along continental Table 3. Summary of Bedded Chert Sedimentation Rates and Possible Barite Sedimentation Rates Age/Location Sedimentation rate (g cm_2/103 yr) Upper Jurassic/Austria-Germany .3-3.0 Upper Jurassic/Italy-Switzerland 1.0 Upper Jurassic/California 1.5 Upper Triassic/Japan 8 Bedded barite .05-.8 Source. Chert sedimentation rates from data of Jenkyns and Winterer (1982); barite sedimentation rates from Dymond et al. (1992) and figure 10.162 PAUL W. JEWELL margins within 30° of the proposed Middle De­vonian to Late Mississippian paleo-equator (e.g., paleogeographic reconstructions of Scotese and McKerrow 1990). The chemistry of these tropical/ subtropical waters may have become oxygen- depleted and nutrient-rich in the same manner as modern waters from similar latitudes. In both the modern Pacific and Atlantic Oceans, equatorial undercurrents move water from west-to-east be­neath productive waters of equatorial upwelling. In a single large, mid-Paleozoic ocean, this equatorial "nutrient-trap" phenomenon may have been more pronounced than it is in modern ocean basins. Upon intersecting continental margins, nutri­ent- rich equatorial waters of the modern ocean are deflected north and south, where coastal upwelling causes even higher productivity (figure 6), adds ad­ditional nutrients, and further depletes available oxidants. If the coastal upwelling system were lo­cated next to a steep, inactive carbonate bank dur­ing the Late Devonian, additional factors would come into play. The most intense portion of the upwelling zone would overlie deep water, and sinking organic matter would be remineralized in the deep portions of the water column. Onshore- offshore mixing in these deep waters would be minimal, further depleting the oxygen and nitrate concentrations. Given the low (02 + NO^) con­centrations in modern upwelling systems such as the one off Peru (<30 |xmoles/L; Hafferty et al. 1978), sulfate reduction in a late Late Devonian upwelling system could be brought on by a de­crease in deep water mixing rates, a decrease in the oxygen concentrations feeding into the upwelling zone (figure 6), an increase in upwelling intensity due to increased zonal winds of a cool climate, or a combination of these three factors. ACKNOWLEDGMENTS Robbie Toggweiler clarified my thinking on the role of onshore/offshore mixing processes in up­welling zones. Richard Barber and Judy Parrish challenged and sharpened some of the ideas pre­sented here. Kathy Nichols, Norm Silberling, Brian Witzke, and Lee Kump constructively reviewed the manuscript. Speculations about Devonian paleoge- ography are strictly those of the author, however. Acknowledgment is made to the Donors of The Petroleum Research Fund, administered by the American Chemical Society for partial support of this research. REFERENCES CITED Berg, R. B., 1988, Barite in Nevada: Montana Bur. Mines and Geol. Mem. 61, 100 p. Berger, W. H., 1989, Global maps of ocean productivity, in Berger, W. H.; Smetacek, V. S.; and Wefer, G., eds., Productivity in the Ocean: Present and Past: Dahlem Workshop Reps. (Life Sci. Res. Rep. 44): New York, Wiley, p. 429-455. --------; Smetacek, V. S.; and Wefer, G., 1989, Ocean pro­ductivity and paleoproductivity-an overview, in Berger, W. H.; Smetacek, V. S.; and Wefer, G., eds., Productivity in the Ocean: Present and Past: Dahlem Workshop Reps. (Life Sci. Res. Rept. 44): New York, Wiley, p. 1-34. Bishop, J. K. B., 1988, The barite-opal-organic carbon association in oceanic particulate matter: Nature, v. 332, p. 341-343. Brockmann, C.; Fahrbach, E.; Huyer, A.; and Smith, R. L., 1980, The poleward undercurrent along the Peru coast: 5 to 15°S.: Deep Sea Res., v. 27A, p. 847-856. Burnett, W. C.; Veeh, H. H.; and Soutar, A., 1980, U- series, oceanographic and sedimentary evidence in support of recent formation of phosphate nodules off Peru, in Bentor, Y., ed., Phosphorites: Soc. Econ. Pa- leont. Mineral. Spec. Pub. 29, p. 61-71. Caputo, M. V., and Crowell, J. C., 1985, Migration of glacial centers across Gondwana during Paleozoic era: Geol. Soc. America Bull., v. 96, p. 1020-1036. Chan, L. H.; Edmond, J. M.; Stallard, R. F.; Broecker, W. S.; Chung, Y. C; Weiss, R. F.; and Ku, T. L., 1976, Radium and barium at GEOSECS stations in the At­lantic and Pacific: Earth Planet Sci., v. 32, p. 258-267. Church, T. M., 1979, Marine barite, in Burns, R. G., ed., Marine minerals (Reviews in Mineral. 6): Mineral. Soc. America, p. 170-210. Clark, S. H. B., and Mosier, E. L., 1989, Barite nodules in Devonian shale and mudstone of western Virginia: U.S. Geol. Survey Bull. 1880, 30 p. Codispoti, L. A., 1983, On variability and sediments in upwelling regions, in Suess, E., and Thiede, J., eds., Coastal Upwelling: Its Sediment Record. Part A: Responses of the Sedimentary Regime to Present Coastal Upwelling: New York, Plenum Press, p. 125­145. Coles, K. S., 1991, Depositional provinces in Nevada dur­ing the Famennian, in Cooper, J. D., and Stevens, C. H., eds., Paleozoic paleogeography of the western United States-II: Pacific SEPM, v. 67, p. 767-782. --------, and Snyder, W. S., 1985, Significance of lower and middle Paleozoic phosphatic chert in the To- quima Range, central Nevada: Geology, v. 13, p. 573­576. Journal of Geology OR IGIN OF BEDDE D BAR ITES 163 --------, and Varga, R. J., 1988, Early to middle Paleozoic phosphogenic province in terranes of the southern Cordillera, western United States: Am. Jour. Sci., v. 288, p. 891-924. Conant, L. C., and Swanson, V. E., 1961, Chattanooga Shale and related rocks of central Tennessee and nearby areas: U.S. Geol. Survey Prof. Paper 357, 91 p. Dube, T. H., 1988, Tectonic significance of Upper De­vonian igneous rocks and bedded barite, Roberts Mountains allochthon, Nevada, USA, in McMillan, N. J.; Embry, A. F.; and Glass, D. J., Devonian of the World, Vol. 2: Sedimentation (Proc. 2nd Int. Sym. Devonian system): Canadian Soc. Petrol. Geol. Mem. 14, p. 235-247. Dugdale, R. C.; Georing, J. J.; Barber, R. T.; Smith, R. L.; and Packard, T. T., 1977, Denitrification and hydro­gen sulfide in the Peru upwelling region during 1976: Deep Sea Res., v. 24, p. 601-608. Dymond, J.; Suess, E.; and Lyle, M., 1992, Barium in deep-sea sediment: a geochemical proxy for paleopro- ductivity: Paleoceanography, v. 7, p. 163-181. Ettensohn, F. R., 1985, Controls on development of Cats- kill Delta complex basin facies, in Woodrow, D. L., and Sevon, W. D., eds., The Catskill Delta: Geol. Soc. America Spec. Paper 201, p. 65-77. Falkner, K. K.; Klinkhammer, G. P.; Bowers, T. S.; Todd, J. F.; Lewis, B. L.; Landing, W. M.; and Edmond, J. M., 1993, The behavior of barium in anoxic waters: Geochim. Cosmochim. Acta, v. 57, p. 537-554. Goldberg, E. D., and Arrhenius, G. O. S., 1958, Chemis­try of Pacific pelagic sediments: Geochim. Cosmo­chim. Acta, v. 13, p. 153-212. Graber, K. K., and Chafetz, H. S., 1990, Petrography and origin of bedded barite and phosphate in the Devonian Slaven Chert of central Nevada: Jour. Sed. Petrol., v. 60, p. 879-911. Hafferty, A. J.; Codispoti, L. A.; and Huyer, A., 1978, JOINT-n, R/V Melville Legs I, II, and IV; R/V Iselin Leg II bottle data, March-May, 1977: Dept. Oceanogr. Data Rept. 45, University of Washington, Seattle, 779 p. Hagen, E., 1981, Mesoscale upwelling variations off the west African coast, in Richards, F. A., ed., Coastal upwelling: Am. Geophys. Union Coastal and Estua- rine Sci. Series, v. 1, p. 72-78. Fleckel, P. H., and Witzke, B. J., 1979, Devonian world paleogeography determined from distribution of car­bonate and related lithic, paleoclimatic indicators, in House, M. R., et al., eds., The Devonian system: Spe­cial Papers in Paleontology 23, p. 99-123. Holden, W. F., and Carlson, E. H., 1979, Barite concre­tions from the Cleveland Shale in north-central Ohio: Ohio Acad. Sci., v. 1979, p. 227-232. Howard, K. W., and Hanor, J. S., 1987, Compositional zoning in the Fancy Hill stratiform barite, Ouachita Mountains, Arkansas, and evidence for the lack of associated massive sulfides: Econ. Geol., v. 82, p.1377-1385. Jenkyns, H. C., and Winterer, E. L., 1982, Paleoceanogra­phy of Mesozoic ribbon radiolarites: Earth Planet. Sci. Letters, v. 60, p. 351-375. Jewell, P. W., and Stallard, R. F., 1991, Geochemistry and paleoceanographic setting of central Nevada bedded barites: Jour. Geology, v. 99, p. 151-170. Johnson, J. G.; Klapper, G.; and Sandberg, C. A., 1985, Devonian eustatic fluctuations in Euramerica: Geol. Soc. America Bull., v. 96, p. 567-587. Johnson, J. A., and Rockliff, N., 1986, Shelf break circula­tion processes, in Mooers, C. N. K., ed., Baroclinic processes on continental shelves: Am. Geophys. Union Coastal and Estuarine Sci. Series, v. 3, p. 33­62. Ketner, K. B., and Smith, J. F., Jr., 1982, Mid-Paleozoic age of the Roberts thrust unsettled by new data from northern Nevada: Geology, v. 10, p. 298-303. Lea, D. W.; Shen, G. T.; and Boyle, E. A., 1989, Coralline barium records temporal variability in equatorial up­welling: Nature, v. 340, p. 373-376. Lees, A., 1975, Possible influence of salinity and temper­ature on modern shelf carbonate sedimentation: Mar. Geol., v. 19, p. 159-198. Lydon, J. W.; Goodfellow, W. D.; and Jonasson, I. R., 1985, A general model for stratiform baritic deposits of the Selwyn Basin, Yukon Territory and District of MacKenzie: Geol. Surv. Canada Paper 85-1 A, p. 651­660. Maynard, J. B., and Okita, P. M., 1992, Bedded barite deposits in the United States, Canada, Germany, and China: two major types based on tectonic setting: Econ. Geol., v. 86, p. 364-376. Millero, F. J., 1991, The oxidation of H2S in the Fram- varen Fjord: Limnol. Oceanogr., v. 36, p. 1007-1014. Mullins, H. T., and Neumann, A. C., 1979, Deep carbon­ate bank margin structure and sedimentation in the northern Bahamas, in Doyle, L. J., and Pilkey, O. H., Jr., eds., Geology of continental slopes: Soc. Econ. Pa- leontol. Mineral., v. 27, p. 165-192. Nuelle, L. M., and Shelton, K. L., 1986, Geological and geochemical evidence of bedded barite potential in Devonian rocks, Valley and Ridge Province, Appala­chians: Econ. Geol., v. 81, p. 1408-1430. Najjar, R. G., 1992, Marine biogeochemistry, in Tren- berth, K. E., ed., Climate System Modeling: Cam­bridge, Cambridge University Press, p. 241-280. Packard, T. T.; Garfield, P. C.; and Codispoti, L. A., 1983, Oxygen consumption and denitrification below the Peruvian upwelling, in Suess, E., and Theide, J., eds., Coastal Upwelling: Its Sedimentary Record. Part A: Responses of the Sedimentary Regime to Pres­ent Coastal Upwelling: New York, Plenum Press, p. 147-173. Papke, K. G., 1984, Barite of Nevada: Nevada Bur. Mines Geol. Bull. 98, 125 p. Parrish, J. T., 1982, Upwelling and petroleum source beds, with reference to the Paleozoic: Am. Assoc. Pe­trol. Geol. Bull., v. 66, p. 750-774. --------, and Curtis, R. L., 1982, Atmospheric circulation, upwelling, and organic-rich rocks in the Mesozoic and164 PAUL W. JEWELL Cenozoic eras: Paleogeog., Paleoclimat., Paleoecol., v. 40, p. 31-66. Pepper, J. F.; Clark, S. H. B.; and de Witt, W.; Jr., 1985, Nodules of diagenetic barite in Upper Devonian shales of western New York: U.S. Geol. Survey Bull. 1653, 11 p. Pickard, G. L., and Emery, W. J., 1983, Descriptive Physi­cal Oceanography: Oxford, Pergamon Press, 249 p. Poole, F. G., 1988, Stratiform barite in Paleozoic rocks of the western United States, in Zachrison, E., ed., Proc. 7th IAGOD Sym., Stuttgart, E. Schweizer- bart'sche Verlagsbuchhandlung, p. 309-319. --------, and Claypool, G. E., 1984, Petroleum source-rock potential and crude-oil correlation in the Great Basin, in Woodward, J.; Meissner, F. F.; and Clayton, J. L., eds., Hydrocarbon Source Rocks of the Greater Rocky Mountain Region: Rocky Mountain Assoc. Geol., p. 179-229. Roberts, R. J.; Hotz, P. E.; Gilluly, J.; and Ferguson, H. G., 1958, Paleozoic rocks of north-central Nevada: Am. Assoc. Petrol. Geol. Bull., v. 42, p. 2813-2857. Rye, R. O.; Shawe, R.; and Poole, F. G., 1978, Stable isotope studies of bedded barite at East Northumber­land Canyon in Toquima Range, Central Nevada: U.S. Geol. Survey Jour. Res., v. 6, p. 221-229. Sandberg, C. A.; Poole, F. G.; and Johnson, J. G., 1988, Upper Devonian of western United States, in McMil­lan, N. J.; Embry, A. F.; and Glass, D. J., eds., Devo­nian of the world, Vol. 1: Regional Syntheses: Proc. 2nd Int. Sym. Devonian System: Can. Soc. Petrol. Geol. Mem. 14, p. 183-220. Schopf, T., 1983, Paleozoic black shales in relation to continental margin upwelling, in Suess, E., and Thiede, J., eds., Coastal Upwelling: Its Sediment Rec­ord. Part B: Sedimentary Records of Ancient Coastal Upwelling: New York, Plenum Press, p. 579-596. Schlanger, W., and Camber, O., 1986, Submarine slope angles, drowning unconformities, and self-erosion of limestone escarpments: Geology, v. 14, 762-765. Schmitz, B., 1987, Barium, equatorial high productivity, and the northward wandering of the Indian continent: Paleoceanog., v. 2, p. 63-77. Scotese, C. R., and McKerrow, W. S., 1990, Revised world maps and introduction, in McKerrow, W. S., and Scotese, C. R., eds., Paleozoic paleogeography and biogeography: Geol. Soc. (London) Mem. 12, p. 1-21. Shawe, D. R.; Poole, F. G.; and Brobst, D. A., 1969, Newly discovered bedded barite in East Northum­berland Canyon, Nye County, Nevada: Econ. Geol., v. 64, p. 245-254. Sheldon, R. P., 1980, Episodicity of phosphate deposition and deep ocean circulation-a hypothesis, in Bentor, Y., ed., Phosphorites: SEPM Spec. Pub. 29, p. 239­247. Shimmield, G. B.; Price, N. B.; and Kahn, A. A., 1988, The use of Th-230 and Ba as indicators of paleopro- ductivity over a 300 Kyr timescale-evidence from the Arabian Sea: Chem. Geol., v. 70, p. 112. Smith, J. F., Jr., and Ketner, K. B., 1975, Stratigraphy of Paleozoic rocks in the Carlin-Pinon Range area, Ne­vada: U.S. Geol. Survey Prof. Paper 867-A, p. A1-A87. Smith, R. L., 1981, Descriptions and comparisons of spe­cific upwelling systems, in Richards, F. A., Coastal upwelling systems: Am. Geophys. Union Coastal and Estuarine Sci. Ser., v. 1, p. 107-118. --------, 1992, Coastal upwelling in the modern ocean, in Summerhayes, C. P.; Prell, W. L.; and Emeis, K.-C., eds., Upwelling systems: evolution since the Early Miocene: Geol. Soc. (London) Spec. Pub. 64, p. 9-28. Southam, J. R., and Peterson, W. H., Transient response of the marine carbon cycle, in Sundquist, E. T., and Broecker, W. S., eds., The carbon cycle and atmo­spheric C02: natural variations Archean to present: Am. Geophys. Union Mon. Series, v. 32, p. 89-98. Stewart, J. H., 1980, Geology of Nevada: Nevada Bur. Mines Geol. Spec. Pub. 4, 136 p. --------, and Poole, F. G., 1974, Lower Paleozoic and up­permost Precambrian miogeocline, Great Basin, west­ern United States, in Dickinson, W. R., ed., Tectonics and Sedimentation: SEPM Spec. Pub. 22, p. 28-57. Stumm, W., and Morgan, J. J., 1981, Aquatic chemistry: New York, Wiley, 780 p. Suess, E.; Kulm, L. D.; and Killingley, J. S., 1987, Coastal upwelling and a history of organic-rich mudstone de­position off Peru, in Brooks, J., and Fleet, A. J., eds., Marine petroleum source rocks: Geol. Soc. (London) Spec. Pub. 26, p. 181-197. Takahashi, T.; Broecker, W. S.; and Langer, S., 1985, Redfield ratio based in chemical data from isopycnal surfaces: Jour. Geophys. Res., v. 90, p. 6907-6924. Toggweiler, J. R.; Dixon, K.; and Broecker, W. S., 1991, The Peru upwelling and the ventilation of the South Pacific thermocline: Jour. Geophys. Res., v. 96, p. 20,467-20,479. Turekian, K. K., and Johnson, D. G., 1966, The barium distribution in seawater: Geochim. Cosmochim. Acta, v. 30, p. 1153-1174. von Breymann, M. T.; Emeis, K.-C.; and Camerlenghi, A., 1990, Geochemistry of sediments from the Peru upwelling area: results from sites 680, 682, 685, and 688, in Suess, E., et al., eds., Proc. ODP Sci. Results 112: Ocean Drilling Program, College Station, TX, p. 491-503. Walsh, J. J., 1975, A spatial simulation model of the Peru upwelling ecosystem: Deep Sea Res., v. 22, p. 201-236. Wilde, P., 1987, Model of progressive ventilation of the late Precambrian-early Paleozoic ocean: Am. Jour. Sci., v. 287, p. 442-459. Wyrtki, K., 1962, The oxygen minimum in relation to ocean circulation: Deep Sea Res., v. 9, p. 11-23.