The National Cave and Karst Research Institute is pleased
to publish Dr. Doug Kirkland's monograph on the role of
hydrogen sulfide on speleogenesis in the Guadalupe Mountains
and western Delaware Basin. Dr. Kirkland's work builds on his
many years of research in southeastern New Mexico and west
Texas, and provides the most comprehensive overview of cave
and karst phenomena in the greater Delaware Basin region in
almost 20 years. His work incorporates and summarizes decades
of research by previous workers, combined with new ideas he
has developed on speleogenesis in the Guadalupe Mountains. We
feel confident that this publication will serve as an
important source book and milestone for future research in
the Delaware Basin region for many years to come. Lewis Land
Managing Editor February 19, 2014
NATIONAL CAVE AND KARST RESEARCH INSTITUTE SPECIAL PAPER 2 ROLE OF HYDROGEN SULFIDE IN THE FORMATION OF CAVE AND KARST PHENOMENA IN THE GUADALUPE MOUNTAINS AND WESTERN DELAWARE BASIN, NEW MEXICO AND TEXAS DOUGLAS W. KIRKLAND
Published and distributed by National Cave and Karst Research Institute Dr. George Veni, Executive Director 400-1 Cascades Ave. Carlsbad, NM 88220 USA www.nckri.org Peer-review: Harvey DuChene, Carol Hill, Lewis Land, Arthur Palmer, Victor Polyak, and George Veni. The citation information: Kirkland, D.W. Formation of Cave and Karst Phenomena in the Guadalupe Mountains and Western Delaware Basin, New Mexico and Texas. Carlsbad (NM): National Cave and Karst Research Institute. ISBN 978-0-9910009-1-3 MANAGING EDITOR Lewis Land New Mexico Bureau of Geology and Mineral Resources and the National Cave and Karst Research Institute Produced with the assistance of the University of South Florida Tampa Library. Cover Photo: Room of Carlsbad Cavern, apparently represents a period of water table stability during one of the latest episodes of cavern enlargement (Palmer et al., 2009) (Photo by A.N. Palmer).
The National Cave and Karst Research Institute is pleased to publish Dr. Doug Guadalupe Mountains and western Delaware Basin. Dr. Kirklands work builds on his many years of research in southeastern New Mexico and west Texas, and provides the most comprehensive overview of cave and karst phenomena in the greater Delaware Basin region in almost 20 years. His work incorporates and summarizes decades of research by previous workers, combined with new ideas that this publication will serve as an important source book and milestone for future research in the Delaware Basin region for many years to come. Lewis Land Managing Editor February 19, 2014 FOREWORD i
CONTENTS Foreword .................................................................................................................................... i Abstract ..................................................................................................................................... 1 Introduction .............................................................................................................................. 3 Objectives and Purpose of Investigation ...................................................................... 4 ........................................... 5 Geographic and Stratigraphic Setting of Caves .......................................................... 5 Peculiar Qualities and Unusual Origin of Caves ........................................................... 6 Role of Sulfuric Acid in Cave Formation ........................................................................ 7 ............................................ 9 Curious Sulfur Isotopic Composition ......................................................................... 9 Microbial Derivation ................................................................................................. 11 Immense Quantity .................................................................................................... 12 ................................. 12 The Existing Shelfal Model ........................................................................................ 13 ......................................................................................... 13 .......................................................................... 14 The Existing Basinal Model ....................................................................................... 15 ........................................................................ 15 .................... 17 .............................................. 18 and its Pathway to Caves .............................................................................................. 19 Beginning of Intense Cavern Formation ................................................................. 20 Formation of Basinal Hydrologic Pathways ........................................................... 21 ................................................ 21 ii
............................................ 23 ........................ 24 ............................................................................... 25 ................ 26 ......................................... 28 .............................................. 29 ................................. 33 .............................................................................. 33 ...................................... 34 Generation and Migration of Methane in Late Tertiary of Western Delaware Basin .......................................................................................... 35 .................................................................................................... 35 .................................................... 36 ............................................................. 37 .................................................. 37 Reaction between Methane and Sulfate Anions in Late-Tertiary of Western Delaware Basin ................................................................ 38 ..................................... 38 ............................... 39 ................................................................................ 39 .................... 40 ................................................................................................. 41 Conduits within Castile Halite into Capitan Formation ......................................... 43 Southwest to Northeast along the Cave Belt and from Higher to Lower Elevations within Individual Caves ............................................. 44 iii
Dissolution of Castile and Salado Halite and Gypsum by Epigenic Groundwater .................................................................................................. 46 .................................................................... 49 Similarities and Differences in Genetic Processes between the Caves and the Major Sulfur Deposits .................................................................... 49 The Hydrologic Pathways .............................................................................................. 49 The Oxidizing Agent ....................................................................................................... 51 Source of Aqueous Oxygen .......................................................................................... 51 Descent of Epigenic Groundwater and Aqueous Oxygen into Late Permian Evaporites ................................................................................................ 52 Precipitation of Native Sulfur ......................................................................................... 52 .................................................................................... 55 Microbial Foodstuff at the Subsurface Masses of Biogenic Limestone .................... 55 Negative Carbon Isotopic Values ......................................................................... 55 Crude Oil: An Unimportant Reductant of Sulfate .................................................. 57 Anaerobic Reduction of Sulfate Anions by Methane in Limestone-Hosted Sulfur Deposits Elsewhere .............................................................. 58 Anaerobic Reduction of Sulfate Anions by Methane in Marine Diagenetic Environments .............................................................................................. 58 Anaerobic Reduction of Sulfate Anions by Methane in Terrestrial Diagenetic Environments Elsewhere .......................................................... 59 Summary and Conclusions .............................................................................................. 61 Acknowledgments .............................................................................................................. 65 ................................................................................................................. 66 iv
LIST OF FIGURES Figure 1 .................................................................................................................... 3 Figure 2 .................................................................................................................... 3 Figure 3 .................................................................................................................... 4 Figure 4 .................................................................................................................... 4 Figure 5 .................................................................................................................... 5 Figure 6 .................................................................................................................... 5 Figure 7 .................................................................................................................... 6 Figure 8 .................................................................................................................... 7 Figure 9 .................................................................................................................... 8 Figure 10 .................................................................................................................. 9 Figure 11 ................................................................................................................ 10 v
Figure 12 ................................................................................................................ 10 34 34 Figure 13 ................................................................................................................ 11 Figure 14 ................................................................................................................ 12 Figure15 ................................................................................................................. 15 Figure 16 ................................................................................................................ 17 Figure 17 ................................................................................................................ 17 Figure 18 ................................................................................................................ 18 Figure 19 ................................................................................................................ 20 Figure 20 ................................................................................................................ 22 Figure 21 ................................................................................................................ 23 Figure 22 ................................................................................................................ 23 Figure 23 ................................................................................................................ 24 vi
Figure 24 ................................................................................................................ 27 Figure 25 ................................................................................................................ 28 Figure 26 ................................................................................................................ 31 Figure 27 ................................................................................................................ 31 Figure 28 ................................................................................................................ 32 Figure 29 ................................................................................................................ 40 Figure 30 ................................................................................................................ 42 Figure 31 ................................................................................................................ 43 2 Figure 32 ................................................................................................................ 50 Figure 33 ................................................................................................................ 56 13 13 vii
1 until the voids contacted the smooth, intact base of a bed of Castile anhydrite (solubility ~1/140 of that of halite), all beds of which dipped uniformly eastward by <1-2. The conduits, except for their anhydritic ceiling, westward by convective dissolution directly up the slight slope of the paleo-Guadalupe tectonic block. Many conduits eventually terminated at the nearly vertical face of the youngest-most Capitan paleo-reef or at the steepto-shallow face of the youngest-most paleo-forereef, both of which were in side-by-side contact with beds of Castile anhydrite and halite. Basinal stratal temperatures transiently increased shortly before and as the conduits were forming resulting in generation of billions of cubic meters of methane (CH 4 ). Much gaseous CH 4 ascended into the Castile evaporites at the same localities at which groundwater convectively rose and sank. The gas progressively dissolved within ambient water beneath a thick (~1 km) sealing cover of reacted with SO 4 2derived from dissolution of Castile anhydrite. The reaction, aided by enzymes of anaerobic microbes, generated many millions of metric tons of both aqueous H 2 S and aqueous CO 2 The CO 2 reacted instantaneously with Ca 2+ liberated as CaSO 4 dissolved, replacing laminated, nodular, massive, and brecciated Castile anhydrite with permeable limestone. The anhydrite-encased limestone bodies, commonly with dimension, in plan, >30 m, formed at ~1000 scattered localities. Pressurized artesian groundwater transported the H 2 S from the carbonate bodies into the conduits within the homoclinal slope and by forced convection moved through fractures and pores of the Capitan Formation and adjacent shelfal carbonates, and descended to low levels because of a relatively high density imparted by dissolved halite. The H 2 throughout the late Miocene and early Pliocene within a basin-margin carbonate aquifer that formed a narrow (~6 km) northeast-trending belt across the eastward-dipping paleo-Guadalupe tectonic block. The highest part of the belt, therefore, was to the far southwest. Here, west-toThis monograph provides a theory for multiple stages of speleogenesis related to production, transportation, and Basin and along the margin of the basin in the vicinity of the Guadalupe Mountains (southeastern New Mexico, USA). Large caves in the Guadalupe Mountains formed during the late Miocene and early Pliocene (~12-4 Ma ago). They originated dominantly from sulfuric acid (H 2 SO 4 ), a powerful cave-forming agent that dissolved both limestone of an ancient sponge-algal reefthe Capitan Formation (Middle Permian; ~270-260 Ma ago)and limestone and dolomite of age-equivalent, near-backreef (shelfal) strata. The reef-front formed the boundary between the Guadalupe Mountains to the northwest and the Delaware Basin to the southeast. The strong acid was produced as dissolved oxygen (O 2 ) from the earths atmosphere reacted with aqueous 2 S) from the adjacent basin. The H 2 S originated within and migrated from the Castile Formation (earliest Late Permian, ~260.0-259.8 Ma Tertiary events: Delaware Basin, and eastward uniform tilting (ultimately by 1 to 2) of the huge paleo-Guadalupe tectonic block (including the mountains and much of the basin). The Castile before extensive Late Tertiary dissolution extended throughout the basin and consisted of thick (tens of meters), remarkably persistent, alternating beds of halite (NaCl) and anhydrite (CaSO 4 ). Permian aquifers eastward down the sporadically rising tectonic block. Much groundwater then rose along fractures generated or regenerated during the tilting and dissolved Castile evaporites at hundreds of local sites vehicle of its own dissolution. To replace the sinking brine, the least dense, least saline, most solutionally aggressive groundwater persistently rose to the highest accessible elevation. The groundwater dissolved chambers vertically upward through thick-bedded halite ABSTRACT
2 east-trending erosion initially removed the impermeable cover of mainly Salado and Rustler evaporitic strata (Late Permian; ~259.8-250.0 Ma ago) and groundwater initially fell allowing atmospheric O 2 to enter the uppermost level of incipient caves. H 2 S-H 2 SO 4 speleogenesis occurred when H 2 S degassed from cave pools and when atmospheric O 2 moved into the caves. The gaseous O 2 probably entered permeable carbonates that cropped out in southwestern highlands; it then descended laterally through fractures beneath sealing evaporites. The H 2 S and O 2 dissolved within subaerial water on carbonate wall rocks and reacted completely (aided by bacterial enzymes) to form H 2 SO 4 Then, over a span of ~8 Ma, each episodic uplift of the tectonic block resulted in further erosion of the cover, deeper descent of the groundwater table, further progression of speleogenesis southeastward along the belt, and deeper penetration of speleogenesis within carbonates of the cave belt. Within 12 to ~50 km southeast of the cave belt, genetically related karstic processes formed deposits of native sulfur dispersed within biogenic limestone and encased within Castile and Salado anhydrite. The caves and the sulfur deposits owe their origin to a coincidence of essentially the same stratigraphic, tectonic, thermal, and biogenic events. The sulfur deposits occur along graben-bounding faults that breached both the Castile (~30% halite; ~0.5 km thick) and the directly overlying Salado (~85% halite; ~0.5 km thick) and extended to the surface. The faults guided hypogenic groundwater upward by forced convection, and during subsequent free convection, the returning brine locally increased the permeability of the steep fracture pathways through bedded anhydrite. Gaseous CH 4 migrated upward along the same pathways. It dissolved within water and reacted with SO 4 2to generate porous CaCO 3 and, within at least three deposits, > 1,000,000 metric tons of H 2 S. pathways dissolved Salado halite (and gypsum) into which overlying Permian and Mesozoic strata collapsed forming large (up to many hectares), closed, karstic depressions. The dolines focused enormous volumes (up to many cubic kilometers) of saline, O 2 -saturated (~2 to >4 mg/l) groundwater into the subsurface. The brine descended through the fault-tracking pathways along an inverted density gradient and discharged into 2 bearing groundwater sinking along one course contacted relatively fresh H 2 S-bearing groundwater rising along an adjacent course, elemental sulfur precipitated.
3 Capitan Formation ( Fig. 2 ). The reef escarpment extends ~65 km southwestward from near the city of Carlsbad, southeastern New Mexico, elevation ~960 m, to Guadalupe Peak, Trans-Pecos Texas, elevation 2,667 m, the highest summit in Texas (see DuChene and Martinez, 2000) ( Fig. 3 ). The escarpment formed mainly during the last few million years as Upper Permian gypsum and halite in the Pecos River Valley just southeast of the reef eroded faster than the limestone reef and its partially dolomitized forereef. Directly northwest of the reef-escarpment in the Guadalupe Mountains of New Mexico are about 400 caves (Queen, 2009) of which more than 30 are major (Ford and Williams, 2007) including two of the worlds largest, deepest, and most spectacular, Carlsbad Cavern (~50 km long, 315 m deep), and 5-6 km west of it, Lechuguilla Cave (~223 km long, 490 m deep). They are all relict having formed primarily in the late Miocene and early Pliocene, ~12 to ~4 Ma (million years) ago (Polyak et al., 1998) ( Fig. 4 ). The Guadalupe Mountains are located in southeastern New Mexico and west Texas ( Fig. 1 ). Bounding the southeastern-facing front of the mountains, Middle Permian paleo-reef and its forereefthe Figure 1. Figure 2. boundary. INTRODUCTION
4 Objectives and Purpose of Investigation Primary objectives are: to formulate a model for genesis of the hydrogen Guadalupe Mountains; to formulate a model for genesis of the geographically and genetically associated huge subsurface karstic deposits of native sulfur, westcentral Delaware Basin; to consider the similarities and differences between the two models; and to review evidence and arguments supporting aqueous CH 4 as the microbial foodstuff that A major purpose of the models is to stimulate further deliberations particularly those with a regional bent about the origin of both the caves and the sulfur deposits. The models can be considered both as working hypotheses and as postulations, the consideration of which may stimulate debating, challenging, and reasoning that results in improved knowledge of these extraordinary cave and karst features. Southeast of the mountain front in the adjacent westcentral Delaware Basin several hundred meters beneath the prospects of native sulfur (Smith, 1980)three of which have a few million to many tens of millions of metric tons of original reserves. Exploration guidelines are few (Smith, 1980), and undiscovered deposits likely remain. One deposit is only 12 km southeast of Carlsbad Cavern. The native sulfur is dispersed within bodies of secondary limestone encased within Permian anhydrite. Karstic processes were an intrinsically accompanying process of the sulfur deposition (Klimchouk, 2007, p. 89). A geographic and a genetic relationship exist between the caves of the Guadalupe Mountains and the large deposits of native sulfur. Both probably formed at about the same time; both probably owe their existence to a coincidence of essentially the same stratigraphic, thermal, biogenic, and tectonic events; and both probably owe their existence to great volumes of migrating methane (CH 4 ) that reacted with a virtually unlimited supply of sulfate anions (SO 4 2). A by-product of the reactionhydrogen sulfuric acid within the vadose environment of the caves and native sulfur within the phreatic environment of the sulfur deposits. Major differences between the genetic history of the caves and those of the sulfur deposits from its environments of formation to its environments of oxidation and the mechanism by which the hydrogen Figure 3. Figure 4. 2 S2 4
5 a surrounding shelf from the marine embayment, which coincided geographically with the structural Delaware Basin ( Fig. 1 ). The reef grew upward and basinward as relative sea level rose (King, 1948). The upper forereef talus dipped steeply (maximum >50; Mruk and Bebout, 1993) and the lower forereef talus dipped gently into the deep (hundreds of meters) marine embayment. Three Middle Permian shelfal formations (in ascending order, the Seven Rivers, Yates, and Tansill) grade into the Capitan Formation ( Fig. 6 ). Near the escarpment, the Capitan reef and forereef together have a maximum thickness of about 600 m (King, 1948, p. 61). The timeequivalent, marine, shelfal strata accumulated in shallow water (mostly <50 m). The Seven Rivers Formation consists primarily of bedded dolomite, and it is possibly 600 ft (183 m) thick near the reef (Hayes, 1964). The formation trends along an approximately 10-km-wide band parallel to and directly northwest of the reef. Near the reef, the Yates Formation consists of siliciclastics as Geographic and Stratigraphic Setting of Caves The locations of several of the more prominent caves of the Guadalupe Mountains are shown in Figure 5 Geographically, most large caves occur within a 6-km-wide band, referred to herein as the cave belt, parallel to and directly northwest of the reef escarpment, and almost all caves reside within a 12-km-wide band (Hill, 1999) ( Fig. 5 ). No long cave systems are known in the far western part of the Guadalupe Mountains ( Fig. 1 ) (DuChene and Martinez, 2000). Stratigraphically, most cave passages occur within the Capitan reef and within an adjacent, correlative, shelfal carbonatethe Seven Rivers Formationand most caves are close to the contact between these formations (Hill, 1987; DuChene and Martinez, 2000) ( Fig. 6 ). A Middle Permian sponge-algal reef, now represented by the Capitan Formation, extended for ~600 km around the perimeter of an ancient marine embayment (e.g., Adams and Frenzel, 1950; Newell et al., 1953). During its growth, the reef separated a lagoonal province situated on Figure 5. Figure 20 Figure 25 Figure 6. SULFIDIC ORIGIN OF CAVES OF THE GUADALUPE MOUNTAINS
6 Peculiar Qualities and Unusual Origin of Caves Caves of the Guadalupe Mountains have strange morphologies (e.g., Davis, 1980; DuChene, 1986; Hill, 1987, p. 22-23; 1996, p. 279; Palmer, 2006). They contain large rooms, many being >15 m in height and passages with abrupt and large-scale changes in crosssectional area (Palmer and Palmer, 2000; Palmer, 2006). The immense size of rooms distinguishes caves in the Guadalupe Mountains from most other caves (Moore, 1960a). The Big Room of Carlsbad Cavern, with an area of ~3.3 hectares and a maximum height of nearly 100 m (Palmer et al., 2009), is among the largest chambers in the world ( Fig. 8 ). Cave rooms and passages end abruptly without breakdown or major passage extensions and without relationship to surface topography; and cave points (e.g., Hill, 1996; p. 279; Hill, 1999). These various morphologic features are unlike those within caves created by aqueous carbonic acid (H 2 CO 3 ), the acidic solvent that usually operates within carbonate terrain. Moreover, caves of the Guadalupe Mountains harbor a strange suite of minerals unlike those found within the great majority of caves within carbonates (see Hill and Forti, 1986). Three caves contain native sulfur; particularly Lechuguilla Cave with several multi-ton deposits that amount to more sulfur than that within all well as carbonates, and the Tansill Formation consists primarily of dolomite. Farther back from the reef the Seven Rivers, Yates, and Tansill consist of red beds and shallow-water evaporites, particularly gypsum. A cross section transversally through the reef ~10 km north of Whites City, New Mexico ( Fig. 7 ), shows the stratigraphic relationship of the Capitan reef and forereef to the shelfal formations and to the timeequivalent Bell Canyon Formation. The cross-section also shows the stratigraphic relationship of the Capitan Formation to the youngest Upper Permian evaporites of the basinthose of the Castile Formation. The Castile consists predominantly of calcite-laminated anhydrite (CaSO 4 ) and anhydrite-laminated halite (NaCl), and the conformably underlying Bell Canyon consists dominantly of sandstone, siltstone, and intermittent tongues of limestone. Along the line of the cross section ( Fig. 7 ), Castile evaporites are laterally juxtaposed against the steep-to-shallow face of the Capitan forereef and the steep face of the Capitan reef. The juxtaposition resulted from depositional onlap of the Late Permian Castile evaporites onto the Middle Permian Capitan reef and Capitan forereef. Although the Capitan is considered Middle Permian and the Castile Late Permian, the Castile evaporitesgypsum, halite, and calcite (or aragonite) began to precipitate only a short time (probably within several thousand years) after the reef died from exposure and/or elevated salinity (e.g., Kirkland, 2003). Figure 7. Figure 13
7 1980; Hill, 1981, 1990, 2000; Kirkland, 1982; DuChene and McLean, 1989; Palmer and Palmer, 2000; Polyak and Provencio, 2001; Palmer 2006). Native sulfur and sulfuric acid both formed by reaction between the precursors O 2 and H 2 S (e.g., Jagnow et al., 2000; Engel et al., 2004), but in the caves, almost all oxidation continued past the intermediate elemental-sulfur stage to yield H 2 SO 4 Role of Sulfuric Acid in Cave Formation A general model has been formulated to explain how the caves of the Guadalupe Mountains formed dominantly from sulfuric acid (e.g., Egemeier, 1971; Buck et al., 1994; Engel et al., 2004; Hose and Macalady, 2006; Palmer, 2006). The model is based largely on investigations of active H 2 S caves elsewhere in the world. A synopsis follows: Many passages and rooms fractures, and pinch out at depth (Palmer et al., 2009) ( Fig. 9 ). H 2 S in solution within groundwater moved other known caves of the world combined (Cunningham et al., 1993; Davis, 2000). Five caves contain the hydrated aluminosilicate mineral endellite [Al 2 Si 2 O 5 (OH) 4 H 2 O], and the rare sulfate minerals alunite [KAl 3 (SO 4 ) 2 (OH) 6 ] and natroalunite [NaAl 3 (SO 4 ) 2 (OH) 6 ] (e.g., Polyak and Provencio, 2000). Two caves contain the unstable mineral epsomite [MgSO 4 H 2 O] (e.g., Hill, 1987, p. 131-132), and fourteen caves contain the common mineral gypsum [CaSO 4 H 2 O], many deposits of which are massive (e.g., Hill, 1987, p. 43). These unusual cave minerals formed either from reactions between sulfuric acid (H 2 SO 4 ) and clay, dolomite, or limestone (e.g., Davis, 1980; Hill, 1987; Queen, 1994; Polyak and Gven, 1996; Palmer, 2006), 2 S) and oxygen (Spirakis and Cunningham, 1992; Cunningham et al., 1993). Furthermore, speleologists have concluded that rooms and passageways of the caves were dissolved not by the weak acid, H 2 CO 3 but by the strong acid, H 2 SO 4 (e.g., Egemeire, 1971; Jagnow, 1977; Davis, Figure 8.
8 Elemental sulfur, however, was a short-lived intermediary. The atmosphere of the caves supplied oxygen to the high partial pressures (somewhat < 0.19 atmospheres), thus, additional oxidation, bio-catalyzed by sulfuroxidizing bacteria, took place probably concomitantly with precipitation of sulfur to form sulfuric acid (e.g., Hose and Pisarowicz, 1999). 2S + 3O 2 + 2H 2 O 2H 2 SO 4 A second reaction pathway was possible. Certain sulfuroxidizing bacteria form H 2 SO 4 directly from aqueous H 2 S and aqueous O 2 and they may have allowed the intermediate step (formation of native sulfur) to be bypassed (Palmer, 2009, p. 217). exposed to the cave atmosphere, the more acidic they became (Palmer, 2006). The strong acid reacted wherever possible with limestone and dolomite. The reaction dissolved carbonates of cave walls and ceilings, precipitated gypsum, and released CO 2 which within water formed carbonic acid (H 2 CO 3 ). This weak acid further enhanced speleogenesis by reacting with limestone to form calcium bicarbonate [Ca(HCO 3 ) 2 ] (see Palmer and Palmer, 2000), a substance that exists naturally only in aqueous solution. Expressed as equations, reaction of the acids with limestone and reaction of the by-products with water are: H 2 SO 4 + CaCO 3 + H 2 4 2H 2 O + CO 2 CaSO 4 2H 2 O + H 2 2+ + SO 4 2+ 3H 2 O CO 2 + H 2 2 CO 3 H 2 CO 3 + CaCO 3 3 ) 2 2+ + 2HCO 3 Where a coat of gypsum (or less commonly, clay, silica, or walls and ceilings, aqueous sulfuric acid dropped onto cave and dolomite commonly replacing the carbonate bedrock with gypsum (Queen, 1973; Queen et al., 1977; Hill, 1987; Buck et al., 1994; Palmer, 2006). The replacement occurred in a delicate balance with carbonate dissolution (Palmer and Palmer, 2000). Figure 10 shows vertical channels in a limestone block that formed by reaction of CaCO 3 with evolving caves of the ancestral Guadalupe Mountains (e.g., Palmer and Palmer, 2000; Kosa and Hunt, 2006b; DuChene and Cunningham, 2006). The groundwater table gradually fell, allowing atmospheric O 2 the other precursor of H 2 SO 4 to enter the upper, subaerial parts of incipient caves through restricted pathways to the surface (see Hose and Macalady, 2006; Palmer, 2006). The H 2 S, which was supplied to cave pools from below the overlying cave atmosphere, and moved toward cave walls and ceilings by diffusion, thermal convection, and barometric winds (Hose and Macalady, 2006; Palmer, 2006). Gaseous H 2 S and gaseous O 2 within the of water on gypsum-coated cave walls and ceilings (Palmer, 2006). The dissolved gases reacted. One reaction pathway involved precipitation of native sulfur within the aqueous microenvironments. The reaction re-charged with H 2 S and O 2 from the cave atmosphere allowing more sulfur to form. 2H 2 S + O 2 2S + 2H 2 O Figure 9.
9 aqueous H 2 SO 4 dripping from the cave ceiling; a product of the reaction, gypsum, partially replaced the block (Palmer et al., 2009). Most gypsum, however, formed within cave pools not by replacement or by precipitation, but by recrystallization of gypsum that fell off cave walls into cave pools (Palmer, 2009, p. 219). The water table within the cave belt progressively fell, and masses of cave gypsum became exposed, most of which were, in turn, dissolved within fresh groundwater 2000). The ions Ca 2+ SO 4 2, and, HCO 3 were carried from the caves within groundwater. Speleogenesis at any particular cave or cave level probably lasted for many tens-to-hundreds of thousands of years. H 2 SH 2 SO 4 speleogenesis essentially ceased by mid-Pliocene, but vadose water continued to enlarge the caves by dissolving Pleistocene between 600,000-20,000 years ago, as estimated from 234 U/ 238 U ratios, groundwater from near the earths seeping water released CO 2 into the cave atmosphere, calcite precipitated and decorated the caves with speleothems. Caves of the Guadalupe Mountains formed mainly above the water table, where water droplets formed (potassium aluminum sulfate hydroxide) occurs within Carlsbad Cavern, Lechuguilla Cave, Cottonwood Cave ( Fig. 5 ), and several nearby caves (Polyak et al., 2006). It requires a pH < ~4 to form. Such an acidic condition would have been virtually impossible to achieve in cave pools in contact with carbonate rock, but it could have been achieved readily within subaerial walls and ceilings (Palmer, 2006). In addition, the sulfur isotopic composition of the cave gypsum ( Fig. 12 ) generally falls within a range restricted to gypsum whose sulfate anions originated only from biogenic processes. The isotopic signature of the cave gypsum is consistent with cave formation above the water table by condensation-corrosion, a process in which the source of sulfur atoms would have been only from biogenic H 2 S. The isotopic signature is inconsistent with cave formation beneath the water table in which the sulfur isotopic signature would have been altered rocks (see Brown, 2006). The source of the adulterating sulfur (as sulfate anions) would have been a small fraction from dissolution of nearby Permian carbonates (see Staudt and Schoonen, 1995) and a large fraction from dissolution of nearby marine sulfate evaporites. Cave formation above the water table is also supported by the large amount of gypsum in the caves compared to a relatively minor amount of native sulfur. Oxidation of H 2 S above the water table was usually complete (yielding H 2 SO 4 and subsequently, CaSO 4 H 2 O), whereas oxidation of H 2 S below the water table (i.e., within cave pools) was probably seldom complete, resulting in the intermediate oxidation productnative sulfur. Oxidation may have also 2 was overwhelmed by H 2 S or where O 2 was in limited supply because passages and/or openings to atmospheric O 2 were restricted (personal communication, A.N. Palmer, 2013). The native sulfur shown in Figure 11 precipitated when the concentration of O 2 was somehow severely limited. Transported to the Caves Curious Sulfur Isotopic Composition Speleologists working in Carlsbad Cavern before about 1980 believed that the voluminous gypsum in the cave came from nearby beds of marine gypsum Figure 10.
10 Kirkland, 1982). The sulfur isotopic compositions were reported as the difference in parts per thousand from a standard (parts per thousand being equivalent to per mil, tenths of a percent and the symbol, ); the 34 S. Permian marine gypsum in deposits near the caves ( Fig. 13 ) are enriched in the 34 S isotope compared to sulfur within a standarda sulfur-bearing iron meteorite found near the ghost town of Canyon Diablo between Flagstaff and Winslow, Arizona. The samples of marine gypsum have positive per mil values, and are described as isotopically heavy; in marked contrast, gypsum samples from Carlsbad Cavern are impoverished in the 34 S isotope, they have negative per mil values, and are described as isotopically light ( Fig. 12 ). The sulfur-bearing minerals within the caves are depleted in 34 S by several percent (several tens of parts per thousand) compared to the 34 S composition of Late Paleozoic marine calcium sulfate ( Fig. 12 ). The on such a wide divergence, the two suites of gypsum samplesthose from Carlsbad Cavern and those from nearby Permian marine strataare deemed unrelated. (e.g., Bretz, 1949; Black, 1954; Good, 1957; Hayes, 1964; Bullington, 1968). These workers hypothesized that surface water and shallow groundwater dissolved Upper Permian marine gypsum of the nearby Delaware Basin and/or Middle Permian marine gypsum of the nearby inner shelf (far back reef) of the western Guadalupe Mountains ( Fig. 13 ). Groundwater transported sulfate anions (SO 4 2) and calcium cations (Ca 2+ ) from one or both of these sources into Carlsbad Cavern where gypsum precipitated within local pooling (Bretz, 1949) during temporary conditions and as the waters cooled (Good, 1957). Early students of Carlsbad Cavern communicated this account with hesitation, but they offered no other explanation (see Jagnow et al., 2000). The massive gypsum deposits in Carlsbad Cavern did not form as early speleologists had envisioned. The primary evidence that invalidated this early hypothesis was analysis of samples of cave gypsum for the ratio between the number of atoms of 34 S and 32 S (Hill, 1981; Figure 11. Figure 12. 34 34
11 values for the sulfur-bearing cave minerals does not 34 S range of nearby Upper or Middle Permian, marine gypsum ( Fig. 12 ). In fact, the range 34 S range of any deposit of Phanerozoic, marine-derived gypsum (or marine-derived anhydrite [CaSO 4 ], which generally replaces its hydrous sister mineral in the shallow subsurface). All cave minerals bearing sulfur in the Guadalupe Mountains formed because of H 2 S-H 2 SO 4 speleogenesis biased selecting of 32 S or 34 S as these minerals formed (e.g., Goldhaber, 1993; Ziegenbalg et al., 2012), therefore, the distinctive sulfur isotopic signature of the sulfur-bearing cave minerals must have been inherited from their precursor, H 2 S, that moved into the caves and that participated in chemical reactions. Microbial Derivation The different origins of the two classes of gypsum are 34 S values. Samples of Upper Permian, marine, Castile gypsum from near Carlsbad Cavern ( Fig. 13 34 S values and a narrow range, +11.3 to +12.0 (n=36) ( Fig. 12 ) (Kirkland et al., 2000). Similarly, samples of Middle Permian marine gypsum from the nearby evaporitic facies of the Seven Rivers Formation ( Fig. 13 ) have positive values and a narrow range, +8.7 to + 10.2 (n=8) ( Fig. 12 ) (Sarg, 1981). On the other hand, samples of gypsum from Carlsbad Cavern have highly negative 34 S values and a wide range, -25.6 to -13.9 (n=13) (Kirkland, 1982; Hill, 1987) ( Fig. 12 ). In addition, samples of gypsum from Lechuguilla Cave, Cottonwood Cave, and McKittrick Hill Cave, and samples of other sulfur-bearing minerals from caves of the Guadalupe 34 S values similar to those of samples of gypsum from Carlsbad Cavern (e.g., Polyak and Gven, 1996) ( Fig. 12 34 S Figure 13. Fig. 15 Figure 7 The distinctive sulfur isotopic signature 34 S values having a wide rangeindicates that its sulfur-bearing precursor, H 2 S, almost certainly formed by a redox (red[duction] + ox[idation]) reaction mediated by anaerobic microbes. In near-surface Phanerozoic environments, the activity of sulfate-reducing microbes is probably the only way that H 2 S 34 S values (i.e., << 0 ) can form (e.g., Dessau et al., 1962; Holser and Kaplan, 1966). Many types of organic matter, assisted by microbial enzymes, reduce 32 SO 4 2to H 2 S at a slightly faster rate than they reduce 34 SO 4 2to H 2 S, the 32 SO bonds being slightly easier to break than the 34 SO bonds. Thus, sulfate-reducing microbes generate H 2 S enriched in 32 S and, if the system were open, they would leave behind sulfate anions enriched in 34 S. The difference between the maximum and 34 S values for the sulfur-bearing cave minerals is broad ( Fig. 12 ), e.g., 11.7 for gypsum from Carlsbad Cavern. Such a broad range compared, for example, to a
12 Also supporting introduction of a great weight of H 2 S into the caves is the enormous weight of carbonate rock removed as the caves formed; more than 3,000,000 metric tons of limestone from the Big Room of Carlsbad Cavern alone (Hill, 1987, p. 79); and with its large galleries and extensive passageways, a truly immense tonnage from Lechuguilla Cave. Much, if not most, of this removal is attributable to H 2 S-H 2 SO 4 speleogenesis (Polyak and Provencio, 2001; Palmer, 2006), and because dissolution of one cubic meter of limestone requires 918 kg of H 2 S (Palmer and Palmer, 2000), an immense weight of H 2 S (millions of metric tons) apparently moved into developing caves of the Guadalupe Mountains. Sources and Pathways of Hydrogen The source area from which microbial H 2 S implicated in creating the caves formed is disputed (e.g., Brown, 2006). Two nearby source areas have been proposed. narrow range of about 0.7 for nearby Upper Permian, basinal, marine gypsum, suggests microbial derivation (e.g., Lein, 1974). The broad range results from variations in microbial strains, in rates of microbial reduction, in ambient temperatures during microbial activity, and in the degree of isolation of the aqueous sulfate being metabolized (e.g., Harrison and Thode, 1958; Coleman, 1985; Machel, 1992; Goldhaber, 2003). Immense Quantity H 2 S transported into caves of the Guadalupe Mountains can be calculated from the weight of cave gypsum. It takes ~0.2 metric tons of H 2 S to produce one metric ton of cave gypsum. Judging from wide blocks of gypsum up to 10-m thick (e.g., Hill, 1987, p. 44-45; Spirakis and Cunningham, 1992; Davis, 2000) having a density of ~2.3 metric tons/m 3 the total weight of gypsum presently within the caves is large. Lechuguilla Cave contains thousands of tons of massive or laminated gypsum (Davis, 2000), and along more than 220 km of surveyed passageways, gypsum appears in many forms including thick coatings on the passage walls, spectacular stalagmites and stalactites, delicate hairs, An example is the remarkable gypsum chandeliers ( Fig. 14 ). They formed late in the history of the cave as vadose seepage dissolved secondary gypsum from an overlying level and re-precipitated it as crystalline masses in an underlying level in which water tends to evaporate (personal communication, A.N. Palmer, 2013). As mentioned above, much gypsum, which is highly soluble in water (maximum ~2.5 g/l), has gone into solution and has been removed within groundwater (Hill, 1987, p. 48-49; Polyak and Provencio, 2001; Hose and Macalady, 2006). Remaining blocks of gypsum in the Big Room of Carlsbad Cavern usually lie in protected alcoves or under overhanging ceilings (Black, 1954). Water dripping from the ceiling has dissolved precise, vertical, cylindrical tubes through blocks of gypsum (Quinlan and Smith, 1968) (one hole is ~4 m long and only ~9 cm in diameter (Bretz, 1949)). The weight of gypsum in the caves before dissolution was huge (Hill, 1987, p. 87); from this original large weight, we can infer that many metric tons of H 2 S moved into the caves, and reacted with O 2 to form H 2 SO 4 which, in turn, reacted with either limestone or dolomite to form gypsum. Figure 14.
13 The Existing Shelfal Model Partially in response to the perceived paleohydrologic (or gaseous) H 2 S updip from the basin to the evolving caves, a model was proposed for transporting aqueous H 2 S downdip within groundwater from the western, inner shelf (high elevations) to the evolving caves of the outer shelf (lower elevations) (e.g., DuChene and Cunningham, 2006; Brown, 2006; DuChene, 2009) ( Fig. 5 ). (For this study, that part of the shelf within 6 as the outer shelf ( Fig. 5 ), and that part beyond 6 km, the far back reef, is designated as the inner shelf.) About 75 shelfal occurrences of sulfur are known within inner shelfal Middle Permian carbonates north and northwest of Carlsbad, New Mexico (Hinds and provide direct evidence of the past presence of H 2 S. Most shows occur within Permian formations older than those that are time equivalents of the Capitan Formation. Some cores of San Andres Limestone (Middle Permian, lower Guadalupian series; Fig. 6 ), for example, show thin coatings of sulfur on fractures and on bedding planes and small crystals of sulfur in vugs and in fractures. Southwest, west, and/or northwest of caves of the Guadalupe Mountains similar accumulations of native sulfur indicative of its precursor, H 2 S, may have existed (or may exist) within Middle Permian strata of the inner shelf (i.e., within higher elevations of the ancestral (or present) Guadalupe Mountains) (e.g., DuChene, 2009) ( Fig. 5 ). deposits occur northwest of the reef escarpment in the (FeS 2 Sulfur combined within these minerals is isotopically 34 S, -1 to -15) (Hill, 1996, her appendix 2). These minor mineral deposits demonstrate that at least modest amounts of microbial H 2 S were available to react support the past presence of H 2 S on the inner shelf. The possibility that the inner-shelf was the source of H 2 S transported to the caves is based in part on analogy with current conditions on the shelf north and northeast of Carlsbad, New Mexico. Here, in Middle Permian The H 2 S is thought to have originated by a microbially mediated redox reaction either southeast of the cave belt within Upper Permian calcium sulfate strata of the adjacent Delaware Basin, or northwest of the cave belt within Middle Permian calcium sulfate strata of the adjacent inner shelf (i.e., strata within the western Guadalupe Mountains) ( Fig. 5 ). While the caves were forming, both proposed source areas contained bedded anhydrite and near the surface bedded gypsum. Figure 13 shows the location of outcrops of Middle Permian Seven Rivers gypsum, which during the late Miocene and early Pliocene were more extensive; it also shows the general location of Upper Permian Castile gypsum either as scattered outcrops or situated just below a thin gypsiferous soil. Dissolution of the calcium sulfate mineralsgypsum and anhydritecreated sulfate anions that were both a potential oxidant and a potential source of the sulfur atoms within the H 2 S molecules. The basin contained abundant petroleum, particularly natural gas, while the caves were forming, and to the northwest, well behind the present-day reef escarpment, the inner shelf ( Fig. 5 ) also possibly contained abundant petroleum while the caves were forming. If we consider the vast quantity of metabolizable organic matter required as a reductant, hydrocarbons (whether crude oil or natural gas) were probably the only viable contenders as reducing agents. They were both a potential reductant and a potential source of the hydrogen atoms within the H 2 S molecules. Which of these tectonic elementsthe basin or the shelfwas the source of the H 2 S has been earnestly contested. Proponents of a basinal source model include Davis (1980), Hill (1987, 1990), Polyak et al., (1998), and Palmer (2006), with C. A. Hill being the principal advocate. Proponents of a shelfal source model include DuChene (1986, 2009), DuChene and McLean (1989), Brown (2006), DuChene and Cunningham (2006), Stafford et al. (2008b; 2009), and Stafford and Nance (2009), with H. R. DuChene being the principal advocate. In addition, both basinal and shelfal proponents agree that the H 2 S apparently entered the caves from below through Hunt, 2006b; DuChene and Cunningham, 2006), but the pathways from where it originated to where it entered 2006; DuChene, 2009).
14 crystals or nodules in micritic dolomites. The calcite is the youngest major diagenetic shelfal mineral (Scholle et al., 1992), and it probably formed primarily in the Miocene and Pliocene. Samples of the sparry calcite usually have isotopic signatures characteristic of genesis in part from organic matter (e.g., -12.8) (Scholle et al., 1992). The sparry calcite apparently formed from the microbially mediated reaction between fractions of oil and sulfate anions (probably derived from Middle Permian marine anhydrite), and the reaction would have generated a 2 S some of which would have been transported downdip within groundwater. Despite the appeal of the shelfal source model, the quantity of H 2 S required for speleogenesis in the Guadalupe Mountains (many millions of tons) was probably inadequate. Even if shallow accumulations of oil like those northeast of Carlsbad, New Mexico, were present on the inner shelf northwest of the present-day escarpment, the quantity of associated H 2 S (see Brown, quantity of H 2 SO 4 required to dissolve the caves. The biogenic H 2 S implicated in genesis of the metallic 388) to have come from the basin rather than from the inner shelf. These deposits of the inner shelf extend for several kilometers farther to the northwest than the cave belt, but sphalerite and pyrite can form from much lower concentrations of aqueous H 2 S than that required for effective H 2 S-H 2 SO 4 speleogenesis. Considering the amount of H 2 S required for cave formation on the outer shelf (i.e., within the cave belt), if H 2 S for cave formation came from the inner shelf, more evidence for past generation of H 2 S might be expected within Middle Permian evaporitic strata of the far back reef. Such evidence might include strata of gypsum and/or anhydrite that contained scattered castile-like structures of secondary limestone (tens of meters across) highly depleted in 13 C. In addition, we might expect on the inner shelf, as well as major accumulations of native sulfur. Such evidence is wanting. A possible reason for its absence, however, is that much of the Guadalupian series has been stripped from the inner shelf (e.g., Boyd, 1958, p. 43; Sarg, 1981), and with erosion, critical evidence may have been lost. back-reef reservoir rocks, 0.9-1.2% H 2 S is associated with shallow (mostly <1000 m) accumulations of microorganisms apparently used fractions of the crude (i.e., the alkanes), to reduce sulfate anions within associated pore water to yield the metabolic by-products CO 2 and H 2 S. Similar accumulations of oil may have been present on the inner shelf west and northwest of the cave belt and may have contributed H 2 S to groundwater that transported it to the outer shelf. On the shelf northwest of the reef escarpment, however, present-day accumulations of oil within Middle Permian strata have not been discovered. Within several-toseveral-tens of kilometers north and east of the Capitan reef vast accumulations (billions of barrels) of crude oil, now largely exploited, were trapped within Middle Permian shelfal carbonates (including carbonates of the Tansill, Yates, and Seven Rivers). Why crude oil was apparently not trapped within these same strata in the Guadalupe Mountains is uncertain (i.e., Hill, 1996, p. 354). Oil may have been swept away by hydrodynamic DuChene, 2009); it may have moved updip and escaped (Hill, 1996, p. 354-356); it may have been removed as reservoir rocks of the inner shelf were eroded; and/or it may have been displaced from traps by natural gas that subsequently escaped because of inadequate sealing. From hydrologic principles alone, Brown (2006) concluded that H 2 S that entered the caves must have direction and, therefore, from higher elevations within inner shelfal strata west or southwest of the caves. He hypothesized that the necessary reductant was either dissolved or particulate organic matter in water; falling within his dissolved category are certain components of crude oil. The presence of crude oil within the western Guadalupe Mountains is indicated by the odor of petroleum from freshly broken limestone of the Grayburg Formation (Middle Permian, lower Guadalupian series; Fig. 6 ) (DuChene, 2009). Furthermore, within sparry calcite, abundant hydrocarbon inclusions occur within Middle Permian strata of mid-shelf settings (Scholle et al., 1992). The calcite, which also contains abundant anhydrite or gypsum. Sparry calcite is disseminated
15 They usually consist of isolated masses of limestone ( Fig. 15 ); relatively few, however, consist of laterally extensive horizons or limestone sheets up to 2 m thick (e.g., Stafford et al., 2008b) (the calcite ridges of Miller, 1992). The castiles have a mean surface area of ~2,300 m 2 the largest having a surface area of ~40,000 m 2 (Stafford et al., 2008b). These scattered bodies of diagenetic limestone while Ochoan evaporitic strata (mainly the upper Castile and the Salado) were sites of generation of large quantities of microbial H 2 S (Kirkland and Evans, 1976). The castiles and their buried counterparts are a crucial element of basinal source models (e.g., Hill, 1987, 1990), therefore, a summary of aspects of their nomenclature, morphology, distribution, and stratigraphy follows; in a later section, biochemical origin is considered. Smith (1980) objected to Adams use of the term castile or castiles for the carbonate masses because of possible confusion with the formation name, Castile. Kirkland and Evans (1976), with similar objections in mind, termed H 2 S was generated on the inner shelf northwest of the cave belt by reaction of fractions of oil with sulfate anions derived from Middle Permian anhydrite and gypsum, the best appraisal, however, appears to be that the quantity generated was less than the colossal amount required for speleogenesis in the outer shelf of the Guadalupe Mountains. The Existing Basinal Model Dotting the western outcrop area of the Castile Formation (lower Ochoan Group), throughout much of the western Delaware Basin, are hundreds of discrete masses of secondary limestone (Adams, 1944; Kirkland and Evans, 1976; Stafford et al., 2008b) ( Figs. 13 and 15 ), many of which have only been partly exhumed out of surrounding Castile gypsum. Adams (1944) labeled these bodies castiles and he noted that they are diagenetic, a term that (excluding weathering and metamorphism) includes all chemical, physical, and biological changes occurring Castile anhydrite ~245 Ma after its inception. The castiles, in plan, commonly have maximum dimensions >30 m. Figure 15. Fig. 13
16 conversion of gypsum to anhydrite but before nearsurface conversion (by hydration) of anhydrite back again to gypsum (Moore, 1960b, p.74; Brown and Loucks, 1988). The depositional and structural fabric of Castile anhydrite is commonly preserved during the replacement, and the steep exposed limestone walls of laminations and microfolds of Castile gypsum and anhydrite (Kirkland and Anderson, 1970; Kirkland and Evans, 1976; Stafford et al., 2008b) ( Fig. 16A ). Almost all castiles before being exhumed in the latest Tertiary and Quaternary were probably fully encased within lower Castile anhydrite. The Castile Formation near the eastern side of the Delaware Basin consists predominantly of thick (tens of meters) alternating sections of halite and anhydrite ( Fig. 17 ) that collectively have a total thickness of 450-550 m. Here, in a narrow band, the highly soluble Castile evaporites have largely escaped the effects of dissolution ( Fig. 18 ), which has otherwise greatly Castile halite in the west-central basin, for example, have completely dissolved ( Fig. 18 ). Without the intervening from which a particular castile rises cannot usually be technique of varve correlation (Anderson and Kirkland, 1966)). The castiles, however, are concentrated in the western more elevated part of the Gypsum Plain where erosion has removed, at least, the Anhydrite IV Member and possibly all or much of the Anhydrite III Member. Just south in Texas, erosion has removed much of the Anhydrite II Member, and, just east of the erosional pinch out of the Castile ( Fig. 18 ), much of the Anhydrite I Member. Most castiles, judging from their present geographic distribution, probably originated from replacement of the Anhydrite I and Anhydrite II members (Kirkland and Evans, 1976; Stafford et al., 2008b). Cropping out on the northernmost part of the Gypsum Plain in New Mexico and on the easternmost part of the Gypsum Plain in Texas and New Mexico, are outliers of dolomite of the Rustler Formation, gypsum of the gypsum) remaining after dissolution of halite within the Salado Formation. Castiles are nearly absent in these peripheral regions ( Fig. 13 ), but beneath these outcropping stratigraphic units, within the lower Castile the structures limestone buttes. Adams (1944), however, had noted that, a more appropriate name than castiles can hardly be imagined. The appropriateness was the similarity to the term Castile and to their physiographic expression as castellated peaks (although not all castiles stand in castle-like relief). Adams terms castile and castiles, after many decades of usage, are well ingrained, are widely used in geologic publications, are nearly always written with a lower case c, and are apparently seldom confused with the name of their host formation. Therefore, I use castiles, or its singular, for the many outcropping bodies of diagenetic limestone surrounded by and partly encased by Castile gypsum or by gypsiferous soil. For their subsurface equivalentsmasses of limestone still fully encased within Castile gypsum or anhydrite, or bothI use diagenetic limestone, or its equivalents, secondary limestone and biogenic limestone. About 1000 castiles crop out in clusters on nearly ( Fig. 13 ) (Stafford et al., 2008b). Erosion of gypsum softer and more soluble than calcitehas unearthed the limestone bodies and has left many standing in high relief ( Fig. 15 ), some by as much as 40 m. From near the entrance to Carlsbad Cavern, a few castiles can be seen several kilometers to the southeast towering above gypsiferous soil (gypsite). The large (~1,800 km 2 ) geographic region in New Mexico and Texas from which the castiles rise has been designated the Gypsum Plain ( Fig. 13 ). It consists mainly of gypsite-mantled Castile bedrock, dissolutioninduced landforms (e.g., sinkholes, solution-subsidence troughs, and caves), of which there are an estimated 9,000 (Nance and Stafford, 2009), and sporadic outcrops of Castile gypsum that comprises ~8 % of the area in which the Castile is at or near the surface (Stafford et al., 2008a). The Gypsum Plain in New Mexico occurs southeast of the reef escarpment and northwest of the dolomitic Rustler Hills ( Fig. 13 ); just to the south in Texas, the Gypsum Plain occurs east of the low-lying Delaware Mountains and west of the Rustler Hills (which stand in relief because of erosional resistant beds of dolomite) ( Figs. 7 and 13 ). The castiles are large-scale replacement features, calcite having replaced laminated, nodular, massive, and brecciated Castile anhydrite by a process termed calcitization. The process occurred after subsurface
17 Most diagenetic masses of Castile limestone beneath the surface of the Gypsum Plain, despite having produced copious amounts of H 2 S, are barren of native sulfur (see Zimmerman and Thomas, 1969, p. 18). One example, located ~5 km west of the Culberson sulfur deposit ( Fig. 32 ), is the one-well Rustler Hills oil field. Its reservoir consists of secondary Castile limestone at a depth of ~140 m (Davis and Kirkland, 1970). Diagenetic masses of limestone beneath the Gypsum Plain barren of both native sulfur and crude oil have likely been encountered during exploration drilling, but, having no economic value, there was little-to-no incentive for documentation. Nevertheless, judging from the number of castiles exposed by erosion that completely lack native sulfur, that contain insignificant amounts of native sulfur, or that show no evidence of ever having contained significant accumulations of native sulfur, many sulfur-free limestone counterparts probably exist beneath the Gypsum Plain. Hill (1987) advanced the following explanation for the source and migration of H 2 S from the Delaware Basin into the caves of the Guadalupe Mountains. During the Late Tertiary, masses of microbial limestone encapsulated anhydrite member, equivalent masses of diagenetic limestone are probably present. Major known deposits of native sulfur occur >100 m beneath the surface of the Gypsum Plain. The sulfur occurs within a carbonate lithology like that of the castiles except that the host rock contains elemental sulfur (e.g., Madsen and Raup, 1987; Smith, 1980). These rare deposits, discussed in a latter section, are apparently essentially buried castiles that enclose disseminations, Figure 16. A. B. Figure 17.
18 her hypothesized migration pathways within the Bell Canyon Formation and with her hypothesized phase of H 2 S. According to Hills model (1987, 1990), updip migration of gaseous H 2 S occurred along permeable beds, and/or along northwest trending joints. However, gaseous H 2 S, because of its buoyancy, would not have migrated downward from microbial loci within Castile anhydrite into the Bell Canyon sandstone (Brown, 2006). Even if gaseous H 2 S were present in the upper Bell Canyon, it would not have migrated into the reef and into carbonate strata of the northwestern shelf through the upper permeable sands, because these within thick, bedded anhydrite of the lower Castile Formation were sites of generation of large quantities of gaseous H 2 S that then migrated into underlying, widely extending beds of sandstone of the Bell Canyon Formation (the upper member of the Delaware Mountain Group) ( Fig. 6 ). The gaseous H 2 S subsequently migrated updip through permeable pathways into caves of the reef and adjacent shelfal carbonates. Challenges to Hills model are concerned not with generation of large quantities of H 2 S within the Castile, for which there is much support, but with Figure 18.
19 strong ties to paleohydrology (Palmer and Palmer, 2000; Klimchouk, 2007, p. 75). Apparently, neither aqueous H 2 S nor gaseous H 2 S moved updip into reefal and shelfal carbonates through either Bell Canyon fractures or beds of sandstone. Caves During their formation, Carlsbad Cavern, Lechuguilla Cave, and other large caves within the Guadalupe Mountains were extraordinary sinks for H 2 S, the weight of H 2 S oxidized to H 2 SO 4 within the Capitan and Seven Rivers formations, as mentioned above, amounted to millions of metric tons. At the same time, the adjacent of H 2 S, millions of metric tons having been generated at subsurface, microbial loci. Furthermore, the two principal formations, one harboring the generators and the other the sinks, were proximal, the Castile Formation being laterally contiguous with the Capitan forereef and with the precipitous face of the Capitan reef ( Figs. 7 and 20 ). While the generators of H 2 S were active, they were sealed beneath hundreds of meters of Salado halite and Rustler anhydrite); and, while the sinks of H 2 S were active, the evaporites were gradually being stripped (over million of years) of these same thick sealing beds. generation and transportation of H 2 S to the caves of the stages: formation of hydrologic pathways within anhydrite and halite of the Castile Formation; generation and migration of methane (CH 4 ) and its reaction with Castile sulfate anions (SO 4 2) to form 2 S); transportation of aqueous H 2 S through hydrologic pathways to the reef; uppermost Permian evaporitic strata; and descent of the water table progressively from southwest to northeast, and concurrent oxidation of H 2 S within caves along the cave belt progressive caves from high elevation to low elevation). normal to it (DuChene, 1986; Brown, 2006). In addition, beds were possible (although unlikely) migration pathways for gaseous H 2 S, but farther out into the basin such beds (e.g., the Lamar Member of the Bell Canyon Formation ( Fig. 6 low permeability. Gaseous H 2 sandstone and siltstone, and through northwesttrending joints. However, why would H 2 S, many times more soluble than CH 4 or CO 2 have persisted in a gaseous state within an environment in which its aqueous solubility was further enhanced by the hydrostatic pressure of hundreds of meters of burial? Conceivably, H 2 S dissolved within Bell Canyon pore water may have diffused into gaseous CH 4 and the mixture of gases may have migrated updip through the Bell Canyon Formation and into the Capitan where the H 2 S was stripped from the CH 4 However, major problems remain: Why would an immense quantity of H 2 S have resided in the upper Bell Canyon? What was its source? Moreover, could the modest quality of migration pathways within the upper Bell Canyon have conveyed the necessary volume of H 2 S to the gaseous H 2 S did not move updip into the Capitan and Seven Rivers carbonates through either Bell Canyon fractures or beds of sandstone. The H 2 S, wherever it was generated, was apparently transported to the caves not as a gas, but as a dissolved component within groundwater (Palmer and Palmer, 2000; Brown, 2006). While the caves were forming, groundwater within the Bell Canyon Formation, in response to the hydraulic gradient created by relief of the ancestral Guadalupe Mountains, moved slowly downdip though pores and through fractures probably at a rate of < 1 m per year (see Hiss, 1975, 1980; Wiggins et al., 1993; Lee and Williams, 2000). Within Bell Canyon sandstone, groundwater in which H 2 S was dissolved could not have moved updip to the shelf edge, i.e., to the Capitan Formation, counter to this of groundwater bearing dissolved H 2 S through the Bell Canyon Formation is considered to be a view without
20 During speleogenesis in the Guadalupe Mountains, the Castile Formation of the Delaware Basin (which accumulated rapidly, ~260.0-259.8 Ma ago) had an extraordinarily uniform stratigraphic framework. Superimposed on this framework were two nearly concurrent Late Tertiary events. Heating of the crust, which caused abundant CH 4 to be generated and source beds to be overpressured. Tilting of the Guadalupe Mountains and most of the Delaware Basin to form the paleo-Guadalupe tectonic block; uplift of the block caused Permian strata to fracture or to re-fracture and provided the potential energy that on its release allowed karstic pathways to form. (I referred to this ancient structural entity as the paleoGuadalupe tectonic block or ancestral Guadalupe tectonic block; during the late Miocene, for example, it was shallower and probably more expansive (see DuChene and Cunningham, 2006) than its present-day tectonic descendant (the Guadalupe tectonic block)). The effect of these two major geologic events (heating and tilting) on Castile evaporites of the Delaware Basin probably led indirectly to cave formation in the Guadalupe Mountains. In addition, I consider the formation of curious, easterly trending, karstic, solution-subsidence troughs on the Gypsum Plain; and I compare and contrast the hypothesized conduits within Castile halite with a closely related karstic modeldevelopment of a void Then, in a following section, I describe relatively recent removal of Rustler, Salado, and Castile strata. Beginning of Intense Cavern Formation Directly northwest of where the Capitan escarpment now trends ( Fig. 3 ) Middle Permian carbonates of the reef and correlative carbonates of the outer shelf (near back reef) underwent only minor, sporadic speleogenesis for ~245 Ma (Hill, 1996, p. 276278). Then, throughout ~8 Ma, from probably early late Miocene (or possibly latest middle Miocene) to early Pliocene (~12 to ~4 Ma ago), the carbonates experienced intense speleogenesis (e.g., Polyak et al., 1998; Polyak and Provencio, 2000). The Tertiary age assignments, which represent minimum ages (Palmer, 2006), are based on 40 Ar/ 39 Ar dating of the potassium-bearing cave mineral, alunite (e.g., Polyak and Provencio, 2000). Figure 19.
21 Formation (e.g., Anderson et al., 1972; Babcock, 1977) and (Leslie et al., 1997; Hovorka, 2000), a complete absence of desiccation surfaces (Hovorka, 2000), and a tendency for (Dean and Anderson, 1978). There are abundant shallowwater and ephemeral saltpan fabrics within the Salado evaporites ( Fig. 6 ) (e.g., Lowenstein, 1988), fabrics that are completely absent in the directly underlying Castile evaporites (Hovorka, 2000). The dominant sedimentological control on the Castile or runoff of fresh water into the basin from the surrounding facies (Kirkland et al., 2000). With only modest regional variability, temperature and humidity affected evaporation over the entire brine body. Such external climatic factors entire depositional environment within a short time (e.g., <1 yr); thus, across the deep-water basin most major and minor lithologic boundaries within the Castile sequence are nearly isochronous (see Kendall, 1988). Regional changes in Castile sedimentation, from calcium carbonate to calcium sulfate, and back again, and from calcium sulfate to sodium chloride, and back again, were abrupt, generally occurring within a season (Anderson et al., 1972; 1978). Such changes within both millennial cycles (Dean and Anderson, 1978) and seasonal cycles (e.g., Kirkland, 2003) produced evaporitic beds and laminae, respectively, that are one of the few examples of true layer cake stratigraphy (Warren, 2006, p. 335), a stratigraphy in which Walthers Law does not apply (Kendall, 1988). Most Castile evaporitic beds, whether millimeter thick laminae or meter thick beds, and whether halite, anhydrite, or calcite, extended throughout, at least, the northern basin. In support of such pervasive climatic control, Castile laminae from many drill cores from the northern half of the basin have been correlated precisely (to a faction of a millimeter) (e.g., Fig. 19 ) (see Anderson and Kirkland, 1966; Dean, 1967; Anderson et al., 1972; Dean and Anderson, 1982). The principle basis for correlation is the unique pattern formed by groups of laminae, each lamina usually having a slightly different thickness and uncommonly a slightly different lithologic character. Beds of Castile halite once on-lapped the Capitan reef and forereef where they are now exposed. Presently, Formation of Basinal Hydrologic Pathways Hydrologic pathways within the Castile Formation probably began to form in the western Delaware Basin near the beginning of the late Miocene. The distinctive character of Castile stratigraphy was a pivotal factor in the formation of the aquifers. Castile evaporites in the western basin before tilting of the ancestral Guadalupe tectonic block and before the ensuing dissolution consisted by volume of ~60% CaSO 4 (~6,000 km 3 ) and ~30% NaCl (~3,000 km 3 ) (see Hayes, 1964; Snider, 1966, his table 3; Anderson, 1978, his table 1). In the eastern basin, where Castile halite has been largely preserved ( Fig. 18 ), the formation consists of eight members: a thin (<1 m) basal limestone, four thick (tens of meters) anhydrite members, and three thick (tens al., 1972) ( Fig. 17 ). The eight-member succession before extensive dissolution in the Late Tertiary was virtually basin wide. The physical setting and the climatic conditions during Castile sedimentation resulted in remarkable lateral persistence of Castile beds and laminae. The rate of tectonic subsidence in the Delaware Basin during the Middle Permian exceeded the rate of accumulation of sediment, and, hence, the tectonic basin, as mentioned above, coincided with a deep marine embayment. Near the end of the Middle Permian, it was rimmed by the living Capitan reef, which eventually grew across the mouth of the embayment (e.g., Kirkland, 2003) transforming it into a large enclosed lagoon (Kendall and Harwood, 1989; Anderson, 1993; Anderson and Dean, 1995; Leslie et al., 1997). It was deep, initially ~550 m (Newell et al., 1953), and extensive, ~25,000 km 2 with 3 /yr) of marine groundwater (Kirkland et al., 2000); but a channel (strait) connecting it to the Permian ocean, as envisioned by R. H. King (1947), was absent (e.g., Kendall, 1988). During the early part of the Late Permian, Castile evaporites rapidly halite, and a lesser volume (~10%) of calcite (or possibly initially aragonite). Much evidence supports the deep basin-deep water interpretation for the Castile. It includes an extensive correlation network of undisturbed laminae (e.g., Fig. 19 ), a transitional contact between the deep-water Bell Canyon
22 south of the latitude of Carlsbad Cavern, Castile halite has dissolved for a few kilometers to many kilometers into the basin ( Fig. 18 ), but Castile gypsum, with a solubility ~140 times less than that of halite (Klimchouk, 2000), has been subjected to much less dissolution. Consequently, adjacent to the escarpment, beds of Castile gypsum presently come right up to the reef talus slopes (written communication, D. B. Smith, 1970); to within twenty feet of the reef (Black, 1954); and, to within as little as a few tens of feet (Kelley, 1971). Like the beds of gypsum, the now missing beds of halite, with the exception of a few beds within the uppermost halite member (Anderson et al., 1972; Hovorka, 1990, p. 283), once extended to the northwestern margin of the basin (Anderson, 1978, p. 5; Kirkland, 2003). Figure 20 a diagrammatic cross-section speleogenesis in the Guadalupe Mountains and before tilting of the tectonic block, shows the face-to-face Early in the Tertiary, the boundary between each bed of Castile halite and each directly overlying bed of Castile anhydrite was persistent, smooth, and, nearly horizontal. Late in the Tertiary, each halite-anhydrite boundary, which represented an instant in geologic time, was persistent, smooth, and slightly inclined. most beds of Castile halite, as mentioned, are missing from the westernmost Delaware Basin, but lithologic evidence of their past presence persists. Where dissolution has occurred, laminae of anhydrite once incorporated within the halite ( Fig. 21A ) remain as a jumbled insoluble residue, and form a thin, distinctive, micro-breccia ( Fig. 21B ) termed a blanket breccia (Anderson et al., 1972; Anderson et al., 1978; Hentz et al., 1989, p. 42). The anhydritic micro-breccia survived dissolution because groundwater that dissolved the NaCl was saturated or nearly so with CaSO 4 Every salt bed recognized on acoustical logs in the eastern side of the Delaware Basin has an equivalent bed of dissolution breccia in the western side of the basin Anderson et al., 1972; 1978). This determination is based anhydrite laminae occurring just below beds of halite in the eastern basin with calcite-anhydrite laminae occurring just below beds of micro-breccia in the western basin. Thicknesses of the dissolution breccias range from a few centimeters to several meters, and the thickness of each bed of breccia is approximately proportional to the thickness of its correlative salt bed (Anderson et al., 1978). Extensive dissolution of Castile halite moved eastward from the Guadalupe Mountain front (the Capitan reef complex) in the latest Tertiary and Quaternary. Within the subsurface, Figure 20. Figure 5
23 Figure 22. 3 and surrounding area. The regional subsidence was augmented by the huge, rapidly deposited (~209,000 yrs; Anderson, 2011) load of Castile evaporites (a weight of ~ 27 18 metric tons, based on a volume of ~10,000 km 3 and a mean density of ~2.7 g/cm 3 ). The great mass of Castile evaporitic sediments helped to depress the crustal surface, and the area in which the evaporites accumulated and beyond was isostatically depressed (part of the earths crust literally sank to upper mantle depths). As the crust slowly subsided, upper Ochoan halite, gypsum, dolomite, limestone, potassium-magnesium salts, and red beds ( Fig. 6 shallow-water sediments, which eventually exceeded that of the Castile, contributed to continued crustal subsidence. The of the time of accumulation (~7 Ma) being represented by nondeposition and minor erosion. Near where the cave belt now trends, these latest Permian strata (upper Ochoan Group: the Salado, Rustler, and Dewey Lake formations) buried both the basinal Castile evaporites (lower Ochoan Group) of the basin and the marginal Middle Permian carbonates of the reef, forereef, and shelf by ~1 km ( Fig. 20 ) (see Crysdale, 1987; Lowenstein, 1988; Garber et al., 1989; Ulmer-Scholle et al., 1993; Klimchouk, 2007, p. 76). In the latest Permian, following precipitation of the Castile evaporites, a huge area subsided (>150,000 km 2 ; see Lowenstein, 1988) including the Castile depositional basin, the fringing, extinct Capitan reef, and a wide many tens of kilometersshelfal area bordering the Delaware Basin (Lowenstein, 1988). The subsidence, including the area of the present Guadalupe Mountains, made space available for uppermost Permian strata accommodation space constitute the upper Ochoan Group (in ascending order, the Salado, Rustler, and Dewey Lake formations ( Fig. 6 )). Figure 22 shows the inferred depositional extent of the thickest of these formationsthe Salado (dashed boundary)and, for comparison, the depositional extent of the directly underlying Castile Formation. Figure 21. A. B. Castile dissolution micro
24 (Hayes, 1964; Kelley, 1971). Castile strata throughout most of the Delaware Basin, except for such localized and generally minor structural disturbances, had (and still have) a similar gradient. Such a structural unit in which strata persistently exhibit the same dip is termed a homocline, and the present-day basin has been subdivided into a homoclinal province and an anticlinal province (Grauten, 1965, his southwestern and far eastern parts. A structural contour map on the base of the Castile east of the Guadalupe Mountains and east of the lower-lying Delaware Mountains ( Fig. 23 ) The homocline encompassed not only the basin, but the Delaware Mountains, the Guadalupe Mountains, and along the mountain front, the Capitan reef and forereef. During the late Miocene, the eastward tilt was not as pronounced, the homoclinal slope probably extended further to the west, departures from the uniform upward dips were fewer, and strata of the ancestral Guadalupe tectonic block probably more closely approached a classic homocline. The uniform dip of the homocline, however, was disturbed near where the cave belt now trends (e.g., Hunt et al., 2003). The disturbance resulted primarily from differential subsidence that occurred during Middle Permian time as the reef prograded into the basin over its own forereef debris and over basinal siliciclastics of the Bell Canyon Formation ( Fig. 20 ). The progradation resulted in downto-the-basin faults and associated folds (e.g., Hunt et al., 2003; Kosa and Hunt, 2006a, 2006b) on which the Late The southern Rio Grande rift, an active thermo-tectonic system of central New Mexico, extends into far west Texas (Seager and Morgan, 1979). The rift experienced renewed activity at 11 Ma (early late Miocene) and 6-4 Ma (latest Miocene-earliest Pliocene) (Lueth et al., 2005), times that coincide with pulses of intense speleogenesis in the Guadalupe Mountains (Polyak et al., 1998). These correlations imply that the major faults that episodically and uniformly tilted strata of the ancestral Guadalupe tectonic block were active at the same time as caves were forming in the mountains (Polyak, 1998; Polyak et al., 2006). Each episode of intense speleogenesis, of which there were at least three (Polyak et al., 2006), was apparently caused by a major increase in eastward tilting of the ancestral Guadalupe tectonic block. With each increase, the accessibility of H 2 S, O 2 or both, to the developing caves increased; sulfuric acid production improved, and the intensity of speleogenesis in the Capitan Formation, Seven Rivers Formation, and, uncommonly, other shelfal units strengthened (see Polyak et al., 2006). Tilting of the paleo-Guadalupe tectonic block culminated in its present 1-2 eastward dip (Hayes and Gale, 1957; Olive, 1957; Hentz et al., 1989; Hill, 1996, p. 219; DuChene and Martinez, 2000), the tilted province rising to the west by 17-22 m/km (King, 1948; Grauten, 1965) ( Fig. 23 ). Strata of the ancestral Guadalupe Mountains and most strata of the Delaware Basin responded as a single structural unit, and with each episode of uplift the accrued gradient of the strata increased slightly. Episodes of tectonic deformation persisted through the late Miocene and early Pliocene, but by the mid-Pliocene they had ceased (DuChene and Cunningham, 2006, and references therein). The uniform dip of Upper Permian strata throughout most of the basinal segment of the paleo-Guadalupe places. Perturbations affecting the consistent eastward dip of Castile strata were principally sparse faulting (e.g., Smith, 1978, 1980; Hentz et al., 1989), minor folds (Kirkland and Anderson, 1970), and elongated anticlines adjacent to the northern Capitan reef (Anderson and Powers, 1978; Hill, 1996, p. 240). In addition, near Slaughter Canyon Cave ( Fig. 5 ), a northwest-trending monocline underlain by a Pennsylvanian or older thrust fault extended into the basin Figure 23.
25 gypsum); CaSO 4 goes into solution, the aqueous solvent increases in density, it becomes gravitationally unstable, and it sinks (e.g., Kempe, 1996). Such anhydritic and halitic dissolution processes can be incorporated into the either from an increase (or decrease) in temperature or from an increase (or decrease) in solute concentration, in either hypogenic or epigenic groundwater, and it may result from a relatively large increase in density, as when halite dissolves, or it may result from a relatively small increase in density, as when anhydrite and gypsum dissolve. Only a small relative increase or decrease in density (>0.01%) is enough to cause water to sink or (In Gulf Coast sediments, dissolution and mass transfer of NaCl occurs even if sediments beneath halite have permeabilities as low as 0.01 md (Sarkar et al., 1995).) Where hypogenic groundwater actively dissolves a salt (e.g., gypsum, anhydrite, halite), the term free convective dissolution is applicable. In laboratory experiments involving dissolution of halite fresh (solutionally aggressive) water and the descending through a simulated fracture. The pathways were close, but the ascending pathway was separate and distinct from almost as immiscible liquids, exhibited little interaction. Natural fractures contain micro-conduits that provide separate pathways for similar ascending and descending turbulence, would probably be only marginally altered by diffusion and by commingling. In addition, densitygroundwater contacts halite or anhydrite liberates potential energy. The liberation is manifest by the kinetic energy of halite and within anhydrite. If the hydrologic system were not dynamic, dissolution would cease. Both halite and anhydrite commonly fuel natural hypogenic-induced convection, with these salts, in a sense, becoming the vehicle of their own annihilation. Aggressive water that dissolves halite increases in Tertiary regional uplift of the Guadalupe tectonic block was superimposed. The Late Tertiary tilting reactivated ancient Middle Permian joints and syndepositional faults and possibly created new ones. The fractures served as pathways for groundwater, as pathways for atmospheric oxygen, and as guides for cave development (e.g., Jagnow, 1977; Kosa and Hunt, 2006b). anhydrite, and haliteunderlie the earths surface in many depositional basins. Such evaporites commonly interact with two principal categories of groundwater: One category, groundwater that descends from recharge surfaces above sedimentary strata is designated epigene. This is the groundwater, for example, that descends from the surface into karstic features such as modern sinkholes. Another category, groundwater that ascends from underlying sedimentary strata is designated hypogene. recharge surfaces (Klimchouk, 2007, p. 3). Hypogenic groundwater, usually artesian, may occur at substantial depth (hundreds of meters), and, under pressure, may rise via forced convection through fractures crossing bedding of non-evaporitic strata to contact evaporitic strata. Such hypogenic groundwater then initiates speleogenesis by free convection without direct connection to the surface (e.g., Anderson and Kirkland, 1980; Klimchouk, 2007; Stafford et al., 2008b). Most epigenic groundwaters are fresh, and in such waters halite is exceedingly soluble 360 g/l at 20C and 370 g/l at 50C, with temperature differences having only slight effect. Even brines are strongly undersaturated with respect to NaCl (Ford and Williams, 2007, p. 45). As the salinity increases, however, the rate of dissolution decreases (Stiller et al., 2007). On contacting halite, undersaturated hypogenic groundwater approaches saturation, its density increases (commonly substantially), it descends, and a type of free convection begins: dense brines sink and simultaneously less dense groundwaters rise. Such convection has been termed A similar process of free convection in the Castile, albeit without formation of brine, occurs when solutionally aggressive, hypogenic groundwater contacts anhydrite (or
26 increases as the slope of pathways (<1 to 90) increases through which aggressive water rises and through which saline water drains. In early studies in the western Delaware Basin, artesian (hypogenic) groundwater was hypothesized to have played a critical role in forming present-day and ancient geomorphologic features as well as mineral deposits on and beneath the Gypsum Plain, namely: east-west trending solution-subsidence troughs (Olive, 1957), sulfur deposits (Hinds and Cunningham, 1970), castiles (Kirkland and Evans, 1976), and many karstic features within the Castile and Salado evaporites (Anderson, 1978; Anderson and Kirkland, 1980). These suppositions have been supported by recent work. Many karstic features within the Castile formed during the Quaternary and Late Tertiary by hypogenic convective groundwater; free convective dissolution Klimchouk, 2007; Stafford et al., 2008a; Stafford et al., 2008b, 2009; Nance and Stafford, 2009; Melville, 2009). Stafford (2008b), for example, states, that the Castile Formation exhibits a diagenetic history that has been and hydrocarbons being delivered upward from permeable clastic units of the Delaware Mountain Group (the group that consists of the Middle Permian Bell Canyon, Cherry Canyon, and Brushy Canyon formations ( Fig. 6 )). Inlet risers, wall channels, ceiling half tubes, and outlet cupolas provide unequivocal evidence of dissolution driven by (mixed convection being forced and free convection occurring together). The magnitude of the hypogenic processes is great: more than half (55%) of all sinkholes in Castile evaporites, for example, of which there are many hundreds, are the result of upward stoping of subsurface voids (Stafford et al., 2008a). In the Late Tertiary in the western basin, a reservoir substantial solutional aggressiveness for anhydrite and halite occurred several meters below the base of the Castile evaporites. The reservoir comprised an essential element density by up to 20%, whereas aggressive water that dissolves anhydrite, which under normal temperature conditions has a solubility equivalent to that of gypsum (Klimchouk, 2000), increases in density by up to only ~0.1% (see Klimchouk, 1997a). However, where hypogenic, solutionally aggressive groundwater, such as most artesian groundwater, contacts bedded anhydrite (or gypsum), the relatively slight increase in density due to incorporation of Ca 2+ and SO 4 2can easily result form large caverns within anhydrite (Kempe, 1996). mechanism for enlarging caves within limestone (e.g., Curl, 1966), although such enlargements, compared to those within anhydrite and gypsum, would occur at exceedingly slow rates. The solubility of anhydrite increases by up to three times in the presence of a NaCl-rich brine (e.g., Klimchouk, 2000). Thus, as a brine substantially undersaturated with respect to CaSO 4 descends through fractures and through voids within bedded anhydrite, it is an effective solvent. Fresh hypogenic groundwater rising within a fracture pathway through a bed of anhydrite could, for example, 4 whereas NaClrich groundwater sinking through the same pathway could potentially dissolve > 4 g/l of CaSO 4 Halite is commonly associated with anhydrite; thus, hypogenic groundwater that dissolves halite forms a solvent for CaSO 4 that enhances permeability during its descent through anhydrite strata. long as solutionally aggressive hypogenic groundwater the ascending and descending groundwater, however, depends on the character of the reservoir of the solutionally aggressive groundwater, the hydraulic pathway, concentration, viscosity), and the reservoir for the sinking, nearly saturated brine. descending brine increases because of a reciprocal relationship with viscosity (Anderson and Kirkland,
27 water within these siliciclastic beds, unlike movement of hydrocarbons, is unencumbered by capillary forces. crude oil to move through sandstone of the upper Bell Canyon Formation, than it was for water (Nottingham, 1960); yet oil migrated through the upper Bell Canyon Formation, and is trapped in sandstone reservoirs in the Bell Canyon include, an eastward dipping potentiometric chemistry over short distances (Davies, 1983). Clearly, within sandstone of the upper Bell Canyon Formation. Artesian pressures during the Late Tertiary probably exceeded present-day artesian pressures (Lindsay, 1998), Where permeability existed, hydraulic pressure during the Late Tertiary drove groundwater (by forced convection) from the Bell Canyon sandstone upward into Castile evaporites. Initial entry points into the lower anhydrite memberthe Anhydrite I Memberwere along joint and fault planes (e.g., Kirkland and Evans, 1976; Hill, 1990; Stafford et al., 2008b), and probably commonly at junctions of joint sets. Furthermore, artesian groundwater for prolonged and persistent convective dissolution within the Castile. The holding reservoir for the aggressive groundwater was porous sandstone of the underlying upper Bell Canyon Formation ( Fig. 6 ). The sandstone, although limited in its transmissibility, provided an (22-27%), weakly cemented, silty, arkosic sandstone (Williamson, 1977). An upper sandstone unit of the Bell Canyon Formationthe Ramsey, a channel sandstone has an average permeability of 39 md (Dutton, 2008). The hydraulic conductivity of the Bell Canyon is variable on local and on sub-regional scales (Davies, 1983). Facies of the Bell Canyon with the greatest potential as a reservoir Dutton, 2008). Such reservoirs trend northeast-southwest, range from less than 0.5 km to more than 6 km in width, 1 m to more than 35 m in thickness, and up to 70 km in length Beds of upper Bell Canyon sandstone presently constitute few stock wells drilled into the Delaware Mountain Group, probably into the upper Bell Canyon, have well yields ranging from 5 to 20 gallons per minute (0.3 to 1.2 l/sec) (Nielson and Sharp, 1990). Movement of pore Figure 24. A. direction markedly, and directly below an B.
28 13 C values (0.0 to +2.5 (Magaritz et al., 1983, their (Stafford et al., 2008b), but by sedimentation. A second barrier, 6-9 m below the base of the Castile (Anderson et al., 1972; Wilde et al., 1999; Tyrrell et al., 2006), the Lamar Limestone Member of the Bell Canyon Formation ( Fig. 6 ), has a thickness of 30 m or more near the reef escarpment. It thins progressively into the basin and within 13 to ~30 km southeast of the escarpment pinches out (Tyrrell et formidable, was the thick (~50 m) basal anhydrite of the Castile Formation (the Anhydrite I Member). Before being fractured during uplift, it was virtually impermeable. With tilting of the Guadalupe block, fractures breached the aquitards, and in the northwestern and west-central Delaware Basin solutionally aggressive groundwater into the Anhydrite I Member. Replacement of anhydrite by gypsum may have occurred along fractures, and the increase in volume of the gypsum may have closed micro-conduits and reduced or eliminated permeability along with overpressured water (forced out of shale) (Lee & Williams, 2000) rose into the upper Bell Canyon from micro-openings along fault surfaces (slip faces). These waters helped re-supply both hypogenic groundwater that rose from upper Bell Canyon siliciclastics into the Castile evaporites (and ultimately moved upward and out of the basin) and nonaggressive brines that sank into Bell Canyon sandstone and ultimately moved downward and out of the basin or downward and into its depths. Presently, much water within the upper Bell Canyon aquifer is highly saline (McNeal, 1965; Hiss, 1975). This was probably not the condition before inception of the Late Tertiary tilting and fracturing when beds of Bell Canyon sandstone, despite residing beneath a thick (>800 m) sequence of halite-rich evaporites, probably contained pore water that was only brackish-to-slightly saline. Three nearly stratigraphically adjacent, lithologic barriers of extremely low permeability (aquitards) prevented brine that may have originated within the overlying halitic section from sinking into Bell Canyon sandstone. One barrier, situated at the base of the Castile Formation, was a thin (< 1 m), basin-wide, laminated carbonatethe Basal Limestone Member (King, 1948; Anderson et al., 1972; Cys, 1978) ( Fig. 17 ). Its petrography (Anderson et al., 1972) and its Figure 25. Figure 5 2 2 S
29 by upward leakage of groundwater from the karstic San Andres limestone (Land, 2006). Most anhydrite breccias and most associated karstic features within the Castile and Salado formations of the Delaware Basin, however, owe their origin, not to dissolution of anhydrite (CaSO 4 ), but to dissolution of the much more soluble evaporite mineral, halite (NaCl). Following tilting of the ancestral Guadalupe tectonic block, rising artesian groundwater at many localities in the northwestern and west-central basin eventually breached the Anhydrite I Member and contacted the base of the Halite I Member of the Castile Formation ( Fig. 17 ). This latter member, free of anhydrite layers > ~3 mm thick, is, at ~125 m, the thickest of the three Castile halite members ( Fig. 17 ). The rising solutionally aggressive artesian groundwater dissolved halite, and chambers grew vertically upward, a consequence of the most aggressive water available for dissolving NaClthe freshest, leastdense watercontinually rising directly to the very top of the growing void where dissolution took place ( Fig. 24 ). The thin intercalated anhydrite laminae within the halite member provided essentially no impedance to upward dissolution. The resulting brine sank into the underlying sandstone beds of the upper Bell Canyon, removing the solute from the Castile. The directly upward growing dissolution chambers within the Halite I Member were eventually blocked. Blockage occurred when advancing voids contacted the intact lower boundary of the thick (~30 m) overlying Anhydrite II Member ( Fig. 24 ), a rock unit of relatively poor solubility and low permeability. The freshest, leastdense, most aggressive water, however, continued to rise; it turned a sharp angle directly beneath the anhydritic ceiling, and dissolved a void within halite directly up the slight tilt of the homoclinal block ( Fig. 24 ). Advancement progressed within the bedded Halite I Member by free convective dissolution, the voids growing laterally westward and slightly upward just below the smooth base of the Anhydrite II Member, which dipped uniformly eastward by < 1 and which extended over thousands of square kilometers. The boundary of the Castile anhydrite members with underlying halite members did not differ greatly from a smooth, slightly sloping plane. Collapse of the westward advancing conduits was impeded by documented for the Castile (Anderson and Kirkland, 1966, their pl. 4), but elsewhere it has seldom been substantiated (R. Evans, personal communication, 1990; Klimchouk, 2000), and deep within the Castile, uncommon. However, if micro-conduits within fractures of the Anhydrite I Member were closed by expanding gypsum, they would have been reopened readily by free convective dissolution. Most castiles contain a central core of calcitized anhydrite breccia (Hayes, 1964; Brown and Loucks, 1988; Stafford, 2008a, p. 166; 2008b) ( Fig. 16B ). The brecciated core apparently formed as bedded anhydrite collapsed into caves that formed as hypogenic to CaSO 4 rose through fractures, contacted bedded anhydrite, dissolved CaSO 4 increased in density, and 4 may have migrated into the voids, ephemerally displaced water, diminished support of directly overlying anhydrite, and induced brittle failure (i.e., by fracturing) (see Bgli, 1980, p. 213). Displacement of the ambient groundwater removed both the buoyant force of the groundwater and the support provided by artesian-pressured and overpressured groundwater. The roof became unstable, and with its collapse into the void, directly overlying beds of anhydrite deformed by stoping, and in a anhydrite. Before cementation and/or compaction, the breccia bodies had substantial permeability, the void space essentially equaling the volume of anhydrite removed (Davies, 1983). Analogous caves (exceeding 200 m in dimension) within Zechstein gypsum or anhydrite of the Sangerhausen and Mansfeld districts, Germany, formed by this same process of free convective dissolution (Kempe, 1996). About 100 cavities of this type are known in the region, encountered through the centuries in the course of mining operations at depths of up to 400 m at the base of the Zechstein gypsum (Klimchouk, 2007, p. 26). In the Guadalupian backreef facies of southeastern New Mexico, numerous hypogene carbonates and evaporite rocks (Stafford et al., 2008b; Stafford et al., 2009). The sinkholes at Bottomless Lakes State Park, ~23 km southeast of Roswell, New Mexico are the product of subsurface dissolution of gypsum
30 conduits (i.e., the upside down topography) controlled precise directions followed by the advancing conduits. points on the ceiling provided a track for the most aggressive groundwaterthe freshest, least denseto anhydrite ( Fig. 24A ). As groundwater with solutional aggressiveness for NaCl paleo-Guadalupe tectonic block ceased in the mid-Pliocene, for example, at a slope of ~20 m/km, its gradient was >1000 times greater than the gradient of the Amazon River from Manaus, Brazil, 1,610 km downstream to Belm, Brazil (online Columbia Encyclopedia). Some conduits departed slightly from linearity because of slight irregularities in the anhydritic ceiling, and because, as NaCl dissolved, gas may have rarely salted out to form temporary obstructive pockets. Furthermore, the path of conduits extending upward from different points of origin commonly coalesced because of chance alignment, and as conduits approached the margin of the basin, they may have received aggressive groundwater from several centers, each having different The morphology of hypothesized conduits within Castile halite must be inferred. Halite because of its high solubility fails to crop out except in extremely arid climates, thus, accessible cave systems within halite are uncommon, and the body of knowledge about such cave systems is limited. Inferences about the morphology of caves within halite based on analogy are untrustworthy compared to those made for the morphology of caves within limestone or gypsum. Conduits within Castile halite are hypothesized to have had because groundwater with maximum aggressiveness for halite moved upward as a stream directly beneath an anhydrite cap following connected subtle highs with only slight tendency for lateral departure, and, hence, for lateral dissolution. Furthermore, the conduits are high-density concentrated brine probably mantled their dissolution ( Fig. 24 A and B ); and they are hypothesized to have had a length that extended for up to several tens of the probably modest width of the voids, the presence of pressured water, and the strength of the anhydritic ceiling. Above the Halite I Member, nearly vertical fractures probably intermittently and transversely cross the slightly tilted Anhydrite II Member. They are probably spaced sparsely because during uplift the intercalated, incompetent beds of Castile halite incorporated most strain. A fracture within the anhydrite ceiling intersected by a conduit provided incipient permeability that may have been slowly enhanced by free convective dissolution. The near-vertical fracture pathway may have eventually allowed pressurized groundwater, upward through the capping bed of anhydrite and to contact a directly overlying bed of halite ( Fig. 25 ). The Anhydrite II, III, and IV members were less prone to dissolution by convecting hypogenic groundwater than the lower part of the Anhydrite I Member because groundwater dissolved much CaSO 4 during its upward passage through lower anhydrite strata and, thus, had limited potential for dissolving more. Once groundwater transversally breached a bed of anhydrite, free convective dissolution once again created a nearly vertically trending chamber through the overlying bed of Castile halite until the void contacted the next intact bed of anhydrite. Then, an up-slope-trending conduit at the top of the bed of halite advanced by the same convective dissolution process directly beneath the eastward dipping and capping bed of anhydrite ( Fig. 25 ). Solutionally aggressive groundwater for halite, for example, that rose transversely through the Anhydrite II Member formed vertical, upward trending chambers through the Halite II. Then, at the top of the Halite II, directly beneath the Anhydrite III, aggressive groundwater dissolved linear conduits up the slight slope of the paleo-Guadalupe tectonic block. Similarly, aggressive groundwater rising transversely within fractures across the Anhydrite III Member may have formed vertically trending chambers within the Halite III Member, followed by up-slope-trending conduits (beneath an anhydrite ceiling) within the member ( Fig. 25 ). Linear conduits within Castile halite propelled forward by free convection advanced up the slight grade of the homocline ( Fig. 25 ). Each began at the widely distributed localities at which a buried body of permeable biogenic limestone was forming, or would soon form. The detailed
31 was presumably slower. With each episode of tectonic uplift, the rate of advancement of conduits increased. In laboratory experiments, distilled water dissolved a block of salt hypogenically through a capillary tube (1.5 mm in diameter) at a rate of one gram per minute in pulsating rate of descent of about 5 cm/sec (Anderson and Kirkland, 1980). These experiments support rather rapid dissolution through natural conduits. The rate of dissolution of Castile halite was persistent and probably always exceeded the rate of ductile closure of halite bounding conduits on their bottom and sides. The natural sinking brine, however, was relatively viscous, as considered above, which slowed its rate of descent, and, in turn, the rate of ascent of solutionally aggressive groundwater. Furthermore, friction between Fig. 24B ) and between the wall rock and the persistent need for aggressive water, and the persistent need for removal of brine from the system, many conduits probably advanced up the homocline at a rate of probably hundreds-to-thousands of years per kilometer. Conduits within halite grew upward until many eventually contacted the Capitan reef or the steep Capitan forereef. Some conduits, however, contacted an anhydritic barrier that mantled parts of the forereef. This obstruction originated from Ca 2+ -bearing Aggressive water within a conduit most actively acquired NaCl at a wedge-shaped dissolution apex ( Fig. 24 B ). It occurred at the westernmost part of a growing conduit, at that part with the highest elevation. Here, conduits probably thinned to less than a centimeter and aggressive groundwater propelled by free convection contacted Castile halite directly, dissolved NaCl, and moved toward saturation. The solvent then reacted to gravity, its direction down the homocline in a direction diametric to that of the rising aggressive water ( Fig. 24 ). Thus, a stream of moderately saline, medium-density, aggressive groundwater within each advancing conduit probably (or nearly saturated), high-density, nonaggressive brine equaling the volume of water descending. A thin residue of conduits (anhydrite laminae constituting 5-10-vol % of Castile halite). Free convective dissolution of halite probably resulted in growth of conduits mainly during the early part of the multi-million-year interval in which the ancestral Guadalupe tectonic block was being episodically uplifted and tilted. The rate of growth of conduits is dissolution of halite advanced vertically within the Halite I Member to create chambers, the rate at which dissolution of halite advanced laterally up a slight slope within the Halite I Member to create conduits Figure 26. Figure 27. exaggerated.
32 up a decreased slope (e.g., < 0.5; illustrated by the slope of the southwestern trend of the Capitan reef on the eastward dipping Guadalupe tectonic block ( Fig. 26 )). Such southwesterly-trending conduits within the halite members, probably relatively commonplace within the Halite I Member, advanced by convective dissolution parallel to the reef until the pressurized hypogenic groundwater within them encountered a pathway into the ancient reef. Many conduits within the Upper Permian Castile halite terminated laterally against the Middle Permian Capitan reef. Near where the Capitan escarpment now trends, and before and during uplift of the Guadalupe block, the Halite III Member, for example, was probably in direct contact with the reef ( Fig. 25 ), which had a dip that exceeded 80 (B. L. Kirkland et al., 1999). Furthermore, the Halite II Member of the Castile was in direct contact with the steep upper forereef, which had a maximum dip of ~65 (Mruk and Bebout, 1993). Its face was unable to retain spring-derived gypsum. Where conduits within halite members of the Castile forereef, the H 2 convection through both fractures and pores into the basin-fringing rocks of limestone and dolomite, rocks that would eventually constitute the cave belt ( Fig. 25 ). spring water, and at places the anhydrite barrier was positioned between shallow-to-moderately dipping beds of Capitan forereef and slightly dipping beds of Castile halite. The upper surface of the Capitan reef was exposed during deposition of Castile evaporites (e.g., McKee, Oriel, et al., 1967; Garber et al., 1989), and it from the lifeless reef into the Castile brine body. The volume of spring water that discharged annually, however, was probably small; only minor amounts of rainfall fell onto the reef both because the surrounding, and because a seasonal, relatively cool, near-surface atmospheric convection (see Kirkland, 2003). Aridity was at a peak during precipitation of Castile halite, temperatures were unusually high, droughts were unusually long, and discharge from the springs was particularly low. Where spring water discharged during the earliest Late Permian into the Castile brine body, it resulted in a thin barrier of gypsum consisting, in part, of gravity deposits that mantled the lower forereef. Following a minor pluvial event, and while Castile halite was the evaporite facies being precipitated, spring water near discharge points diluted the surface of the Castile brine causing halite to cease precipitating within a narrow marginal area. It also introduced Ca 2+ (derived from dissolution by groundwater of reef and back-reef carbonates and backreef gypsum). The Ca 2+ reacted with excess SO 4 2within the near-surface, marine-derived Castile brine, causing calcium sulfate to supersaturate, and inducing gypsum to precipitate adjacent to the reef. Gypsum accumulated where the slope of the forereef was below the angle of repose. With burial in the latest Permian by strata of the upper Ochoan Group ( Fig. 6 ), the gypsum was replaced by anhydrite. Conduits within Castile halite that advanced directly to the lower-to-middle forereef may have been blocked at some places by the spring-derived, anhydritic cover; with their upward and westward advancement obstructed, growth of conduits changed direction. Still within the uppermost part of a bed of Castile halite, they advanced by densitydriven, free convective dissolution southwesterly Figure 28.
33 of thousands of years. Within this extended interval, solutes that resulted from dissolution of Castile halite and anhydrite (removal of which formed the chambers and conduits) could have easily been accommodated by Bell Canyon sandstone (e.g., Anderson, 1978, p. 56). Lambert (1983) concluded that the Bell Canyon was an ineffective repository of sinking Castile brine, but his assertion was effectively countered by Davies (1983) the Bell Canyon. A large volume of epigenic groundwater having NaCl in solution moved into Bell Canyon sandstone after H 2 S-H 2 SO 4 speleogenesis in the Guadalupe Mountains had ceased. The groundwater may have been derived, in part, from the western shelf as well as from the western Delaware Basin. Even today, epigenicand hypogenicderived brines from dissolution of Castile and Salado slope within Bell Canyon sandstone. These later phases of brine generation (latest Tertiary and Quaternary) were primarily responsible for the present-day, basinal distribution of chlorinity (a stand in for salinity) that increases gradually eastward from fewer than 10 g/l in the western part of the basin to about 150 g/l along the eastern margin of the basin (Hiss, 1975). Diffusion brine-drainage smoothed heterogeneities of the salinity gradient. A series of solution-subsidence troughs on the western Gypsum Plain trend parallel to regional dip and extend from near the latitude of Cottonwood Cave ( Fig. 5 ) south for several tens of kilometers into Texas. These karstic features, which are abundant on the western Gypsum Plain of Texas, are commonly straight, narrow, shallow, and east-northeast ( Fig. 28 ). They are typically a few meters deep, 0.01 to 1.6 km in width, and 0.8 to ~16 km in length (Olive, 1957; Hill, 1996, p. 312). The troughs have been mapped in Texas and New Mexico by King (1949) and in Texas by Hentz et al. (1989, their pl. 1), Plain is remarkably well displayed in Google Earth. Their eastward (down-dip) limit does not extend beyond the updip limit of either sub-eroded Castile halite or suberoded Salado halite ( Fig. 18 ) (Hinds and Cunningham, Groundwater within the Bell Canyon Formation during artesian) upward into the halite members of the Castile Formation where it was transformed into dense brine conduits, and discharged into Bell Canyon sandstone. Relatively fresh, low-density, solutionally aggressive groundwater entered the Castile from the upper part of the underlying beds of sandstone, whereas solutionally nonaggressive, high-density groundwatercommonly saturated, or nearly so, with both NaCl and CaSO 4 sank from the Castile into the lower part of underlying beds of Bell Canyon sandstone ( Fig. 27 ). The discharging brine then streamed down the homoclinal slope within Bell Canyon sandstone following the lowest, most permeable across the basin, then moved upward under pressure into the Capitan aquifer, and near present-day Hobbs, New Fig. 6 ), and ultimately discharged into paleo-streams that extended to the ancestral Gulf of Mexico (see Hiss, 1975, 1980). Some brine that descended into the Bell Canyon probably by-passed the Middle Permian easterly and northeasterly escape routes and moved directly downward through any available connected voids (e.g., interstitial pores and bladed cracks) into rocks within and beneath the Bell Canyon where it displaced less-dense water. Such deep descending brines are expected. They are sequestered within the depths of probably most of the earths major sedimentary basins, and in some basins, deep brine is the sole record of evaporitic deposition that has otherwise vanished (by dissolution) from the is either pore water with a greater density than that of the descending brine or an impermeable lithologic barrier, which beneath some sedimentary basins occurs thousands of meters below the contact of sedimentary strata within fractured igneous and metamorphic basement rocks (e.g., Mller et al., 1997). Bell Canyon was persistent, long lasting, and effective. from discharge points without interruption for many thousands of years and perhaps for many hundreds
34 Formation. The Ochoan-derived brine had substantial solutional aggressiveness for CaSO 4 through remnant voids and through permeable breccias of anhydrite and gypsum within previous dissolution conduits, it may have put substantial CaSO 4 into solution and it may have formed solutionally enhanced linear restricted to a calcium sulfate lithology, support for the caves weakened, and overlying anhydrite and/or gypsum roof-rock collapsed. With settling and compaction, linear troughs possibly formed on the western Gypsum Plain, troughs that mimic the older dissolution conduits. The hypothesized dissolution conduits within Castile halite have few recognized karstic counterparts. The nearest analogue is possibly the nearly horizontal, commonly broad chamber that forms at the crest of domes. Such chambers opened and closed repeatedly with a period of several thousand years, a record preserved in overlying anhydrite caprock as bizarre sedimentary beds termed, katatectic layers (Goldman, 1933, p. 84; 1952, p. 6; Taylor, 1938, p. 12). At many domes, the cyclic karstic process is apparently still active. Crestal dissolution chambers form entirely within diapiric halite directly below the nearly horizontal base of anhydrite caprock. During their maximum open phase, chambers (~1 m high (e.g., Goldman, 1933, p. 92)) probably extend laterally for up to several kilometers. They grow laterally by convective dissolution that proceeds from the annular margin of a dome toward its central axis. Scattered pillars of anhydrite, yet-to-bedissolved halite, and high-pressure water that rises from deep overpressured strata along the steep, permeable, annular, domal margin support the broad voids. The fragmentsa residue of anhydrite laminae (~2-8 wt%) once intercalated within the halite (Taylor, 1938, p. 110). (The fragments are analogous to those that resulted from dissolution of Castile halite (see Fig. 21B )). The central part of salt diapirs, in plan, rises slightly faster than the peripheral part (Goldman, 1952, p. 19). Thus, The troughs were investigated by Olive (1957) who hypothesized that Castile gypsum was dissolved hypogenically along underground drainage channels following joints that extended parallel to the direction of regional dip, and that when the roofs above channels could no longer be supported, collapse ensued. Rather than being structurally controlled, could these modern solutionsubsidence troughs ( Fig. 28 ) be a surface manifestation of earlier formed Castile dissolution conduits? Such a karstic formation, which may have taken place during the Pliocene, would explain their lineation, their abundance, and especially their easterly dip directly down the slope of the Guadalupe tectonic block. South of the latitude of Carlsbad Cavern in New Mexico, the Anhydrite III and IV Members of the Castile and Quaternary alluvium may have obscured some solution-subsidence troughs (as well as some castiles). The unusual troughs probably exist because of dissolution of gypsum or anhydrite, or both, not because of dissolution of halite, which by the beginning of the Pleistocene, at the latest, had been pervasively removed from areas now encompassed by the troughs (Stafford et al., 2008a). In the early late Miocene, a thick sequence (hundreds of meters) of primarily Rustler and Salado strata covered the western Delaware Basin the paleo-Guadalupe Mountains, and, in Texas, the area represented by the Delaware Mountains (and beyond). A prolonged phase of erosion, which extended through Pliocene time, followed that removed the covering strata. Runoff resulting from erosional dissolution of the thick section of Salado halite was rich in NaCl. During the latest Tertiary, some runoff brine possibly gained access into subsurface dissolution conduits within the Castile Formation. At that point, the conduits possibly became artesian escape routes for dense, saline meteoric water that originated in elevated ground. The older phase of hypothesized convection, which was forced and which propelled groundwater up-gradient from east-to-west, may have ceased in the Pliocene as artesian pressures waned. A new phase of hypothesized convection (which was also forced) may have now propelled saline groundwater down-gradient from within earlier-formed conduits and ultimately into the alluvium, and into channel sandstone of the Bell Canyon
35 crest of salt domes tend to be ring-shaped and broad. Castile never approached horizontality, those within chambers at the crest of salt domes did so periodically. Generation and Migration of Methane in Late Tertiary of Western Delaware Basin A great volume of CH 4 was generated in the Delaware Basin just before and during speleogenesis in the Guadalupe Mountains. These two eventsCH 4 generation in the basin and speleogenesis in the Guadalupe Mountainsare closely of CH 4 generation, neither the caves of the Guadalupe Mountains nor the large subsurface deposits of native sulfur of the western Delaware Basin would have formed. In this section, I consider formation of the CH 4 its migration upward into the Bell Canyon, and its further migration upward into the lower Castile Formation. through the late Miocene (Barker and Pawlewicz, 1987, 1993). Compared to a geothermal gradient during the Holocene of 18-21C/km (1.0-1.2F/100 ft) (Mazzullo, 1986) and to an estimated paleo-geothermal gradient during the Paleozoic of 30C/km (1.6F/100 ft) (Barker the Miocene resulted in a paleo-geothermal gradient of 40-50C/km (2.2-2.7F/100 ft) (Barker and Haley, 1986; Barker and Pawlewicz, 1987, 1993). Support for the high late Miocene geothermal gradient analysis of a myriad of microscopic particles of vitrinite. These vitrinite particles were derived from woody tissues of higher plants and they commonly reside abundantly as viewed under immersion oil with a microscope increases logarithmically with their level of maturation (essentially with the extent of their cooking; preeminent factors being time and, especially, temperature (e.g., Tissot and Welte, 1984, p. 222-223)). The vitrinite was extracted from its mineral matrix using strong acids in a technique termed acid maceration. Strong acids dissolved mineral matter within samples of cores and well cuttings from about 50 wells, and insoluble particles of vitrinite within the residue were prepared for analysis. perimeter of a dome at a low angle (estimated at 1-3). Groundwater most aggressively dissolves halite at that part of a chamber closest to the central vertical axis of the highest elevation. The resulting saturated (or nearly slowly down the slight incline to the outer margin of the dome. Forced convection of groundwater from outside the annular margin replaces departing brine; the inwardare preferentially dissolved because the inward moving groundwater within the uppermost part of the thin, ringshaped chamber (directly below the anhydrite caprock) because downward dissolution of halite highs operates horizontality through persistent dissolution, is called a salt mirror by Fulda (1938), a solution table by and without free convective circulation, a stagnant layer of nonaggressive, saturated brine covers the horizontal diapiric crest. Downward dissolution of halite virtually ceases, and by deformation, the diapir moves slowly thousand years, the chamber closes. During closure, the layer of uncemented, micro-fragments of anhydrite on anhydrite caprock forming a katatectic layer (in reverse superposition; i.e., the lowest layer in the sequence of caprock layers being the youngest). Then, at the crest of the halite dome, beginning at the annular margin, a new cycle of karstic dissolution begins. Both types of dissolution voids, the hypothesized channels within bedded Castile halite and the poorly documented anhydrite-capped salt domes, had (and have) a nearly uniformly smooth anhydritic ceiling, a slightly dipping free and forced convection. The evaporitic karst systems, however, have morphologies that differ: although both systems probably had (and have) a low height, developing dissolution voids within the Castile were probably linear and narrow, whereas developing dissolution voids at the
36 decompose systematically into a variety of petroleum components of lower molecular weight plus graphitic carbon. Two principal episodes of generation of petroleum occurred within source strata of the Delaware Basin, a primary episode mainly during the Middle and Late Permian, and a secondary episode during the Late Tertiary. Paleozoic strata were nearly at maximum depth of burial and at (or nearly at) maximum temperature by the end of the Permian Period, and basinal source strata had generated huge volumes of crude oil (e.g., Hills, 1984; Hill, 1996, p. 351) much of which migrated out of the basin and into traps on the surrounding shelf. Heat the Mesozoic and Early Tertiary, and additional burial of Paleozoic strata was meager. Additional maturation of kerogen was likely to have been slight, thus, only relatively small volumes of new petroleum would likely have been generated. This situation changed with the strata along especially the western side of the basin were subjected to higher temperatures than those achieved during the near-maximum burial of the Late Permian (Barker and Pawlewicz, 1987, 1993). The ephemeral, Late Tertiary heating event resulted not in generation of huge volumes of additional oil, but, importantly, in generation of vast volumes of dry gas. The CH 4 of the Late Tertiary was generated in part from cracking of oil generated in the western basin more than two hundred million years earlier. The generation processes of the Late Paleozoic left a profusion of droplets of oil dispersed within Lower Permian and within older addition, sporadic accumulations of trapped oil most of which were minor. With increasing Miocene stratal temperatures, the top of the principal zone of generation of dry gas moved upward. As it did so, oil within disseminated droplets and oil trapped within reservoirs were transferred into the zone of generation of dry-gas where it cracked systematically into progressively lowerand, ultimately, into large volumes of CH 4 the terminal hydrocarbon product. Paleozoic kerogen dispersed within strata below a depth of >2 km in the western basin also decomposed during the Late Tertiary episode of maturation further generating The well samples were from widely distributed localities in and surrounding the Delaware Basin and throughout much of the represented geologic section (Pawlewicz et al., 2005). For the western Delaware Basin, using the of vitrinite particles, Barker and Pawlewicz (1987, 1993) reconstructed the Miocene paleo-geothermal gradient, the Miocene zone of generation of crude oil, and the underlying Miocene zone of generation of dry gas (i.e., natural gas consisting of >95% CH 4 ). Brown (2004) attributed the anomalously high, thermal condition of the western Delaware Basin to Cenozoic volcanics; Barker and Pawlewicz (1987), on the other hand, attributed it to magmatic bodies and thinning of the crust. Each of these heating events probably played a role. Extensive volcanism, which began in early Oligocene, occurred near the southwestern margin of the Delaware Basin (the Davis Mountain area) (Anderson, 1968; Parker and McDowell, 1987; Henry et al., 1994), but by the late Miocene its heating effects were reduced. The past presence of a deeply lying Late Tertiary body (or bodies) of magma in the northwestern Delaware Basin are inferred from ~30 Late Tertiary igneous sills and dikes, whichalong with probably others undetectedwere injected along faults into Paleozoic strata (Kelley, 1971; Calzia and Hiss, 1978). Furthermore, a late phase of Basin and Range deformation in the Miocene stretched both the crust and the uppermost mantle, and the consequent thinning extended westward into the Delaware Basin (Barker and Pawlewicz, 1987; Hentz and Henry, 1989; Hentz et al., 1989). Here, and in shelfal areas west of the basin, the crustal thinning allowed deep, hot material to move to shallower depths, and, thus, to contribute to Great volumes of petroleum (gas and oil) have been generated in the Delaware Basin. It has been generated primarily from maturation of a type of insoluble organic matter, termed kerogen, dispersed within marine carbonates. Molecules of kerogen commonly have remarkably high molecular weight, widely varying composition, and exceedingly complex
37 unable to do so at a fast enough rate, and, therefore, pressures within source strata increased substantially beyond their normally expected hydrostatic pressure. CH 4 and water generally escaped slowly. Faults, however, may have been generated or regenerated suddenly during episodes of tilting of the paleo-Guadalupe tectonic block or during the Basin and Range deformation, and CH 4 and overpressured water may have moved precipitously along micro-conduits of bladed fracture surfaces between high-pressured geologic sections and overlying lower-pressured geologic sections. A second pressured hydrologic regime in the westcentral and northwestern basin was established in the early Late Tertiaryan artesian system, as mentioned driven by the topographic elevation of recharge areas high in the western ancestral mountains. The primary artesian aquifer, according to Lee and Williams (2000, Bell Canyon Formation, but the thin, continuous third sand aquifer of the underlying Bone Springs Formation (Lower Permian; Leonardian series) ( Fig. 4 ) (see Montgomery, 1997). On the other hand, as the primary invoked sandstone of the Cherry Canyon Formation (Middle Permian; Guadalupian series ( Fig. 6 )). Artesian water, overpressured water, and CH 4 within some of these underlying aquifers probably moved upward along faults across hundreds of meters of Paleozoic section into the overlying Bell Canyon Formation. The pressurized Upper Permian evaporites within beds of Bell Canyon sandstone where they mixed with artesian groundwater moving eastward within the same beds of Bell Canyon Strain associated with tilting of the ancestral Guadalupe tectonic block, as considered above, created and/or rejuvenated steep joints and steep basinal faults (e.g., Anderson, 1981). Furthermore, crustal extension during the associated Late Tertiary Basin and Range deformation created steeply dipping, northeast-trending, normal faults (Smith, 1978; Hentz and Henry, 1989; Crawford and Wallace, 1993). These deformations created fractures within outcrops near the western edge of the copious volumes of natural gas (Lee & Williams, 2000). Residual Lower-Permian-and-older kerogen, which had been nearly non-reactive since Permian time, was transferred into the rising zone of generation of dry gasrising in response to the increasing stratal temperatures. The kerogen cleaved further than it had during the Late Paleozoic generating progressively smaller molecular fragments (those with several carbon atoms) and, ultimately, generating residual graphite and large volumes of CH 4 Finally, increased stratal temperatures caused the top of the terminal stage of maturation, that of metagenesis, to move upward. Residual Paleozoic kerogen transferred into this zonewhich was situated deep within the sedimentary sectionwas subjected to intense maturation. Particulate organic matter of all types, including vitrinite, trended toward graphite, and, concurrently, it generated remarkably large volumes of natural gas, the hydrocarbon fraction being exclusively CH 4 (e.g., Kopp et al., 2000). throughout much of the Delaware Basin within deep Mississippian, Pennsylvanian, and Permian shales (Lee and Williams, 2000). The excess pressures provide a mechanism for moving water and CH 4 into younger strata. Fluid pressure within overpressured strata exceeds normally expected hydrostatic pressure, which by an imaginary column of water extending vertically from the earths surface to the depth in question (if fresh, a gradient of 9.74 kPa/m (~0.43 psi/ft)). The transient, high-stratal temperatures of the Late Tertiary initiated the abnormally high pressures, which coincide with the principal zone of generation of natural gas (Lee and from the generation of the natural gas (Lee and Williams, 2000; Hansom et al., 2003). The basinal CH 4 as discussed above, was generated Paleozoic strata. A unit volume of such organic matter on being subjected to the severe Late Tertiary maturation generated many unit volumes of natural gas (e.g., Lee and Williams, 2000). To reduce the excess volume, water along with natural gas, dominantly CH 4 attempted to move out of the source rocks through micropores and
38 Delaware Basin. Much about these microbes however, remains a puzzle. In fact, the microbially mediated reaction between CH 4 and SO 4 2was once thought to be impossible (e.g., Ivanov, 1968, p. 13). In the laboratory, working with various cultures of sulfate-reducing microorganisms, for example, Sorokin (1957) was unable to detect a reaction between CH 4 and SO 4 2. Using pure cultures of sulfate-reducing bacteria, Davis and Yarbrough (1966), and much later, Harder (1997), were able to oxidize radioactive CH 4 ( 14 CH 4 ) by SO 4 2, but at or taxa of sulfate-reducing microbes that can effectively oxidize CH 4 are unknown (e.g., Skyring, 1987; Widdel, 1988). That CH 4 can be the microbial foodstuff seems quite remarkable: This small moleculethe simplest, lightest, and most abundant of hydrocarbonshas a carbon-hydrogen covalent bond among the strongest of the hydrocarbons, and of all possible reactive organic compounds, it is the most stable (personal communication, W. L. Orr, 1989). Furthermore, compared to anaerobic oxidation of other metabolizable organic substrates, anaerobic oxidation of CH 4 provides only small amounts of energy for microbial functions (i.e., Wake et al., 1977; Valentine, 2002; Hinrichs and Boetius, 2002). The perplexing issues surrounding this puzzle still hold, and the process remains a geochemical and microbiological enigma (Valentine and Reeburgh, 2000; Alperin and Hoehler, 2009). Although researchers recognized the process, known as anaerobic methane oxidation, about 35 years ago, it remains poorly establish the reaction mechanism, fully understand the factors that control oxidation rates, or isolate responsible organisms (Alperin and Hoehler, 2010). Nevertheless, within anoxic sediments of most present-day marine environments, within sediments associated with marine CH 4 seeps, and within, at least, some present-day saline lacustrine environments, sulfate anions are clearly reduced anaerobically; and it has been demonstrated unequivocally that the microbial reduction consumes CH 4 (Harder, 1997). The same or similar microbial process, although seldom recognized, also occur within anoxic terrestrial strata (e.g., Kirkland et al., 1995). The sum of the evidence for a microbial redox reaction involving oxidation of CH 4 and reduction of SO 4 2within anaerobic marine Gypsum Plain (just south of the Guadalupe Mountains in the Delaware Mountains) ( Fig. 13 ). They can be seen near the exposed contact between the Bell Canyon Formation and the Anhydrite I Member of the Castile Formation (King, 1948; Olive, 1957; Dietrich et al., 1983; Hentz et al., 1989) ( Fig. 23 ). To the east, an extensive gypsite mantle and various karstic features obscure most fracture traces on the Gypsum Plain (Hentz et al., 1989, p. 36). A great volume of gaseous CH 4 probably many billions of cubic meters, migrated within the Delaware Basin during the late Miocene and early Pliocene. CH 4 moved out of Permian source beds of Wolfcampian, Leonardian, and early Guadalupian age ( Fig. 4 ), and probably out of Pennsylvanian and older source strata into carrier beds and into both new and reactivated fractures. Driven by its abnormally high pressure (Lee and Williams, 2000) and by its buoyancy, gaseous CH 4 wherever possible, moved persistently upward. Eventually, much CH 4 resided within beds of sandstone in the upper Bell Canyon Formation. Upward migration of earlier generated natural gas and crude oil had been blocked by the impermeable limestone barriers; but the barriers that protected the overlying anhydrite and halite from dissolution and the overlying anhydrite from reaction were now breached, and the newly formed fracture pathways allowed both CH 4 and fresh-to-brackish groundwater to rise into the Anhydrite I Member of the Castile Formation ( Fig. 17 ). Reaction between Methane and Sulfate Anions in Late Tertiary of Western Delaware Basin Aqueous CH 4 has the thermodynamic potential of reacting with sulfate anions (SO 4 2). The reaction is activated either thermally or enzymatically (i.e., catalytically). In the area of the Gypsum Plain, microbial enzymes caused the activation within the Castile and Salado formations. In this section, I consider the enigmatic biogenic process and the fate of the biogenic by-products: CO 2 and H 2 S. In addition, I present sulfur isotopic data that support activation by microbes, and I argue that beneath the Gypsum Plain, thermochemical sulfate reduction of Castile anhydrite by CH 4 During the Late Tertiary, particular strains of anaerobic microbes were the agents that allowed CH 4 to reduce sulfate anions in the northwestern and west-central
39 microbes can vary with local environmental conditions. 34 S value for the samples of near-surface 34 S of +11.6 (n=36) exhibited by Castile anhydrite and gypsum (e.g., Kirkland et al., 2000). This is true as well for three samples each from a different large deposit of native sulfur several hundred meters beneath the Gypsum 34 S values of -4.7, -0.3 (Hill, 1996; her appendix 2), and +6.7 (Davis and Kirkland, 1970). The sulfur isotopic signatures of these 34 S values of the H 2 S oxidized to form the native sulfur. Not only do the samples of native sulfur have relatively isotopically 34 S values (21.8), characteristics that support a microbial origin for the H 2 S. Residual anhydrite associated with sulfur mineralization at the Pokorny deposit has an 34 S of +26.6), which is also consistent with a microbial origin. Stafford et al. (2008b) hypothesized that calcitization of Castile anhydrite and the accompanying generation of H 2 S beneath the area delimited by the Gypsum Plain resulted from thermochemical sulfate reduction. Sulfate anions and organic matterdominantly fractions of oil but also possibly CH 4 (Worden and Smalley, 2004) react during this abiotic process to generate H 2 S and CO 2 probably failed to occur within the Castile of the western Delaware Basin. This conclusion is based on the inferred ambient temperature during the postulated reaction, on 34 S values of samples of native sulfur from the minor 34 S values of samples of native sulfur from the major subsurface deposits. Almost all estimations of the temperature at which thermochemical sulfate reduction is initiated are >120C (e.g., Claypool and Mancini, 1989; Heydari and Moore, 1989; Worden et al., 1995; Machel et al., 1995; Rooney, 1996; Heydari, 1997; Worden et al., 2000; Cai et al., 2004). Lower Castile evaporites were buried during calcitization by about 1 km of overburden (Barker and Halley, 1986; Crysdale, 1987; Luo et al., 1994, this depth, well below 120C, were inadequate for a but they were probably nearly optimal for a high rate sediments is compelling (Valentine, 2002), and the a major factor in global carbon cycling (Strous and Jetten, 2004). At a pH less than about seven, the overall reaction is expressed as: SO 4 2+ CH 4 H 2 S + H 2 O + CO 3 2The marine microbes that mediated the modern reaction are probably the same as or related closely to those active within the Castile during the Late Tertiary. The microbes that promote the redox reaction are probably archaea and sulfate-reducing bacteria working as symbiotic aggregates (e.g., Hinrichs et al., 1999; Boetius et al., 2000; Valentine, 2002). There is, however, a possibility that some archaea oxidize CH 4 without the need for a syntrophic partner bacterium (Valentine and Reeburgh, 2000). Archaea, a separate domain of living organisms (along with Bacteria and Eukarya) exist within a variety of terrestrial, freshwater, and marine habitats (DeLong, 2003), but they are renown for surviving, and commonly thriving, within extreme environmentsthriving, in part, because they usually completely lack competition. Like archaea, sulfate-reducing bacteria can tolerate wide variations in salinity, temperature, pressure, and pH (e.g., H 2 O, CH 4 and SO 4 2within their Castile habitats, growth of these puzzling CH 4 -oxidizing bacteria and/or archaea would have confronted few ecological barriers; the principal ones being an unusually high (toxic) concentration of dissolved H 2 S (e.g., Reis et al., 1991), a dearth of trace amounts of critical nutrients (NO 3 -1 PO 4 3, etc.) (e.g., Ehrlich, 1990), a temperature > ~85C (e.g., Machel, 1987), and a trace or more of dissolved O 2 (e.g., Pfennig et al., 1981). Samples of native sulfur from major and minor deposits of the western Delaware Basin have sulfur isotopic values that are isotopically light. Seven samples of native sulfur from minor occurrences associated with 34 S values that range from -15.1 to +9.2 with a mean of +1.6 and a median of +3.0 (data from Kirkland and Evans, 1976; Stafford, 2008a, 34 S values for samples of native sulfur within the caves of the Guadalupe Mountains ( Fig. 12 ); this is not unusual since sulfur isotopic fractionation imparted by sulfate-reducing
40 For their cellular carbon, the microbes assimilate dissolved CO 2 a by-product of the reaction, and/or fatty acids such as acetic acid (CH 3 COOH) dissolved in trace amounts within ambient water (e.g., Jansen et al., 1984). Microbes that inoculated reaction sites within Castile anhydrite were likely introduced in the Late Tertiary from an extraneous source. Probably not until after the tectonic block had been tilted and fractured did the lower Castile evaporites have the abundant living space and the required nutrients necessary for vigorous microbial growth. Modern sulfate-reducing bacteria are obligate anaerobes (Atlas, 1997, p. 990), and this was true of sulfate-reducing microbes within the Castile, whether they were bacteria and archaea working symbiotically, or archaea working alone (see Pfenning et al., 1981). Following the initial tectonic deformation and its attendant fracturing, microbes were probably transported within groundwater from anoxic niches within the Bell Canyon Formation upward through fractures into anoxic niches within the Castile Formation. Microbial loci were scattered geographically within the subsurface of the western Delaware Basin. The locations of many microbial loci are presently represented by the castiles ( Figs. 13 and 15 ). At the microbial loci, aqueous CH 4 and SO 4 2of microbial sulfate reduction. Furthermore, within the upper Rustler Formation at the Culberson sulfur deposit ( Fig. 32 ) (Crawford and Wallace, 1993, their figs. 5-7) and ~15 km south of the Culberson deposit at the Dutch Draw sulfur deposit (Salisbury, 1992), calcium sulfate reacted with CH 4 to yield sulfur and calcite at a depth of 0.5 km or less and at a temperature <<120C. If thermochemical sulfate reduction had operated, the basinal samples of native sulfur, rather than having isotopically light values and a broad range, would have had, in all likelihood, isotopically heavy values and a narrow range. H 2 S generated via thermochemical sulfate reduction, with rare exception (e.g., AlonsoAzcrate et al., 2001), has an isotopic signature nearly identical to that of its parent anhydrite (e.g., Krouse, 1977; Orr, 1986; Goldhaber, 1993; Machel et al., 1995; Worden et al., 1995; Worden and Smalley, 1996; Worden et al., 2000; Cai et al., 2004). If the reaction between sulfate anions and hydrocarbons were caused by elevated temperatures instead of by microbial 34 S values of native sulfur within the limestone host rock would probably have clustered 34 S value for sulfur atoms combined within primary Castile anhydrite and 34 S values have a wide range, a mean of +1.6 for near-surface sulfur, and a mean of +0.6 for subsurface sulfur. Reaction between CH 4 and SO 4 2probably occurred in the western basin during about an eight-million-year interval (~12 to ~4 Ma ago) primarily within the moderately buried (by ~0.8-1.0 km) lower Castile Formation. The reaction, intense at times, was mediated by sulfatereducing microorganisms. The overall diagenetic redox reaction was identical to that which occurs within modern marine sediments; but the setting within the Castile Formation was terrestrial, and it took place about evaporitic sediments of the Castile Formation. Microbial enzymes greatly accelerate the rate of the reaction between CH 4 and SO 4 2and allowed microbes to take advantage of energy released as the reaction proceeds. The onset of the reaction is nearly instantaneous and in most geologic settings, the rates are extremely high compared to most inorganic geological processes (Machel, 2001). Figure 29.
41 in at least one large mass of diagenetic limestone were in place before calcitization (Crawford and Wallace, 1993), and all calcite breccias within the castiles and within the subsurface carbonate masses probably had this same paragenesis: brecciation of anhydrite followed by calcitization of clasts. Many samples of limestone from the castiles and from permeability ( Fig. 29 ). The brecciated cores of castiles at the time of their formation had a porosity that equaled the volume of mineral matter dissolved. Furthermore, replacement of CaSO 4 by CaCO 3 commonly created void space; the calculated quantities of compounds involved (the stoichiometry) indicate that the replacement reaction yields calcite with a porosity of 20-25% (e.g., Davis and Kirkland, 1970; Kreitler and Dutton, 1983; Machel, 2001). Similarly, where complete reduction of gypsum occurred and where H 2 S failed to escape but there are 3-4 parts of calcite to each part (by weight) of sulfur (Ivanov, 1968, p. 104). Within individual castiles, an intermediate zone between the brecciated core and the peripheral zone commonly exhibits porous replacement calcite; it is vuggy and shows original lamination, although laminae are commonly distorted ( Fig. 29 ) (Stafford et al., 2008b). The peripheral zone of castiles is unbrecciated and is not macroscopically porous (Stafford et al., 2008b). Here, calcite appears to have replaced anhydrite volume-for-volume (Brown and Loucks, 1988), a replacement process that would seemingly not have created porosity. Enough porosity and permeability must have been created, however, to allow water, CH 4 and microbes to move to the reaction front, i.e., the boundary between biogenic calcite and primary Ochoan anhydrite. Reaction ceased at such a boundary when there was inadequate living space for the microbes, when CH 4 in solution was unable to move to the anhydrite-calcite boundary at a fast enough rate to maintain growth of the microbes, or when H 2 S was unable to move away at a fast enough rate to avoid poisoning the microbes. Calcitization would have failed to occur at the subsurface microbial loci unless the aqueous H 2 S generated as a metabolic by-product was oxidized (to elemental sulfur) or unless pathways allowed it to escape. Its oxidation or anhydrite members of the Castile Formation. As the two constituents reacted, ambient waterin a sense, acting as a second catalystbecame a refreshed solvent for gaseous CH 4 and for local anhydrite. The same water was used repeatedly as a solvent during the redox reaction, in fact, the reaction created water as a byproduct. Importantly, there were few limits on growth of the microbes, and probably only a small fraction of CH 4 that migrated into their habitats escaped consumption, carbon and hydrogen atoms within the CH 4 combining nearly wholly with oxygen and sulfur (from SO 4 2anions) to form microbial biomass, and, in addition, CO 2 H 2 O, and H 2 Sthe metabolic by-products expelled from the microbial cells. Anaerobic microbial agents were apparently actively forming H 2 S in the widely distributed buried masses of biogenic limestone of the northwestern and west-central Delaware Basin at about the same time as aerobic microbial agents were actively forming H 2 SO 4 in the caves of the Guadalupe Mountains. Furthermore, the limestone masses, which on differential erosion became castiles ( Figs. 13 and 15 ), probably formed contemporaneously with and immediately after formation of the linear dissolution conduits within the uppermost parts of beds of Castile halite. The diagenetic carbonate masses were the habitats of the sulfate-reducing microbes. It is where they generated H 2 S and CO 2 the microbial by-product CO 2 forming HCO 3 and CO 3 2ions within ambient water. As the microbes used SO 4 2in their metabolism, anhydrite (CaSO 4 ) continuously dissolved to the limit of its solubility, as it did so, Ca 2+ became available, and reacted with the CO 3 2to precipitate CaCO 3 Calcitization resulted in nearly simultaneous solution of anhydrite and precipitation of calcite, a process that commonly accurately preserved the fabric of Castile anhydrite ( Fig. 16A ). Calcitization occurred dominantly at interfaces between calcite and anhydrite. It progressed, for example, at the boundaries of the porous limestone masses outward into anhydrite bedrock at right angles to fracture planes along dual solution-precipitation fronts, and within breccia clasts at the boundary between anhydrite and biogenic calcite. The angular clasts of anhydrite-collapse breccias were particularly susceptible to calcitization because they provided an abundant surface area for dissolution and for microbial substrate. Clasts of anhydrite breccia
42 al, 2008b). Furthermore, calcitized evaporites are usually observed within caves that consist primarily of laminated or massive gypsum (Stafford, 2008a, p. 118). The largest of such caverns, Dead Bunny Hole, with a length of ~440 m, is a hypogenic maze cave developed in both laminated gypsum and calcitized evaporites (Stafford et al., 2008a). Another large void within a castile, described originally as a natural cavern (Porch, 1917), is actually a 37-meter-deep vertical mineshaft dug in the early 1900s in search of sulfur (Stafford et al., 2008a). The mineshaft at three separate depths contains small voids (<10 m 3 each). Stafford et al. (2008a) attributed these minor voids to sulfuric acid dissolution of the biogenic limestone. H 2 S is still issuing from this mine (Richardson, 1905; Kirkland and Evans, 1976; Smith, 1980; Hill, 1996, p. 306), and the small isolated voids may represent late-stage dissolution; the strong acid being generated by near-surface oxidation (by aqueous O 2 ) of H 2 S to native sulfur within shallow groundwater, and its further oxidation to H 2 SO 4 During the late Miocene and early Pliocene, when all castiles (i.e., all bodies of biogenic limestone) were buried, most H 2 S generated in situ failed to react locally with O 2 to form native sulfur. Reaction failed because a thick (~1 km) cover of mainly Salado and Rustler evaporites (upper 2 -bearing meteoric water and because hypogenic (largely artesian) water was an inadequate source of O 2 Brown (2006) concluded, most H 2 S gas from the Castile Formation is likely to have been vented to the atmosphere and Bodenlos (1973) concluded that many limestone masses were not sealed bioepigenetic systems. Similarly, Hentz et al., (1989) its escape prevented it from being concentrated to toxic levels (see Reis et al., 1991, 1992; Klimchouk, 1997b; proliferated, with the rate of sulfate reduction being determined primarily by both the rate of introduction of CH 4 (the foodstuff) and the rate of removal of H 2 S (the detrimental waste product). Major occurrences of native sulfur are absent with the castiles (e.g., Hentz et al., 1989), but minor Quaternary occurrences occur within and near several castiles (e.g., Kirkland and Evans, 1976; Hill, 1996, p. 360-362; Stafford, 2008a, p. 139; Stafford et al., 2008b). These minor castile-associated sulfur deposits formed by reaction between H 2 S and O 2 (like the native sulfur within major subsurface deposits and like that within Lechuguilla Cave). Could major deposits of associated sulfur have existed during the buried phase of castiles (i.e., before erosion exposed the limestone masses)? Probably not, as erosion removed the cover of evaporites, near-surface O 2 -bearing water would have intruded, and dissolved O 2 within the shallow meteoric water would have reacted with the hypothesized large deposits of associated elemental sulfur eventually oxidizing them to H 2 SO 4 The abundantly produced strong acid, would have reacted vigorously with the intimately associated limestone host rock to form caverns of substantial size within the carbonate masses, caverns that would be manifest within present-day castiles. Yet, Stafford (2008a, p. 92), during his comprehensive investigation of Castile karst, seldom found caves within calcitized evaporites; only eight such caves have been reported (Hill, 1996, p. 306; Stafford et Figure 30.
43 been carried forward (up the slope of the homocline) (resulting from both artesian pressure and overpressure) dominated, and it resulted in the transportation of H 2 Sbearing saline groundwater upward and westward. Free had previously moved brine downward and eastward, greatly diminished or entirely ceased. Most H 2 S-bearing brine within Castile conduits probably moved into the Capitan Formation of the A system of joints trends approximately perpendicular to the Capitan reef escarpment (e.g., King, 1948; Hayes, 1964; Jagnow, 1977). Although these joints are less developed than associated joints that trend parallel to the escarpment, the northwest-trending joints probably served as principal pathways into the Capitan reef and forereef. The permeability of these joints (including those within the partially dolomitized upper forereef (Melim and Scholle, 2002)) would have been enhanced by the NaCl-rich groundwater (see Palmer, 2009, p. 121) and by the acidity of the same H 2 S-bearing groundwater. Furthermore, fracture openings and reef cavities in the Capitan were commonly occluded by anhydrite and halite (Darke and Harwood, 1990; Harwood et al., 1991), which would have been removed by under-saturated brine transported within the Castile conduits. concluded that the castiles formed in a ground-water/ hydrocarbon circulatory system that lacked the seal (trap) necessary to cause extensive sulfur mineralization. In the late Miocene and early Pliocene, however, almost all basinal H 2 S-generating systems (i.e, the buried masses of secondary limestone) were probably capped. H 2 S in solution in the lower Castile, with rare exceptions, failed to escape through vertically trending, karstic pathways within the thick overlying Ochoan strata (upper Castile, Salado and Rustler) to be vented into the atmosphere. Instead, much H 2 S was transported within groundwater through the lateraland slightly upward-trending conduits within Castile halite into the Capitan Formation and, eventually, into the slowly enlarging caves of the Guadalupe Mountains. Water through Conduits within Castile Halite into Capitan Formation Late Tertiary H 2 S-bearing groundwater within the dissolution conduits, which was under a pressure up the homocline and into the Capitan Formation. The Bell Canyon siliciclastics through the subsurface masses of biogenic Castile limestone, which occurred dominantly within the Anhydrite I and Anhydrite II members. At these dissolved the by-product, H 2 S. (H 2 S on a molar basis is nearly three times as soluble in water as CO 2 and ~75 times as soluble as CH 4 (Palmer and Palmer, 2000)). The rising nearly fresh-to-moderately saline groundwater held a greater concentration of H 2 S in solution than any sinking NaCl-saturated brine with a molar concentration of about six (see Barrett et al., 1988; Duan et al., 2007). The ascending groundwater moved upward by forced convection through the porous microbial limestone, fractured anhydrite, and dissolution chambers into the low, narrow, linear conduits in the uppermost part of Castile halite members. Then under substantial pressure, it migrated up to several tens of kilometers westerly directly up the slightly dipping homocline to the reef or to the forereef. The H 2 S-bearing groundwater, acquired probably seldom became saturated. Any H 2 S that might have been forced out of solution by increasing salinity (i.e., that salted out into a gaseous phase) would have Figure 31. 2
44 gradient, parallel to bedding through the dissolutionproduced aquifers of low gradient until eventually Fig. 30 ). from Southwest to Northeast along the Cave Belt and from Higher to Lower Elevations within Individual Caves Groundwater within the paleo-Capitan aquifer during the eight-million-year span during which the primary episode of speleogenesis took place (Polyak et al., 1998) transported H 2 S in solution slowly to the northeast. The trend of the aquifer turned, and groundwater continued discharging into the underlying San Andres Limestone (see Hiss, 1977, 1980) ( Fig. 31 ). Compared to the to average about 1.5 m/day (Hiss, 1975, p. 198), the the late Miocene and early Pliocene was greatly reduced. Factors causing the reduction were: Recharge areas were only partially stripped of their impermeable cover (i.e., upper Ochoan evaporitic strata). Permeability was only modestly enhanced (cavernous passageways were in the process of forming). absent (the Pecos River had not yet breached the aquifer, which it did about 0.6 Ma ago (Hiss, 1975)). Evaporites possibly partially plugged pores of the paleo-aquifer (e.g., Harwood et al., 1991). The dip of the paleo-aquifer through most of its history, at least, was less than that of the modern aquifer. Incised submarine canyons normal to the northern a lower permeability than the adjacent reef (Hiss, the aquifer was inhibited. 2 S. Rather, near where the reef escarpment now resides within the lower part of the Capitan paleo-aquifer, joints and faults (those parallel to the escarpment) as well as connected pores allowed a H 2 S-bearing Castile brine was forced into the northwestfrom current hydrologic conditions (Hiss, 1975), the potentiometric surface of the Castile brine was greater than the potentiometric surface of groundwater within the Capitan reef. The hydrologic pressure of the Castile brine was more than adequate to move saline water into the Capitan; only a ~5-20% greater pressure would have been required to move various concentrations of saline water up the conduits than it would have taken to move fresh water up the conduits. The saline groundwater transported H 2 S in solution and moved into a basin-margin aquiferthe Capitan paleo-aquiferthat coincided with much of the cave belt. Eventually the aquifer sequestered an abundance of H 2 S-rich groundwater. The Capitan reef, the older Goat Seep reef, and the adjacent, permeable outer shelfal carbonates presently constitute a modern aquifer that extends, in part, parallel to and just northwest of the Guadalupe Mountain front (e.g., Motts, 1968; Hiss, 1975, 1980). In the late Miocene and early Pliocene, the Capitan paleo-aquifer, the precursor of the modern-day aquifer, consisted of westerly into the paleo-aquifer from the basinal Castile easterly into the paleo-aquifer from the back-reef, shelfal strata of the ancestral Guadalupe Mountains was fresh. On entering the ancient aquifer, the more saline, H 2 Scharged, basinal-derived groundwater, because of its greater density, descended to a bottommost position. It then moved into the lower level of evolving caves within reefal carbonates and within adjacent shelfal carbonates. Artesian groundwater that transported H 2 S to the Capitan Formation within the cave belt had an unusual Fig. 30 ). From topographic areas of eastward down the paleo-Guadalupe tectonic block within shelfal aquifers. Beneath the Castile Formation, it to bedding within basinal aquifers of low gradient. The markedly, and for a short distance (< 1 km), it moved upward approximately vertically through fracturedirected and solutionally enhanced pathways transverse to bedding where it acquired H 2 S. It then, once again, markedly changed direction, and in conduits within the upper part of Castile halite members, backtracked, and
45 limit would have taken place during the earliest history of uplift of the paleo-Guadalupe tectonic block. Later, during Basin and Range deformation, slip on the southsoutheast trending border faults that now bound the Salt Basin graben (e.g., King, 1948) would have downdropped the hypothetical caves and their Middle Permian host carbonates to be buried beneath Late Tertiary and Quaternary alluvium. A primary control for speleogenesis in the Guadalupe Mountains (and for genesis of the large sulfur deposit beneath the Gypsum Plain) was not so much acquisition of H 2 S, of which there was probably a large relative abundance, but, rather, acquisition of O 2 In fact, availability of O 2 was a limiting factor in cave development (Palmer and Palmer, 2000). Intense speleogenesis by H 2 SO 4 demanded an abundant supply of O 2 ; on a molar basis, each metric ton of sulfuric acid that was generated required approximately two metric tons of molecular oxygen (see Palmer and Palmer, 2000). The reservoir for such an abundant supply of O 2 was one that atmospheric oxygen could best provide, which it did part of caves. With each minor episode of uplift (of which there were almost certainly many), the water table descended progressively and intermittently from southwest to northeast along the cave belt. Similarly, within individual caves, the water table, responding to each episode of uplift, descended; and with major episodes of intense uplift, atmospheric O 2 entered the upper part of a new group of developing caves (Polyak, 1998; DuChene and Cunningham, 2006). Gaseous H 2 S and gaseous O 2 within the atmosphere of the caves repeatedly dissolved within the subaerial water of condensation, the habitat of sulfur-oxidizing bacteria and the site of H 2 SO 4 formation. The amount of CaCO 3 dissolved by the H 2 SO 4 depended on a relatively high concentration of CO 2 in the cave atmosphere; if aeration were too strong, the partial pressure of CO 2 would have dropped, and the H 2 SO 4 would have lost much of its cave-forming potency (Palmer, 2006; 2009, p. 218-219). Entrances to nearly all large caves of the Guadalupe Mountains were (and are) few and small in relation to the size of the caves (e.g., Hill, 1999; Palmer, 2006), and they were completely absent until the largely evaporitic upper Ochoan cover was removed by erosion. The initial source of atmospheric oxygen for many caves was probably not through the small, sparse overlying entrances, which repository of moderately-to-highly saline groundwater charged with H 2 repository of the H 2 S-charged groundwater coincided with the developing cave belt. Incipient caves must have initially been present for large caves to form from subaerially aqueous H 2 SO 4 residing on cave surfaces. The incipient caves may have resulted from solutional enlargement of fractures by hypogenic artesian groundwater without involvement of H 2 S (DuChene, 2009). Aqueous H 2 S, a weak acid, however, conceivably played a role in dissolving the carbonates. Possibly, the incipient caves formed at a halocline separating saline, dense groundwater containing H 2 S from overlying, fresh, less-dense groundwater containing O 2 (see Queen, 1994); the dissolved constituents may have reacted microbially form H 2 SO 4 that, in turn, reacted with carbonates to form small subaqueous caves. With tilting of the tectonic block, the Permian (Upper Ochoan) and Cretaceous sedimentary coverconsisting and evaporitesbegan to erode. Erosion was probably initially most intense in the far west at high elevations of the ancestral Guadalupe Mountains. It then probably progressed gradually to the east to lower elevations of the mountains, most likely it tended to progress directly down the homocline. The narrow belt in which major caves of the Guadalupe Mountains evolved, consisting of, in particular, the Capitan Formation and the near-reef part of the Seven Rivers Formation, trended in a southwestnortheast direction across the eastward dipping Guadalupe homocline ( Fig. 26 ancestral Guadalupe tectonic block intermittently rose the highest elevation of the cave belt was its southwestern part (as it remains today) ( Fig. 26 ). This elevated part, which would probably have been subjected to more intense cave belt from which the virtually impermeable upper Ochoan, lithologic cover was removed. Intense speleogenesis may have begun about 10 km northeast of where Guadalupe Peak, Texas, now rises ( Fig. 3 ) (Polyak et al., 1998). Conceivably, however, sulfuric-acid caves may have been dissolved in a more elevated part of the cave belt, a part that may have once extended for several kilometers into Texas. Formation of such hypothesized caves beyond their present known
46 of the basin to create the prominent escarpment that forms the Guadalupe Mountain front ( Fig. 2 ). The epigenic dissolution markedly changed the paleogeography of Castile halite. It has vanished (by dissolution) adjacent to the Guadalupe and Delaware mountains except in the subsurface just southwest of Carlsbad, New Mexico ( Fig. 18 ); here the elevation of the Capitan Formation is low compared to elsewhere along the cave belt, and expression of the escarpment is slight. Presently, Castile halite south of Whites City, New Mexico, is mostly several-kilometers-to-several-tens-ofkilometers downdip from the mountain front ( Fig. 18 ). Collapse breccia of Castile anhydrite in the northwestern Delaware Basin commonly lies directly upon blanket beds of micro-breccia. The clasts of the micro-breccia consist of millimeter-to-centimeter fragments of anhydrite ( Fig. 21B ) remaining after dissolution of Castile halite (Anderson et al., 1978; Anderson and Kirkland, 1980). Coarse collapse breccia, on the other hand, consists of much larger angular fragments mostly of laminated anhydrite ranging up to more than 30 cm in maximum dimension (Anderson et al., 1978). The clasts are now closely spaced in a tight, interlocking (Hentz et al., 1989). Most coarse collapse breccias within the Castile originated following removal of halite by the epigenic groundwater. Anhydritic roofs collapsed by brittle failure as air displaced groundwater. Many, but not all, beds of dissolution breccia are overlain by collapse breccia (Anderson et al., 1978). Some bedded anhydrite apparently subsided gradually without extensive fracturing either by sagging or by settling into voids formed by halite dissolution. Where collapse has occurred, stoping usually diminished upward into Castile anhydrite, and most collapse breccias are overlain by intact beds of Castile anhydrite (and near the surface Castile gypsum). Rarely, however, breccia pipes of Castile anhydrite and gypsum stoped through 1; Wallace and Crawford, 1992; Crawford and Wallace, 1993; Stafford et al., 2008 a and b). Extensive Quaternary fracturing of beds of Castile anhydrite (due to widespread removal of Castile halite) allowed H 2 S to move upward from reaction sites. Near the surface, it reacted with aqueous O 2 within groundwater to form scattered minor deposits of native sulfur (e.g., Porch, 1917). probably developed late in the history of the caves, but rather from southwestern areas of the cave belt initially freed of the Upper Ochoan lithologic cover. Because of the slight dip of the aquifer (0.5 or less), a drop of a few meters in the water table opened several kilometers of the cave belt to aeration. Atmospheric oxygen may through permeable fractures within and parallel to the trend of the reef, and, hence, to the upper subaerial part of expanding caves. Such restricted pathways would have reduced aeration substantially. Along the cave belt over a span of about eight million years, the groundwater table fell in progressive steps by a cumulative ~1,100 m, thus, caves to the northeast are generally younger than caves to the southwest (Polyak et al., 1998; Polyak and Provencio, 2000). Within particular caves, the water table fell with each episode of uplift. New cave levels apparently formed whenever Cavern and Lechuguilla Cave have three or four principal levels between an elevation of 1100 and 1370 m, the higher being ~6 Ma in age, the lower ~3.5 Ma (Ford and Williams, 2007, p. 246). Dissolution of Castile and Salado Halite and Gypsum by Epigenic Groundwater Both hypogenic karst systemsthat within the Castile and that within the reef and adjacent shelflost their and an epigenic karst system became superimposed on the older hypogenic system (see Stafford, et al., 2008a). Erosion removed Lower Cretaceous sandy limestone (e.g., Lang, 1947), Upper Permian Dewey Lake red beds (if present), and most evaporites and carbonates of the Rustler, Salado, and, in addition, within the basin, the upper Castile and part of the lower Castile (e.g., Anderson and Kirkland, 1980). Erosion of the Salado Formation, for example, has resulted in complete removal of its dominant lithologic fractionhalite from the northwestern shelf, from the reef, and from much of the western basin ( Fig. 18 ). Denudation during the epigenic erosional phase exposed the castiles and In addition, solutional denudation of Castile evaporites during the Pleistocene and Holocene exposed the Capitan reef and forereef along much of the northwestern margin
47 The great volume of water that dissolved Salado and Castile (Ochoan) halite and, to a lesser extent gypsum and anhydrite, drained into the Pecos River and into major sinks just east of bodies, up to 460 m thick, are related to dissolution of halite within the lower Salado and, secondarily, to dissolution of halite within the uppermost Castile (Anderson et al., 1978). These centers of collapse into which Tertiary alluvium accumulated occur at the eastern part of a front or wedge of expanded dissolution that in places has extended downdip beneath overlying upper Salado halite and Rustler carbonates and evaporites for distances of up to 30 km (Anderson, 1978, 1981, 1982; Anderson and Kirkland, is inversely proportional to the thickness of halite remaining within the Salado and uppermost Castile. The thick the caves of the Guadalupe Mountains and the major sulfur deposits beneath the Gypsum Plain. The accommodation space in which the alluvial bodies accumulated was created as Ochoan halite (primarily Salado halite) dissolved to form brine that descended through permeable pathways within the underlying Castile evaporites and drained into channel
49 Principal genetic events that were the same were: Hypogenic groundwater under pressure intruded Upper Permian basinal evaporites at various localities and convectively dissolved anhydrite and halite. Gaseous CH 4 intruded the evaporites at these same localities. SO 4 2oxidized aqueous CH 4 biogenically resulting in large quantities of both permeable CaCO 3 and aqueous H 2 S. Large quantities of aqueous O 2 oxidized the H 2 S. Principal genetic events that were different were: At the developing sulfur deposits, groundwater transported aqueous H 2 S up steep grades (~90) and laterally for < ~2 km, whereas at developing caves, groundwater transported aqueous H 2 S up slight O 2 at developing sulfur deposits was transported to reaction sites within saline, epigenic groundwater, whereas at the evolving caves O 2 was transported from the surface to aqueous reaction sites as a gas. Aqueous O 2 at evolving sulfur deposits reacted, probably mainly abiotically, within the phreatic zone with aqueous H 2 S to generate native sulfur, whereas aqueous O 2 at evolving caves reacted biogenically with aqueous H 2 S mainly within the vadose zone to generate H 2 SO 4 The Hydrologic Pathways Hydrologic pathways at the major sulfur deposits differed markedly from hypothesized hydrologic pathways that extended up the homocline within Castile halite. At rare localities on the Gypsum Plain, micro-conduits within the steep, graben-bounding faults guided fresh, solutionally aggressive, hypogenic groundwater directly upward. This buoyant groundwater bypassed thick (tens of meters) barriers of bedded anhydrite (the Anhydrite members of the Castile hypogenically-derived groundwater drained down the steep fault surfaces along adjacent pathways into the Bell Canyon aquifer. The density-driven, free convective dissolution enhanced the permeability of the fault conduits through beds of Castile anhydrite and formed voids within Castile halite. With dissolution of the halite, collapse breccias formed consisting predominantly of Castile anhydrite. ORIGIN OF MAJOR DEPOSITS OF NATIVE SULFUR Major subsurface deposits of sulfur are associated with graben-bounding faults (e.g., Smith, 1978; Hentz et al., 1989; Hentz, 1990; Hill, 1996, p. 366). The grabens clearly have an origin distinct from that of the solutionsubsidence troughs (e.g., Hentz et al., 1990). Parallel faults form graben systems as much as 0.8 km wide and ~6.5 km long (Hentz et al., 1989, p. 39). The faults dip at 50-90, are displaced by 15-75 m, and strike to the northeast (Hentz et al., 1989; Wallace and Crawford, 1992; Crawford and Wallace, 1993; Guilinger and Nestlerode, 1992). The graben-bounding faults probably extend (or extended) transversely through, at least, the upper Paleozoic section, and through a thin Mesozoic section now almost entirely removed by erosion. Native sulfur occupies voids within massive replacement limestone, voids within and between replacement limestone clasts, and planar voids lined occur several hundred meters beneath the Gypsum Plain within the northern and eastern part of the Rustler Springs sulfur district, Texas, an area of ~3,100 km 2 deposits of native sulfur (e.g., Porch, 1917, his plate The locations of four important deposits of native sulfur relative to the cave belt are shown in Figure 32 Reserves of the Leonard Minerals deposit are unreported (see Hill, 1996, p. 365, for a description of the deposit). Original reported reserves of native sulfur at the Pokorny deposit (Klemmick, 1992) and at the Phillips Ranch deposit (Guilinger and Nestlerode, 1992) each probably exceed two million metric tons, and original reserves at the giant Culberson deposit probably exceed 75 million metric tons (Crawford and Wallace, 1993) ( Fig. 32 ). Similarities and Differences in Genetic Processes between the Caves and the Major Sulfur Deposits The subsurface deposits of sulfur probably originated at about the same time as the caves, and the comprehensive genetic events that occurred at the major sulfur deposits and those that occurred at the caves demonstrate both similarities and differences.
50 The chemical and biological processes that formed H 2 S and limestone at the major deposits of sulfur were the same as those that operated at diagenetic masses of exposed as castiles). The primary locale of microbial CH 4 consumption at the sulfur deposits, whether it was stratigraphically low (Castile) or stratigraphically high (Salado, Rustler) within the evaporite succession, was determined, in particular, by the rates of introduction of CH 4 and groundwater, and by the permeability of the steep, fault-directed pathway. In addition, migration of sparse crude oil into the developing sulfur deposits may have thwarted upward stratigraphic movement of sulfur deposition by forming an oil cap. Microbes at each of three major sulfur deposits ( Fig. 32 ) consumed, at least, 1,000,000 metric tons of CH 4 (and at the Culberson deposit vastly more), and microbes at each of the major deposits, generated, at least, 1,000,000 metric tons of H 2 S (and at through caverns, breccias, and permeability-enhanced fault conduits. Eventually, an effective fault-tracking hydrologic pathway extended from the Bell Canyon sandstone directly upward through Castile anhydrite and halite into the overlying lower Salado anhydrite and halite. Such hydrologic pathways provided a migration route for gaseous CH 4 which moved into the Ochoan evaporite section as the pathways were forming or shortly afterward. Huge volumes of gaseous CH 4 eventually moved into what was, in effect, a large, robust, fermentation chamber where the gas dissolved within water and reacted with SO 4 2. Judging from the volume of CH 4 consumed, some CH 4 may have migrated upward into the Ochoan evaporites along faults from deep (> 2 km), Lower Paleozoic (Devonian and older), unusually rich source beds. Figure 32. 2
51 beneath the Gypsum Plain was, in all likelihood, the same as that which operated at both the minor (near-surface) sulfur deposits and the caves of the Guadalupe Mountains, namely aqueous O 2 Bacteria may have partly mediated the reaction between aqueous O 2 and aqueous H 2 S. Ivanov (1968) used radioactive H 2 S at the Shor-Su deposit, Uzbekistan, to show that bacterial enzymes catalyzed ~40% of the elemental sulfur. However, the reaction between aqueous O 2 and aqueous H 2 S proceeds readily inorganically, and at the deposits beneath the Gypsum Plain sulfur oxidizing bacteria possibly failed to play an important role in precipitating elemental sulfur. Source of Aqueous Oxygen Artesian groundwater rising from the Bell Canyon sandstone was probably not the source of the aqueous O 2 that oxidized the H 2 S. Several reasons depreciate this source: The quantity of O 2 required was so vast that artesian groundwater could not have supplied enough aqueous O 2 to restricted localities of major sulfur deposition. For example, within a reasonable time frame, could artesian (hypogenic) groundwater diluted by overpressured anoxic water forced out of shale have transported the ~35,000,000 metric tons of required O 2 to the restricted locality of the Culberson sulfur deposit? During its transportation eastward from mountainous recharge areas, dissolved oxygen within artesian water would have been diminished by reacting with dispersed particles of organic matter, pyrite, and siderite (FeCO 3 ). These constituents of the Bell Canyon occur predominantly within the siltstone lithofacies, which constitutes about 35% of the formation (Bozanich, 1979), and which has a hydraulic conductivity not much lower than the sandstone (Davies, 1983, p. 14). Late Tertiary pore water within Bell Canyon sandstone and within underlying Permian aquifers may have pervasively contained dissolved CH 4 Dissolved O 2 within artesian groundwater of the Bell Canyon might have reacted with the CH 4 The reaction, which is mediated by aerobic bacteria, forms CO 2 and H 2 O. Apparently, the source of dissolved O 2 was not hypogenic artesian groundwater. As the oxidizing agent of H 2 S, a source of dissolved O 2 within epigenic meteoric-derived water was more likely than a source of dissolved O 2 within hypogenic meteoric-derived water. the Culberson deposit vastly more). Equivalent weights of H 2 S were probably generated by larger masses of biogenic limestone, many now represented by larger castiles. The Oxidizing Agent Aqueous oxygen was clearly the oxidizing agent of H 2 S at caves of the Guadalupe Mountains; dispute exists, however, as to the oxidizing agent at the major sulfur deposits. Was it O 2 or was it some other agent? To form the major sulfur deposits, prodigious weights of O 2 would have been required; ~35 million metric tons, for example, at the Culberson deposit ( Fig. 32 ). This is perplexing: How could such a great weight of O 2 be transported for possibly as much as 1 km beneath the earths surface where at a redox boundary (or redox boundaries) it reacted with H 2 S? The O 2 must have been transported within groundwater, and because the maximum solubility of O 2 in groundwater is ~10 mg/l (e.g., Winograd and Robertson, 1982), huge volumes (cubic kilometers) of groundwater would have been required. Furthermore, the aqueous O 2 would seemingly have poisoned the obligate anaerobes that formed the H 2 S. In addition, why would surface or near-surface, oxygen-bearing meteoric water sink through a transverse hydrologic pathway that extended for hundreds of meters from the surface to the Bell Canyon aquifer? More likely, it would seem, groundwater within the underlying Bell Canyon aquifer would have risen within the hypothetical transverse pathways (due mainly to artesian pressure) possibly to discharge at ancient springs such it does today at Rustler Spring and Castile Spring (for which the Castile Formation is named (Richardson, 1905)). proposed other oxidizing agents for the major sulfur deposits, oxidizing agents that would have been readily available within the subsurface. The problem would be solved, for example, if either sulfate anions (SO 4 2) or carbon dioxide (CO 2 ) were the oxidizing agent (Feely and Kulp, 1957; Davis and Kirkland, 1970; Ruckmick et al., 1979; Guilinger and Nestlerode, 1992; Guilinger, 1993; Miller, 1992; Klemmick, 1992, 1993). Little evidence, however, supports these agents (e.g., Davis et al., 1970). Moreover, most geologists who have investigated biogenic sulfur deposits have invoked aqueous O 2 as the oxidant (e.g., Ivanov, 1968; Smith, 1978, 1980; Davis and Kirkland, 1979; Hill, 1992; Machel, 1989, 1992, 2001; Niec, 1992; Spirakis and Cunningham, 1992; Wallace and Crawford, 1992; Crawford and Wallace, 1993; Klimchouk, 2007, p. 92). The agent that oxidized H 2 S at major sulfur deposits
52 oxygen-bearing brine sank (by forced convection) fresh hypogenic groundwater continued to rise (also by forced convection) from Bell Canyon sandstone along separate, adjacent, and parallel pathways. The dense, saline, epigenic groundwater driven by gravity transported O 2 deeply into the Castile evaporite of seawater can transport (at ~22C) ~4 milligrams of O 2 per kilogram of brine, and a marine brine with a salinity ten times that of seawater (i.e., a brine at saturation) can transport (at ~22C) ~2 milligrams of O 2 per kilogram of brine (Kinsman et al., 1974). The dissolved-oxygen capacity of the dissolution brines closely approximates the dissolved-oxygen capacity of marine brines (Sherwood et al., 1991). Thus, the salty groundwater that descended from the Salado into the Castile could have transported about one-fourth to more than about one-half as much dissolved O 2 as a similar volume of fresh groundwaterif it were able to sink. Within the sinking brine, the deleterious effect on solubility of O 2 brought on by increasing temperature diminished as the salinity increased until at a salinity of ~250 g/kg it nearly disappeared (see Sherwood et An effective drainage system was required for disposal of the huge volume of sinking brine. Major Ochoan sulfur deposits probably formed only where grabensandstone of the upper Bell Canyon, such as channel sandstone that apparently underlies the giant Culberson sulfur deposit (Smith, 1980; Crawford and Wallace, 1993; Hill, 1996, p. 365). Brine that discharged into the the slight grade (probably < 0.5) to the northeast and discharged into the Capitan and San Andres aquifers, the potentiometric pressure within the sandstone being greater than the potentiometric pressure within these aquifers (see Hiss, 1975). Precipitation of Native Sulfur Aqueous H 2 S moved, in part by diffusion, away from sites of active microbial growth, and, at the same time, aqueous CH 4 moved, in part by diffusion, towards the same sites. Microbial growth occurred at contacts between anhydrite (the wall or fractured rock) and permeable diagenetic limestone. The limestone formed Descent of Epigenic Groundwater and Aqueous Oxygen into Late Permian Evaporites The locales of major sulfur deposits apparently had a tracking pathway that extended from the Bell Canyon upward through both the Castile evaporites (~30% halite; ~0.5 km thick) and the directly overlying Salado evaporites (~85% halite; ~0.5 km thick). Hypogenic groundwater, therefore, could move directly upward along the fault-directed pathway bypassing potential barriersnamely, interbeds of Castile anhydrite (~60 % of the formation) and interbeds of Salado anhydrite (~12% of the formation). Hypogenic groundwater that eventually reached the earths surface and discharged at springs dissolved atmospheric O 2 Then, on dissolving upper Ochoan halite, the previous hypogenic groundwater became epigenic groundwater, and descended along the same fault-directed pathway in which it had ascended. Furthermore, epigenic, oxygenbearing groundwater from overlying Rustler aquifers entered the pathways, dissolved Salado halite, and descended. Similarly, epigenic groundwater may have entered pathways from topographic structural troughs a possible surface manifestation of the grabens (e.g., Salado halite followed, and dissolution formed a cavern into which overlying Permian, Mesozoic, and Cenozoic strata collapsed or subsided (see Miller, 1992; Crawford and Wallace, 1993). Collapse resulted in heterolithic breccias. Fractures radiated to the surface, allowing additional water to descend. A large (many hectares) catchment basin, a doline, formed that channeled huge volumes of water into the Salado Formation where it dissolved additional NaCl, in addition to CaSO 4 and CaSO 4 2H 2 O. The resulting saline groundwater several-to-many cubic kilometersdescended by Canyon. A huge volume of Salado halite (and a much lesser volume of calcium sulfate) went into solution, also possibly measured in cubic kilometers. With dissolution of NaCl, the resulting brine became gravitationally unstable. It sank for hundreds of meters along a steep, transverse, pipe-to-chimneylike conduit through the evaporite succession. The concentration of the oxidantdissolved oxygen within the saline groundwater was low, but the volume of sinking saline groundwater was huge. The epigenic,
53 virtually in place as dissolved CO 2 excreted from the enigmatic microbes reacted almost instantaneously with Ca 2+ freed as CaSO 4 dissolved. Additional CaSO 4 persistently dissolved within ambient water as the Ca 2+ reacted, and as SO 4 2(also freed as CaSO 4 dissolved) reacted with CH 4 Rising buoyant hypogenic groundwater and sinking 2 S was sequestered within the relatively, fresh, hypogenic groundwater being propelled upward by artesian pressure, overpressure, and buoyancy. Where epigenic, O 2 -bearing, saline groundwater sinking along one course contacted hypogenic, H 2 S-bearing, brackish groundwater rising along an adjacent course, native sulfur precipitated within the pores of the secondary limestone. Both aqueous O 2 and aqueous H 2 S, each harmful to the eliminated at reaction interfaces separating the twoof the limestone host rock, openings were obstructed resulting in redox interfaces shifting over time. At some deposits, sulfur deposition possibly occurred along a horizontal surface corresponding to a zone in changed markedly (a pycnocline). Oxygen, a crucial component for production of H 2 SO 4 in caves of the Guadalupe Mountains, was in unlimited (although restricted) supply and it had a high partial pressure. to support thriving colonies of sulfur-oxidizing bacteria. Oxidation of one metric ton of H 2 S to H 2 SO 4 required about four times more O 2 (by weight) than was required to oxidize one metric ton of H 2 S to sulfur. Elemental sulfur was deposited within environments in which aqueous O 2 was relatively sparse and in environments in which sulfuroxidizing bacteria that formed H 2 SO 4 were apparently deposited within phreatic groundwater within the Ochoan basinal evaporite section (i.e., Castile and Salado) (some sulfur was also deposited within phreatic groundwater in caves of the Guadalupe Mountains, primarily, within cave pools of Lechuguilla cave (Cunningham et al., 1993; DuChene and Cunningham, 2006)). The second stage of the oxidation process at these locales, i.e., from native sulfur to sulfuric acid (H 2 SO 4 ), failed to occur.
55 basinal reservoirs support CH 4 as having been the primary microbial foodstuff. (All isotopic analyses are reported as parts per thousand deviation from the 13 C/ 12 C ratio of a specimen of Belemnitella americana an extinct, marine cephalopod mollusk that served as an early standard.) 13 C mode of -37.0; the most negative value being -39.2 (Kirkland and Evans, 1976). A sample of calcite from 13 C of -38.0 (Davis and Kirkland, 1970) and a sample of calcite from the 13 C of -49.0 (Crawford and Wallace, 1993) ( Fig. 33 ). If oil were the dominant microbial foodstuff, we would be unable to account for these highly negative values. None of the fractions of the sparsely associated crude oil has such highly negative isotopic signatures. The benzene-soluble fraction of oil 13 C of and aromatic fractions of oil extracted from secondary limestone within the Castile section of the Culberson sulfur deposit ( Fig. 32 13 C values of -26.7 and -27.7, respectively (Crawford and Wallace, 1993). In general, oils within Permian reservoirs beneath the 13 C range of -27.2 to -28.2 (McNeal and Mooney, 1968) ( Fig. 33 ). Samples of CH 4 from Lower Permian and deeper Paleozoic reservoirs beneath the Castile, on the other 13 C isotopic signatures in the range of -35.0 to -51.0 (n>18) (Stahl and Carey, 1975; written communication, R. S. Squires, 1980; Hill, 1996, her appendix 5) ( Fig. 33 ). This range for CH 4 13 C range of -30.0 to -55.0 for CH 4 produced during late stages of kerogen evolution (Tissot and Welte, 1984, their table II.6.1). Thus, thermogenic CH 4 easily accounts for 13 C values of samples of secondary limestone from both the castiles and the sulfur deposits, whereas crude oil does not. Ten samples of calcite out of 20 from the castiles have 13 C values more positive than -30 (and 18 of the 20 13 C values more negative than -9) (Kirkland and Evans, 1976, their table 1) ( Fig. 33 ). Because CH 4 is proposed as the primary reductant, we must account for these moderately elevated values. Carbon atoms within calcite of the castiles are not entirely Microbial Foodstuff at the Subsurface Masses of Biogenic Limestone Anaerobic reaction between CH 4 and SO 4 2is common within modern marine sediments, but within sedimentary strata on land, the diagenetic reaction is uncommonly recognized and may be rare. However, the diagenetic reaction between CH 4 and SO 4 2within Castile anhydrite in the northwestern and west-central Delaware Basin was paramount, the redox reaction of the Late Tertiary being indirectly responsible for the caves of the Guadalupe Mountains, for the castiles of the Gypsum Plain, and for the large native-sulfurbearing limestone masses beneath the Gypsum Plain. The volume of CH 4 consumed by microbes was huge billions of cubic metersand some CH 4 from deep (thousands of meters) within the basinal sedimentary 4 and SO 4 2may still be occurring within the presentday Ochoan succession beneath the Gypsum Plain, but languidly because of diminished migration of thermogenic CH 4 In addition, where halite has been removed, oxygenated epigenic watersdeadly to the functioning microbeshave now penetrated deeply and extensively through fractures within gypsum and anhydrite evaporites of the Salado and Castile formations. Support for the CH 4 fraction of natural gas as having been the dominant reductant of sulfate anions within Upper Permian strata of the Delaware Basin is summarized in the following sections. Highermolecular-weight homologues within natural gas (i.e., ethane (C 2 H 6 ), propane (C 3 H 8 ), and other gases in the alkane series) probably also reacted with SO 4 2; compared to CH 4 however, they were trivial reductants, their volumes being too small (e.g., <~5%). Furthermore, whereas both CH 4 and higher-molecularweight natural gases were generated during the Late Tertiary within the deeply buried stratigraphic section (Lower Permian, and Middle and Lower Paleozoic), the higher-molecular-weight homologues may have been cracked predominantly to CH 4 Negative Carbon Isotopic Values Carbon isotopic analyses of samples of calcite from the castiles, samples of calcite from the major sulfur deposits, and samples of both crude oil and CH 4 from METHANE AS MICROBIAL FOODSTUFF
56 a mixing line between the isotopic compositions of these two dominant carbon sources. Samples adulterated with even small amounts (e.g., 15 wt%) of isotopically heavy evaporitic CaCO 3 13 C values that 4 Eleven samples of calcite from the Culberson sulfur 13 C of -49 to -20) than samples of calcite from the castiles ( Fig. 33 ). This is probably because primary evaporitic calcite and dolomite are relatively sparse within Salado residual anhydrite; therefore, relatively little carbon having a markedly 13 13 C of samples of the biogenic Salado calcite. Two additional potential sources of relatively isotopically heavy carbon are HCO 3 and CO 3 2residing within pore water of sandstone beds directly beneath the Castile (Lee and Williams, 2000). Such anions, with a probable 13 C range of -2.0 to -3.0 (Dutton, 2008), may have moved from the Bell Canyon into the overlying Castile, and may have transferred some of their carbon to the biogenic, but consist of a mixture of varying proportions 13 CH 4 and isotopically very heavy carbon from primary sedimentary CaCO 3 13 C ~ +6.0; typical range, 5.56.5; n=140), which constitute the characteristic dark calcite laminae of the Castile Formation ( Fig. 19 ). The isotopically heavy fraction of carbon within Castile sedimentary calcite (usually a minor component) originated in the Late Permian as CO 2 dissolved within seawater that had been concentrated by evaporation. The dissolved CO 2 reacted with Ca 2+ to precipitate CaCO 3 that formed sedimentary laminae (possibly originally aragonite, but now calcite and rarely dolomite). The associated laminae of sedimentary calcite ( Fig. 16A left) persisted or recrystallized as biogenic calcite replaced 13 C the castiles. Sedimentary evaporitic calcite within the Castile ranges from ~3 to >80 wt% with a mean of ~10 wt% (see Anderson and Dean, 1995; Kirkland, 2003). Samples from limestone masses within the Castile, therefore, have carbon isotopic compositions that fall on Figure 33. A. 13 13 C values more negative B. 13 C. Variation in 13 A C C
57 produced) indicates that to have served as the primary reductant of sulfate anions, fractions of the migrant crude oil had an inadequate volume. The total estimated weight of the reductant (the foodstuff) that entered sites of sulfur deposition within the Castile Formation (and Culberson sulfur deposit) was clearly many thousands of times greater than the weight of crude oil observed at the various subsurface sulfur deposits. Middle and Lower Paleozoic strata of the Delaware Basin contain vast quantities of CH 4 (e.g., Hills, 1984) and relatively small quantities of crude oil, whereas equivalent strata on the surrounding shelf north and east of the basin contain relatively little CH 4 and vast quantities of crude oil; hence, for the basin, it is easier to envision CH 4 rather than fractions of crude oil as the primary reductant. Could crude oil be sparse within the sulfur-bearing and sulfur-barren calcitized bodies because microbes consumed it? Probably not, if this were true, an obvious display of asphaltic compounds, refractory to microbes, would have remained. If oil had been the foodstuff at the giant Culberson deposit, an estimated 32 million m 3 (~200 million barrels!) of oil would have been required as a reductant (Smith, 1978; Ruckmick et al., 1979). Most containing N, S, and Onamely, resins (low-molecularweight) and asphaltenes (high-molecular-weight)that are neither readily dissolved within water nor effectively metabolized by microbes (e.g., Tissot and Welte, 1984, p. 467). Oil within Bell Canyon reservoirs contains about 10-wt% of such asphaltic compounds (K. A. Kvenvolden, written communication, 1966). If the cited values are representative and if oil were the foodstuff at the Culberson sulfur deposit, then at least 3.2 million m 3 (~20 million barrels) of viscous asphaltic materialthe unmetabolizable fractionwould be expected within pores of the deposit. Yet, Smith (1980) reports, Only a little asphaltic material and minor amounts of oil have been found. CH 4 to have served as principal reductants, therefore, the paucity of asphaltic residue and, in addition, the minor volume of oil encountered within the secondary deposits support the interpretation that CH 4 was the primary microbial foodstuff. The Pokorny Sulfur Deposit, located beneath the Gypsum Plain about 30 km southeast of Carlsbad Cavern, secondary limestone masses. Another possible source of relatively isotopically heavy carbon was gaseous CO 2 (see Holmquest, 1965) that may have accompanied gaseous CH 4 that moved into the Castile. Such CO 2 may 13 C of about -3.0 (Ballentine et al., 2001, their table 1), and its carbon may have been incorporated into the secondary limestone masses. Crude Oil: An Unimportant Reductant of Sulfate Several subsurface, sulfur-bearing, limestone masses contain crude oil (Davis and Kirkland, 1970; Smith, 1978, 1980; Crawford and Wallace, 1993; Klemmick, 1993; Guilinger, 1993). The large carbonate-sulfur body that forms the Culberson sulfur deposit ( Fig. 32 ), which extends from the lower Castile Formation upward into the overlying Salado and Rustler formations, is practically free of oil except for the basal Castile (Crawford, 1990). Sulfur-bearing limestone at the Philips Ranch sulfur deposit ( Fig. 32 Anhydrite I Member, can be very oily (Guilinger and Nestlerode, 1992; Guilinger, 1993). A mass of Castile limestone apparently lacking native sulfur forms the (Davis and Kirkland, 1970). The nearby Screw Bean oil deposit, has also produced minor volumes of oil from the Castile (Clark, 1990) (probably from porous diagenetic limestone). Oil within the Castile Formation migrated from the directly underlying Bell Canyon Formation, but it probably originated within deeper formations (Crawford and Wallace, 1993). Once within the Castile, fractions of the oil were degraded anaerobically by microbes to generate heavy oil, and, among other byproducts, CO 2 and H 2 S. The diagenetic masses of limestone within the Castile Formation are genetically analogous to the limestone caprock of salt domes. The limestone caprock of salt domes was initially thought to form by reduction of the underlying anhydrite caprock of salt domes by crude oil (Thode et al., 1954; Feely and Kulp, 1957). However, it now seems likely that CH 4 that migrated into the limestone caprock was the primary reductant (e.g., Posey, 1986; Saunders and Swann, 1994). Crude oil was not a major reductant in the Castile. Its minor volume within the buried Castile limestone
58 1977; Barker et al., 1979; Jassim et al., 1999). Samples 13 C values that range from -32 to -65, a range considered a diagnostic feature of bioepigenetic calcite recognized in major sulfur deposits around the world (Klimchouk, 2007, p. 89). Such isotopically light carbon incorporated within the huge masses of secondary limestone that host these sulfur deposits could probably have been derived only from CH 4 That CH 4 was the likely reductant at major biogenic deposits elsewhere in the world supports CH 4 as having been the reductant of SO 4 2at the large sulfur deposits beneath the Gypsum Plain. Anaerobic Reduction of Sulfate Anions by Methane in Marine Diagenetic Environments Anaerobic reduction of SO 4 2by CH 4 is pervasive in modern marine sediments. An ever-available supply of SO 4 2(~0.27 wt%) diffuses from overlying seawater into the sediments. In addition, on continental shelves, CH 4 both biogenic and thermogenicis abundant, and commonly rises as seeps through oceanic sediments by buoyancy, by other sources of pressure, and by diffusion. The reaction between CH 4 and SO 4 2within marine sediments, as mentioned above, is identical to that which occurred during the Late Tertiary within Castile and Salado anhydrite. The reaction in both diagenetic environments is (and was) mediated by microbes, and the enigmatic, biologic agents that operate today within modern oceanic sediments may be the same or related microbial species (or strains) as those that operated within anhydrite of the Ochoan group. Stakes, 1999; Aloisi et al., 2002; Orcutt et al., 2005). Associated near-surface components are hydrogen and commonly massive accumulations of limestone. Ancient marine deposits of diagenetic limestone having lithologic and paleo-biologic characteristics similar to active, modern marine cold-seeps also rarely crop out on land (e.g., Peckmann et al., 1999; Clari and Martire, 2000). Limestone of both modern and ancient cold-seep deposits is depleted in 13 C (i.e., it is isotopically light) (e.g., Jrgensen, 1992; Suess et al., 1999; Peckmann et al., 1999; Kotelnikova, 2002) attesting to its origin from organic matter. Furthermore, samples of the 13 C values and about 32 km northwest of the Culberson deposit ( Fig. 32 ), consists of a diagenetic mass of limestone about 200 m beneath the Gypsum Plain that contains dispersed elemental sulfur (Davis and Kirkland, 1970; crude oil were encountered in exploratory wells at the deposit (Klemmick, 1992). Oil occurs as stains on thin (< ~1 m) limestone intervals (Davis and Kirkland, 1970, in diameter are unstained (Klemmick, 1992). (These borehole (W. E. Dean, personal communication, 1967)). Here too, the paucity of asphaltic residue and the minor volume of oil support CH 4 as having been the primary microbial foodstuff. Asphaltic residue is unreported from the castiles, which again supports CH 4 as having been the primary reductant. If fractions of oil had been the primary reductant, a residue of resins and asphaltenes would be expected within pores of the castiles, and rather than being entirely or essentially devoid of asphalt, the limestone might fall into the category of a tar-rich carbonate. Because asphalt is absent, fractions of oil were not primary reductants, leaving CH 4 as the only reasonable candidate. Furthermore, castiles of the Gypsum Plain are genetically related to the subsurface deposits of Ochoan sulfur and limestone beneath the Gypsum Plain (e.g., Kirkland and Evans, 1976; Smith, 1980), their carbonate petrologies, for example, being strikingly similar (Madsen and Raup, 1987). Since CH 4 was apparently the primary microbial foodstuff at the Ochoan sulfur deposits, it was apparently the primary foodstuff at the castiles as well. Anaerobic Reduction of Sulfate Anions by Methane in Limestone-Hosted Sulfur Deposits Elsewhere Elsewhere in the world, CH 4 is likely the primary reductant of sulfate anions at most, if not all, major deposits of biogenic sulfur (Mamchur, 1969). Examples include: caprock deposits of the US Gulf Coast (e.g. Posey, 1986), Miocene stratabound deposits of Poland stratabound deposits of eastern Ukraine (Andrejchuk and Klimchouk, 2001), and late Miocene stratabound deposits of Sicily (Ziegenbalg et al., 2012). CH 4 was probably also the primary reductant at middle Miocene stratabound deposits of northern Iraq (see Al-Sawaf,
59 Anaerobic Reduction of Sulfate Anions by Methane in Terrestrial Diagenetic Environments Elsewhere Conspicuous evidence of microbially mediated reduction of sulfate anions (SO 4 2) by CH 4 or, for that matter, reduction of molecular oxygen (O 2 ) by CH 4 is uncommonly recognized within the Earths terrestrial environments. In rare instances, aerobic bacterial cells (living and dead) are preserved ephemerally at and near direct evidence that CH 4 has reacted with O 2 (Davis, 1952). Lithologic evidence of anaerobic microbial oxidation of CH 4 in terrestrial diagenetic environments is apt to be preserved prominently only where CH 4 has leaked into shallow strata having an absence of dissolved oxygen, a favorable concentration of SO 4 2(e.g., >1000 ppm, Oehler and Sternberg, 1984, their ~40 to ~85C). Evidence of such anaerobic oxidation consists of bleaching of redbeds by H 2 S (such as occurs in some Mesozoic strata of the western USA (E. F. McBride, written communication, 1996) and above the 1974)), but especially from a combination of carbonate, revealing isotopic signatures. The redox reaction between CH 4 and SO 4 2at several a magnitude that rivals that which occurred at the Culberson sulfur deposit (Kirkland et al., 1995). At the CH 4 migrated upward from complex Pennsylvanian structures across an angular unconformity into gently folded, gypsum-bearing Permian red beds. Sulfateconsumed an estimated 37 billion m 3 (1.3 trillion ft 3 ) of CH 4 and in the process generated millions of metric tons of the metabolic by-products, CO 2 and H 2 S. The by2+ Fe 2+ and Mg 2+ to form chimneys of bleached red and lesser amounts of pyrite and marcasite (FeS 2 ), and, in addition, trace amounts of sphalerite, (Zn,Fe)S and 13 C values as 34 S values as low as -37.9 (Kirkland et al., 1995). Other terrestrial examples in which anaerobic reduction of SO 4 2by clearly derived directly or indirectly from CH 4 (either thermogenic, biogenic, or both). The H 2 S generated at modern cold seeps supports an impressive mat of H 2 Soxidizing bacteria as well as an associated seep fauna, including tubeworms. The ecology of the seep fauna is built around bacteria, especially Beggiatoa sp. that obtain their energy from reaction between aqueous O 2 and aqueous H 2 S (e.g., Orcutt et al., 2005). The comprehensive processes that occur at modern cold-seep deposits of the worlds oceans were duplicated, in many respects, by the comprehensive processes that occurred in southeastern New Mexico; these include microbial generation of H 2 S at the calcite masses and its oxidation at the caves and at the sulfur deposits. Furthermore, within shallow anaerobic sediments of the worlds oceans, SO 4 2oxidizes CH 4 The oxidation is apparently nearly pervasive in shallow oceanic sediments, but is most vigorous in sediments of continental shelves. Dissolved within seawater, O 2 diffuses downward through marine pore water for a few millimeters to exhausted (e.g., Jrgensen, 1982). Similarly, SO 4 2within seawaterless reactive and in much greater concentration (>300 times)diffuses downward even farther. Below its limit of diffusion (typically <1 m), a varied assemblage of anaerobic microbes (including methanogenic archaea) decomposes particulate organic matter by hydrolysis and by fermentation to generate CH 4 (e.g., Barker, 1956; McCarty, 1964). The upward diffusing CH 4 eventually contacts the downward diffusing SO 4 2 (see Valentine, 2002). At the interface, anaerobic microbes (probably a consortium of archaea and bacteria) then bio-catalyze oxidation of CH 4 and the coupled reduction of SO 4 2to form the metabolic waste products, H 2 S and CO 2 (e.g., Boetius et al, 2000; Hinrichs and Boetius, 2002). Most of the sulfur and carbon atoms are incorporated into pyrite and calcite, respectively. Little CH 4 escapes anaerobic oxidation to enter either the overlying O 2 -bearing sediments or the O 2 -bearing water column, the sulfate-dependent methane oxidation acting as a barrier (Valentine and Reeburgh, 2000). The microbially mediated reaction between CH 4 and SO 4 2occurs collectively on a vast scale within the worlds oceans, an estimated 100 trillion grams of CH 4 per year (Reeburgh, 1989), the amount of CH 4 consumed being approximately equivalent to 4 to the atmosphere (Hinrichs and Boetius, 2002).
60 CH 4 has played an important role are calcite-cemented (Enos and Kyle, 2002), possibly the Beeri sulfur deposit, southern Israel (e.g., Druckman et al., 1994), and many low-temperature mineral deposits of copper, iron, lead, uranium, vanadium, and zinc within sedimentary strata.
61 temperatures in Permian and older petroleum-source strata, which, in turn, cracked dispersed crude oil and further decomposed dispersed kerogen to generate copious volumes of CH 4 In addition, the episodic and uniform easterly tilting during the late Miocene and early Pliocene (by a cumulative 1-2) along with a nearly contemporaneous late-phase of Basin and Range crustal extension, created and rejuvenated fractures within the Paleozoic sedimentary section. In the Delaware Basin near the beginning of the late Miocene, pressurized, nearly fresh artesian groundwater, which originated in the ancestral Guadalupe Mountains, moved upward from sandstone beds of the upper Bell Canyon Formation (Middle Permian) through the new and rejuvenated fractures into the directly overlying Anhydrite I Member (thickness ~50 m) of the Castile Formation. The artesian groundwater dissolved CaSO 4 the density of the solvent increased, it became gravitationally unstable, and Ca 2+ and SO 4 2-bearing groundwater sank back into the Bell Canyon. Taking its place, under artesian pressure, the freshest, least-dense water available rose inherently to the highest accessible elevation. The continuous process of free convection created dissolution voids, and, subsequently, fractures and collapse breccias within the Anhydrite I Member through which CH 4 aggressive groundwater (rising), and nonaggressive groundwater (sinking) moved freely. Hypogenic groundwater ascended through the solutionally enhanced, transverse pathways through the Anhydrite I Member and eventually contacted the base of the directly overlying Halite I Member (thickness ~125 m) of the Castile Formation. The rising groundwater readily dissolved the bedded NaCl, and the resulting brine sank. Simultaneously, groundwater with the greatest solutional aggressiveness for NaCl (that with the lowest density and the lowest concentration of solutes) rose continuously to the solution front where it, in turn, dissolved additional Castile halite. The free convective process resulted in chambers being dissolved vertically upward within the Halite I Member until they contacted the intact base of the next overlying bed of anhydrite (namely, the base of the Anhydrite II Member), which dipped uniformly eastward over An immense weight (millions of metric tons) of 2 S) moved into caves of the Guadalupe Mountains during the late Miocene and early Pliocene (~12-4 million years ago). The H 2 S reacted with O 2 form sulfuric acid (H 2 SO 4 )the primary cave-forming agent in the mountains. The caves formed within Middle Permian reefal limestone (the Capitan Formation) and within adjacent, time-equivalent, shelfal carbonates (particularly, the Seven Rivers Formation). Pathways previously proposed for transporting the precursor, H 2 large quantities of gaseous H 2 S nor aqueous H 2 S apparently migrated from the northwestern Delaware Basin updip into the evolving caves through siliciclastics of the Bell Canyon Formation (Middle Permian; Guadalupian series). Furthermore, large quantities of H 2 S dissolved within artesian groundwater apparently did not migrate from elevated (mountainous) shelfal strata northwest of where the modern Capitan reef escarpment now exists downdip through permeable Middle Permian strata into the evolving caves. Instead, the H 2 S involved in speleogenesis was probably transported into the evolving caves from the adjoining Delaware Basin through upwardinclined pathways within Castile halite (Upper Permian; lower Ochoan Group). The Castile Formation, a thick (~0.5 km) evaporite unit (originally ~30% halite, ~60% Each Castile member, bed, and laminawhether halite, anhydrite (probably initially gypsum), or calcite (possibly initially aragonite)that formed by deposition and diagenesis in the Late Permian had, with few exceptions, an extraordinarily uniform thickness, lithology, and contact relationships over many thousands of square kilometers. Two approximately coeval Late Tertiary events superimposed on the consistent Castile stratigraphic framework resulted in intense H 2 S-H 2 SO 4 speleogenesis in the Guadalupe Mountains. These events were Basin, and eastward tilting of the paleo-Guadalupe tectonic block, a huge homocline that included the Guadalupe Mountains and much of the Delaware Basin. SUMMARY AND CONCLUSIONS
62 artesian groundwater. Forced convection transported the groundwater upward through the Castile dissolution conduits and into the Capitan Formation with a velocity construction of the conduits. The saline, H 2 S-laden groundwater on reaching the ancient reef and forereef moved, wherever possible, through interconnected pores (particularly those along fractures) into the limestone reef, where, because of the relatively high density of the introduced saline groundwater, it descended to a low level. The cave belt of the Guadalupe Mountains, a sixkilometer-wide band parallel to and including the Capitan reef, has a northeast-southwest trend across the uniformly eastward dipping, Guadalupe tectonic block. Because of block the highest elevation of the cave belt was to the southwest. Erosion, which generally progressed down the tectonic block from west to east, probably initially removed the stratal cover consisting primarily of Rustler and Salado evaporites (Upper Permian, upper Ochoa Group) from the most elevated southwestern part of the cave belt. In step with intermittent uplifts, erosional removal of the evaporitic cover from the cave belt probably progressed to the southeast over millions of years. A primary control over both H 2 S-H 2 SO 4 speleogenesis in the Guadalupe Mountains and H 2 S-S genesis in the Delaware Basin was availability of O 2 Abundant H 2 S during the late Miocene and early Pliocene charged the sluggishly moving groundwater in the lower part of the ancestral Capitan aquifer, which over much of its extent coincided (in plan) with the cave belt. O 2 was dissolved within groundwater in the upper part of the aquifer in meager concentrations, and generation of minor quantities of aqueous H 2 SO 4 at a pycnocline may have resulted in incipient caves. Intense speleogenesis began only when atmospheric O 2 became available to the ancestral Guadalupe tectonic block and descent of the water table, gaseous O 2 from the earths atmosphere initially entered the upper part of the most elevated incipient caves. The O 2 probably migrated laterally as a gas within northeast-trending fracture pathways from high ground to the southwest where the evaporitic cover was probably initially breached. Mediated by aerobic bacteria H 2 S and O 2 from the cave atmosphere reacted within subaerial water of condensation, to form aqueous H 2 SO 4 The strong acid, in turn, reacted with reefal and thousands of square kilometers. Then, by the same process of convective dissolution, but in an abrupt change in dip (from ~90 to < ~1) and in direction of dip (from upward to westward), anhydrite-capped voids advanced up the slight homoclinal slope for up to several tens of kilometers. The solvent, nearly saturated with CaSO 4 no longer readily dissolved anhydrite. The conduits, except halite. The width of dissolution conduits is inferred to and their length, long (up to tens of kilometers). Halite dissolved most actively at the most elevated, thinnest, and most western part of growing conduits where solutionally aggressive groundwater directly contacted abruptly and diametrically changed direction. Just beneath rising aggressive water (in a two-way stream), directly down the slight slope of the homocline, and passed through fractures, breccias, and voids within the Anhydrite I Member and drained into sandstone of the Bell Canyon Formation. Growing conduits continuously advanced westerly up the homoclinal slope as the ascending, aggressive groundwater dissolved halite. Many conduits eventually contacted the steep face of the reef or against Capitan carbonates. Within the basin, millions of cubic meters of gaseous CH 4 migrated from Lower Permian and deeper source strata upward into upper Bell Canyon sandstone just before, during, and just after formation of the conduits. Gaseous CH 4 moved upward through fractures or through anhydrite breccias into the lower anhydrite members of the Castile following the same pathways as the hypogenic, freshto-brackish groundwater. Within anoxic ambient water, sulfate anions (freed as anhydrite dissolved) reacted with aqueous CH 4 The reaction catalyzed by microbial enzymes formed H 2 O, H 2 S, and CO 2 Almost all CH 4 that invaded the Castile was probably transformed. The dissolved CO 2 (as CO 3 2) reacted instantaneously with Ca 2+ (which like SO 4 2formed as anhydrite dissolved) to form diagenetic masses of CaCO 3 most ultimately having a maximum dimension, in plan, >30 m. Many limestone masses, exposed by later erosion, constitute the present-day castiles of the Gypsum Plain. The H 2 S generated at the porous masses of subsurface limestone dissolved readily within the rising, pressurized, largely
63 At the sulfur deposits, biogenic H 2 S was supplied to reaction sites within hypogenic, relatively fresh groundwater that moved directly upward along steep, permeability-enhanced, fault-tracing pathways for meters to many tens of meters by buoyancy, artesian pressure, and overpressure. At the developing caves, biogenic H 2 S was supplied to reaction sites from basinal microbial loci (represented by the carbonate masses of the western basin). Within the uppermost part of beds of Castile halite, the H 2 Sbearing groundwater, driven by artesian pressure and into the Capitan reef and forereef, eventually into cave pools, degassed into cave atmospheres, and, eventually, moved to reaction sites on cave walls and ceilings. Carbon-isotopic values of samples of the castiles support CH 4 as the microbial foodstuff. Half of 20 samples of calcite from the secondary masses of Castile limestone masses exposed at the earths surface (the castiles) have 13 C values in the range -39 to -28 (Kirkland and Evans, 1976). The carbon within these samples must have been derived entirely or largely from CH 4 Their values 13 C range exhibited by thermogenic CH 4 but they fall outside the range exhibited by crude oil. To form the castiles, the subsurface masses of limestone, and the caves of the Guadalupe Mountains the basinal microbial agents required a huge volume of foodstuff (i.e., a reductant). Besides CH 4 probably only crude oil Permian crude oil contains an asphaltic fraction (~10%) that microbes are unable to devour. The castiles and the buried sulfur-bearing limestone masses are virtually devoid of an asphaltic residue, an absence that supports CH 4 as the primary microbial foodstuff. Samples of calcite 13 C values more positive than -35 probably dominantly from minor amounts of Castile evaporitic 13 C of about +6.0. The presence of this fraction probably resulted in ten samples (out of 20) calcite generated strictly or predominantly by CH 4 CH 4 is the primary reductant at most, if not all, large biogenic sulfur deposits elsewhere in the world (e.g., US Gulf Coast, Poland, Ukraine, Sicily). In addition, these deposits are all associated with evaporites, and they all shelfal carbonates initiating intense speleogenesis. Then, during an interval of ~8 Mawith additional episodes of uplift, descent of the water table, and erosion of the coverspeleogenesis descended progressively in steps both within the slightly inclined (<0.5) northeasttrending cave belt and within individual caves. A genetic and geographic relationship exists between caves of the Guadalupe Mountains and large deposits of native sulfur (a few million to many millions of metric tons) beneath the adjacent Gypsum Plain. The deposits and the caves probably formed at about the same time, and they probably both owe their existence to a coincidence of essentially the same stratigraphic, thermal, biogenic, and tectonic events. Similarities in the genetic history of both include: Hypogenic groundwater convectively dissolved Castile anhydrite and halite. Anaerobic microbes within the evaporite sequence mediated a reaction between CH 4 and SO 4 2generating immense quantities of H 2 S. Oxidation of the aqueous H 2 S required immense quantities of aqueous O 2 Processes forming the caves and those forming the major deposits of sulfur required a lengthy period (probably many hundreds of thousands of years). Differences in their genetic history include: At the sulfur deposits, O 2 oxidized H 2 S to form native sulfur probably primarily inorganically within the phreatic realm, whereas at the caves, O 2 oxidized H 2 S to form H 2 SO 4 biogenically mainly within the vadose realm. At the major sulfur deposits, epigenic groundwater bearing O 2 dissolved halite of the Salado Formation (~85% halite; ~0.5 km thick); the resulting boost in density caused oxygen-bearing brine to sink directly downward along an inverted density gradient through a permeability-enhanced faulttracking pathway to aqueous reaction sites within brecciated Salado anhydrite and brecciated and bedded Castile anhydrite. At the caves, however, gaseous O 2 from the earths atmosphere probably moved laterally and slightly downward under a stratal cover by diffusion, thermal convection, and barometric winds to aqueous reaction sites within basin-margin carbonates.
64 have petrologic and isotopic characteristics remarkably similar to those of the castiles of the Gypsum Plain and to those of the large, limestone-hosted, sulfur deposits beneath the Gypsum Plain. These analogous traits support CH 4 as having been the dominant substrate (foodstuff) at microbial loci within the Castile Formation, loci that are now represented by masses of biogenic limestone. A biogenic reaction between CH 4 and SO 4 2, the same overall reaction that took place within the Castile during modern, shallow marine sediments and within marine sediments associated with seeps of CH 4 Bio-enzymes that catalyze the redox reaction are derived either from archaea or from a consortium of sulfate-reducing bacteria and archaea. The marine microorganisms that mediate the modern reaction are probably the same as or related closely to those active within the Castile during the Late Tertiary. Furthermore, many modern, cold-seep deposits of the worlds oceans display processesincluding SO 4 2reduction by CH 4 and O 2 reduction by H 2 S duplicated ~12 to ~4 Ma ago by terrestrial processes operating collectively within the Castile, Capitan, and Seven Rivers formations.
65 contributed to my knowledge of the Capitan reef and sections of the text. To help its readability, Thane McCulloh, Peggy Kirkland, and Melodye Rooney, reviewed all, and Maggie George, David Kirkland, and Henry Snider, parts of the manuscript. I thank them all for their help and encouragement. Harvey DuChene, Carol Hill, Lewis Land, Arthur Palmer, and George Veni, also reviewed the manuscript. all generous with their time, and they were helpful in many ways. Their constructive comments substantially improved the text. Arthur Palmer, in addition, added color Cavern and Lechuguilla Cave. I greatly appreciate their help and their enthusiastic support. Special thanks also to Lewis Land for managing the editorial process and to Julie Fielding for formatting the manuscript. A poster session by Michael Queen in 1979 initially piqued my curiosity about the origin of caves within the Guadalupe Mountains. In 1980, Ronal Kerbo and David Kirkland generously helped me collect samples of gypsum from Carlsbad Cavern for isotopic analysis. Much more recently, in 2006 during a tour through Carlsbad Cavern, Harvey DuChene reignited my interest in and graciously shared his wide knowledge of Carlsbad Cavern and Lechuguilla Cave. In addition, he sent me several hard to come by articles on the caves of the Guadalupe Mountains. Robert Evans, over many years, helped to formulate ideas on dissolution chambers at the crests of salt diapirs. Thane McCulloh Miocene of the western Delaware Basin, Richard Koepnick reviewed the sections on microbial and thermochemical sulfate reduction, and Melodye Rooney reviewed aspects of the geochemistry. Rodger Andersons experiments on dissolution of halite and his shared knowledge of dissolution processes have been ACKNOWLEDGMENTS
66 Adams, J.E., Cheney, M.G., DeFord, R.K., Dicky, R.I, Dunbar, C.O., Hills, J.M., King, R.E., Lloyd, E.R., Miller, A.K., and Needham, C.E., 1939, Standard Permian section of North America: American Association of Petroleum Geologists, v. 2, p. 1673-1681. Adams, J.E., 1944, Upper Permian Ochoa series of Delaware basin, west Texas and southeast New Mexico: American Association of Petroleum Geologists, v. 28, p. 1596-1625. Adams, J.E., and Frenzel, H.N., 1950, Capitan barrier reef, Texas and New Mexico: Journal of Geology, v. 58, p. 289-312. http://dx.doi.org/10.1086/625749 Aloisi, G., Bouloubassi, I., Heijs, S.K., Pancost, R.D., Pierre, C., Sinninghe Damst, J.S., Gottschal, J.C., Forney, L.J., and Rouchy, J., 2002, CH 4 -consuming microorganisms and the formation of carbonate crusts at cold seeps: Earth and Planetary Science Letters, v. 203, p. 195-203. http://dx.doi.org/10.1016/S0012-821X(02)00878-6 Alonso-Azcrate, J., Bottrell, S.H. and Tritila, J., 2001, Sulfur redox reactions and formation of native sulfur veins during low-grade metamorphism of gypsum evaporites, Cameros Basin (NE Spain): Chemical Geology, v. 174, no. 4, p. 389-402. http://dx.doi.org/10.1016/S0009-2541(00)00286-2 Alperin, M., and Hoehler, T., 2009, Anaerobic methane oxidation by archaea/sulfate-reducing bacterial aggregates: 1. Thermodynamic and physical constraints: American Journal of Science, p. 869-957. http://dx.doi.org/10.2475/10.2009.01 Alperin, M., and Hoehler, T., 2010, The ongoing mystery of http://dx.doi.org/10.1126/science.1189966 Al-Sawaf, F.D.S., 1977, Sulfate reduction and sulfur deposition in the Lower Fars Formation, Northern Iraq: Economic Geology, v. 72, p. 608-618. http://dx.doi.org/10.2113/gsecongeo.72.4.608 Anderson, J.E., 1968, Igneous geology of the central Davis Mountains, Jeff Davis County, Texas: Bureau of Economic Geology, text accompanying Quadrangle Map 36, 18 p. Anderson, R.Y., 1978, Deep dissolution of salt, northern Delaware Basin, New Mexico: Report to Sandia Laboratories, Albuquerque, New Mexico, 106 p. Anderson, R.Y., 1981, Deep-seated salt dissolution in the Delaware Basin, in Wells, S.G., and Lambert, W., eds., Environmental Geology and Hydrology in New Mexico, New Mexico Geological Society, Special Publication 10, p. 133-145. Anderson, R.Y., 1982, Deformation-dissolution potential of bedded salt Waste Isolation Pilot Plant Site, Delaware Basin, New Mexico: Fifth International Waste Management, Berlin, Germany, 10 p. Anderson, R.Y., 1993, The Castile as a nonmarine evaporite, in Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M., eds., Carlsbad Region, New Mexico and West Texas, New Mexico Geolological Society Guidebook, Forty-Fourth Annual Field Conference, p. 12-13. REFERENCES CITED Anderson, R.Y., 2011, Enhanced climate variability in the tropics: a 200,000 yr annual record of monsoon variability from Pangeas equator: Climate of the Past, v. 7, p. 757-770. http://dx.doi.org/10.5194/cp-7-757-2011 Anderson, R.Y., and Dean, W.E., 1995, Filling the Delaware Basin: Hydrologic and climatic controls on the Upper Permian Castile Formation varved evaporite, in Scholle, P.A., Peryt, T.M. and UlmerScholle, D.S., eds., The Permian of Northern Pangea, Sedimentary Basins and Economic Resources, v. 1, New York, Springer-Verlag, p. 61-78. http://dx.doi.org/10.1007/978-3-642-78590-0_4 Anderson, R.Y., Dean, W.E., Kirkland, D.W. and Snider, H.I., 1972, Permian Castile varved evaporite sequence, west Texas and New Mexico: Geological Society of America Bulletin, v. 83, p. 59-86. http://dx.doi.org/10.1130/00167606(1972)83[59:PCVESW]2.0.CO;2 Anderson, R.Y., Kietzke, K.K., and Rhodes, D.J., 1978, Development of dissolution breccias, northern Delaware basin, New Mexico and Texas, in Austin, G.S., ed., Geology and Mineral Deposits of Ochoan Rocks in Delaware Basin and Adjacent Areas: Socorro, NM, New Mexico Bureau of Geology and Mineral Resources Circular 159, p. 47-52. Anderson, R.Y., and Kirkland, D.W., 1966, Intrabasin varve correlation: Geological Society of America Bulletin, v. 77, p. 241-256. http://dx.doi.org/10.1130/00167606(1966)77[241:IVC]2.0.CO;2 Anderson, R.Y., and Kirkland, D.W., 1980, Dissolution of salt http://dx.doi.org/10.1130/00917613(1980)8<66:DOSDBB>2.0.CO;2 Anderson, R.Y., and Powers, D.W., 1978, Salt anticlines in Castile-Salado evaporite sequence, northern Delaware basin, New Mexico, in Austin, G.S., ed., Geology and Mineral Deposits of Ochoan Rocks in Delaware Basin and Adjacent Areas: Socorro, NM, New Mexico Bureau of Geology and Mineral Resources Circular 159, p. 78-83. Andrejchuk, V.N., and Klimchouk, A.B., 2001, Geomicrobiology and redox geochemistry of Ukraine: The study from Zoloushka Cave: Geomicrobiology Journal, v. 18, p. 275-295. http://dx.doi.org/10.1080/01490450152467796 Atlas, R.M., 1997, Principles of microbiology, 2 nd ed., Boston, Massachusetts, McGraw-Hill, 1298 p. Babcock, L.C., 1977, Life in the Delaware Basin: The paleoecology of the Lamar Limestone, in Hileman, M.E., and Mazzullo, S.J., eds., Upper Guadalupian Facies Permian Reef Complex, Guadalupe Mountains, New Mexico and West Texas Society of Economic Paleontologists and Mineralogist, Permian Basin Section, Publication 77-16, 1977, Field Conference Guidebook, p. 357-387.
67 Bgli, A., 1980, Karst hydrology and physical speleology: Berlin, Springer-Verlag, 284 p. http://dx.doi.org/10.1007/978-3-642-67669-7 carbonates in a native sulfur deposit: Stable isotope and trace element discrimination related to the transformation of aragonite to calcite: Isotopes in Environmental and Health Studies, v.33, p. 177-190. http://dx.doi.org/10.1080/10256019708036345 Boyd, D.W., 1958, Permian sedimentary facies, central Guadalupe Mountains, New Mexico: New Mexico Bureau of Mines and Mineral Resources, Bulletin 49, 100 p. Bozanich, R.G., 1979, The Bell Canyon and Cherry Canyon Formations, Eastern Delaware Basin, Texas: lithology, environments and mechanisms of deposition, in Sullivan, N.M., ed., Guadalupian Delaware Mountain Group of West Texas and Southeast New Mexico, 1979 Symposium and Field Conference Guidebook, Permian Basin Section, Society for Sedimentary Geology (SEPM), Publication 79-18, p. 121-141 Bretz, J.H. 1949, Carlsbad Caverns and other caves of the Guadalupe Block, New Mexico: Journal of Geology, v. 57, p. 447-463. http://dx.doi.org/10.1086/625660 in cratonic and foreland basins: A proposed methodology with application to the Delaware Basin area, west Texas and New Mexico (abs.): American Association of Petroleum Geologists, Annual Meeting, Dallas, Texas. Brown, A.A., 2006, Unresolved problems with sulfate speleogenesis of Carlsbad Caverns, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds. Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 36-38. Brown, A.A., and Loucks, R.G., 1988, Castiles in the Castile Formation, in Reid, S.T., Bass, R.O., and Welch, P., eds., Guadalupe Mountains Revisited, Texas and New Mexico, 1988 Field Seminar: Midland, TX, West Texas Geological Society Publication 88-84, p. 114-116. Buck, M.J., Ford, D.C. and Schwarcz, H.P., 1994, from oxidation of H 2 S, in Sasowsky, I.D., and Palmer, M.V., eds., Breakthroughs in Karst Geomicrobiology and Redox Geochemistry, Special Publication no. 1, p. 5-9. Bullington, N.R., 1968, Geology of the Carlsbad Caverns: West Texas Geological Society Guidebook 68-55, p. 20-23. Cai, C., Xie, Z., Worden, R.H., Hu, G., Wang, L., and He, H., 2004, Methane-dominated thermochemical sulphate reduction in the Triassic Feixianguan Formation East Sichuan Basin, China: towards prediction of fatal H 2 S concentrations: Marine and Petroleum Geology, v. 21, p. 1265-1279. http://dx.doi.org/10.1016/j.marpetgeo.2004.09.003 Ballentine, C.J., Schoell, M., Coleman, D. and Cain, B.A., 2001, 300-Myr-old magmatic CO 2 in natural gas reservoirs of the west Texas Permian basin: Nature, v. 409, p. 327-330. http://dx.doi.org/10.1038/35053046 Barker, C.E. and Halley, R.B., 1986, Fluid inclusion, the thermal history of the Bone Spring Limestone, southern Guadalupes, Texas, in Gautier, D.L., ed., Roles of Organic Matter in Sediment Diagenesis: Tulsa, OK, Society for Sedimentary Geology (SEPM), Special Publication no. 38, p. 189-203. Barker, C.E., and Pawlewicz, M.J., 1987, The effects thermal maturity of Leonardian and younger rocks, western Delaware Basin, Texas, in Cromwell, D.W., and Mazzullo, L., eds., The Leonardian Facies in W. Texas and S.E. New Mexico and Guidebook to the Glass Mountains, West Texas: Midland, TX, Permian Basin Section-Society for Sedimentary Geology (SEPM) Publication 87-27, p. 69-83. Barker, C.E., and Pawlewicz, M.J., 1993, Post-tectonic reheating of portions of the Permian Basin as structural sections, in Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M., eds., Carlsbad Region, New Mexico and West Texas: New Mexico Geological Society, 44 th Annual Field Conference Guidebook, p. 29-30. Barker, H.A., 1956, Biological formation of methane: Industrial and Engineering Chemistry, v. 48, p. 1438-1442. http://dx.doi.org/10.1021/ie51400a023 Barker, J.M, Cochran, D.E. and Semrad, R., 1979, Economic geology of the Mishraq native sulfur deposit, northern Iraq: Economic Geology, v. 74, p. 484-495. http://dx.doi.org/10.2113/gsecongeo.74.2.484 Barrett, T.J., Anderson, G.M. and Lugowski, J., 1988, The at 25-95 C and one atmosphere: Geochimica et Cosmochimica Acta, v. 52, p. 807-811. http://dx.doi.org/10.1016/0016-7037(88)90352-3 Berg, R.R., 1979, Reservoir sandstones of the Delaware Mountain Group, southeast New Mexico, in Sullivan, N.M., ed., Guadalupian Delaware Mountain Group of West Texas and southeast New Mexico, 1979 Symposium and Field Conference Guidebook, Permian Basin Section, Society for Sedimentary Geology (SEPM), Publication 79-18, p. 75-95. Black, T.H., 1954, The origin and development of the Carlsbad Caverns: New Mexico Geological Society, 5 th Field Conference, p. 136-142. Bodenlos, A.J. 1973, Sulfur, in United States Mineral Resources, U.S. Geological Survey Professional Paper 820, p. 605-618. Boetius, A., Ravenschlag, K., Schubert, G.J., Rickert, D., Widdel, F., Gieseke A., Amann, R., Jrgensen, B.B., Witte, U., and Pfannkuche, O., 2000, A marine microbial consortium apparently mediating anaerobic oxidation of methane: Nature, v. 407, p. 623-626. http://dx.doi.org/10.1038/35036572
68 Darke, G., and Harwood, G., 1990, Time constraints on sulfate-related diagenesis, Capitan reef complex, West Texas and New Mexico (Abstract): AAPG Annual Convention, San Francisco, California (available on Internet). Davies, P.B., 1983, Assessing the potential for deepseated salt dissolution and subsidence at the Waste Isolation Pilot Plant (WIPP): State of New Mexico Environmental Evaluation Group, Conference Proceedings, WIPP Site Suitability for Radioactive Waste Disposal, Carlsbad, New Mexico, 62 p. Davis, D.G., 1980, Cave development in the Guadalupe Mountains a critical review of recent hypotheses: National Speleological Society Bulletin, v. 42, no 3, p. 42-48. Davis, D.G., 2000, Extraordinary features of Lechuguilla Cave, Guadalupe Mountains, New Mexico: Journal of Cave and Karst Studies, v. 62, no. 2, p. 147-157 dirt bed: American Association of Petroleum Geologists Bulletin, v. 36, p. 2186-2188. Davis, J.B., and Kirkland, D.W., 1970, Native sulfur deposition in the Castile Formation, Culberson County, Texas: Economic Geology, v. 65, p. 107-121. http://dx.doi.org/10.2113/gsecongeo.65.2.107 Davis, J.B., and Kirkland, D.W., 1979, Bioepigenetic sulfur deposits: Economic Geology, v. 74, p. 462-468. http://dx.doi.org/10.2113/gsecongeo.74.2.462 Davis, J.B., Stanley, J.P., and Custard, H.C., 1970, by sulfate ions to produce elemental sulfur in salt domes: American Association of Petroleum Geologists Bulletin, v. 54, no. 12, p. 2444-2447. Davis, J.B., and Yarbrough, H.F., 1966, Anaerobic oxidation of hydrocarbons by Desulfovibrio desulfuricans : Chemical Geology, v. 1, p. 137-144. http://dx.doi.org/10.1016/0009-2541(66)90012-X Dean, W.E., 1967, Petrology and geochemical variations in the Permian Castile varved anhydrite, Delaware Basin, Texas and New Mexico [Ph.D. Dissertation]: University of New Mexico, 318 p. Dean, W.E., and Anderson, R.Y., 1978, Salinity cycles: Evidence for subaqueous deposition of Castile Formation and lower part of Salado Formation, Delaware Basin, Texas and New Mexico, in Austin, G.S., ed., Geology and Mineral Deposits of Ochoan Rocks in Delaware Basin and Adjacent Areas: Socorro, NM, New Mexico Bureau of Geology and Mineral Resources Circular 159, p. 15-20. Dean, W.E., and Anderson, R.Y., 1982, Continuous subaqueous deposition of the Permian Castile evaporites, Delaware Basin, Texas and New Mexico, in Hanford, C.R. and Loucks, R.G., and Davies, G.R., eds., Depositional and Diagenetic Spectra of EvaporitesA Core Workshop: SEPM Core Workshop 3, p. 324-353. DeLong, E.F., 2003, Oceans of Archaea: The American Society for Microbiology, v. 69, no. 10, p. 503-511. Dessau, G., Jensen, L., and Nakai, N., 1962, Geology and isotopic studies of Sicilian sulfur deposits: Economic Geology, v. 57, p. 410-438. http://dx.doi.org/10.2113/gsecongeo.57.3.410 Calzia, J.P., and Hiss, W.L., 1978, Igneous rocks in the northern Delaware Basin, New Mexico and Texas, in Austin, G.S., ed., Geology and Mineral Deposits of Ochoan Rocks in Delaware Basin and Adjacent Areas: Socorro, NM, New Mexico Bureau of Geology and Mineral Resources Circular 159, p. 39-46. Clari, P.A., and Martire, L., 2000, Cold seep carbonates in the Tertiary of northwestern Italy: Evidence of bacterial degradation of methane, in Riding, R.E., and Awramik, S.M., eds., Microbial Sediments, New York, Springer-Verlag, p. 261-269. Clark, K.F. 1990, Road log, in Kyle, J.R., ed., Industrial Mineral Resources of the Delaware Basin, Texas and New Mexico: Society of Economic Geology, Guidebook, no. 8, p. 1-83. Claypool, G.E., Holser, W.T., Kaplan, W.T., Sakai, H., and Zak, I., 1980, The age curves of sulfur and oxygen isotopes in marine sulfate and their mutual interpretation: Chemical Geology, v. 28, p. 199-260. http://dx.doi.org/10.1016/0009-2541(80)90047-9 Claypool, G.E., and Mancini, E.A., 1989, Geochemical relationships of petroleum in Mesozoic reservoirs to carbonate source rocks of Jurassic Smackover Formation, southwestern Alabama: American Association of Petroleum Geologists, v. 73, no. 7, p. 904-917. Coleman, M.L., 1985, Geochemistry of diagenetic non-silicate minerals: kinetic considerations: Philosophical Transactions, London, Royal Society, v. A315, p. 39-56. http://dx.doi.org/10.1098/rsta.1985.0028 Crawford, J.E., 1990, Geology and Frasch-mining operations of the Culberson sulfur mine, Culberson County west Texas: Society of Economic Geologists, Guidebook no. 8, p. 141-162. Crawford, J.E. and Wallace, C.S., 1993, Geology and mineralization of the Culberson sulfur deposit, in Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M, eds., Carlsbad Region New Mexico and West Texas, New Mexico Geological Society Guidebook 44 th Field Conference, p. 301-316. Crysdale, B.L., 1987, Fluid inclusion evidence for the origin, diagenesis and thermal history of sparry calcite cement in the Capitan Limestone, McKittrick Canyon, west Texas [MSc. thesis]: University of Colorado, 78 p. Cunningham, K.I., DuChene, H.R., and Spirakis, C.S., 1993, Elemental sulfur in caves of the Guadalupe Mountains, New Mexico, in Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M. eds., Carlsbad Region, New Mexico and West Texas, New Mexico Geological Society, Guidebook 44 th Annual Field Conference, p. 129-136. Curl, R.L., 1966, Cave conduit enlargement by natural convection: Cave Notes (USA), v. 8, no. 1, p. 4-8. the Castile-Bell Canyon contact, in New Mexico Bureau of Geology and Mineral Resources Circular 159, p. 53-56.
69 Engel, A.S., Stern, L.A., and Bennett, P.C., 2004, Microbial contributions to cave formation: New insights into sulfuric acid speleogenesis: Geology, v. 32, no. 5, p. 369-272. http://dx.doi.org/10.1130/G20288.1 Enos, J.S., and Kyle, J.R., 2002, Diagenesis of the Carrizo Sandstone at Butler Salt Dome, East interaction near halokinetic structures: Journal of Sedimentary Research, v. 72, no. 1, p. 68-81. http://dx.doi.org/10.1306/061101720068 Feely, H.W., and Kulp, J.L., 1957, Origin of Gulf Coast salt dome sulfur deposits: Geological Society of America Bulletin, v. 41, p. 1801-1853. Ford, D.C., and Hill, C.A., 1989, Dating results from Carlsbad Cavern and other caves in the Guadalupe Mountains, New Mexico: Isochron/West, v. 54, p. 3-7. Ford, D. and Williams, P., 2007, Karst Hydrology and Geomorphology: Chichester, England, John Wiley and Sons, Ltd., 553 p. Fulda, E., 1938, Teil 2: Steinsalz und Kalisalze, in Beyschlag, F., Krush, P., and Vogt, J.H., eds., Lagersttten der Nutzbarn Minerakien und Gesteine nach Form, Inhalt und Entsehung, pt. 2, v. 3, p. 1-240. Garber, R.A., Grover, G.A., and Harris, P.M., 1989, Geology of the Capitan shelf marginsubsurface data from the northern Delaware Basin, in P.M. Harris, and G.A. Grover, eds., Subsurface and Outcrop Examination of the Capitan Shelf Margin, Northern Delaware Basin, Society for Sedimentary Geology (SEPM) Core Workshop, v. 13, p. 3-269. http://dx.doi.org/10.2110/cor.89.13.0003 Goldhaber, M.B., 1993, Thermochemical sulfate reduction as a source of sedimentary H 2 S: Society of Sedimentary Geology (SEPM), International Geological Congress, Annual Meeting, Boston, Massachusetts, p. A-21. Goldhaber, M.B., 2003, Sulfur-rich sediments, in Holland, H.D., and Turekian, K.K., eds., Treatise on Geochemistry, 7, Sediments, Diagenesis, and Sedimentary Rocks, F.T. Mackenzie, ed., New York, Elsevier, p. 257-288. Goldman, M.I., 1933, Origin of the anhydrite cap rock of American salt domes: U.S. Geological Survey, Professional Paper 175-D, p. 83-114. Goldman, M.I., 1952, Deformation, metamorphism, and mineralization in gypsum-anhydrite cap rock Sulfur Salt Dome, Louisiana: Geological Society of America Memoir 50, 169 p. Good, J.M., 1957, Noncarbonate deposits of Carlsbad Caverns: National Speleological Society Bulletin, v. 19, p. 11-23. Grauten, W.F., 1965, Fluid relationships in Delaware Mountain sandstone, in Young, A., and Galley, J.E., eds., Fluids in Subsurface Environments: Tulsa, OK, American Association of Petroleum Geologists Memoir 4, p. 294-307. Guilinger, J.R., 1993, The geology and development of the Phillips Ranch sulfur deposit, in Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M., eds., Carlsbad Region New Mexico and West Texas, New Mexico Geological Society Guidebook 44th Field Conference, p. 21-23. Dietrich, J.W., Owen, D.E., and Shelby, C.A., 1983, Van Horn-El Paso sheet: Austin, University of Texas Bureau of Economic Geology, Geologic Atlas of Texas, scale 1:250,000. Donovan, T.J., 1974, Petroleum microseepage at Cement, Oklahoma: evidence and mechanism. American Association of Petroleum Geologists Bulletin, v. 58, p. 429-446. Druckman, Y., Weissbrod, T., and Aharon, P., 1994, venting imprinted on a Quaternary eolianite from southern Israel: Geo-Marine Letters, v. 14, p. 170176. http://dx.doi.org/10.1007/BF01203728 Duan, Z., Sun, R., Liu, R., and Zhu, C., 2007, Accurate thermodynamic model for the calculation of H 2 S solubility in pure water and brines: Energy & Fuels, v. 21, p. 2056-2065. http://dx.doi.org/10.1021/ef070040p DuChene, H.R., 1986, Observations on previous hypotheses and some new ideas on cavern formation in the Guadalupe Mountains, in Jagnow, D.H., ed., Geology Field Trip, National Speleological Society Convention Guidebook, Tularosa, New Mexico, p. 96-100. (Guadalupian) Artesia Group to sulfuric acid speleogenesis in the Guadalupe Mountains, New Mexico and Texas, USA, in Land, L., and Veni, G., eds., NCKRI Symposium 1, Advances in Hypogene Karst Studies, National Cave and Karst Research Institute, p. 111-120. DuChene, H.R., and Cunningham, K.I., 2006, Tectonic Mountains, New Mexico and Texas, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 195-202. DuChene, H.R, and Martinez, R., 2000, Postspeleogenetic erosion and its effects on caves in the Guadalupe Mountains: National Speleological Society Bulletin, v. 62, p. 75-79. DuChene, H.R., and McLean, J.S., 1989, The role of Guadalupes of southeastern New Mexico, in Harris, P.M., and Grover, G.A., eds., Subsurface and Outcrop Examination of the Capitan Shelf Margin, Northern Delaware Basin: Tulsa, OK, Society for Sedimentary Geology (SEPM), Core Workshop 13, p. 475-481. Dutton, S.P., 2008, Calcite cement in Permian deep-water sandstones, Delaware Basin, west Texas: Origin distribution and effects on reservoir properties: American Association of Petroleum Geologists Bulletin, v. 92, no. 6, p. 765-787. Egemeier, S.J., 1971, A comparison of two types of solution caves: Unpublished report to Carlsbad Caverns National Park, April 12, 7 p. Ehrlich, H.L., 1990, Geomicrobiology, New York, N.Y., Marcel Dekker, 646 p.
70 Heydari, E., 1997, The role of burial diagenesis in hydrocarbon destruction and H 2 S accumulation, Upper Jurassic Smackover Formation, Black Creek Field, Mississippi: American Association of Petroleum Geologists Bulletin, v. 81, no. 1, p. 26-45. Heydari, E., and Moore, C.H., 1989, Burial diagenesis and thermochemical sulfate reduction, Smackover Formation, southeastern Mississippi salt basin: Geology, v. 17, p. 1080-1084. http://dx.doi.org/10.1130/00917613(1989)017<1080:BDATSR>2.3.CO;2 Hill, C.A., 1981, Speleogenesis of Carlsbad Caverns and other caves in the Guadalupe Mountains, in B.F. Beck, ed., Proceedings of Eight International Congress of Speleology, Bowling Green, Kentucky, v. 1, p. 143-144. Hill, C.A., 1987, Geology of Carlsbad Caverns and Other Caves in the Guadalupe Mountains, New Mexico and Texas: New Mexico Bureau of Mines and Mineral Resources Bulletin 117, 150 p. Hill, C.A., 1990, Sulfuric acid speleogenesis of Carlsbad Caverns and its relationship to hydrocarbons, Delaware Basin, New Mexico and Texas: American Association of Petroleum Geologists Bulletin, v. 74, p. 1685-1694. Hill, C.A., 1992, Isotopic values of native sulfur, barite, celestite, and calcite: their relationship to sulfur deposits and to the evolution of the Delaware basin (preliminary results), in Wessell, G.R., and Wimberley, B.H. eds., Native Sulfur, Developments in Geology and Exploration: Society of Minerals, Metallurgy, and Exploration (Phoenix, AZ Meeting, Feb. 1992), p. 147-157. Guadalupe Mountains, New Mexico and West Texas: New Mexico Geology, v. 15, no. 3, p. 56-65. Hill, C.A., 1996, Geology of the Delaware Basin, Guadalupe, Apache, and Glass Mountains, New Mexico and West Texas, Midland, TX, Permian Basin Section Society for Sedimentary Geology (SEPM), Publication 96-39, 480 p. Hill, C.A., 1999, Origin of Caves in the Capitan, in Saller, A.H., Harris, P.M., Kirkland, B.L., and Mazzullo, S.J., eds., Geologic Framework of the Capitan Reef: Permian Basin SectionSociety for Sedimentary Geology (SEPM), Special Publication 65, p. 211222. http://dx.doi.org/10.2110/pec.99.65.0211 Hill, C.A., 2000, Sulfuric acid, hypogene karst in the Guadalupe Mountains of New Mexico and West Texas (U.S.A.), in Klimchouk, A.B., Ford, D.C., Palmer, A.N., and Dreybrodt, W., eds., Speleogenesis: Evolution of Karst Aquifers, Huntsville, AL, National Speleological Society, p. 309-316. Hill, C.A., and Forti, P., 1986, Cave minerals of the world, Huntsville, AL, National Speleological Society, 238 p. Hills, J.M., 1984, Sedimentation, tectonism, and hydrocarbon generation in Delaware Basin, west Texas and southeastern New Mexico: American Association of Petroleum Geologists Bulletin, v. 68, p. 250-267. Guilinger, J.R., and Nestlerode, E.L., 1992, The geology and development of the Phillips Ranch sulfur deposit, in Wessel, G.R. and Wimberly, B.H. eds., Native SulfurDevelopments in Geology and Exploration, Society Mining and Metallurgy, Exploration Proceedings, Phoenix, Arizona, Chapter 14, p. 125-134. Haigler, L., 1962, Geologic notes on the Delaware Basin: New Mexico Bureau of Mines and Mineral Resources, Circular 63, 14 p. Hansom, J., Lee, M., Doser, D.I., and Feng, Y., 2003, The hydrodynamics and related biogeochemical processes in the Permian Basin, western Texas: Abstracts with Program Geological Society of America, v. 35, no. 6, p. 573. Harder, J., 1997, Anaerobic methane oxidation by bacteria employing 14 C-methane uncontaminated with 14 C-carbon monoxide: Marine Geology, v. 137, p. 13-23. http://dx.doi.org/10.1016/S00253227(96)00075-8 Harrison, A.G., and Thode, H.G., 1958, Mechanism of the bacterial reduction of sulphate from isotope fractionation studies: Transactions Faraday Society, v. 58, p. 84-92. http://dx.doi.org/10.1039/tf9585400084 Harwood, G.M., Darke, G., and Kendall, A.C., 1991, Fluid conduits through the Capitan reef in the Delaware Basintheir scale, timing and availability, in Candelaria, M.P., ed., Permian Basin PlaysTomorrows Technology Today: Midland, TX, West Texas Geological Society Publication 91-89, p. 165-168. Hayes, P.T., 1964, Geology of the Guadalupes, New Mexico: U.S. Geological Survey Professional Paper 446, 69 p. Hayes, P.T., and Gale, B.T., 1957, Geology of the Carlsbad Caverns East quadrangle, New Mexico: U.S. Geological Survey Geologic Quadrangle Map GQ-98. Henry, C.D., Kunk, M.J., and McIntosh, W.C., 1994, 40 Ar/ 39 Ar chronology and volcanology of silicic volcanism in the Davis Mountains, Trans-Pecos Texas: Geological Society of America, v. 106, no. 11, p. 1359-1376. http://dx.doi.org/10.1130/00167606(1994)106<1359:AACAVO>2.3.CO;2 Hentz, T.Z., 1990, Native sulfur deposits in the western Delaware Basin, Trans-Pecos Texas: Origin, distribution, and exploration strategy, in Kyle, J.R., ed., Industrial Mineral Resources of the Delaware Basin, Texas and New Mexico, Guidebook Series, v. 8, Society of Economic Geologists, p. 115-149. Hentz, T.F., Price, J.G., and Gutierrez, G.N., 1989, Geologic occurrence and regional assessment of evaporite-hosted native sulfur, Trans-Pecos Texas: The University of Texas at Austin, Report of Investigations 184, 70 p. Hentz, T.F., and Henry, C.D., 1989, Evaporite-hosted native sulfur in Trans-Pecos Texas: relation to latephase Basin and Range deformation: Geology, v. 17, p. 400-403. http://dx.doi.org/10.1130/00917613(1989)017<0400:EHNSIT>2.3.CO;2
71 Hunt, D.W., Fitchen, W.M., and Kosa, E., 2003, Syndepositional deformation of the Permian Capitan reef carbonate platform, Guadalupe Mountains, New Mexico, USA: Sedimentary Geology, v. 105, no. 3-4, p. 89-126. http://dx.doi.org/10.1016/S0037-0738(02)00104-5 Ivanov, M.V., 1968, Microbiological processes in the formation of sulfur deposits: English Translation, no. 1850, U. S. Department of Commerce, 298 p. Jagnow, D.H., 1977, Cavern development in the Guadalupe Mountains: Albuquerque, New Mexico, Adobe Press (Second Printing, 1986), 55 p. Jagnow, D.H., Hill C.A., Davis, D.G., DuChene, H.R., Cunningham, K.I., Northup, D.E., and Queen J.M., 2000, History of the sulfuric acid theory of speleogenesis in the Guadalupe Mountains, New Mexico: Journal of Cave and Karst Studies, v. 62, no. 2, p. 54-59. Jansen, K., Thauer, R.K., Widdel, F., and Fuchs, G., 1984, Carbon assimilation pathways in sulfate reducing bacteria: Formate, carbon dioxide, carbon monoxide, and acetate assimilation by Desufovibrio baarsii : Archives of Microbiology, v. 138, no. 3, p. 257-262. http://dx.doi.org/10.1007/BF00402132 Jassim, S.Z., Raiswell, R., and Bottrell, S.H., 1999, Genesis of the middle Miocene stratabound sulfur deposits of northern Iraq: Journal of the Geological Society, London, v. 156, p. 25-39. http://dx.doi.org/10.1144/gsjgs.156.1.0025 Jrgensen, B.B., 1982, Mineralization of organic matter in the sea bedthe role of sulphate reduction: Nature, v. 296, p. 643-645. http://dx.doi.org/10.1038/296643a0 Jrgensen, N.O., 1992, Methane-derived carbonate cementation of marine sediments from the Kattegat, Denmark: Geochemical and geological evidence: Marine Geology, v. 103, p. 1-13. http://dx.doi.org/10.1016/0025-3227(92)90006-4 Kelley, V.C., 1971, Geology of the Pecos country, southeastern New Mexico: New Mexico Bureau of Mines and Mineral Resources, Memoir 24, 75 p. Kempe, S., 1996, Gypsum karst of Germany, in Klimchouk, A., Lowe, D., Cooper, A., and Sauro, U., eds., Gypsum Karst of the World: International Journal of Speleology, v. 25, no. 3-4, p. 209-224. Kendall, A.C., 1988, Aspects of evaporite basin stratigraphy, in Schreiber, B.C., ed., Evaporites and Hydrocarbons, New York, Columbia University Press, p. 11-65. Kendall, A.C., and Harwood, G.M., 1989, Shallow and implications, in Subsurface and Outcrop Examination of the Capitan Shelf Margin, Northern Delaware Basin, Harris, P.M. and Grover, G.A. eds., Society for Sedimentary Geology (SEPM), Core Workshop 13, p. 451-457. King, P.B., 1948, Geology of the southern Guadalupes, Texas: U.S. Geological Survey Professional Paper 215, 183 p. Hinds, J.S., and Cunningham, R.R., 1970, Elemental sulfur in Eddy County, New Mexico: U.S. Geological Survey Circular 628, 13 p. Hinrichs, K.-U., Hayes, J.M., Sylva, S.P., Brewer, P.G., and DeLong, E.F., 1999, Methane-consuming archaebacteria in marine sediments: Nature, v. 398, p. 802-805. http://dx.doi.org/10.1038/19751 Hinrichs, K.-U. and Boetius, A., 2002, The anaerobic oxidation of methane: New insights in microbial ecology and biogeochemistry, in Wefer, G., Billett, D., Hebbeln, D., Jrgensen, B.B., Schlueter, M., and van Weering T.C E., eds., New York, SpringerVerlag, Ocean Margin Systems, p. 457-477. Hiss, W.L., 1975, Stratigraphy and ground water hydrology of the Capitan aquifer, southeastern New Mexico and western Texas [Ph.D. dissertation]: University of Colorado, 396 p. Hiss, W.L., 1977, Movement of ground water in Permian Guadalupian aquifer systems, southeastern New Mexico and west Texas (abs.), in Hileman, M.E. and Mazzullo, S.J., eds., Upper Guadalupian facies, Permian Reef Complex, Guadalupe Mountains, New Mexico and West Texas, 1977 Field Conference Guidebook, Midland, TX, Permian Basin SectionSociety for Sedimentary Geology (SEPM), Publication 77-16, p. 487. Hiss, W.L., 1980, Movement of ground water in Permian Guadalupian aquifer systems, southeastern New Mexico and western Texas, in Dickerson, P. W., and Hoffer, J.M., eds., Trans-Pecos Region, Southeastern New Mexico and West Texas: 31 st Field Conference Guidebook: Albuquerque, NM, New Mexico Geological Society, p. 289-294. Holmquest, H.J., 1965, The origin and distribution of gases in the Delaware and Val Verde Basins, in Young, A. and Galley, J.F., eds., Fluids in Subsurface Environments, A Symposium, American Association of Petroleum Geologists Memoir 4, p. 257-279. Holser, W.T., and Kaplan, J.R, 1966, Isotope geochemistry of sedimentary sulfates: Chemical Geology, v. 1, p. 93-135. http://dx.doi.org/10.1016/0009-2541(66)90011-8 Hose, L.D., and Macalady, J.L., 2006, Observations the key to understanding Guadalupe Mountains speleogenesis?, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 185-194. Hose, L.D., and Pisarowicz, J.A., 1999, Cueva de Villa Luz, Tabasco, Mexico: Reconnaissance study of an active sulfur spring cave and ecosystem: Journal of Cave and Karst Studies, v. 61, no. 1, p. 13-21. Hovorka, S.D., 1990, Sedimentary processes controlling halite deposition, Permian Basin, Texas [Ph.D. Dissertation]: University of Texas at Austin, 393 p. Hovorka, S.D., 2000, Deep-water to shallow-water transition in evaporites in the Delaware Basin, Texas, in Lindsay, R.F., Smith, A.H., Trentham, R.C., and Ward, R.F., West Texas Geological Society Publication 00-108, p. 273-299.
72 Klimchouk, A., 1997a, Speleogenetic effects of water density differences, in 12 th International Congress of Speleology, La Chaux-de-Fonds, proceedings, v. 1, p. 161-164. Klimchouk, A., 1997b, The role of karst in the genesis of sulfur deposits, Pre-Carpathian region, Ukraine: Environmental Geology, v. 31, issue 1/2, p. 1-20. http://dx.doi.org/10.1007/s002540050158 Klimchouk, A., 2000, Dissolution and conversion of gypsum and anhydrite, in Klimchouk, A., Ford, D., Palmer, A., Dreybrodt, W., eds., Speleogenesis, Evolution of Karst Aquifers, Huntsville, Alabama, National Speleological Society, Inc., p.160-168. Klimchouk, A., 2007, Hypogene speleogenesis: Hydrological and morphogenetic perspective: National Cave and Karst Research Institute Special Paper 1, Carlsbad, NM, National Cave and Karst Research Institute, 106 p. Kopp, O.C., Bennett III, M.E., Clark, C.E., 2000, Journal of Coal Geology, v. 44, p. 69-84. http://dx.doi.org/10.1016/S0166-5162(99)00069-5 and fractures. Upper Permian Capitan platform, New Mexico, U.S.A.: Journal of Sedimentary Research, v. 76 no. 1, p. 131-151. http://dx.doi.org/10.2110/jsr.2006.08 Kosa, E., and Hunt, D.W., 2006b, The effect of syndepositional deformation within the Upper Permian Capitan Platform on the speleogenesis and geomorphology of the Guadalupe Mountains, New Mexico, USA.: Geomorphology, v. 78, p. 279-308. http://dx.doi.org/10.1016/j.geomorph.2006.01.038 Kotelnikova, S., 2002, Microbial production and oxidation of methane in deep subsurface: EarthScience Reviews, v. 58, p. 367-395. http://dx.doi.org/10.1016/S0012-8252(01)00082-4 Kreitler, C.W., and Dutton, S.P., 1983, Origin and diagenesis of cap rock, Gyp Hill and Oakwood salt domes, Texas: Univ. of Texas, Bureau of Economic Geology, Report of Investigation 131, 58 p. Krouse, H.R., 1977, Sulfur isotopic studies and their role in petroleum exploration: Journal of Geochemical Exploration, v. 7, p. 189-211. http://dx.doi.org/10.1016/0375-6742(77)90081-4 Kupfer, D.H., 1976, Shear zones inside Gulf Coast salt stocks help to delineate spines of movement: American Association of Petroleum Geologists Bulletin., v. 60, p. 1434-1447. Land, L., 2006, Hydrology of Bottomless Lakes State Park, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 95-96. Lang, W.B., 1947, Occurrence of Comanche rocks in Black River Valley, New Mexico: American Association of Petroleum Geologists Bulletin, v. 31, no. 8, p. 1472-1478. King, P.B., 1949, Regional geologic map of parts of Hudspeth and Culberson counties, Texas: U.S. Geological Survey Oil and Gas Investigations Preliminary Map 90. King, R.H., 1947, Sedimentation in the Permian Castile Sea: American Association of Petroleum Geologists Bulletin, v. 31, p. 470-477. Kinsman, D.J.J., Boardman, M., and Borcsik, M., 1974, An experimental determination of the solubility of oxygen in marine brines, in Proceedings 4 th International Symposium on Salt, v. 1, Cleveland, Northern Ohio Geological Society, p. 325-327. Kirkland, B.L., Longacre, S.A., and Stoudt, E.L., 1999, The dynamic Capitan Reef: An image of an ancient reef and suggestions for future research, in Saller, A.H., Harris, P.M., Kirkland, B.L., and Mazzulo, S.J., eds., Geologic Framework of the Capitan Reef, Society for Sedimentary Geology (SEPM), Special Publication, v. 65, p. 161-173. http://dx.doi.org/10.2110/pec.99.65.0161 Kirkland, D.W., 1982, Origin of gypsum deposits in Carlsbad Caverns, New Mexico: New Mexico Geology, v. 4, no. 2, p. 20-21. Kirkland, D.W., 2003, An explanation for the varves of the Castile evaporites (Upper Permian), Texas and New Mexico, USA: Sedimentology, v. 50, p. 899-920. http://dx.doi.org/10.1046/j.1365-3091.2003.00588.x Kirkland, D.W., and Anderson, R.Y., 1970, Microfolding in the Castile and Todilto evaporites, Texas and New Mexico: Geological Society of America Bulletin, v. 81, p. 3259-3282. http://dx.doi.org/10.1130/00167606(1970)81[3259:MITCAT]2.0.CO;2 Kirkland, D.W., Denison, R.E., and Dean, W.E., 2000, Parent brine of the Castile evaporites (Upper Permian), Texas and New Mexico: Journal of Sedimentary Research, Section A: Sedimentary Petrology and Processes, v. 70, p. 749-761. http://dx.doi.org/10.1306/2DC40935-0E47-11D78643000102C1865D Kirkland, D.W., Denison, R.E., and Rooney, M.A. 1995, of south central Oklahoma, USA: Marine and Petroleum Geology, v. 12, no. 6, p. 629-644. http://dx.doi.org/10.1016/0264-8172(95)98089-N Kirkland, D.W., and Evans, R., 1976, Origin of limestone buttes, Gypsum Plain, Culberson County, Texas: American Association of Petroleum Geologists Bulletin, v. 60, p. 2005-2018. Klemmick, G.F., 1992, Geology and mineralization of the Pokorny sulfur deposit, Culberson County, Texas, in Wessel, G.R. and Wimberly, B.H., eds., Native SulfurDevelopments in Geology and Exploration, Society Mining and Metallurgy, Exploration Proceedings, Phoenix, Arizona, Chapter 8, p. 109123. Klemmick, G.F, 1993, Geology of the Pokorny sulfur deposit, Culberson County, Texas, in Love, D.W., Hawley, J.W., Kues, B.S., Adams, J.W., Austin, G.S., and Barker, J.M., eds., Carlsbad Region New Mexico and West Texas, New Mexico Geological Society Guidebook 44 th Field Conference, p. 18-19.
73 Machel, H.G., 1992, Low-temperature and hightemperature origins of elemental sulfur in diagenetic environments, in Wessel, G.R., and Wimberly, B.H., eds., Native SulfurDevelopments in Geology and Exploration, Littleton, Colorado, p. 3-22. Machel, H.G., 2001, Bacterial and thermochemical sulfate reduction in diagenetic settingsOld and new insights: Sedimentary Geology, v. 140, p. 143-175. http://dx.doi.org/10.1016/S0037-0738(00)00176-7 Machel, H.G., Krouse, H.R., and Sassen, R., 1995, Products and distinguishing criteria of bacterial and thermochemical sulphate reduction: Applied Geochemistry, v. 10, p. 373-389. http://dx.doi.org/10.1016/0883-2927(95)00008-8 Madsen, B.M., and Raup, O.B., 1987, Petrography of a sulfur occurrence in the Castile Formation, southwest Texas: U.S. Geological Survey, OpenFile Report 87-479, 13 p. Magaritz, M., Anderson, R.Y., Holser, W.T., Saltzman, E.S., and Garber, J., 1983, Isotope shifts in the Late Permian of the Delaware Basin, Texas, precisely timed by varved sediments: Earth and Planetary Science Letters, v. 66, p. 111-124. http://dx.doi.org/10.1016/0012-821X(83)90130-9 and evaporite solution in Delaware basin, Texas and New Mexico: Geological Society of America Bulletin, v. 64, no. 5, p. 539-546. http://dx.doi.org/10.1130/00167606(1953)64[539:CFAESI]2.0.CO;2 Mamchur, G.P., 1969, Isotopic composition of carbon in calcite paragenetic with sulfur: Geochemistry International, v. 6, p. 660-670. lower Permian dolomites, Delaware Basin, Texas implications for basin evolution: Geological Society of America Bulletin, v. 70, no. 8, p. 943-952. McCarty, P.L., 1964, The methane fermentation, in Principles and Applications in Aquatic Microbiology, Heukelekian, H. and Dondero, N.C., eds., New York, John Wiley and Sons, Inc., p. 314-343. McKee, E.D., Oriel, S.S., et al., 1967, Paleotectonic maps of the Permian System: U.S. Geological Survey, Miscellaneous Geological Investigations Map I-450, 164 p. McNeal, R.P., 1965, Hydrodynamics of the Permian Basin, in Young, A. and Galley, J.E., eds., Fluids in Subsurface Environments, American Association of Petroleum Geologists, p. 309-326. McNeal, R.P., and Mooney, T.D., 1968, Relationships of oil composition and stratigraphy of Delaware, in Basins of the Southwest, v. 2, The Composition and Stratigraphic Relationships of Permian Basin Oils, Texas and New Mexico, American Association of Petroleum Geologists SW Section 10 th Annual Meeting, Wichita Falls, Texas, West Texas Geological Society, p. 68-75. Melim, L.A., and Scholle, P.A., 1995, The forereef facies of the Permian Capitan Formation: The role of sediment supply versus sea-level changes: Journal of Sedimentary Research, B65, p. 107-118. Lambert, S.J., 1983, Dissolution of evaporites in and around the Delaware Basin, southeastern New Mexico and west Texas: Sandia National Laboratories, Albuquerque, New Mexico, Report Sand-82-0461, 96 p. Lee, M. and Williams, D.D., 2000, Paleohydrology of the Delaware Basin, western Texas: Overpressure development, hydrocarbon migration, and ore genesis: American Association of Petroleum Geologists Bulletin, v. 84, no. 7, p. 961-974. Lein, A. Yu., 1974, Origin of native sulfur in Mishrah deposit (Iraq): International Geologic Review, v. 17, no. 8, p. 881-885. http://dx.doi.org/10.1080/00206817509471608 Leslie, A.B., Harwood, G.M., and Kendall, A.C., 1997, Geochemical variations within a laminated evaporite deposit: Evidence for brine composition during formation of the Permian Castile Formation, Texas and New Mexico, USA, Sedimentary Geology, v. 110, p. 223-235. http://dx.doi.org/10.1016/S0037-0738(96)00087-5 Lindsay, R.F., 1998, Meteoric recharge, displacement of oil columns and the development of residual oil intervals in the Permian basin, in DeMis, W.D., and Nelis, M.K., eds., The Search Continues into the 21st Century, Midland, West Texas Geological Society, p. 98-105. Lowenstein, T.K., 1988, Origin of depositional cycles in a Permian saline giantthe Salado (McNutt zone) evaporites of New Mexico and Texas: Geological Society of America Bulletin, v. 100, p. 592-608. http://dx.doi.org/10.1130/00167606(1988)100<0592:OODCIA>2.3.CO;2 Lucas, S.G., 2006, Ochoa Group, not series or stage, Upper Permian of west Texas and southeastern New Mexico, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 62-63. Lueth, V.W., Rye, R.O., and Peters, L., 2005, Sour gas hydrothermal jarosite: ancient to modern acidsulfate mineralization in the southern Rio Grande Rift: Chemical Geology, v. 215, p. 339-360. http://dx.doi.org/10.1016/j.chemgeo.2004.06.042 Luo, M., Baker, M.R., and LeMone, D.V., 1994, Distribution and generation of the overpressure system, eastern Delaware Basin, western Texas and southern New Mexico: American Association of Petroleum Geologists Bulletin, v. 78, p. 1386-1405. Machel, H.G., 1987, Some aspects of diagenetic sulphatehydrocarbon redox reactions, in Marshall, J.D., ed., Diagenesis of Sedimentary Sequences, Geological Society Special Publication 36, p. 15-28. Machel, H.G., 1989, Relationships between sulphate reduction and oxidation of organic compounds to carbonate diagenesis, hydrocarbon accumulations, salt domes, and metal sulphides deposits: Carbonates and Evaporites, v. 4, p. 137-151. http://dx.doi.org/10.1007/BF03175104
74 Nielson, P.D., and J.M. Sharp, Jr., 1990, Tectonic controls on the hydrogeology of the salt basin, Trans-Pecos Texas, in Hydrology of Trans-Pecos Texas, Kreitler, C.W. and Sharp, J.M. Jr., eds., Bureau of Economic Geology, Austin, Texas, Guidebook 25, p. 110-104. Nottingham, M.W., 1960, Recent Bell Canyon exploration in the north Delaware Basin, in Natural Gas of the Southwest, Transactions, v. 1, Southwestern Federation of Geological Societies, Abilene, Texas, Third Annual Meeting, p. 139-153. Oehler, D.Z., and Sternberg, B.K., 1984, Seepageinduced anomalies, false anomalies, and implications for electrical prospecting: American Association of Petroleum Geologists Bulletin, v. 68, no. 9, p. 1121-1145. OFlaherty, V., Mahony, T., OKennedy, R., and Colleran, E., 1998, Effect on pH on growth of methanogenic, syntrophic and sulphate-reducing bacteria: Process Biochemistry, v. 33, p. 555-567. http://dx.doi.org/10.1016/S0032-9592(98)00018-1 Olive, W.W., 1957, Solution-subsidence troughs, Castile Formation of Gypsum Plain, Texas and New Mexico: Geological Society of America Bulletin, v. 68, p. 351-358. http://dx.doi.org/10.1130/00167606(1957)68[351:STCFOG]2.0.CO;2 Orcutt, B.N., Boetius, A., Elvert, M., Samarkin, V., and Joye, S.B., 2005, Molecular biogeochemistry of sulfate reduction, methanogenesis and the anaerobic oxidation of methane at Gulf of Mexico cold seeps: Geochimica et Cosmochimica Acta, v. 69, no. 17, p. 4267-4281. http://dx.doi.org/10.1016/j.gca.2005.04.012 Orr, W.L., 1986, Kerogen/asphaltenes/sulfur relationships in sulfur-rich Monterey oils: Organic Geochemistry, v. 10, no. 1-3, p. 499-516. http://dx.doi.org/10.1016/0146-6380(86)90049-5 Palmer, A.N., 2006, Support for a sulfuric acid origin for caves in the Guadalupe Mountains, New Mexico, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 195-202. Palmer, A.N., 2009, Cave Geology: Dayton, Ohio, Cave Books, 454 p. Palmer, A.N., and Palmer, M.V., 2000, Hydrochemical interpretation of cave patterns in the Guadalupe Mountains, New Mexico: Journal of Cave and Karst Studies, v. 62, no. 2, p. 91-108. Palmer, A.N., Palmer, M.V., Queen, J.M., DuChene, H., and Cunningham, K.I., 2009, The Guadalupe Mountains, New MexicoTexas, in Palmer, A.N. and Palmer, M.V., eds., Caves and Karst of the USA: Huntsville, Alabama, National Speleological Society, Inc., p. 272-285. Parker, D.F., and McDowell, F.W., 1987, K-Ar geochronology of Oligocene volcanic rocks, Davis and Barrilla Mountains, Texas: Geological Society, London, Special Publications, v. 30, p. 415-431. Pawlewicz, M., Barker, C.E., and McDonald, S., 2005, Texas and southeast New Mexico: U.S. Geological Survey Open-File Report 2005-1171, 25 p. Melville, T.T., 2009, Using ground penetrating radar to characterize karst features in Eddy County, New Mexico and Culberson County, Texas [MSc. thesis], Stephen F. Austin State University, 126 p. Miller, L.J., 1992, Sulphur ore controls within the Salado and Castile formations of west Texas, in Wessel, G.R., and Wimberly, B.H., eds., Native Sulfur Developments in Geology and Exploration, Society Mining and Metallurgy, Exploration Proceedings, Phoenix, Arizona, Chapter 14, p. 165-192. Mller, P., Weise, S.M., Althaus, E., Bach, W., Behr, H.J., Borchardt, R., Bruer, K., Drescher, J., Erzinger, J., Faber, E., Hansen, B.T., Horn, E.E., Huenges, E., Kmpf, H., Kessels, W., Kirsten, T., Landwehr, D., Lodemann, M., Machon, L., Pekdeger, A., Pielow, H.-U., Reutel, C., Simon, K., Walther, J., Weinlich, the German Continental Deep Drilling Program (KTB): Journal of Geophysical Research, v. 102, no. B8, p. 18,233,254. http://dx.doi.org/10.1029/96JB02899 Montgomery, S.L., 1997, Permian Bone Spring Formation; sandstone play in the Delaware Basin; Part II, Basin: American Association of Petroleum Geologists Bulletin, v. 81, no. 9, p. 1423-1434. Moore, G.W., 1960a, Geology of Carlsbad Caverns, New Mexico, in P.E. Spangle ed., A Guidebook to Carlsbad Caverns National Park: Washington, D.C., National Speleological Society Guidebook No. 1, p. 10-17. Moore, G.W., 1960b, Origin and chemical composition of evaporite deposits [Ph.D. Dissertation]: Yale University, 174 p. Motts, W.S., 1968, The control of ground-water occurrence by lithofacies in the Guadalupian reef complex near Carlsbad, New Mexico: Geological Society of America Bulletin, v. 79, p. 283-298. http://dx.doi.org/10.1130/00167606(1968)79[283:TCOGOB]2.0.CO;2 Mruk, D.H., and Bebout, D.G., 1993, Slope, in Bebout, D.G., and Kerans, C., eds., Guide to the Permian Reef Trail, McKittrick Canyon, Guadalupe Mountains National Park, West Texas, Guidebook 26: Austin, Bureau of Economic Geology, The University of Texas, p. 14-22. Nance, R., and Stafford, K.W., 2009, Karst development in the Castile Formation of Eddy County, New Mexico, and Culberson County, Texas; a study of multiple models for regional speleogenesis: Proceedings of the 15 th International Congress of Speleology, v. 15, no. 2, p. 923-929. Newell, N.D., Rigby, J.K., Fischer, A.G., Whiteman, A.J., Hickox, J.E., and Bradley, J.S., 1953, The Permian Reef Complex of the Guadalupes Region, Texas and New Mexico: San Francisco, W.H. Freeman and Co., 236 p. Niec, M., 1992, Native sulfur deposits in Poland, in Wessel, G.R. and Wimberly, B.H., eds., Native SulfurDevelopments in Geology and Exploration, Phoenix, AZ, American Institute of Mining, Metallurgical and Petroleum Engineers, p. 23-50.
75 Queen, J.M., 2009, Geologic setting, structure, tectonic history, and paleokarst as factors in speleogenesis in the Guadalupe Mountains, New Mexico and Texas, USA.: Proceedings of the International Congress of Speleology, v. 15, no. 2, p. 952-957. Queen, J.M., Palmer, A.N., Palmer, M.V., 1977, Speleogenesis in the Guadalupe Mountains, New Mexicogypsum replacement of carbonate by brine mixing, in 7 th International Congress Kingdom, p. 332-336. Quinlan, J.F., and Smith, A.R., 1968, Drip pockets in gypsum (abs.): National Speleological Society Bulletin, v. 30, no. 2, p. 35-36. Reeburgh, W.S., 1989, Coupling of the carbon and sulphur cycles through anaerobic methane oxidation, in Brimblecombe, P., and Lein, A.Y., eds., Evolution of the Global Biochemical Sulphur Cycle, New York, Wiley, p. 149-159. Reis, M.A.M., Lemos, P.C., Almeida, J.S., Carrondo, M.J.T., 1991, Evidence for the intrinsic toxicity of H 2 S to sulphate-reducing bacteria: Applied Microbiology and Biotechnology, v. 36, p. 145-147. http://dx.doi.org/10.1007/BF00164716 Reis, M.A.M., Almeida, J.S., Lemos, P.C. and Carrondo, of sulfate reducing bacteria: Biotechnology and Bioengineering, v. 40, p. 593-600. http://dx.doi.org/10.1002/bit.260400506 Richardson, G.B., 1905, Native sulphur in El Paso County, Texas: U.S. Geol. Survey, Bull. 260, p. 589-592. Rooney, M.A., 1996, Carbon isotope ratios of light hydrocarbons as indicators of thermochemical sulfate reduction, in Organic Geochemistry, Developments and Applications to Energy, Climate, Environment and Human History, proceedings of the 17 th International Meeting on Organic Geochemistry, Spain, Donostia-San Sebastian, p. 523-525. Ruckmick, J.C., Wimberly, B.H., and Edwards, A.F., deposits: Economic Geology, v. 74, p. 469-474. http://dx.doi.org/10.2113/gsecongeo.74.2.469 Salisbury B.K., 1992, Geophysical and geochemical surveys in Delaware Basin sulfur exploration, in Wessel, G.R. and Wimberly, B.H., eds., Native SulfurDevelopments in Geology and Exploration, AZ, Phoenix, American Institute of Mining, Metallurgical and Petroleum Engineers, p. 81-90. Saunders, J.A., and Swann, C.T., 1994, Mineralogy and geochemistry of a cap-rock Zn-Pb-Sr-Ba occurrence at the Hazlehurst salt dome, Mississippi: Economic Geology, v. 89, p. 381-390. http://dx.doi.org/10.2113/gsecongeo.89.2.381 Sarg, J.F., 1981, Petrology of the carbonate-evaporite facies transition of the Seven Rivers Formation (Guadalupian, Permian), southeast New Mexico: Journal of Sedimentary Petrology, v. 51, p. 73-95. Sarkar, A., Nunn, J.A., and Hanor, J.S., 1995, Free thermohaline convection beneath allochthonous salt sheets: An agent for salt dissolution and Research, v. 100, p. 18,085-18,092. http://dx.doi.org/10.1029/95JB01857 Peckmann, J., Thiel, V., Michaelis, W., Clari, P., Gaillard, C., Martire, L., and Reitner, J., 1999, Cold seep deposits of Beauvoisin (Oxfordian; southeastern France) and Marmorito (Miocene; northern Italy); microbially induced authigenic carbonates: International Journal of Earth Sciences, v. 88, no. 1, p. 60-75. http://dx.doi.org/10.1007/s005310050246 Pfennig, N., Widdel, F., and Trper, H.G., 1981, The dissimilatory sulfate-reducing bacteria, in The Procaryotes, 1 st edition, Starr, M.P., Stolp, H., Tper, H.G., eds., Berlin, Springer-Verlag, p. 926-940. Polyak, V.J., 1998, Clays and associated minerals in caves of the Guadalupe Mountains, New Mexico [Ph.D. dissertation]: Texas Tech University, 190 p. Polyak, V.J., and Gven, N., 1996, Mineralization of alunite, natroalunite, and hydrated halloysite in Carlsbad Caverns and Lechuguilla Cave, New Mexico: Clays and Clay Minerals, v. 44, p. 843-850. http://dx.doi.org/10.1346/CCMN.1996.0440616 Polyak, V.J., McIntosh, W.C., Gven, N., and Provencio, P., 1998, Age and origin of Carlsbad Caverns and related caves from 40 Ar/ 39 Ar of alunite: Science, v. 249, p. 1919-1922. http://dx.doi.org/10.1126/science.279.5358.1919 Polyak, V.J., McIntosh, W.C., Provencio, P.P., and Gven, N., 2006, Alunite and natroalunite tell a storyThe age and origin of Carlsbad Caverns, Lechuguilla Cave, and other sulfuric-acid type caves of the Guadalupe Mountains, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 203-209. Polyak, V.J., and Provencio, P.P., 2000, Summary of the timing of sulfuric-acid speleogenesis for Guadalupe caves based on ages of alunite: Journal of Cave and Karst Studies, v. 62, no. 2, p. 72-74. Polyak, V.J., and Provencio, P.P., 2001, By-product materials related to H 2 S-H 2 SO 4 of Carlsbad, Lechuguilla, and other caves of the Guadalupe Mountains, New Mexico: Journal of Cave and Karst Studies, v. 63, no. 1, p. 23-32. Porch, E.L., Jr., 1917, The Rustler Springs sulphur deposits: University of Texas Bulletin 1722, 71 p. Posey, H.H., 1986, Regional characteristics of strontium, carbon, and oxygen isotopes in salt dome cap rocks of the western Gulf Coast [Ph.D. dissertation]: University of North Carolina, 248 p. Posgate, J.R., 1979, The Sulfate-Reducing Bacteria: London, Cambridge University Press, 151 p. Queen, J.M., 1973, Large-scale replacement of carbonate by gypsum in some New Mexico caves: National Speleological Society Convention, Bloomington, Indiana, Abstracts, p. 12. Queen, J.M., 1994, Speleogenesis in the Guadalupes: The unsettled question of the role of mixing, phreatic in Sasowsky, I.D., and Palmer, M.V., eds., Breakthroughs in Karst Geomicrobiology and Redox Geochemistry: West Virginia, Charles Town, Karst Waters Institute, Special Publication 1, p. 64-65.
76 Stafford, K.W., Ulmer-Scholle, D., and RosalesLagarde, L., 2008b, Hypogene calcitization: Evaporite diagenesis in the western Delaware Basin: Carbonates and Evaporites, v. 23, no. 2, p. 89-103. http://dx.doi.org/10.1007/BF03176155 Stafford, K.W., Land, L., Klimchouk, A.B., and Gary, M.O., 2009, The Pecos River hypogene speleogenetic province: A basin-scale karst paradigm for eastern New Mexico and west Texas, USA., in Land, L., and Veni, G., eds., NCKRI Symposium 1, Advances in Hypogene Karst Studies, National Cave and Karst Research Institute, p. 121-135. Stafford, K.W., and Nance, R., 2009, Ascending water of the Delaware Basin southeastern New Mexico, and far West Texas, in Proceedings of the 15th International Congress of Speleology, v. 15, no. 2, p. 991-997. Stakes, D.S., Orange, D.I, Paduan, J.B., Salamy, K.A., and Maher, N., 1999, Cold-seeps and authigenic carbonate formation in Monterey Bay, California: Marine Geology, v. 159, p. 93-109. http://dx.doi.org/10.1016/S0025-3227(98)00200-X Stahl, W.J. and Carey, B.D., 1975, Source-rock West Texas: Chemical Geology, v. 16, p. 257-267. http://dx.doi.org/10.1016/0009-2541(75)90065-0 Staudt, W.J. and Schoonen, M.A.A., 1995, Sulfate incorporated into sedimentary carbonates, in Vairavamurthy, M.A. and Schoonen, M.A.A., eds., Geochemical transformations of sedimentary sulfur: American Chemical Society Symposium Series 612, p. 332-345. Stiller, M., Yechieli, Y., and Gavrieli, I., 2007, The rate of dissolution of halite in dilute Dead Sea brines: Geological Survey of Israel, Report GSI/01/2007, 19 p. Strous, M. and Jetten, M.S.M., 2004, Anaerobic oxidation of methane and ammonium: Annual Review of Microbiology, v. 58, p. 99-117. http://dx.doi.org/10.1146/annurev. micro.58.030603.123605 Suess, E., Torres, M.E., Bohrmann, G., Collier, R.W., Greinert, J., Linke P., Rehder, G., Trehu, A., Wallmann, K., Winckler, G., and Zuleger, E., 1999, Gas hydrate destabilization: enhanced dewatering, benthic material turnover and large methane plumes at the Casadia convergent margin: Earth and Planetary Science Letter, p. 170, 1-15. Taylor, R.E., 1938, Origin of the cap rock of Louisiana salt domes: Louisiana Geological Survey Bulletin v. 11, 191p. Thode, H.G., Wanless, R.K., and Wallouch, R., 1954, The origin of native sulfur deposits from isotope fractionation studies: Geochemica et Cosmochimica Acta, v. 5, p. 286-298. http://dx.doi.org/10.1016/0016-7037(54)90036-8 Tissot, B.P., and Welte, D.H., 1984, Petroleum Formation and Occurrence: New York, Springer-Verlag, 699 p. http://dx.doi.org/10.1007/978-3-642-96446-6 Scholle, P.A., Ulmer, D.S., and Melim, L.A., 1992, Latestage calcites in the Permian Capitan Formation and its equivalents, Delaware Basin margin, west Texas and New Mexico: evidence for replacement of precursor evaporites: Sedimentology, v. 39, p. 207-234. http://dx.doi.org/10.1111/j.1365-3091.1992.tb01035.x Seager, W.R., and Morgan, P., 1979, The Rio Grande rift in southern New Mexico, west Texas, and northern Chihuahua, in Riecker, R.E., ed., Rio Grande Rift, Tectonics and Magmatism, Washington D.C., American Geophysical Union, p. 87-106. http://dx.doi.org/10.1029/SP014p0087 Sherwood, J.E., Stagnitti, F., Kokkinn, M.J., and Williams, W.D., 1991, Dissolved oxygen concentrations in hypersaline waters: Limnology and Oceanography, v. 36, no. 2, p. 235-250. http://dx.doi.org/10.4319/lo.1991.36.2.0235 Skyring G.W., 1987, Sulfate reduction in coastal ecosystems: Geomicrobiology Journal, v. 5, no. 3/4, p. 295-374. http://dx.doi.org/10.1080/01490458709385974 Smith, A.R., 1978, Sulfur deposits in Ochoan rocks of southeast New Mexico, in New Mexico Bureau of Geology and Mineral Resources Circular 159, p. 71-77. Smith, A.R., 1980, Sulfur deposits of Ochoan rocks of the Gypsum Plain, southwestern New Mexico and west Texas, in Dickerson, P.W. and Hoffer, J.M., eds., Trans-Pecos Region, New Mexico Geological Society, Guidebook, 31 st Field Conference, p. 277-283. Snider, H.I., 1966, Stratigraphy and associated tectonics of the Upper Permian Castile-Salado-Rustler evaporite complex, Delaware Basin, west Texas and southeast New Mexico [Ph.D. Dissertation] University of New Mexico, 196 p. Sorokin, Yu.I., 1957, On the question of the ability of sulphate reducing bacteria to use methane for the Akademii Nauk SSSR, v. 115, p. 816-818. Spirakis, C.S., and Cunningham, K.I., 1992, Genesis of sulfur deposits in Lechuguilla Cave, Carlsbad National Park, New Mexico, in Wessel, G.R. and Wimberly, B.H., eds., Native SulfurDevelopments in Geology and Exploration, AZ, Phoenix, American Institute of Mining, Metallurgical and Petroleum Engineers, p. 139-145. Stafford, K.W., 2008a, Hypogene karst and sulfate diagenesis of the Delaware Basin: southeastern New Mexico and far west Texas [Ph.D. dissertation]: New Mexico Institute of Mining and Technology, 307 p. Stafford, K.W., 2008b, Cavernous porosity and associated sulfate diagenesis in the Castile Formation, Eddy County, New Mexico, and Culberson County, Texas: Bulletin of West Texas Geological Society, v. 47, no. 6, p. 14-17, p. 20-26. Stafford, K.W., Nance, R., Rosales-Lagarde, L., and Boston, P., 2008a, Epigene and hypogene gypsum karst manifestations of the Castile Formation: Eddy County, New Mexico and Culberson County, Texas, USA: International Journal of Speleology, v. 37, no. 2, p. 83-98. http://dx.doi.org/10.5038/1827-806X.37.2.1
77 Wilde, G.L., Rudine, S.F., and Lambert, L.L., 1999, Formal designation: Reef Trail Member, Bell Canyon the Guadalupian-Lopingian boundary, in Saller, A.H., Harris, P.M., Kirkland, B.L., and Mazzullo, S.J., eds., Geologic Framework of the Capitan Reef: Tulsa, OK, SEPM Special Publication 65, p. 63-83. http://dx.doi.org/10.2110/pec.99.65.0063 Williamson, C.R., 1977, Deep-sea channels of the Bell Canyon Formation (Guadalupian), Delaware Basin, Texas-New Mexico, in Hileman, M.A. and Mazzulo, S.J., eds., Upper Guadalupian Facies, Permian Reef Complex, Guadalupe Mountains, guidebook: Permian Basin SectionSEPM Publication 77-16, p. 409-431. Winograd, I.J., and Robertson, F.N., 1982, Deep oxygenated ground water: Anomaly or common occurrence? Science, v. 216, p. 1227-1230. http://dx.doi.org/10.1126/science.216.4551.1227 Worden R.H., and Smalley, P.C., 1996, H 2 S-producing reactions in deep carbonate gas reservoirs: Khuff Formation, Abu Dhabi: Chemical Geology, v. 133, p. 157-171. http://dx.doi.org/10.1016/S0009-2541(96)00074-5 Worden, R.H. and Smalley, P.C., 2004, Does methane react during thermochemical sulphate reduction? Proof from the Khuff Formation, Abu Dhabi in Wanty, R. B. and Seal II, R. R., eds., Water-Rock Interaction, London, Taylor and Francis Group, p. 1049-1053. Worden, R.H., Smalley, P.C., and Cross, M.M., 2000, on thermochemical sulfate reduction. Khuff Formation, Abu Dhabi: Journal of Sedimentary Research, v. 70, no. 5, p. 1210-1221. http://dx.doi.org/10.1306/110499701210 Worden, R.H., Smalley, P.C., and Oxtoby, N.H., 1995, Gas souring by thermochemical sulfate reduction at 140C: American Association of Petroleum Geologists Bulletin, v. 79, p. 854-863. Ziegenbalg, S.B., Birgel, D., Hoffmann-Sell, L., Pierre, C., Rouchy, J.M., Peckmann, J., 2012, Anaerobic oxidation of methane in restricted hypersaline Messinian environments revealed by 13 C-depleted molecular fossils: Chemical Geology, v. 292-293, p. 140-148. http://dx.doi.org/10.1016/j.chemgeo.2011.11.024 Zimmerman, J.B., and Thomas, E., 1969, Sulfur in West Texas: its Geology and Economics: Austin, TX, Bureau of Economic Geology, Geological Circular 69-2, 35 p. Zobell, C.E., 1958, Ecology of sulfate reducing bacteria: Producers Monthly, v. 22, p. 12-29. Tyrrell, W.W., Diemer, J.A., Bell, G.L. Jr., and R.J. Bichsel, 2006, Thickness variations in the Lamar Limestone and Reef Trail Members of the Bell Canyon Formation, northwestern Delaware Basin, New Mexico and west Texas, in Land, L., Lueth, V.W., Raatz, W., Boston, P., Love, D.L., eds., Caves and Karst of Southeastern New Mexico, New Mexico Geological Society Guidebook 57 th Field Conference, p. 67-70. Ulmer-Scholle, D.S., Scholle, P.A., and Brady, P.V., 1993, back-reef carbonates of the Delaware Basin, west Texas and New Mexico: Journal of Sedimentary Petrology, v. 63, no. 5, p. 955-965. Valentine, D.L., 2002, Biogeochemistry and microbial ecology of methane oxidation in anoxic environments: a review: Antonie van Leeuwenhoek, v. 81, p. 271-282. http://dx.doi.org/10.1023/A:1020587206351 Valentine, D.L., and Reeburgh, W.S., 2000, New perspectives on anaerobic methane oxidation: Environmental Microbiology, v. 2, no. 5, p. 477-484. http://dx.doi.org/10.1046/j.1462-2920.2000.00135.x Wake, L.V., Christopher, R.K., Rickard, P.A.D., Andersen, J.E., and Ralph, B.J., 1977, A thermodynamic assessment of possible substrates for sulphate-reducing bacteria: Australian Journal of Biological Sciences, v. 30, p. 155-172. Wallace, C.S.A. and Crawford, J.E., 1992, Geology of the Culberson ore body, in Wessel, G.R. and Wimberly, B.H., eds., Native SulfurDevelopments in Geology and Exploration, AZ, Phoenix, American Institute of Mining, Metallurgical and Petroleum Engineers, p. 91-105. Ward, R.F., Kendall, C.G.St.C., and Harris, P.M., 1986, Upper Permian (Guadalupian) facies and their association with hydrocarbons, Permian basin, west Texas and New Mexico: American Association of Petroleum Geologists Bulletin, v. 70, p. 239-262. Warren, J.K., 2006, Evaporites, Sediments, Resources, and Hydrocarbons: New York, Springer, 1036 p. http://dx.doi.org/10.1007/3-540-32344-9 Widdel, F., 1988, Microbiology and ecology of sulfateand sulfur-reducing bacteria, in Zehnder, A.J., ed., Biology of Anaerobic Microorganisms, New York, John Wiley & Sons, p. 469-586. Wiggins, W.D., Harris, P.M. and Burruss, R.C., 1993, Geochemistry of post-uplift calcite in the Permian Basin of Texas and New Mexico: Geological Society of America Bulletin, v. 105, p. 779-790. http://dx.doi.org/10.1130/00167606(1993)105<0779:GOPUCI>2.3.CO;2