New Mexico Bureau of Mines & Mineral Resources Bulletin 117 Geology of Carlsbad Cavern and other caves in the Guadalupe Mountains, New Mexico and Texas

The results of the sulfur-isotope study are crucial to understanding the process of speleogenesis which produced the large cave passages in the Guadalupe Mountains. The cave gypsum could not possibly have derived from the Castile anhydrite beds as suggested by the local pooling model of Bretz (1949) or the mixing model of Queen et al. (1977a). The average isotopic composition of the Castile anhydrite is +10.3 (Table 23); if the cave gypsum originated by non-biological precipitation from Castile brines, then the cave gypsum and Castile anhydrite should have almost identical osotopic compositions. The whole-rock analyses, that show the cave gypsum not to be related chemically to the evaporites of the Gypsum Plain (Table 26), provide further evidence against the speleogenesis models of Bretz and Queen.

Egemeier (1971, 1973, 1981) and Maslyn (1979) suggested that the Big Horn Basin, Wyoming, caves dissolved by sulfuric acid. Egemeier, and later Davis (1979a), extended the concept of sulfuric-acid dissolution and replacement-solution to the caves of the Guadalupe Mountains, whereby cave walls are replaced and then enlarged. These authors cited blindly terminating passages of large diameters, native sulfur, and carbonate-free gypsum in Big Horn Basin caves as features mimicking those in Guadalupe caves.

The Big Horn Basin caves do resemble Guadalupe caves superficially, but upon close inspection many important dissimilarities emerge.

(1) Guadalupe caves show little or no evidence of thermal activity such as is associated with the replacement-solution caves of the Big Horn Basin. They do not possess thermal springs or spring slots, features important to Egemeier's replacement-solution model of speleogenesis.

(2) Blind cave-passage terminations do occur in Big Horn Basin caves, but only on their upslope ends, near the input points of thermal springs. Guadalupe caves have no such upslope spacial correlation with respect to pits or fissures.

(3) The gypsum in Guadalupe caves does not form as mounds on the floor corresponding to sluffed-off ceiling crusts; where it has not been dissolved away, the gypsum exists as continuous or segmented floor blocks.

(4) Replacement-solution gypsum crusts in Big Horn Basin caves occur as millimeters or centimeters thick, friable masses overlying cave walls and ceilings. In most Guadalupe caves, thin ceiling or wall crusts are rare or absent.

(5) Gypsum-limestone pairs in Big Horn Basin caves show correspondence in cation ratios, whereas those in Guadalupe caves do not (Fig. 79).

Jagnow (1977, 1979) also suggested that Guadalupe caves dissolved by sulfuric acid, but proposed that pyrite was the source of the acid. According to Jagnow's model, pyrite in the Yates Formation weathered and oxidized to sulfuric acid which moved downdip along backreef bedding planes until it reached the Capitan Limestone, where it dissolved out the large cave passages. A number of objections can be made to Jagnow's pyrite model of speleogenesis.

(1) Not enough pyrite exists in the Yates overburden to explain the immensity of the caves. As pointed out by Davis (1980), large caves like Carlsbad Cavern are not located near pyritic masses in the Yates Formation. Morehouse (1968) found up to 16% pyrite and marcasite in limestone overlying the Dubuque, Iowa, caves, but those caves are of limited extent both vertically and horizontally. Young (1915) described caves associated with pyrite near Battle Mountain, Nevada, but those are also small and are located directly beneath pyrite seams in the rock.

(2) It is highly doubtful that the sulfuric acid derived from pyrite could have remained in an unreactive state while it moved downdip to the cave-forming zone. If bedding surfaces in backreef beds were avenues along which sulfuric acid entered the reef, then why did it not react with the limestone immediately to form caves in the backreef beds (or why did it not dissolve the limestone right at its point of weathering)? Vear and Curtis (1981) reported that for the weathering of pyritic shales in England, more than 99% of the sulfuric acid produced is immediately consumed in carbonate dissolution reactions or in clay-mineral transformations. Keller et al. (1966) measured a pH of 1 in a pyritic shale in Indiana and a pH of 3.5 only 1.2 m below the shale where sulfuric acid reacted with limestone. In the Guadalupe Mountains, immediate reaction of pyrite-derived sulfuric acid with limestone also appears to be the case. On the surface around limonite-after-pyrite crystals one often finds solution cups that are approximately two to three times the diameter of the crystals themselves. This association suggests that, upon weathering and oxidation, an immediate sulfuric-acid reaction etches out a "nest" cup in the limestone around each pyrite-limonite crystal.

(3) When pyrite-produced sulfuric acid reacts with limestone, it produces the SO₄^2- ion in solution. It is the sulfate ion which moves downdip and into the caves, not the sulfuric acid; the sulfate ion either remains in solution or it precipitates as sulfate speleothems in the air-filled part of the cave. If pyrite was the main supplier of dissolved sulfur to Guadalupe caves, then why are there so few sulfate speleothems in the caves (with the exception of Cottonwood Cave, which lies directly below the pyritic Yates sandstone)? As discussed in Part II of this report, sulfate speleothems have sulfur-isotope signatures that suggest they may have derived from pyrite.

(4) The σ³⁴S value of -2.5 for pyrite in the Yates Formation (Table 22) suggests that the cave gypsum (average σ³⁴S = -15.1) may not be genetically related to the pyrite.

(5) Native sulfur in Carlsbad Cavern is found on the undersides of bedrock and speleothems, which suggests a source of sulfur from below, not from above as proposed by Jagnow.

(6) As Davis (1980) pointed out, Guadalupe caves do not possess vertical shafts beneath points of sulfuric acid input. Enlarged fissures and shafts are located below, not above, large rooms. The only cave in the Guadalupe Mountains which perhaps conforms to Jagnow's sulfuric-acid model is the Queen of the Guadalupes, a 60 m deep vertical-shaft complex which underlies limestone containing a gossan mass of hydrated iron oxide.

(7) Jagnow's interpretation of the role of pyrite in speleogenesis may be backward. As discussed in the section on Mississippi Valley-type ore deposits, the pyrite may be the result of cave-forming processes rather than the cause of them.

Recently, DuChene (1986) proposed another source for the hydrogen sulfide that dissolved Guadalupe caves. DuChene agreed with a sulfuric-acid model related to oil and gas as proposed in this study, but he thought that the most likely source area for the H₂S was east of the Guadalupe Mountains and the most likely migration path was updip through the Capitan aquifer. DuChene speculated that the Guadalupe Mountains are hinged near the present location of the Pecos River and that west of this hinge line oil and gas are absent in the reef because they moved updip and out of the system, the H₂S having dissolved out Guadalupe caves during its migration.

DuChene (1986) raised some important objections to the model proposed in this study, in particular hydrologic problems of moving gas from basin to reef, and stratigraphic problems concerned with the non-continuous, lenticular sandstone bodies within the Bell Canyon Formation and with the fact that these sandstones may not intertongue significantly with the Capitan Limestone. It is DuChene's belief that these factors could limit, but not necessarily totally prevent, the migration of gas from the Delaware Basin into the Capitan Limestone. DuChene's model, however, also has its problems.

(1) If hydrocarbons once existed in the reef rock of the Guadalupe Mountains before uplift, as they do today in areas of the uplifted shelf rock (Ward et al., 1986, fig. 14), then why is there a negligible hydrocarbon content in Guadalupe caves and cave deposits (0.015 mg/g. Moran, 1955; 0.5 ppm, Table 15)?

(2) Why does there appear to be a correlation in age between the movement of the halite margin in the basin past Carlsbad Cavern and the time of the last speleogenesis events in the cave?

(3) DuChene's model implies that hydrocarbon migration and cave development occurred early in the uplift of the Guadalupe Mountains, and that all the formational oil and gas once contained in these reef rocks has since escaped out of the system. If this is true, then why do sulfur crystals overlie late-stage speleothems in the Christmas Tree and New Mexico Rooms?

(4) In Carlsbad Cavern sulfur is associated with the Bell Canyon(?) Formation and forereef facies, beds that dip directly toward the basin.