Douglas W. Kirkland

Permian Castile Varved Evaporite Sequence, West Texas and New Mexico

Laminations in the Upper Permian evaporite sequence in the Delaware Basin appear in the preevaporite phase of the uppermost Bell Canyon Formation as alternations of siltstone and organic layers. The laminations then change character and composition upward to organically laminated claystone, organically laminated calcite, the calcite-laminated anhydrite typical of the Castile Formation, and finally to the anhydrite-laminated halite of the Castile and Salado. Laminae are correlative for distances up to 113 km (70.2 mi) and probably throughout most of the basin. Each lamina is synchronous, and each couplet of two laminated components is interpreted as representing an annual layer of sedimentation—a varve. The thickness of each couplet in the 260,000-varve sequence (a total thickness of 447.2 m, 1467 ft) has been measured individually and recorded and provides the basis for subdividing and correlating major stratigraphic units within the basin. The uppermost 9.2 m (30.3 ft) of the Bell Canyon Formation contains about 50,850 varve couplets; the Basal Limestone Member of the Castile about 600; the lowermost anhydrite member of the Castile (Anhydrite I) contains 38,397; Halite I, 1,063; Anhydrite II, 14,414; Halite II, 1,758; Anhydrite III, 46,592; Halite III, 17,879; and Anhydrite IV, 54,187. The part of the Salado collected (126.6 m) contains 35,422 varve couplets. The Bell Canyon-Castile sequence in the cores studied is apparently continuous, with no recognizable unconformities. The dominant petrologic oscillation in the Castile and Salado, other than the laminations, is a change from thinner undisturbed anhydrite laminae to thicker anhydrite laminae that generally show a secondary or penecontem-poraneous nodular character, with about 1,000 to 3,000 units between major oscillations or nodular beds. These nodular zones are correlative throughout the area of study and underly halite when it is present. The halite layers alternate with anhydrite laminae, are generally recrystallized, and have an average thickness of about 3 cm. The halite beds were once west of their present occurrence in the basin but were dissolved, leaving beds of anhydrite breccia. The onset and cessation of halite deposition in the basin was nearly synchronous. The Anhydrite I and II Members thicken gradually across the basin from west to east, whereas the Halite I, II, and III Members are thickest in the eastern and northeastern part of the basin and thicken from southeast to northwest. This distribution and the synchroneity indicate a departure from the classical model of evaporite zonation.

Using Strontium Isotopes to Determine the Age and Origin of Gypsum and Anhydrite Beds

The variation of 87Sr/86Sr in seawater with time can be used to determine the age of calcium sulfate beds precipitated from brine derived from seawater. The determination of a marine origin can be established by the consistency of strontium isotope results from samples at different stratigraphic levels in a single gypsum/anhydrite bed collected at separated sites. The scatter of strontium isotope results in the gypsum/anhydrite samples examined here is interpreted to result from the contribution of meteoric strontium to the salina. Strontium isotope results from three evaporite settings, the Jurassic Todilto Formation from New Mexico, the Permian Blaine Formation from Blaine County, Oklahoma, and the Permian Salado Formation from New Mexico, are used to determine the age and origin of the parent brine for these gypsum/anhydrite beds. The calcium sulfate beds from each of these formations precipitated from salinas that originated with a marine flooding but show, at different times and localities, volume dominance by meteoric water. The marine/meteoric mixing of ancient salinas can be modeled by using strontium concentrations and isotope ratios estimated from modern analogs. A limited comparison of strontium and sulfur isotopes shows that sulfur isotopes in the settings studied are less sensitive to meteoric influx.

Dissolution of salt deposits by brine density flow

The origin of collapse structures and breccias that vertically penetrate or occur within impermeable evaporites has never really been understood. The density of the brine that develops as salt deposits are dissolved can generate continuous gravitational brine movement. If the source of the dissolving water is artesian, or continuous, a flow cycle is developed in which the salt itself supplies the density gradient that becomes the vehicle of its own dissolution. The Delaware Basin in western Texas and southeastern New Mexico provides a particularly good example of how brine density flow can produce dissolution chambers that collapse to form breccias. The potential for dissolution by brine flow is an inherent property of partly exhumed evaporites and may constitute a risk factor in the storage of radioactive waste in evaporite deposits.

Intrabasin Varve Correlation

Statistical correlation techniques have been applied to the stratigraphically correlated laminations in four sedimentary basins to demonstrate the degree of lateral variation of certain types of layers and to help determine their interrelationships. Laminae thicknesses in a varved sequence deposited in a small Pleistocene lake in Texas have a very high correlation coefficient (+0.90) over a distance of 275 m. Laminae in the Jurassic Todilto Formation of New Mexico were correlated stratigraphically over a distance of 5.3 km and would probably be traceable over a much greater distance except for the disturbing effects of currents on the bottom of the basin. Five contiguous Todilto sections over a distance of 3.7 km have correlation coefficients of +0.85 or greater. Couplets of bituminous calcite and anhydrite in the Permian Castile Formation of Texas correlate stratigraphically over a distance of 14.2 km with a coefficient as high as +0.99. In a section of 74 couplets 27.4 m above the base of the Castile, anhydrite laminae thicknesses have a correlation coefficient of +0.97; calcite laminae, +0.72; and organic-rich bands in the calcium carbonate laminae, +0.73. There is no relationship between the thickness of associated calcite and anhydrite laminae. Richter-Bernburg measured and subjectively correlated anhydrite laminae thicknesses in cores of the Permian Zechstein Formation in Germany. Correlation coefficients were obtained using his data which range from +0.85 to +0.32 over distances up to 290 km. Zechstein statistical correlation deteriorates with distance. If Castile correlation changes at the same rate, it should be possible to correlate laminae stratigraphically over the maximum length of the basin.

Origin, varves, and cycles of Jurassic Todilto Formation, New Mexico

The upper Jurassic Todilto formation is a laminated sequence of limestone and gypsum of nonmarine or saline-paralic origin in northwestern New Mexico. The 10-13-year sunspot cycle and 60-, 85-, 170-, and 180-year cycles were recorded in the varved sequence. The presence of the sunspot cycle indicates that the laminations were deposited annually. The limestone member was deposited in about 14,000 years and the gypsum member in about 6,000 years. Deposition began with a two-fold clastic-organic varve and progressed through a sequence of increasing and decreasing varve complexity accompanied by a decrease and increase in clastic content. The end result is a transitional lithologic change from sandstone to limestone to gypsum to shale brought about by changes in individual varve laminae. The progression of changes is related to a cyclic change from clastic to evaporite and back to clastic deposition described here as the varved clastic-organic-evaporite (c-o-e) cycle. The varved cycle is nearly complete in the Todilto sequence and is present in the Green River basin of Colorado and Wyoming, the Delaware basin of Texas and New Mexico, the Paradox basin of Utah and the Four Corners region, and in most other evaporite deposits.

Microfolding in the Castile and Todilto Evaporites, Texas and New Mexico

The Castile laminated calcite-anhydrite of Eddy County, New Mexico, and Culberson County, Texas, and the Todilto laminated calcite of Valencia County, New Mexico, contain small-scale fold chains characterized by varying degrees of buckle shortening in successive laminae. These microfolds are associated with larger folds and have apparently resulted from compression generated by tectonism. In the Castile formation, anhydrite laminae, with an average thickness of about 1.1 mm, alternate with thinner calcite laminae (usually about 0.4 mm) which have thickened and thinned in the deformed zones to accommodate folding of the anhydrite laminae. Castile micro-folds generally have a wavelength of less than 1 cm and occur in hinge areas of “megafolds” with wave-lengths of about 0.5 to 2.5 m. The microfolds are cylindrical and have uniform axial trends. Particular zones in the Castile sequence were predisposed to microfolding, a tendency which apparently existed for tens of kilometers. In these zones, relatively thin anhydrite laminae commonly have been deformed by buckling, whereas thicker anhydrite laminae commonly have reacted to stress by increasing in thickness normal to bedding; laminae of intermediate thickness have been deformed by a combination of these processes. Thicker anhydrite laminae in some places have acted as struts which have prevented or reduced the degree of buckling of adjacent thinner laminae. The over-all style of microfolding has been determined usually by variability in the thickness of successive calcite and anhydrite laminae. In the Todilto limestone member, white laminae of finely crystalline calcite with thicknesses up to about 0.5 mm are intermittently present. These laminae commonly are micro-folded. Usually only a single lamina is involved in the microfolding, and adjacent dark colored, coarser crystalline limestone layers are unfolded. The folds, in profile, are of irregular pattern and usually have wavelengths between about 0.5 to 1.0 cm. In plan, the folds are composed of a complex array of miniature highs and depressions; no trend is apparent. The Todilto limestone at the locality studied apparently reacted to compressive stress by increasing in thickness normal to bedding as well as by buckling. The white, lithologically distinct laminae were subjected to buckle shortening, took up slack, and were microfolded in open spaces generated along bedding surfaces as layer-parallel shortening of adjacent dark limestone beds took place. Many microfolds in other laminated and thin-bedded sequences probably had origins analogous to those of the Castile and Todilto formations.

Method of Predicting Reservoir Quality for Feldspathic Sandstones of Southern California: ABSTRACT

The diagenetic alteration of a sandstone results from the combined effects of many factors. In order to ascertain the role played by a particular diagenetic factor, the remaining diagenetic factors must be held constant. For the sandstones of the basins of southern California, we have found that the effects of almost all of the principal diagenetic factors are essentially uniform, the notable exception being thermal history. Because of this, we have been able to evaluate the diagenetic imprint of temperature upon the sandstones. Measured reservoir property data taken on core samples of reservoir sandstones from 16 fields in the Los Angeles, Ventura, and San Joaquin basins were used to determine the average rate of porosity and permeability loss with depth for each field. straight line appears to be the proper representation for the porosity vs. depth profiles for the interval of interest. The slope of this line is defined here as the porosity gradient. Porosity gradients for the fields investigated range from 1.1% to 5.8%/1,000 ft. A direct relationship exists between the porosity gradient and the present geothermal gradient for the 16 fields which have been examined. As geothermal gradient increases, porosity gradient increases. The correlation coefficient between these variables is +0.916 for geothermal gradients between 1.6°F and 2.2°F/100 ft. A similar relationship also exists between the rate of permeability loss with depth and the geothermal gradient, but the average deviation from the mean permeability value is so great the relationship has little practical significance. End_of_Article - Last_Page 143------------