A sulfur isotopic study of Uinta Basin hydrocarbons

Sulfur isotope analyses of tar from oilimpregnated sandstones, crude oils and solid hydrocarbons from the Uinta Basin (Tertiary), Utah, show wide variations, but all are heavier than the troilite sulfur isotopic standard. The data range from +2.7 to 30.3 permil and sulfur contents are low, usually less than 1 percent. Crude oils and tars from rocks of Mesozoic and Paleozoic age show SS3 4 values within plus or minus about 7 permil of the meteorite standard and exhibit sulfur contents in excess of 1 percent. The positive <5S34 values reflect the progressive enrichment of S34 in waters and sediments of Tertiary Lake Uinta resulting from bacterial reduction of sulfate and loss of hydrogen sulfide that had been enriched in S32 Tars from the large oil-impregnated sandstone deposits at P. R. Spring and Sunnyside, along the south edge of the Uinta Basin, show sulfur contents less than 1 percent and 5S3 4 values in the range +88 to + 22 permil. A detailed outcrop and core-hole study of P. R. Spring samples showed that 5S3 4 values in tars generally increased upward stratigraphically and tended to be correlatable with lithology between drill holes up to 5 miles apart. Within single drill holes, abrupt differences in the 5S3 4 value of sulfur (±5 permil) in the tar were found in separate impregnated sands a few feet apart stratigraphically. These isotopic characteristics are evidently inherited from the time of impregnation. The variation of 5S3 4 of the tars is strong evidence against impregnation by influx of a single well mixed mass of petroleum. The isotopic data are best explained by assuming that impregnation occurred selectively, sand by sand, with hydrocarbons derived from laterally equivalent lacustrine sediments toward the center of the basin. The delta facies of the Douglas Creek Member of the Green River Formation at P. R. Spring may have acted as a paleoaquifer in draining waters of compaction from the basin. Hydrocarbons in solution moved long distances laterally, but did not mix vertically, thus preserving the same isotopic pattern in the oil-impregnated sands as is shown by the source sediments in the interior of the basin. On the north rim of the basin, the large oilimpregnated sandstone deposits at Raven Ridge and Asphalt Ridge showed no consistent pattern in 5S34 with stratigraphic position; individual impregnated sands, however, appear to have consistent values. The Raven Ridge results (+18 to +30 permil) indicate the hydrocarbons were derived from the Green River Formation at or above the level of the Parachute Creek Member. At Asphalt Ridge the results (+7 to +22 permil) indicate some of the hydrocarbons were derived from lower Green River or Wasatch lacustrine sediments. The Rim Rock deposit showed a rather limited spread in 5S3 4 (+4.7 to 6.9 permil; five analyses), indicating impregnation perhaps involved secondary migration of a Wasatch or lower Green River oil pool ruptured by faulting along the structurally complex north rim of the basin. Other deposits along the north rim of the Uinta Basin showed 5S3 4 values typical of Wasatch or lower Green River organic-rich shales. One exception is the Whiterocks deposit where tar (+21 permil) impregnates Jurassic Navajo Sandstone. This is clearly a case of upward migration of hydrocarbons generated in the middle to upper Green River Formation. The isotopic composition of sulfur in Lake Uinta changed markedly with time toward heavier ratios. Thus paraffin-rich and gilsonite-like oils with similar sulfur isotopic compositions may have formed in a common source environment. Baker (1967) proposed that oil accumulates by local precipitation of hydrocarbons from migrating waters expelled by compaction in a sedimentary basin. The pattern of paraffin-rich oils toward the interior of the basin and asphaltic type in the oil-impregnated sandstones along the basin rim is in accord with the known order of solubility. Isotopic data for a single paraffin-rich oil from Red Wash and more asphaltic oils from P. R. Spring and Asphalt Ridge support Baker's hypothesis, but other critical combinations of isotopic and chemical data are not yet available.