ORGANIC GEOCHEMICAL INDICATORS OF DYNAMIC FLUID-FLOW PROCESSES IN PETROLEUM BASINS

A variety of preliminary geophysical, geochemical, and geological obse rvations (Anderson et al., Geophysics, 12-17, April 1991a,b; Anderson, Oil and Gas J. 85-91, 26 April 1993), along with well production data (Schumacher, AAPG Ann. Conv. Abstr., April 1993), have led to the hyp othesis that fluid injection into Eugene Island Block 330 (EI-330) in the U.S. Louisiana Gulf Coast may be a continuing episodic process, wi th the most recent injections having occurred on short geologic time s cales (10,000 yr or less). Organic geochemical evidence presented here suggests these processes may occur on very short time scales (years). A process, called ''dynamic fluid injection'', was proposed to explai n these observations and is envisioned as episodic pressure build-up f ollowed by rapid escape of gas and oil through geopressure and injecti on into overlying reservoirs. In this contribution, gas and oil compos itional data and sidewall core vitrinite reflectance data from EI-330 reservoirs were examined to determine whether they are more consistent with conventional oil and gas alteration processes versus recent dyna mic fluid injection. The following EI hydrocarbon patterns are consist ent with subsurface oil and gas migration-fractionation having altered the EI-330 oils al sometime in the geologic past: (i) Ratios of n-hep tane/methylcyclohexane (F) plotted against those of toluene/n-heptane (B) for EI-330 oils lie in a region of either very high maturity or wi thin a ''migrated C7'' field typical of an evaporative fractionation e vent. High maturity was ruled cut as the cause of the high F values be cause ethane vs propane delta(13)C values were not consistent with hig h maturity. Thus, the high F values are proposed to represent the ''mi grated C7 hydrocarbon'' end of an evaporative fractionation event. In comparison, nearby South Marsh Island Block 128 (SMI-128) oils, locate d on the opposite side of a salt ridge, show higher gas maturities alo ng with a tight clustering of F vs B values consistent with little or no evaporative fractionation. (ii) Vitrinite reflectances are higher i n sidewall cores from a fault system thought to feed these reservoirs than in samples away from the fault. (iii) Biomarkers for EI-330 oils are consistent with a source from the same or very similar marine Lowe r Cretaceous or Jurassic source rocks of almost identical type and mat urity. However, significant fractionations in percent sulfur, oil dens ity (API gravity), and percentage of different sizes of molecules with in the same compound class occur between reservoirs and fault-blocks. Other observations suggest that oil migration-fractionation processes may be occurring on short time scale (possibly as little as years): (i ) Consistent overproduction of oil and gas in the history of productio n of EI-330 reservoirs extending back to 1972. (ii) Abnormally high co ncentrations of light C3 to C9 hydrocarbons in some EI-330 oils. The l ightest n-alkanes diffuse fastest and tend to be lost as reservoirs le ak over time. (iii) Whole oil chromatograms show sharp well-resolved l ight hydrocarbon n-alkane peaks superimposed over a ''humpane'' type b aseline typical of biodegraded oils in the shallowest GA and HE reserv oirs. The presence of the highly biodegradable light n-alkanes togethe r with the biodegraded oils is consistent with oil remigration into th e shallower reservoirs. Assuming that conditions are adequate to suppo rt aerobic bacterial biodegradation in these shallow reservoirs (i.e. adequate supplies of meteoric water containing oxygen and nutrients), then the presence of highly biodegradable light n-alkanes together wit h heavier biodegraded oil suggests very recent (years to 100s of years ) injection of the light n-alkanes into the reservoir. (iv) Comparison of whole oil and C7 hydrocarbon data for EI-330 reservoirs taken in 1 974, 1984 and 1988 suggest temporal changes in both light and heavy hy drocarbons for a number of wells and intervals. One consistent trend i n several intervals is the appearance of enhanced concentration of C15 +n-alkanes in 1984 but not in 1974, 1988, or in the most recent sampl e collected in December 1993. Duplicate analyses of a number of sample s collected 5 days apart appear to rule out sampling and analytical ar tifacts as the cause of these changes. These temporal changes, if conf irmed in future research, suggest that biodegradation and reinjection of n-alkanes into some shallower and cooler EI-330 intervals is an on- going process which must be occurring on very short time scales. An ex tensive new drilling and sampling program was recently carried out at EI-330 in which gases, fluids, and cores were collected. Analyses are in progress; results will be used to test the viability of the ''dynam ic fluid injection'' hypothesis vs other more conventional reservoir f ractionation-migration processes.