Gas reservoirs that water out under moderate to strong water drives are normally abandoned when the expenses associated with salt-water disposal make continued operations uneconomical. Under favorable conditions, however, watered-out reservoirs can continue to produce substantial quantities of gas at competitive prices if operators are prepared to dispose of large volumes of water. Enhanced gas recovery (EGR) techniques can extend production from many reservoirs that are now watering out and will soon be abandoned if conventional practices are followed. The EGR method involves co-production of gas and water. If large volumes of water are produced, the reduced reservoir pressure causes expansion of free gas formerly trapped in the water-invaded zone during the primary production stage. Some of this free gas becomes mobilized and producible. Also, pressure reduction at the surface releases additional but minor amounts of gas dissolved in the formation water. The Port Arthur field, Jefferson County, Texas, contains several watered-out gas reservoirs, thick sandstone aquifers, and gas stringers that collectively make the field ideal for testing the co-production technology. The objective sandstones occur in the lower Hackberry (Oligocene) interval at depths of 10,850 to 11,700 ft. The field covers about 1,900 acres (3 mi2) and originally produced gas condensate from an anticlinal closure on the downthrown side of a major fault that separates the Port Arthur field from the Port Acres field. Some of the lower Hackberry sandstones, interpreted as submarine channel and fan deposits, are laterally extensive and have excellent physical characteristics for producing gas and water. Net-sandstone thickness averages 350 ft. Core data and well log analyses show that porosities average 30 percent, permeabilities average 60 md, salinities average 67,900 ppm sodium chloride, and methane solubility averages 25.7 scf/bbl. Abundant shallow Miocene sands in the area could be used for salt-water disposal. Available well logs were analyzed to determine porosity and other formation characteristics of reservoirs being studied. Water saturations were also calculated from logs to help locate gas-water contacts. Seismic data were acquired and reprocessed to (1) provide structural information to supplement geological interpretations, (2) locate boundaries of aquifers and gas reservoirs, and (3) evaluate seismic response to low saturations of free gas dispersed in water-invaded zones of watered-out gas reservoirs. Attaining these objectives was severely limited by the poor signal-to-noise ratio in the seismic data. The quality of data was adequate for structural interpretation but was not suitable for reservoir delineation or detection of gas zones. Modeling studies showed what kind of seismic response should be expected from known subsurface geology and suggested that better reservoir delineation could be achieved with increased bandwidth, improved signal-to-noise ratio, and a better knowledge of reservoir acoustic impedances; but it remains unclear whether dispersed free gas in a watered-out reservoir can be detected with seismic data. The Hackberry C reservoir was selected for numerical modeling because of its high productivity, high average abandonment pressure gradient (0.67 psi/ft), and excellent physical properties. The original gas in place (OGIP) was estimated to be 56.2 Bcf, 35 percent of which was recovered during primary production. A three-dimensional, two-phase model was used to perform history matches and to predict the amount of fluid that might be produced under natural flow conditions if a new well were drilled. During a projected 8-year production period, the predicted reservoir bottom-hole flowing pressure would decline from 6,600 to 4,200 psi. Predicted production was 5.1 Bcf of gas, 51,000 bbl of condensate, and about 9 million bbl of water. An additional 10 percent of the OGIP was predicted as recoverable if the EGR co-production method were used. These results of the simulation study predicted the reservoir performance if a new well were drilled to a depth of 11,650 ft and located on a site near the Meredith No. 2 Doornbos (well 14). Because the reservoir is still geopressured, artificial lift methods would not be required to produce from this test well. Cash-flow calculations show that the break-even gas price is $2.40/Mcf for a 15-percent rate of return after payment of Federal income tax. The net present worth of the investment is about $968,000 for a gas price of $3.00/Mcf, and it increases substantially at higher gas prices. The economic outlook for the prospect would be even better if production from the C reservoir were commingled with production from other reservoirs in the field.
Gregory, A. R., Lin, Z. S., Reed, R. S., Morton, R. A., and Ewing, T. E., 1984, Enhanced Gas Recovery from Watered-Out Reservoirs--Port Arthur Field, Jefferson County, Texas: The University of Texas at Austin, Bureau of Economic Geology, Report of Investigations No. 142, 58 p.