How many exploration teams make four huge discoveries in four years? This is the story of one such team from ARCO Indonesia Inc.’s new ventures department.
As described in last month’s installment of Historical Highlights,” Resilient: Part I,” the Wiriagar Deep-1 exploration wildcat successfully tested natural gas in 1994, flowing a combined 30 million cubic feet per day from three stratigraphic intervals on ARII’s onshore Wiriagar production sharing contract. The WD-1 pressure/temperature interpretation that John Marcou and I made suggested a more-than-2,000-foot Jurassic gas column, potentially filling the northwest-southeast plunging Wiriagar Deep anticlinorium to the spill point with Occidental Petroleum’s deeper Roabiba and/or Ofaweri structures on their Berau PSC. Larry Casarta, senior explorationist and Sonny Sampurno, staff geologist, led by new ventures manager Suherman Tisnawidjaja, were fully supportive of the revised play concept from a deep oil play to a deep gas play, but what about the home office?
Surprising Results
ARCO hoped for a Bintuni Basin oil discovery that would be easier to commercialize than gas. Pursuing the oil play, I searched along the banks of the Wiriagar River by canoe for oil seeps in early 1994 but found none.
From September to December 1994, I was wellsite geo-supervisor on ARII’s mountaintop East Onin-1 exploration well, targeting Mesozoic deep oil plays on the onshore Bomberai PSC. Utilizing a helicopter-transportable mineral exploration rig, specially modified for both continuous coring and conventual drilling, the wildcat was situated at 3,600-feet elevation on the axial crest of a structural and topographic high delineating the southwest boundary of the Bintuni Basin. The well had numerous drilling problems and some gas shows, but no oil.
In 1995 I searched the Bintuni Bay seafloor for oil seeps or gas chimneys, utilizing Oceaneering Inc.’s manned submersible. But visibility was bad and results inconclusive.
WD-1’s deep gas play concept would require an offshore appraisal well on Oxy’s Berau permit, at the approximate midpoint of the proposed 2,000 to 2,200-foot gas column. If the gas play concept was correct, Oxy’s permit also held most of the Jurassic gas reserves.
Palynologists Paul Waton and Steve Noon had completed the laborious task of analyzing the WD-1 drill cuttings in late 1994. The results stunned us. WD-1’s thinly bedded Jurassic sandstone, which tested 3.8 million standard cubic feet per day wasn’t Callovian-Bathonian-Bajocian like Oxy’s offshore gas discovery wells (Roabiba, Ofaweri, and Wos). The palynomorph assemblage indicated WD-1’s gas-charged reservoir was potentially not the same reservoir as Oxy’s; it was Aalenian, six million years older.
Because WD-1’s palynology was done on drill cuttings, there could be reasonable doubt regarding biostratigraphic dating. After discussions with the palynologists, Larry and I felt the dating of cuttings might be in error. The palynologists used paleo-pollen/spore assemblages for age-dating with a greater margin for error than index-taxa age-dating. Perhaps the samples had been contaminated by poor sample-board collecting; maybe drill cuttings to surface had been lagged wrong, resulting in incorrect depths; or possibly the Jurassic reservoir was barren of spores/pollen with palynomorph assemblages actually from underlying shale cuttings (the few Aalenian sidewall cores were barren). Therefore, the team downplayed the palynological discrepancies when ARCO’s top management came to Jakarta for presentations.
Gas, Not Oil
ARCO’s home office was intrigued by the new gas play concept. Their recollections were described by Jamie Robertson in the Tangguh chapter of the 2005 AAPG Special Publication “Discoverers of the 20th Century: Perfecting the Search,” an excerpt from which is:
“... thoughtful analysis of the well’s pressure data by Casarta, Marcou, and Salo indicated that the gas zones were significantly overpressured and … a gas column height in excess of 2000 feet (610 meters) was a reasonable interpretation ... The pressure data indicated that the discovery could be large enough to anchor an LNG (liquefied natural gas) project even if there was no downdip oil leg below the gas. The WD-1 well data, including the pressure analysis, were used to generate revised prospect maps, and the project team estimated that there was a one-in-three chance that the discovery was commercial, now defined as gas reserves in excess of 6 TCF … If Casarta, Marcou and Salo were right about the size of the accumulation, much of the field lay to the south on the offshore Berau PSC still held by Occidental …”
With ARCO’s approval, ARII negotiated with Oxy in February 1995, and thanks to tight-holing the gas data interpretations by ARII president Leon Codron, Suherman and Sampurno, ARII farmed-in to the Berau PSC as operator with a 60-percent working interest in exchange for paying 100 percent of the drilling cost to satisfy Oxy’s fourth and final well commitment on the permit.
The Wiriagar Deep-2 appraisal well would be drilled 13 kilometers southeast of WD-1, downdip along the plunging axial crest at the estimated midpoint of the proposed more-than-2,000-foot gas column. The well would appraise the Jurassic reservoir for gas content, reservoir pressure, porosity and permeability, net pay, and prove-up connectivity with WD-1. The well was spudded on Sept. 2, 1995, and I landed on the deck of the drillship shortly thereafter.
While drilling a 17 1/2 inch hole through the New Guinea Limestone Group members, circulation losses occurred until eventually it was incurable. Learning from WD-1 experiences, the crew was prepared for this eventuality with a rotating head installed for blind drilling the remainder of the NGLG. I used CCL/gamma tool logging thorough drill pipe to determine the base of the NGLG at depth 5000 ft, where the new casing shoe was set. After drilling the 12 1/2 inch hole through Eocene shales, the Palaeocene was encountered with the sandstones water-wet (no gas shows). Following wireline logging at 6,710 feet, 9 5/8 inch casing was set and drilling the 8 1/2 inch hole began.
After drilling 650 feet of additional Palaeocene shale, WD-2 took a 2021-unit gas kick. Mud weight was increased to 10.8 pounds per gallon, and three 30-foot cores were cut in separate gas-bearing Paleocene sandstones each associated with high gas. But at 7,624 feet, a pronounced drilling break occurred indicating unusually porous lithology. I chose a 60-foot core-barrel to be run in the hole this time, and with a 1,440-unit background gas reading, cut WD-2’s first sandstone core. Upon recovery, I was surprised to see graded bedding of fine to medium to coarse sandstone, followed by conglomerate. Another 60-foot core-barrel was run in hole, with additional graded-bedding intervals containing shale rip-up clasts recovered after coring, so I asked for yet another 60-foot core-barrel.
Following the recovery at surface, I identified massively bedded shale at the bottom of the final core. But I hadn’t seen the basal Cretaceous green glauconitic sandstone marker bed, so I suspected this was a Palaeocene sequence unrelated to WD-1’s Palaeocene reservoirs. At WD-2 we’d cored a stacked turbidite-channel complex, characterized by classic Bouma-sequence graded bedding, with a 160-foot thick, overpressured, gas-charged reservoir.
Conventional drilling continued until we encountered the glauconitic sandstone at 8,363 feet, almost 1,000 feet deeper than prognosed. This time, approval was granted for Jurassic cores. Late Jurassic shale was drilled conventionally until we circulated out a 10-foot drill-break at 8,547 feet, and the coring assembly was run in hole. We cut four cores back-to-back, recovering the entire 120-foot Jurassic reservoir, as well as capturing underlying Early Jurassic shale source rock, the Jurassic/Permian unconformity, and the upper 10 feet of Late Permian sandstone and coaly source rock.
Deep Gas Discovery
WD-2 reached total depth at 9,755 feet on Oct. 26, 1995, with wireline logs run thereafter. Marcou once again designed the testing program and was supervisor witnessing the subsequent drill stem tests, pressure tests and flow rates conducted on the primary Palaeocene and Jurassic gas zones, with testing completed on December 6.
The well tested at a combined rate of 110 million standard cubic feet per day, with the follows test results:
- The Late Permian was tight with minor gassy water.
- The Jurassic Callovian sandstone flowed 30.4 mmscfg/d.
- The Jurassic Bajocian-Bathonian sandstone flowed 36.0 mmscfg/d.
- The Palaeocene turbidite sequence flowed 29.9 mmscfg/d.
- The shallower Palaeocene sandstone flowed 13.8 mmscfg/d.
Laboratory geochemical analyses of the cored reservoirs established conclusively that oil staining and spotty fluorescence in Jurassic cores described in WD-1 and WD-2 was due to residual oil in the pore system. Live oil had previously migrated through the Jurassic sandstones but been flushed out and displaced by late maturity phase lean gas. There was no oil leg present now. The WD-2 appraisal well vindicated the revision of the deep oil play to a deep gas play.
WD-2 pressure data confirmed the existence of a giant Jurassic field with an estimated vertical gas column close to 2,000 feet, but the appraisal well also resulted in the discovery of the new Palaeocene submarine channel/fan complex containing additional significant reserves. Palaeocene reservoirs tested at WD-1 were a deep marine, relatively thin-bedded discontinuous paleofacies. This was a significant unexpected discovery.
Palynological analyses of WD-2 cores confirmed the Jurassic interval contained gas-charged Aalenian, Bajocian, Bathonian reservoirs and minor Callovian sandstone. This tied the WD-1 discovery well reservoir to the Oxy discoveries.
This combined dataset confirmed that an angular unconformity represented erosion of the upper Middle Jurassic at the northern culminating crest with only vestiges of the Aalenian reservoir present over onshore Wiriagar PSC. Data confirmed the Jurassic reservoir discovered at WD-1 was indeed the mid-Jurassic Aalenian and was not the same reservoir drilled on Oxy’s wells. Faults and fractures likely provided reservoir communication through the intervening Jurassic shale separating the Aalenian from Bajocian-Bathonian-Callovian sequence. All the mid-Jurassic intervals were in pressure communication and could be treated as one massively bedded sandstone reservoir for production purposes.
More Drilling
The Jakarta team was ecstatic with the WD-2 results, and the new ventures and drilling teams received ARCO exploration awards for WD-1 and WD-2. By late 1995, plans were made to drill six additional wells. The Suhanah drilling barge would be used to drill northern delineation wells in the Wiriagar Swamp, onshore Wiriagar PSC, while the drillship would be used to drill offshore Berau PSC for flank and downdip delineation/appraisal of the gas field. I would sit the Wiriagar Deep-5, which would test the maximum downdip gas/water contact, and obtain critical aquifer pressures and water samples to complete an appraisal-certification process of the giant Wiriagar Deep accumulation.
Six appraisal/delineation wells were to be fast-track drilled in late 1995 through 1996, but once again obstacles were encountered.
Casarta called Sampurno and me into his office. “The commercial people say the only way to develop gas in Papua, 3,000 kilometers from Jakarta, is by LNG export. They estimated we need more than 6 trillion cubic feet of certified gas reserves for commercial LNG development. I’ve done probabilistic reserve estimates, and I keep coming up with 3P reserves between 6 and 7 TCF,” he said.
I replied, “That’s good, right? We’re in the money for Wiriagar for an LNG development project, right?”
Casarta showed us the probabilistic distribution: “Look, the proved and probable plus possible reserves are not quite 7 TCF. Nobody is going to approve an FID (final investment decision) for remote LNG development on 3P reserves, and no bank will fund the development based on possible reserves. We need 2P, not 3P reserves, of 6 to 7 TCF. Wiriagar Deep is a good LNG anchor, but you guys gotta come up with more Bintuni gas reserves,” he said.
It wasn’t the first obstacle the team faced, but this time around, ARII would have to spend a lot of money drilling eight Wiriagar wells. If the deep gas play didn’t have enough reserves to justify LNG development, then ARII had wasted a lot of capital on a stranded asset. Our “assets” were on the line if we didn’t find more gas.
Serendipity Again
We made structural cross-sections, gross and net interval isopach maps, and regional well correlations for Bintuni Basin regional exploration efforts in a quest to find more reserves. I rechecked cartons sent to ARII from Oxy’s Jakarta office. One box had vintage well-data from Sebyar-1 drilled by Total in 1986, including a tattered, moldy thermal mudlog field-print. The bottom of the mudlog noted gas chromatography results from a “weak-blow” drill stem test sampled from a Permian open-hole test at total depth. The wireline tools hadn’t been able to get to bottom to log this interval, so there were no wireline logs for total depth. Apart from the mudlog there was no other data on this test or interval.
The drill stem test wasn’t mentioned in the Sebyar final well report nor on the completion logs. Total had even reported Sebyar-1 as a dry hole to Pertamina, and every PSC operator in West Papua (then known as Irian Jaya) displayed the dry hole symbol for Sebyar-1.
It pays to look at the original raw data and not rely on other people’s interpretations or conclusions. Total hadn’t encountered any Jurassic sandstone in their 1986 exploration wildcat, but they had flowed some gas in a weak blow from the Late Permian tight sandstones. The Sebyar-1 gas chromatography results from the Permian test were nearly identical to those from our Wiriagar Deep Jurassic drill stem test.
ARCO and Core Lab’s source rock evaluations, together with well-test condensate geochemical analyses, suggested that terrestrial coals and carbonaceous shales of the Late Permian were probably basin wide and mostly gas prone. Early Jurassic marine coaly/carbonaceous shales were unevenly distributed and mixed oil-gas prone. Comingling of Jurassic-Permian hydrocarbons probably occurred during primary migration. Furthermore, a re-analysis of Wiriagar oilfield showed it wasn’t oil being produced from Pertamina’s shallow Kais Limestone after all, but was a retrograde condensate, with the lighter gas fraction having leaked off.
The Sebyar-1 weak blow and gas composition was a eureka! moment for me. As Robertson described, “Salo unearthed from old well files the overlooked fact that the Permian in Sebyar-1 had actually flowed some gas.”
Sebyar-1 sat within Casarta’s mapping of a structural closure labelled “V,” on the migration pathway between Wiriagar Deep and the source-kitchen area, 100 kilometers to the southeast. Any reservoirs south of the Jurassic erosional truncation in Bintuni Basin (Perkins and Livsey, 1993) might be gas-charged. V was probably a mirror image of the Wiriagar Deep play. Sampurno still laughs about the day that I ran down the office hallway shouting, “Sebyar has gas! Sebyar has gas!”
Suherman immediately instructed Sampurno to do percent-gas equivalent molecular conversion of the C1 through C4 chromatography. We examined the data and discussed the implications. Casarta’s 1994 mapping showed Sebyar within a large, pre-Cretaceous structural closure mapped as V-structure. Sebyar-1 was situated at the culminating crest of this slightly deeper, but even bigger, anticline roughly parallel to the Wiriagar Deep structure. Sebyar-1 hadn’t encountered any Jurassic sandstones, because it was north of the mid-Jurassic erosional truncation. Yet it had tested Late Permian moveable gas with the same composition as our WD gas discoveries. The conclusion was inescapable: V-structure’s Jurassic reservoir south of the erosional unconformity was likely gas-charged as well.
Competition
Permo-Jurassic generated hydrocarbons in the basin depocenter to the southeast must have migrated updip along V-structure’s axial fold through the Permian/Jurassic carrier/reservoir strata with only the Permian sandstone north of the Jurassic erosional limit preserved and gas-charged. Geophysicist Stephen Scott (who had joined the NV team December 1994) proposed renaming V-structure as “Vorwata” and proceeded to remap it. The problem now was that Sebyar and half of the structure were located on British Gas’s Muturi PSC (onshore and offshore) just east of Berau PSC, with only half of Vorwata on Berau PSC.
The WD-1 gas-discovery and WD-2 appraisal/discovery results were publicly known by 1996. BG knew Bintuni Basin was a gas play, and rebuffed ARII’s overtures to farm-in to Muturi PSC, so ARII quickly diverted the drillship from Wiriagar drilling to drill the Vorwata-1 exploration well on ARII’s portion of the V-structure in late 1996. Geologist Kevin Perry well-sat the V-1, and cored 140 feet of overpressured, gas-charged Jurassic sandstone, which flowed 31 mmscfg/d on a DST designed and witnessed by Marcou.
The Vorwata-2 appraisal, drilled 18 kilometers southeast of the V-1 discovery well during February-April 1997, encountered a massive 446-foot net pay Callovian-Bathonian-Bajocian-Aalenian sandstone sequence, which tested a whopping 100 mmscfg/d. The Vorwata plunging anticline, parallel to Wiriagar Deep 16 kilometers to the west, was larger than Wiriagar and had even greater Jurassic net-reservoir thickness.
ARII drilled three more appraisal/delineation wells on Vorwata and BG independently drilled two wildcats, Nambumbi-1 and Sakauni-1 on their portion of the structure in 1997. The N-1 well was north of the Jurassic sandstone erosional limit and encountered gas shows in the Permian, however, BG didn’t test the interval. S-1 was drilled off-structure on the eastern flank of Vorwata and was a dry hole.
BG agreed to a collaborative agreement in 1997, leading to a round of unitization negotiations between ARII’s Berau JV and BG’s Muturi JV. Following the drilling of eleven Vorwata ARII/BG wells, the structure was classified a supergiant gas field with reserves of 14.9 trillion cubic feet.
Another Huge Discovery
But the team’s discovery streak wasn’t over yet. The “U-structure” (renamed Ubadari), located 40 kilometers southwest of the WD structure, was drilled in late 1997 and early 1998. Ubadari-1 was another Jurassic gas discovery well, and U-2 was the aquifer delineation well. Ubadari was ARII’s last large Bintuni Basin gas field discovery, with reserves totalling 2 trillion cubic feet.
The team benefited from full support from ARII president Leon Codron and exploration VP John Duncan. From 1995 ARII’s NV team was greatly expanded by adding Tony Lawrence (petrophysicist for all Bintuni wells); sedimentologist/explorationist Phil Lowry; certification manager Tim Verseput; geologists Jim Doyle, Bo Henk, Meizarwin and Tom Shackleton; geophysicist Benny Eza; wellsite geologists Dahlini, Indro, Akbarie, Muzir and Espiritu, among others. The Tangguh project benefitted from their contributions, and the sum of the team effort was greater than the individual contributions.
The project would have never progressed without ARCO’s support of ARII’s endeavours, including backing from ARCO Chief Geologist David Nicklin; ARCO President Marlan Downey and vice presidents Jamie Robertson, Barry Davis and Dodd DeCamp. Many more people were involved but there isn’t space to list everyone, so my apologies to others omitted in this account.
Four gas reservoir discoveries were made by ARII in the Bintuni Basin between 1994 and 1998: Wiriagar Deep’s Palaeocene and Jurassic, Vorwata’s Jurassic and Ubadari’s Jurassic reservoirs. Tangguh LNG commerciality was assured with these four world-class discoveries. The team overcame a lack of geophysical data through open-minded innovation, complemented by each person’s unique geological and reservoir P-T interpretations.
The NV exploration and drilling teams were presented with ARCO Exploration Awards for the Wiriagar Deep (WD-1 and WD-2) and Vorwata (V-1) discoveries by ARCO International. There were six Bintuni Basin gas fields discovered between 1990 and 1998, Oxy’s three small fields and ARII’s supergiant Vorwata, giant Wiriagar Deep (with Palaeocene and Jurassic reserves), and the large Ubadari gas field. LNG development was assured.
In 1998, then-President Suharto of Indonesia named the Bintuni Basin gas fields LNG project “Tangguh,” meaning “resilient” in Bahasa Indonesian. The ARII team had shown resilience in the face of obstacles and setbacks, and it succeeded in the commercialization of the giant and supergiant Bintuni Basin gas fields.
The database from 24 ARII/BG wells in the Bintuni Basin consisted of over 500 pressure readings, more than 50 drill stem tests, more than 6,000 feet of core and a 690 square-mile 3-D seismic survey over the six gas fields with greatly improved subsurface resolution. This dataset led to independent certification of Wiriagar Deep-Vorwata-Ubadari reserves by DeGoyler & MacNaughton in July 1998 of 24 trillion cubic feet, with proved reserves of 14.4 trillion cubic feet. An additional 1.5 trillion cubic feet of 3P reserves were calculated for Oxy’s vintage Roabiba-Ofaweri-Wos gas fields.
BP acquired ARCO in 2000, including ARII’s Tangguh assets, and led a consortium in developing the Tangguh LNG production/export facility, with the first LNG tanker shipment made in 2009. With only 15 years from discovery to first LNG production, that’s resilience.
The author thanks John Marcou for editing suggestions and personal photo contributions; Jamie Robertson for permission to republish quotes and maps; and Larry Casarta, Suherman Tisnawidjaja, Sonny Sampurno, and Matt Silverman for corrections, comments and suggestions.