Samuel T. Pees
Petroleum geologist
Meadville, Pa.
The Oil Creek Valley of Pennsylvania--in production from shallow wells since 1859 and for many hundreds of years before then by collection pits dug by Native Americans--is seeing the pursuit of an oil field phenomenon known as the deeper pool discovery.
A portion of the early North American oil belt is now realizing good gas production from Lower Silurian Medina group sandstones that have thick development on a trend that cuts southeastward across historic Oil Creek in Venango County.
As currently drilled and mapped, the geological medial line of this belt crosses the valley about 1 mile upstream from Pioneer, a ghost town of the first decade of U.S. oil.
This deeper production at that locale is known as Oil Creek pool (Fig. 1). It has a number of favorably interacting geological parameters that are attracting infill drilling. Extension drilling has reached 4 1/2 miles southeast of Pioneer, where the name Cornplanter pool is assigned.
There is a possibility for other similar trends in the Oil Creek area. The best possibility is northern Splane and Tecza gas pools, which new drilling has now shown to coalesce (Fig. 1).
Currently this combined trend is only drilled on the west flank of the valley, but it heads in a southeast direction. It would cross the creek between the Petroleum Centre ghost town and the borough of Rouseville if drilling pursues the thick belt and if it persists as presently interpreted.
A protuberance of thick, clean sandstone juts from Toonerville pool eastward toward Miller Farm, an abandoned railroad settlement on the west bank of Oil Creek. This trend could continue across the creek, a possibility that should be tested.
Two deep wells straddle Oil Creek in the vicinity of the famous Drake Well and are appropriately designated Drake Well pool (Fig. 1). The geology and reservoir characteristics of the beds are variable in this locale, but the sandstones are productive.
Fig. 2 shows the areas drilled for Medina gas in Northwest Pennsylvania. The location of Oil Creek and the position of detailed OGM base map which is used for Figs. 1, 4, and 5 is depicted by a rectangle.
EARLY DRILLING
The shallow reservoirs are Upper Devonian in age and were named since the early 1860s as the Venango sands. They bear numbers in Oil Creek Valley: First sand, Second, Third Stray, and Third.
The Venango Third sand at an average depth of only 450-500 ft in the flats was the best reservoir. It had the greatest pressure and volume, a combination that resulted in spectacular gushers and initial flow rates of 1,000-4,000 b/d of oil in the better wells.
The Third sand is not everywhere present. Where not found, one or more of the other sands may be stratigraphically developed and make a well.
The Venango sands are shallow marine, near-shore sandstone bodies formed as bars, ridges, peninsulas, and transitory beach deposits. Conglomerates consisting of milky quartz pebbles mixed with the sand grains are present in the bodies in some areas.1
The Venango sands' section regionally contains other sandstone than those described by number above. Some of these occur at various places not far east of Oil Creek Valley. The Red Valley and Lytle sands are examples.
Edwin L. Drake struck oil in the thinning edge of a stray sandstone in the Upper Devonian Riceville formation shale, which is above the First Venango sand. It was a lucky find at 69 1/2 ft, especially considering that only 38 1/2 ft of the hole was in bedrock, the overburden being alluvium and glacial outwash.
This thin, stray sandstone is about 2 ft thick. It was given the name Drake sand. It has never been found in outcrop and never positively identified in another well.
In the first 5 years after Drake's 1859 discovery, 900 wells were drilled along Oil Creek in the stretch between Titusville and Oil City. They were crowded on the steep valley walls, the narrow flats, the mouths and lower part of tributaries, islands, gravel bars, and even in the creek.
Drilling had also taken place along the Allegheny River and other streams. The total number of early wells in this oil region would have reached several thousand in short order.
By 1865 drilling had commenced on the plateaus. Depths were greater due to the higher elevations, but exceptional strikes such as at Pithole temporarily justified the economics of the day.
A few historic wells and leases are still producing in the valley. However, for the most part it is an incredible outdoor museum of the drillers' and producers' tools and machinery, lying in the dense forest like materiel de la guerre, degrading and returning to contour.
Oil Creek State Park covers much of the valley, but its mineral rights are largely held by private interests (Fig. 1). The state park staff, the outdoorsmen, fishermen, hunters, birdwatchers, hikers, bicycle trail riders, history researchers, lumbermen and foresters, archaeologists, geologists, and the oilman mingle in this early seat of the oil industry.
DEEP GAS PRODUCTION
These Lower Silurian beds are called the Medina group, a stratigraphic section of shales and sandstones comprising about 190 ft (Fig. 3).
The Medina play has been going on in earnest in Northwest Pennsylvania since the 1950s, but there were earlier wells, too. The success rate for the Medina objective has been an astounding 98%. This refers to completed wells versus those abandoned for mechanical reasons or dry holes.
Fig. 2 shows the Medina group fields in Northwest Pennsylvania. A portion of the blank areas will probably be filled in gradually unless they are inaccessible. Some state game lands are scheduled to be drilled. There are even places in the drilled areas where infill wells are justified.
Most of today's operators are more picky than in distantly past years. Acreage on perceived trends is sought and the well array is programmed accordingly.
Most geology and engineering is put into the projects as higher volumes per well are sought. Like some other sandstone sections distributed over a large region, the Medina group has good, fair, and lean areas depending on the presence or absence of explicit reservoir factors.
In practice lean wells can be produced in a multiwell venture that also includes above average and exceptional wells, the latter two categories holding up the overall ultimate recovery of the total wells in the array.2
The Medina group has produced at least 378,423,288 Mcf of gas in Pennsylvania by yearend 1992.3 It has also produced more than 2 million bbl of oil along with the gas.
There are a few areas that are oil prone, but most of this producing region is dry gas or with slight liquid. It is thought that the reported production data leading to the cumulative amounts indicated above may be lower than the actual volume of gas that could have been produced during a 4550 year span.
Many more Medina wells will be brought into production as new geological leads are followed.
MEDINA STRATIGRAPHY
Upper Medina Unit. The upper 30-50 ft of the Medina group in Oil Creek pool is argillaceous but can contain some shaly sandstones and siltstones (Fig. 3). The depositional environment of this unit is believed to be a tidal flat.4
There are seldom any sandstone beds of adequate reservoir quality within this section. Exceptions are a thin bed or two with qualifying porosity and acceptable saturations, but the occurrence of these is rare and sporadic. If included in the completion, two or three perforations would usually suffice.
Grimsby formation. The Grimsby formation contains the main bulk of the multiple sandstone gas reservoirs in the Medina group. The sandstone bodies are separated by thin, shale beds. Also the bodies can be locally modified by very thin interdigitated shale or shaly sandstone layers. The sandstones in the upper part of this stack of sandstone bodies often comprise the main gas zone in a Medina well in this area.
If fining-upward sandstone features indicate a fluvial channel-fill, then a sandstone body at 5,952 ft depth on the log of Fig. 3 could be an example. However, the Grimsby formation for the most part is shallow marine and littoral. Some of the thicker sandstones in Fig. 3 could be barrier bars.
The footage of all 70% clean sandstone in the Grimsby formation plus that in the underlying upper part of the Cabot Head formation have been lumped together and isopached to show the total aggregate of clean sandstone in this section of the Medina group and to define the build-up trends 5 (Fig. 4).
Upper Cabot Head sand-stones. These sandstones, where present, can contribute significantly to the overall sandstone composition of the Medina group. The Cabot Head formation is normally a distinctive shale of 25 ft to more than 40 ft thick in its lower part, but impure siltstone and very shaly sandstones may be present in the lower section, too. The upper part of what would usually be a continuous shale sequence can locally contain well developed, clean sandstones in this area. These sandstones can add to the overall Cabot Head formation section, increasing it to more than 50 ft.
The sandstone body with the sloping base from 6,030-50 ft on the log Fig. 3 is an Upper Cabot Head formation sandstone body. The sandstone above it may belong to the Cabot Head, also.
Laughrey4 interprets some Cabot Head sandstones as tide and storm-deposited bars as studied m the Creacraft well m nearby Crawford County. These sandstones overlie a shale section that serves as a very effective vertical seal above the Whirlpool formation sandstone, which is the basal unit of the Medina group.
The Cabot Head sandstones pinch out laterally in mudstone as seen by local and regional log correlations. This situation turns them into completely sealed entities. These sandstones are occasionally included in the well completions in this general area.
Whirlpool formation sandstone. The Whirlpool formation sandstone can be an important gas reservoir as has been proven in a large number of wells in Northwest Pennsylvania and western New York. It is sometimes completed singly but usually is commingled with the Grimsby or the Grimsby-Cabot Head sandstones, the latter being the case in a few wells in the Oil Creek area.
The Whirlpool formation sandstone rests on an unconformable surface at the top of the Upper Ordovician Queenston formation shale (Fig. 3). It normally has 7-15 ft of 70% clean sandstone in most of the mapped area.
Where more than 10 ft of qualifying clean sandstone occur (usually in a single body), a build-up is mapped using the 10 ft and 15 ft isopachous lines. These Whirlpool trends can be discerned on Fig. 5. They correspond fairly well with the position of the build-ups of clean sandstone (lumped together) in the Grimsby-Cabot Head formations (Fig. 4).
Even where the Whirlpool formation does not exhibit a 10 ft (or thicker) build-up feature, the clean sandstone is usually thick enough to be an acceptable reservoir, porosity and saturations permitting.
Fining upward, coarsening upward to a point, and then fining to the top of the sandstone, interdigitated shale breaks, and blocky patterns are among the gamma ray motifs seen on logs in the Whirlpool build-up features. The main southeast build-up of Oil Creek pool (Fig. 5) may be a shore-normal or cross-strike depositional feature. The Whirlpool in adjacent Crawford County was interpreted to be deposited in sublittoral, littoral, and subaerial environments with sublittoral dominating in the west of the county.6 The Oil Creek area in Venango County is probably in a transitional littoral zone.
MEDINA RESERVOIRS
The Medina group sandstones show specific buildup trends with a set of parameters that make these trends highly prospective. The salient qualities for Oil Creek pool are shown in Table 1.
The baseline for above-norm porosity in any 70% clean sandstone in the Medina group is 9% DO. Porosity zones are frequently found in the 6% (or less) to 8.5% range.
A build-up of Grimsby-Cabot Head sandstones in Cherrytree Run pool has 20 ft of above-norm density porosity in one well and 10 ft in another. There are a number of high porosity locales exceeding 10 ft of 9% DO. The higher porosity peaks in pools other than Oil Creek pool are in the range of 11-19.9%.
The Whirlpool formation sandstone has numerous small, interrupted build-up areas reaching or exceeding 10 ft of clean sandstone (Fig. 5) in pools other than Oil Creek pool. In Tecza and Cherrytree Run pools the Whirlpool peak density porosities run 11-13% with 3 ft being the greatest amount of qualified above-norm footage (9% DO).
The Whirlpool can be a productive reservoir of import even where the density porosity is much lower, 6% or less. A study7 by the author concluded that a calculated log porosity (x-plot) of the Whirlpool sandstone may be critical at 4.5%. In other words, the sandstone could break down by hydrofracture treatment at that low porosity level and give up gas, but porosities less than that would not be viable.
Shut-in initial pressure. It is noted that the definition of the areas where wells have 1,400 psi to 1,500+ psi shut-in casing pressure/48 hr (or more) shows a reasonably close correspondence to the Grimsby-Cabot Head and Whirlpool trends of clean sandstone build-ups and their above-norm porosity factors.
This pattern does not appear to be markedly influenced by the depth of the reservoir. In a sense, a high pressure anomaly is part of the favorable trend factors in this area.
Total depths. The Medina wells in Oil Creek pool have TDs ranging from 5,812 ft to 6,226 ft. All but two of these wells in this pool are on the east side of the creek.
Estimated ultimate recovery. The author does not have enough production information at hand to fully estimate recoverable reserves in the Oil Creek area. A greater duration of production history is needed to properly formulate the decline curves.
Much of this information is tightly held in a new area, and single well metering is sometimes not available anyway.
Nevertheless, based on the local reservoir parameters and regional production knowledge gained elsewhere in the Medina producing region, some rough estimates are made here. Generally, they are higher than normal, a situation that suits the favorable geological trends described in the article.
The total mapped area in this article has a few wells with estimated potentials capable of exceeding 500 MMcf. A very few could exceed that estimate greatly.
The EUR norm, averaged from all wells on the mapped area of Figs. 1, 4, and 5, may be on the order of 225 MMcf. Included in the total wells are some poor ones that would yield less than 100 MMcf. The latter are off-trend and hampered by poor sandstone development and reduced porosity.
Comparison was made to other areas of detailed study that bore a reasonable geological match to Oil Creek-Cornplanter pools. Oil Creek pool as it now stands has reservoir characteristics that suggest an averaged EUR on a multiwell program basis of 225 MMcf/well. The trend and its neighbors on the west side of the creek appear to have the factors that may allow the norm to exceed this estimate in a multiwell situation on trend.
Since the above are rough estimates for the Oil Creek area, a new operator should not read this to mean that the first and subsequent wells in any program on a trend would all meet this norm. Some will be higher and others lower, a balancing act that will establish the reserves' norm pertinent to the multiwell program that pursues the geological trend.
In calculating reserves over the geographic spread of Medina group fields in Northwest Pennsylvania, the author has found that there are average figures or norms applicable to one field or pool (or to a particular trend) and higher or lower estimates that apply to other areas which have their own set of parameters.
An EUR estimate is not an "across the board" blanket situation. The favorable reservoir parameters vary within the sandstones and from place to place. The geometries of those bodies also vary, sometimes drastically over short distance, meaning that reservoir volume can be restricted or amplified depending on the net effective sandstone present and its porosity.
Examples of estimated norms may vary from 157 Mmcf (part of Conneaut field in Northwest Crawford County), 175 MMcf and 200-225 MMcf () in other Medina areas in Crawford and Venango counties, to mention a few. The production time factor is also important (life of well) as is the production practice, which should be prudent so as not to allow premature pressure depletion.
FUTURE OBJECTIVES
The Oil Creek pool trend and other possible trends mentioned above deserve ongoing drilling (such as the southeast trend between Petroleum Centre and Rouseville and the Miller Bridge lobe).
Even areas somewhat off the medial line of the favorable trends such as some places rest of the creek also merit attention because the recoverable reserves should economically pass muster, even if not at estimated peaks as on the trends.
Filling in of the current blank areas between Tecza and Cherrytree Run pools and the creek seems reasonable.
Fig. 2, which shows the Medina fields and pools, indicates small producing or tested areas east of the detailed maps of Figs. 1, 4, and 5. Production was found n the northeastern sector of Venango County (off the map). East of Venango County is Forest County, where six wells tested the Medina group (shown by dots on Fig. 2).
Results attributed to the Medina in Forest County ranged from saltwater with a show of gas to 25, 100, and 250 Mcfd of gas in the Grimsby-Cabot Head section. The operator chose to complete these wells uphole in Upper Devonian beds. The liberty to spot the Forest County wells in Fig. 2 was taken because of their shows in the Medina (this was not done elsewhere).
The Medina group (Grimsby section) undergoes a facies transition eastward where it becomes the tighter Tuscarora facies. The Whirlpool formation and the Cabot Head shale eventually drop out.
This article does not treat on areas outside of the detailed study area of Fig. 1 as far as drilling prospects are concerned. Suffice to say that more deep Medina opportunities in Venango County will eventually be noticed. Geologic extrapolation already suggests fairways and favorable anomalies that merit testing.
The Medina group objective although called "deep" is only a little over halfway to crystalline basement.
The remaining sedimentary column, untouched in the subject area, should contain carbonates and sandstones in the lower section, beds that have proved to be reservoirs and hydrocarbon bearing in other areas in Northwest Pennsylvania, Southwest New York, and Ohio.
Of course, the situation relative to those deeper beds in this model area is purely exploratory at this time. The point is that there is still some unexplored section at greater depth in this area of interest.
There is also a possibility of local hydrocarbon saturations in beds above the Medina group, already behind pipe. A lower porous zone in the Lockport formation (Silurian) and the Middle to Lower Devonian (Onondaga-Bois Blanc-Oriskany) may be prospective sometimes. These stratigraphic sections should always be scrutinized. Some of the beds have turned out to be productive occasionally at places in Northwest Pennsylvania.
CONCLUDING REMARKS
Although other objectives above and below the Medina group come to light in Northwest Pennsylvania, the main thrust of deep drilling in the northwest counties continues to be for the Lower Silurian sandstones.
The amount of space for infill drilling is ample, even in the heart of the play, and without a doubt future drilling programs will eventually proceed to the edges of the normal Medina facies.
Already the play is moving outside of the heretofore traditional region and southward off the map of Fig. 2. If there is a notion that natural gas supplies are waning in the East, this small study area can be taken as a model and encouragement to seek remaining and new reserves in a Medina producing region.
An important long reach horizontal Medina well was drilled in Rome Township, Crawford County (north of the Oil Creek model area) in 1988. It was completed as a gas well after a record-breaking frac for this objective in this region. A pioneer hightech well, it has shown the way for adjustments in the technique for this type of drilling into the bundle of sandstones which constitute the Medina reservoirs.
Horizontal drilling could be applied at other Medina sites considered to be beneficial, especially where natural fracture swarms have aided permeability in these brittle sandstones. The author holds a favorable but qualified view of the future of horizontal drilling into the Medina, with trend selection of certain specific parameters being the site and drilling direction qualifiers.
Looking at the deep wells now in Oil Creek pool and general area, it is gratifying to see a 135 year old, pressure-depleted, shallow oil belt come back to life by virtue of deeper gas.
Since certain possible trends were mentioned, the author reminds readers that exploring and exploiting the natural resources of the earth is always attended by miner's risk. Available well control (adequate logs, etc.) for this study was limited to only a portion of the wells on the OGM base map.
REFERENCES
- Dickey, P.A., Oil geology of the Titusville quadrangle, Pennsylvania, Pennsylvania Geological Survey Mineral Resource Report. 22, 87 p., 1941.
- Pees, S.T., and Burgchardt, C.R., What to expect from a typical Medina gas well, World Oil, Vol. 200, No. 2, pp. 37-40, Feb. 1, 1985.
- John Harper, Pennsylvania Geological Survey, verbal communication.
- Laughrey, C.D., Petrology and reservoir characteristics of the Lower Silurian Medina group sandstones, Athens and Geneva fields, Crawford County, Pa., Pennsylvania Geological Survey, Mineral Resource Report 85, 126 p., 1984.
- Pees, S.T., Mapping Medina (Clinton) play by lumping clean sandstones, OGJ, Mar. 9, 1987, pp. 42-45.
- Pees, S.T., Geometry and petroleum geology of the Lower Silurian Whirlpool formation, portion of N.W. Pennsylvania and N.E. Ohio, Northeastern Geology, Vol. 8, No. 4, 1986, pp. 171-200.
- Pees, S.T., Model area describes Northwest Pennsylvania's Medina play, OGJ, May 23, 1983, pp. 55-60.
