Shale Gas: Penn State Researchers Rediscover Circular Reasoning

Guest “are you fracking kidding me?” by David Middleton

Researchers unearth cost-effective method for finding shale gas
New method approximates available gas in untapped areas of Marcellus Shale region using well production data

David Kubarek
September 03, 2019

UNIVERSITY PARK, Pa. — A new method for exploring natural gas in the Marcellus Shale, developed by Penn State researchers, shows potential high yield areas can be found more easily and with lower costs.

Traditionally, natural gas hot spots are determined using a combination of current well production data and various geological tests. The researchers, in work published in the SPE Reservoir Evaluation & Engineering journal, detailed a method for approximating available gas in untapped areas using well production data taken from more than 5,600 existing wells.

Researchers used only wells with more than two years of production logs and assigned a decline curve analysis — the amount of production loss over time — for each well. They then applied these decline curves over the entire region of the Marcellus Shale. That allowed researchers to forecast the amount of gas that would be generated over time if a new well were created. Researchers then validated their findings using geological maps, which were created from core samples.

“Rather than looking at these geological proxies for production we’re just looking at production itself,” said Eugene Morgan, assistant professor of petroleum and natural gas engineering in the Penn State College of Earth and Mineral Sciences. “By looking at just production and mapping we see that it agrees really well with these geological variables related to production, which validates our approach. By looking at production alone you’re directly targeting the information you’re after.”

Morgan said the method held strong during validation, showing that it was almost as effective at forecasting natural gas resources as methods that include costly geological data sampling.

[…]

Penn State University

This map accompanied the article and includes the actual caption:

Penn State researchers detail a method for approximating available gas in untapped areas using well production data taken from more than 5,600 existing wells
“Penn State researchers detail a method for approximating available gas in untapped areas using well production data taken from more than 5,600 existing wells in the Marcellus Shale region. That could lead to more economical location of natural gas with less disturbance on the region’s geology. IMAGE: PENN STATE”

That could lead to more economical location of natural gas with less disturbance on the region’s geology.

Penn State

“Less disturbance on the region’s geology”???

The purpose of drilling and frac’ing gas wells is to disturb the region’s geology. That’s how you get the gas to come out of the rocks and go into pipelines If you don’t disturb the geology, you don’t produce oil or gas… You can’t even drill and produce a groundwater well without disturbing the geology.

In fairness to the Penn State researchers, the press release was clearly written by someone who was totally clueless.

The SPE paper is pay-walled, but clearly not written by someone who was totally clueless.

Combining Decline-Curve Analysis and Geostatistics To Forecast Gas Production in the Marcellus Shale

Zhenke Xi (Pennsylvania State University) | Eugene Morgan (Pennsylvania State University)

To estimate the production potential at a new, prospective field site by means of simulation or material balance, one needs to collect various forms of costly field data and make assumptions about the nature of the formation at that site. Decline-curve analysis (DCA) would not be applicable in this scenario, because producing wells need to pre-exist in the target field. The objective of our work was to make first-order forecasts of production rates at prospective, undrilled sites using only production data from existing wells in the entire play. This is accomplished through the co-Kriging of decline-curve parameter values, where the parameter values are obtained at each existing well by fitting an appropriate decline model to the production history. Co-Kriging gives the best linear unbiased prediction of parameter values at undrilled locations, and also estimates uncertainty in those predictions. Thus, we obtained production forecasts at P10, P50, and P90, and we calculated the estimated ultimate recovery (EUR) at those same levels across the spatial domain of the play.

[…]

SPE Reservoir Evaluation & Engineering journal

This is all well and good. In resource plays, like the Marcellus, decline-curve analysis (DCA), all by itself, can pretty well forecast the estimated ultimate recovery (EUR). The production data can be used to forecast production… Not a new concept and basically circular reasoning. Furthermore, the authors clearly state that this is for”first-order forecasts of production rates.” It’s a reconnaissance tool. By interpolating DCA’s play-wide, you can identify areas which may be more prospective than others, in resource plays. This might help high-grade areas on which to focus more detailed geological analyses; although most geologists will want to work the schist out of the entire play. You don’t find things that everyone else missed using this sort of methodology.

However, this is not “a new method for exploring natural gas in the Marcellus Shale”… It’s not a method of exploring for anything. Nor does it show that “potential high yield areas can be found more easily and with lower costs.” You still have to pick a drilling location. And the only way you can do this is to interpret the geological data from nearby wells and incorporate any geophysical data that are available. The drillers have to have a target (x,y and depth).

One of the problems with science communication is that press releases are often far-removed from the scientific publication and then misreported by the media.

The new method of exploration comes at the right time. Demand for natural gas in the United States is growing, but prices have been depressed because supply from the shale patch is growing faster than demand. As a result, in April this year, the benchmark natural gas contract at the Waha Hub slipped into negative territory, at  -$9 per million British thermal units. Just a month later, the benchmark slipped into negative territory once again, at -$4.28 per mmBtu. The average price for the first five months of the year was $0.92 per mmBtu.

At the end of August, natural gas was trading above US$2 per mmBtu, but analysts warned that soon they could slip lower than this as summer ends and temperatures fall, undermining demand, which spiked during the heat waves. Now, there will likely be a lull in demand before it picks up during the winter.

Despite the lower prices, U.S. natural gas production continued to increase in August and set a new daily production record of 92.1 billion cubic feet per day on August 5, the EIA said, citing data from OPIS PointLogic Energy. Between May and August, gas production rose by 2.5 percent, mainly driven by the Northeast.

In this price and supply context, the new exploration method, if applied on a larger scale, could have implications for U.S. LNG exports as well.

Oil Price Dot Com

“The new method of exploration high-grading comes at the right time” for what? With natural gas prices routinely falling into negative territory and averaging less than $1/mmBtu, nothing can make drilling natural gas wells economic, except NGL’s…

Confidence in this fast pace of production growth is well-founded; reported initial production rates and other factors suggest that the break-even gas price for many dry Marcellus producers is only about $2.50/MMBTU, and that—thanks to supplemental returns on natural gas liquids (NGLs) and, in some cases, condensates—the break-even gas price for many wet Marcellus/Utica producers is even lower: about $2/MMBTU. In some cases the effective natural gas breakeven price gets all the way to zero, with fortunate producers achieving breakeven returns from the sale of NGLs and condensates.

RBN Energy

Production data is usually purchased from vendors who specialize in aggregating production data. Well logs and other geological data are also usually purchased from vendors. These sorts of data aren’t expensive, particularly in comparison to seismic surveys. And all of these costs are insignificant compared to the cost of drilling and completing the wells.

Even if this high-grading tool works and is widely employed, it would just increase the volume of natural gas production, putting downward pressure on prices. Re-quoting the Oil Price Dot Com article…

The new method of exploration comes at the right time. Demand for natural gas in the United States is growing, but prices have been depressed because supply from the shale patch is growing faster than demand.

Oil Price Dot Com

A new tool that increases production and barely affects costs is not coming at the right time, when natural gas is ranging from -$10/mmBtu to +$2/mmBtu.

Advertisements

75 thoughts on “Shale Gas: Penn State Researchers Rediscover Circular Reasoning

    • Gas is often produced as a byproduct of crude, or condensate, and torching is not legal in most circumstances.

      However as new gas liquification plants come on line prices are likely to rise. NG is not cheap in the rest of the World.

      • It seems that the Netherlands is trying to wean itself off NG. A dear friend of mine just moved there and decided to buy all electric appliances, including space heating, because NG won’t be available in a few years. Does anyone know if that is true?

        • Dutch resident here, in short yes ng will be fased out. Newly build social housing, isn’t fitted with ng connection anymore at the moment. NG should be gone in 2030 for households.

          • This is crazy. Natural gas delivered directly to homes for heating and cooking is far more efficient than any electric generating scheme, no matter what the source (even nuclear, though that doesn’t generate CO2). How do they justify it?

          • “…How do they justify it?…”

            Justification? We don’t need no stinking justification; besides, this is what Greta wants.

            You want warmer houses, cooked food & lights at night? Move to Saudi Arabia. As a side benefit, you’ll also be relieved of having to participate in all those elections.

          • The Netherlands will halt production at Groningen, Europe’s largest onshore natural gas field, by 2022, eight years earlier than initially planned, the Dutch government said on Tuesday.

        • Looks a bit premature, to say the least. To quote Dutch Wiki: “On January 8, 2012, it was announced that a new gas discovery had been made in Dongeradeel, in the northeast of Friesland. According to the NAM, this is the largest find since 1995: a total of 4 billion Sm3. “

        • But it’s OK because they are going to ban winters so no one goes cold. Really.

          This epic stupidity is scheduled to be repeated later in the U.K., right after MPS finish treating the House of Commons like a Saturday night venue for apparently drunken oafs to have an out of tune karaoke session.

        • Michael, Javert, this is a “physical” question.

          The Dutch have been using the Groningen gas field as a baloon (aka Underground Gas Storage UGS), filling the UGS with gas when demand was low, delivering gas from the UGS when demand was high.

          Earthquakes.

          Heating seems to be replaced by Geothermal. Electricity will have to be replaced by Utilities owned monopoly intermittants (solar, wind like in Denmark), by imports from other countries and from gas-electric plants.

          The Dutch will still use gas (perhaps from Norway and Russia in particular). They will use gas also after 2030 (albeit the Dutch may be moronic enough to ban new installs in residential buildings).

          Oddgeir

  1. Interesting collection of data, David, along with some misguided published stuff of course. Utilizing 3D Seismic we could easily identify a gas chimney coming off the J-K transitional Vaca Muerta black shale in the Dorsal Huincal area of the Neuquen Basin. These gas chimneys were where tectonism fractured the Vaca Muerta and caprock sufficiently to allow a localized escape of gas to higher levels. Adjacent area of VAca Muerta still held the gas in the black shale and could be produced. We were looking for insitu uranium and the mineralizing fluids utilized these same chimneys. Penn State used to have one of the five best collections of Economic Geology Professors in the USA, have no idea where they are going now, probably into Climate Change Geology.

  2. She’s baaaack !

    “Climate Activist Greta Thunberg: Once You ‘Fully Understand’ Climate Emergency You Know What You Can Do”

    https://www.breitbart.com/clips/2019/09/10/climate-activist-greta-thunberg-once-you-fully-understand-climate-emergency-you-know-what-you-can-do/?utm_source=facebook&utm_medium=social&fbclid=IwAR1JSu32EaTRRv8-BUT2LA_fuIYpkh6tLQlD3T-MWFdS1eCy3RSxz-LHno4

    “I tell the world leaders when I speak to them and I tell them how it is … I tell them the science, how it looks.”

    N.U.T.S……..

  3. Mr. Middleton’s article and its accompanying map leaves any sentient reader wondering how many billions of dollars of revenue the citizens of New York State and Maryland are foregoing because of their respective states’ insane anti-fracking policies.

    Thank you.

    • You are making a great point.
      The climate kook policies of the catastrophists not only result in wasteful costly sources of power. They also result in loss of new cheaper power sources and the loss if the wealth and well being that new cheap energy would provide.

  4. This must be the most useless study ever done. Ten years ago any geologist working the area could have told you the best areas in the Marcellus were in southwest PA in Washington and Greene counties and northeast PA in Susquehanna and Bradford counties simply by analyzing geophysical log data for thickness, depth, density (measure of porosity), resistivity (indicative of hydrocarbon content) and gamma ray (proxy for organic content). If the study did not address the proximity of offset wells it is even more useless. If wells are within 1,000 feet of each other, production from each will not average what a single well would and infill wells drilled considerably after an initial well do even worse regardless of the area. Good luck untangling that mess.

    • My mother’s family came from Bradford County (Sayre), where they allow drilling, but unfortunately they didn’t own any land or mineral rights.

      The did, however, have a small farm in Chemung County, NY, where they did own the land and mineral rights, but unfortunately NY does not allow drilling.

  5. I thought fracking was banned in New York State.
    That map appears to show wells in NY.
    And what is with the true/false?

    • There is (or was) conventional oil & gas production in NY State. The true/false thing is weird. Maybe they don’t know what the proper map symbols are.

      • I would assume the “false” legend would mean the well was drilled but no production reported. Pennsylvania only requires production reporting yearly so a lot of producing wells might not show up yet. University professors and students have a way of taking the most mundane information and making it less useful.

      • I’m 100% certain those blue circles are not dry holes and there are no P&A reports for those wells. They are either wells that haven’t been fracked, hooked up to pipelines yet or the state has no production reports. Pennsylvania is notorious for poor record keeping so just because a cursory search shows no production, it doesn’t mean it hasn’t been reported or recorded. Also, some operators may report gross production by pad since not all wells may be individually metered. The university types doing this study appear not to be capable of putting out a map that is clear enough to be of any use to anyone other than themselves.

        • The SPE paper probably explained it better, assuming the map was in the paper.

          Good point about dry holes… In resource plays, it should be just about impossible to drill a dry hole.

    • There were a few vertical Marcellus wells drilled and fracked in NY before the ban on fracking horizontal wells. I think you can still drill a vertical or horizontal well in NY, you just can’t frack it. It’s a distinction the State makes so they can’t be sued for denying the right to access your minerals. You can drill but the State assures you will lose all your money. It’s a state run by demagogues and totalitarians.

  6. I think this line of research might have implications well beyond this particular application. Think of the automobile industry, for example: car-makers could make a note of what colors of cars sell more and actually MAKE MORE cars of those more popular colors!

  7. If only it were easier to get pipelines built. There is an untapped market for NG in the northeast US, and something of a chicken-egg, catch-22 dilemma, along with opposition from anti-frackers, enviro-zealots and carbonistas, who, having had great success in shutting down coal power have now set their sights on NG.
    Many currently using propane for heating and such would switch to NG with some slight alteration to their equipment, as NG would likely be cheaper. Meanwhile, ratepayers suffer needlessly with much higher electric rates than need be.

  8. They did say “less” disturbance, not “no” disturbance. Maybe they were thinking of less exploratory drilling leading to less geological disturbance?

  9. Bruce- I think most of the states in the northeast, from New York and north and east, have firmly turned down several natural gas pipeline projects as bad for the environment. You can tell some people drinking more water is good, but you can’t make them drink.

  10. The problem is as usual the reports in MSM outlets. Non-scientists who think they can write about science. and twist everything into a complete Utopian realm!
    Although the study approach of analysing decline curves is fine – it is very limited in terms of data analysis. They do not consider well length, stratigraphic horizon, completion style, cluster spacing, proppant usage and placement, etc.
    In other words – primitive data analysis that will produce fancy maps good for the fire in the backyard!

  11. The price of natural gas cycles, just like oil. And though the price for gas is down now someday it will climb again. And when it does the cost of generated power will climb too. In a healthy energy economy when the price of gas climbs the power companies switch back to cheaper coal and we as the customer benefit; but that may not be true in the future.

  12. Admittedly although my experience comes from getting an underground silver mine into production just as Nelson Bunker Hunt was getting his fingers slapped and the silver price dropped from $14 to $2, I have learned that if you’re incurring all the expense of drilling wells, doing geophysics, and building production pipelines, and then selling gas for somewhere between the break-even price and paying people as much as you have to for them to take it–you’re going broke. The only thing dumber than the farmer buying hammers for $3 inside a MacLeods store and taking them outside and selling them for $2 because ‘it beats farming’ is the bag of hammers.

    Thankfully we have researchers at Penn State who can get paid for finding ways for others to go broke more efficiently.

  13. Must say that it’s refreshing, and a bit amusing, that a paper on improving natural gas production (even if it’ has issues) came out of Penn State. I would have assumed that that University was adamanntly and mannifestly opposed to anything remotely connected to any carbon compound.

    • There may be sections of Penn State over which Mann and his cronies hold no sway. Also, it is unlikely that the editor of the journal concerned would invite a review by anyone close to the climate scam.

  14. The Cabot General Manager in West Virginia 40 years ago said his favorite drill site geology was “closeology” followed by “betweenology”.

  15. By reducing “geological disturbance” I took it to mean drilling less low production or dry wells (or whatever a gas well is called when it produces little gas).

    This “new” approach seems so obvious I cannot for a moment believe its new. It may be new to the team of scientists who wrote the paper, but I bet and good geologist worth their salt was using this approach in some form, when the data was available. Someone still needs to find the new fields before any data is available.

    My only beef with natural gas is distribution in areas where their are frequent earthquakes. One can use the natural gas to produce centrally located electrical power and then distribute that… Not as efficient (power delivery wise), but a lot easier to maintain. The Netherlands should consider this before throwing away their resources.

    • It’s not really new. Kriging has been used for a long time in mapping and surface modeling; as have various forms of production maps.

      This method doesn’t really avoid drilling any wells. It’s a manner of high-grading areas for the purpose of more detailed geological analysis.

      Old method:
      Map regional production trends to high-grade areas.
      Correlate well logs.
      Make structure and isopach maps.
      Identify prospects.
      Lease prospects.
      Drill prospects.

      New method:
      Map EUR by kriging decline curves to high-grade areas.
      Make structure and isopach maps.
      Identify prospects.
      Lease prospects.
      Drill prospects.

  16. Good production tends to occur proximate to pre-existing good production. Brilliant.

    This concept is about as old as the second well ever drilled and has many limitations. The first thing a prospector or developer does is to look at the relevant existing production.

  17. David Middleton, it’s always worthwhile to puncture the hype balloons, but I think your scorn is wasted here.

    The researchers, in work published in the SPE Reservoir Evaluation & Engineering journal, detailed a method for approximating available gas in untapped areas using well production data taken from more than 5,600 existing wells.

    That says it all: it’s a new method for using extant data. It isn’t circular: actual production from each new well can be compared to the forecast for that well, and an improved expected mean squared prediction error can be calculated eventually.

    Is it the ideal time? How could they improve the time by either producing the technique 10 years ago or keeping it a secret til 10 years hence? Isn’t it always worthwhile to reduce production costs? The method can be adapted to other regions that have enough data, with more or less accuracy. Would you eschew the method on purpose now that it has been developed, if you were investing in new wells in Texas, for example?

    • From the post…

      This is all well and good. In resource plays, like the Marcellus, decline-curve analysis (DCA), all by itself, can pretty well forecast the estimated ultimate recovery (EUR). The production data can be used to forecast production… Not a new concept and basically circular reasoning. Furthermore, the authors clearly state that this is for “first-order forecasts of production rates.” It’s a reconnaissance tool. By interpolating DCA’s play-wide, you can identify areas which may be more prospective than others, in resource plays. This might help high-grade areas on which to focus more detailed geological analyses; although most geologists will want to work the schist out of the entire play. You don’t find things that everyone else missed using this sort of methodology.

      The only thing “new” about this is the use of kriging decline curves to make an estimated ultimate recovery map… And I don’t think it’s actually a new concept. Geologists already employ kriging to make such maps…

      Geostatistical Analysis of Oil Production and Potential Using Indicator Kriging
      Michael Edward Hohn
      Ronald R. McDowell
      West Virginia Geological and Economic Survey Morgantown, West Virginia, U.S.A.

      ABSTRACT
      Heterogeneous reservoirs lead to serious discontinuities in production from well to well, whether caused by sedimentological or structural factors. Geostatistical methods, such as ordinary kriging, are useful both for assessing degree of heterogeneity and for drawing maps of production and initial potential.

      […]

      AAPG, 1994

      It’s using production data to forecast production… Something that’s already being done and a sort of circular reasoning. The areas with the most favorable decline curves (best production) will likely yield higher estimated ultimate recoveries (best production).

      My objections aren’t to to SPE paper or the methodology per se. My objections are to the clueless characterizations of it in the PSU press release and Oil Price article. Describing this as the “ideal time” for such a methodology is 100% idiotic. It’s not a matter of when would be a better time. There is no ideal time for reinventing a wheel. A more opportune time for a new tool that increases production and barely affects costs would be when there isn’t a gas glut.

  18. over the (short) history of ‘fracking’ it appears to me that the biggest, short term issue is water pollution and depletion of surface water. This first came up with the earliest experiences with fracking in Western PA, where waste water pumped OUT of the wells led to massive fish kills due to black algae.

    second point: gas companies would do themselves a favor if they made a push for natural gas as a motor fuel. if every fleet truck was converted, it should result in savings for fleet owners and, more important, a PR win for reducing motor fuel pollution.

    • Frac’ing has been a routine well completion procedure for at least 70 years.

      Wastewater isn’t “pumped out of” gas wells, it flows out with the gas. 90-99% of the wastewater isn’t even related to frac’ing. Almost all oil and gas wells produce wastewater. The pore space in sedimentary rocks is filled with three things: 1) oil, 2) natural gas, and/or 3) water, usually brine. Frac’ing does not deplete or pollute surface water. Wastewater is generally disposed of in two ways: 1) injected into subsurface brine reservoirs, 2) delivered to water treatment facilities. Flowback water from initial production is sometimes allowed to evaporate in lined wastewater disposal pits. It’s never dumped, untreated into lakes, rivers, etc. The only real problem with wastewater disposal, is that improperly cited and/or operated injection wells can cause induced seismicity.

      It’s the oil & gas industry. They aren’t separate industries. Companies that focus almost exclusively on gas plays, like Chesapeake, have been lobbying for CNG for transportation for a long time.

      • The earliest form of fracking was routine well over 100 years ago.
        A “torpedo” was loaded with nitroglycerine, dropped down the hole, and detonated. This was done to fracture the rock and to form a cavity into which the crude could flow.

  19. In my career as a Petroleum Engineer I have experience working as a Reservoir Engineer doing existing field development and infill drilling. I do not see much new here other than the automating of the data gathering process and the screening/high grading of the well location determination process. This will not replace the investigative work that needs to be done to prepare an ultimate drilling candidate which will involve geological investigations, well and completion design, offset well reservoir drainage studies, etc. (the work that would get done regardless of the screening process). David has provided an accurate critique of the information provided by Penn State.
    I too cringed when reading the press release. There are some real boner statements indicating the total lack of understanding of the subject matter by the author. And the end product is obviously not subject to review by the proponent organization. An obvious example of how modern society treats technical matters and thus how it follows that the process can be jiggered by advocates.

    • Did you ever have a boss who would forward every article like this to the engineers and geo’s, asking why we were’t doing this sort of thing?

      • When I worked for Amoco Canada we’d get articles and (mostly) technical papers circulated but never under the banner of “Why aren’t we doing this?”. It was always treated as information that we could choose to pursue if appropriate. Generally, when working on an established oil or gas field, there are norms followed for interventions. Small tweaks to procedures to avoid very expensive and embarrassing screw ups. (the voice of experience here) One time I made a (seemingly) small “correction” to a stimulation program (it was a frac job) when I was supervising the work. As a result, the job was unsuccessful. A week later on the repeat job, it was successful. My boss did NOT fire me. He allowed me to “learn” by my mistake.
        Adding new technology and tweaking existing processes should be done in an environment of circumspection and applicability to the site specifics.

        • Our COO at Chieftain, who was a geologist, would circulate every article he ran across, usually about geophysical “black boxes” that found oil. He didn’t always ask why weren’t we doing that; but it sort of felt that way. Otherwise he was a great manager.

          99% of what I’ve learned in this business has been from mistakes… usually underestimating how unpredictable nature can be. No matter how much I’ve learned or how much the technology has improved, nature hasn’t become more predictable.

  20. I looked at a developed field 20 years ago for a company who had just purchased it and was hoping to find a few more wells. It was stratigraphically trapped oil in a fairly tight Cretaceous sandstone. None of our maps, net pay, total sand, max porosity, nothing corresponded to production. Mapping projected ultimate production showed some clear linear sweet zones. Natural fractures I guess. We drilled a few more wells along those trends until we ran out of sand.

    Anyway, this reminded me of that field, none of our geologic maps were nearly as good as just mapping production.

  21. I think you can break down most progressive thought to close-system, zero-sum gain, circular reasoning.
    No one can rise up without bringing someone else down – no one can be prosperous unless someone else is ruined.
    It’s why they have to destroy everything – they have no concept of creation.

Leave a Reply

Your email address will not be published. Required fields are marked *