Appalachian Basin

Currying Favor with Natural Gas Opponents, Myers “Defies Theories of Hydrogeology”

Natural gas opponents thought, for the zillionth time last year, they had found a game changer in Dr. Tom Myers report indicating it was theoretically possible for fracturing to migrate thousands of feet upward through multiple layers of rock but new research indicates the Myers study essentially defies gravity, science and hydrogeology and seems to have been done to curry favor with the usual suspects. 

Last year Dr. Tom Myers unveiled a study that claimed hydraulic fracturing fluids can migrate vertically through thousands of feet of rock to contaminate water aquifers — within as little as three years’ time.  The only “evidence” supplied by Myers to support this assertion was a series of assumptions, which was used to formulate a computer model that simulated his conclusions.

Respected hydrologists were, for this reason, quick to point out the significant oversights in Dr. Myers’ work. Of course, as with all modeling studies, the quality of what you get out depends entirely on the quality of what you put in. In this case, previous reviews and a new study with input from the Pennsylvania Geological Survey (PGS) and Pennsylvania Council of Professional Geologists (PCPG) show Myers’ assumptions weren’t very good at all.

The latest study critiquing Myers’ work helps, in fact, point out why the Myers study has unsupportable conclusions.  Among other items, the latest review found Myers’ effort involved the “complete misinterpretation” of previous works and “defies theoretical concepts of hydrogeology.”  The research team also noted that overall Myers’ model “is not realistic…and is not consistent with what the scientific community has documented regarding…the history… in the Appalachian basin.”

So with that, let’s take a closer look at some specifics.

Problem #1: Myers Can’t Tell the Difference Between Shale and Sandstone

Myers, in reaching his flawed conclusions, assumes almost the entire geology above the Marcellus Shale (4,921 feet) is “predominantly sandstone.”  This is not true, a point the Pennsylvania research team makes abundantly clear.  Specifically, the technical rebuttal to Myers’ work notes:

…the characterization of the geology overlying the shale as mostly sandstone is not consistent with the data gathered from more than 100 years of subsurface drilling in the basin.

The rebuttal goes further, noting that even in southern New York (the region Myers describes as his modeling area) a significant portion of geology overlying the Marcellus Shale is comprised of shales and siltstones of lower permeability than sandstone.

That’s a crucial point, too, because remember: for Myers’ theory to be relevant, the layers of rock above the Marcellus must allow for the migration (upwards!) of fluids. But, the fact he assumed nearly 5,000 feet of space was significantly more permeable than it actually is shows why, from the get go, his study is inherently flawed.

Problem #2: Wildly Inflated Fracturing Size

Another key assumption relied upon by Myers was the size of the fractures that would be created by the hydraulic fracturing process.  Specifically, Myers assumed a continuous 19.7 foot wide vertical fault with high permeability would extend from the Marcellus formation at depth to the surface.  If that seems a bit unrealistic, it should.  The Pennsylvania research team, in their technical rebuttal, notes (emphasis added):

This fault geometry and hydraulic characterization are unsupported by any empirical data collected by the scientific community for the Appalachian basin (or anywhere else for that matter).

Well that’s that!

The team goes on to note that due to in-site stress and other geologic considerations that are well established, fracture orientations in rock vary in depth and lithology and can’t be presumed vertical throughout an entire sequence of sedimentary rock. Unfortunately, that is exactly what Myers assumed anyway.

Other studies of real data of fracture stimulations bear out the technical rebuttal’s response.  Specifically, a previous paper entitled “Hydraulic Fracture Height Growth: Real Data” examined fracture lengths initiated in the Marcellus and Barnett shale basins. In specific regards to the Marcellus Shale, that study found:

The Marcellus data…The fractures grow upward much taller than what was seen in the Barnett (some fractures grew nearly 1,500 feet), but the shallowest fracture tops are still around 4,800 ft, almost a mile below the surface and thousands of feet below the aquifers in those counties. (page 11)

That can be seen rather easily in the graph below which mapped out the observed fracture propagations in the Marcellus Shale as well as the distance between those fractures and the nearest source of groundwater

image002Problem #3:  Misses the Mark on Subsurface Fluid Movement and Dynamics

The technical rebuttal also noted significant discrepancies with the assumptions Myers made in regards to fluid movement.  First, Myers assumed a continual upward fluid migration in the Appalachian basin.  However, the technical rebuttal notes “there is no evidence to suggest upward fluid migration has been continuously occurring in the Appalachian basin.”  The research team, in fact, noted that if such a situation were actually occurring throughout geologic time (as Myers assumes), then “the fresh groundwater aquifers currently sourcing more than one million private water supplies in Pennsylvania would be unpotable.”

Moreover, the research team observed Myers’ one-dimensional vertical flow system “is not plausible for any aquifer system” because “water mainly moves horizontally…and not vertically.”  These major oversights led the authors to contend the scenario Myers used to reach his conclusions “defies the theoretical concepts of hydrogeology.”

The authors also noted Myers utilized a numerical computer model, specifically USGS MODFLOW-2000, in his groundwater flow and transport simulations.  But, the use of this model to simulate groundwater flow is a significant flaw, as the model is not equipped to account for fracture porosity, multi-phase flow or density.  Additionally, Myers used the model in an attempt to simulate the hydraulic fracturing process, which the rebuttal notes is something the system simply cannot do with any degree of certainty.

Given the significant errors in Myers’ study, it seems reasonable to ask how a researcher could make such large mistakes that don’t seem to be supported by an objective scientific review.   Quite simply, that’s what happens when you develop a hypothesis and then fashion a study to meet that end goal.

It’s a tactic Myers seems to be familiar with, given that his client list includes such well-known shale critics as the Center for Biological Diversity, the Natural Resources Defense Council and Catskill Mountainkeeper, to name a few.  Ironically, it’s exactly this “scientific approach” that led Myers to declare the U.S. Environmental Protection Agency’s findings in Pavillion, Wyoming are “scientifically sound” — even though no other federal agency does.

Quite simply, this technical rebuttal exposes the Myers’ study for what it is: an effort to malign hydraulic fracturing with pre-scripted activist talking points, all carefully hidden behind the guise of “science.”  The problem for Myers, however, is that eventually the truth comes out.

  • Brian Oram
    Posted at 11:35h, 12 May Reply

    I am sorry to say I must agree with the PA Geological Council. We are scientists first and geologists and other disciplines second, we must rely on the facts and not use science to develop the outcome we want. Just my thoughts – Brian Oram, Professional Geologist and Soil Scientist.

    • Anthony Fabina
      Posted at 16:29h, 21 May Reply

      Sorry Brian,

      I am a Geologist first, which makes me a scientist, with a lot of disciplines. Science is where you find the facts and support them with proof.


  • Gerrit van Tonder
    Posted at 08:28h, 13 May Reply

    Myers is correct. In all geological basins the water wants to move upwards and is just looking for a preferential flow path like a faulty well annalus or a fault. Look at the following website:

    • Tom Shepstone
      Posted at 16:41h, 13 May Reply

      Did you actually read the article?

  • scott
    Posted at 18:34h, 13 May Reply

    Thanks John for bringing the latest in a long series of Myers criticism to our attention. Myers work on the subject has been so flawed that it has drawn outrage in the reservoir engineering community (Saiers and Barth, Siegel, Cline, to name just a few, 2012) for not only it’s unrealistic model assumptions but for using such a simplistic modeling software that cannot be used for multiphase flow. The scary part is that Myers was even allowed to attend the recent EPA workshops on subsurface modeling where he continued to embarrass himself with yeat another flawed poster session presentation that also deserves a good debunking.

    As for the previous comment on preferential flow to a faulty annulus or fault …this is nonsencs. The preferential flow is always to the wellbore pressure sink and safely to the surface facilities through tubing and multiple redundant layers of casing and cement.

    Scott Cline, PhD petroleum engineering

    • Robert Nolan
      Posted at 19:24h, 17 May Reply

      I note at the same EPA workshop in North Carolina, in Session 2, who should appear, but another Professor with a PhD in Duplicity and a minor in BS….Mr Debunked himself, Tony Ingraffea

      Wellbore Integrity: Failure Mechanisms, Historical Record and Rate Analysis – Anthony Ingraffea, Cornell University

    • Friedrich Mohs
      Posted at 07:35h, 04 June Reply

      Scott Cline:
      Why so inclinded to included PhD peroleum enginerring? Why is that relevant here? Afterall, it appears Myers himself holds a PhD, perhaps this is not something to boast about? Just saying…


  • Tony
    Posted at 05:21h, 14 May Reply

    Gerrit van Tonder has a vested interest in defending Myers…he parroted the methodology to supposedly “prove” that shale gas development in South Africa would have similar results.

    Note that in his comment here he does not address the scientific criticism of the methodology of Myers.He simply blanket states that Myers is correct. Without proof. A reputable academic worth his salt would immediately be able to reply with a technical discussion or a repudiation of the points adressed in the rebuttal He cannot. His link is to a sensationalist article supporting his work. Some evidence.

    Van Tonder’s own faulty methodology has already been reviewed and debunked :

    It seems his position is as much motivated by spite as by academic logic.

Post A Comment