Drilling Induced Methane Migration: A Potential But Manageable Problem Via Drilling and Inadequate Casing Cementing of Oil & Gas Wells: Examples from Northeastern Pennsylvania

Drilling Induced Methane Migration: A Potential But Manageable Problem Via Drilling and Inadequate Casing Cementing of Oil & Gas Wells: Examples from Northeastern Pennsylvania

Introduction

Naturally occurring methane is present in many aquifers in many areas. This is typically due to adjacent gas-bearing strata, occasionally including the rock that hosts the aquifer itself. In most cases it is naturally occurring methane migration from gas-bearing rock to freshwater-bearing rock. Lakes, landfills, and wetlands are other possible sources of migrating methane. These sources in addition to manure ponds and sewage treatment plants consist of biogenic methane, derived from microbial reactions. Methane from gas wells is predominantly thermogenic, sourced from once deeply buried temperature and pressure altered organic matter. It is not always easy to determine whether gas sources are exclusively thermogenic or mixed with biogenic gas since thermogenic gas can be microbially altered during migration. Isotopic and molecular analyses are the best means to determine gas type(s). It should be noted that methane is not considered toxic, although it could be if it were to collect enough in a confined place to replace oxygen availability. Flammability is probably its biggest danger.

There are quite a few documented cases of methane migration, or stray gas migration, from wells drilled through shallow gas bearing zones to nearby water wells. In most cases the gas zone is very close to the fresh water aquifer, typically just below it. In northeastern Pennsylvania, the main focus area of this post, this is the case and methane migration (mostly thermogenic here) can occur on a limited scale simply through drilling through the shallow gas-bearing zone which can flow into the aquifer between time of drilling and time of casing and cementing. Since drilling creates a conduit for gas to flow upward it will if conditions are right. Methane is lighter than water. It will flow from higher pressure to lower pressure if given a conduit to do so. This is why people living near wells can experience bubbling and frothing water when and just after the aquifer is drilled through. Methane can also migrate up through soil to the atmosphere and sometimes ends up bubbling up through streams. There was one case out in the Rockies where a geologist speculated that drawdown of the water table by agricultural irrigation and oil and gas operators caused gas, or coalbed methane, to bubble out in streams. Studies have confirmed that shallow migrating methane tends to flow toward lower elevations (valleys and streams) due to the lower pressure compared to higher elevations. Worse cases have been caused by improper well-cementing where gas presumably continued to flow from the gas zone into the aquifer. This can be the case also where water wells have been drilled too deep. Some methane migration occurs naturally as evidenced by long bubbling streams and flaming faucets but changes directly after drilling usually indicate drilling related methane migration. In most cases the methane migration problem can be fixed by properly cementing or re-cementing the casing and waiting for the gas to dissipate or bubble out. Accumulated gas could possibly lead to explosions in enclosed areas and aeration systems are recommended to vent the methane or methane-rich water, whatever the source. Radon gas migrating from near-surface organic shale beds is also a significant problem in the area. Inadequately abandoned early gas wells drilled from the 1850’s to the 1930’s may have also contributed to methane migration by providing conduits. Methane may accumulate in the headspace of water wells and/or it may be dissolved in the water. The major methane migration cases in northeast Pennsylvania have been fixed and methane levels in the water in those cases are now acceptable. In one recent case a geothermal well for a shallow geothermal heat pump system was being installed near Owego, New York when the drilling rig caught fire due to migration of very shallow methane. Apparently, in areas with shallow methane it tends to migrate toward the valleys when it gets close enough to surface – due to lower pressure and a better developed fracture system there. That suggests that water wells in valley areas have a better chance of experiencing methane and indeed higher concentrations of dissolved methane have been documented in the valleys of the area.

In northeastern PA the Upper Devonian Catskill Sandstone and the Lock Haven formation below it are likely the main gas sources of the migrating methane. The Catskill is highly fractured naturally with the reservoir fluids flowing along the vertical fractures of this orthogonal fracture system. The fracture system may permit the reservoir fluids to flow far quickly due to the interconnectivity, particularly in valleys, which tend to have greater fracture density. Geochemistry suggests that these gas sources are associated with brackish and saline waters (from the Lock Haven formation) with longer residence times than the fresh water aquifers above.

Adequate Cement Jobs Prevent Continuing Methane Migration

Methane migration is mostly associated with certain areas where shallow gas is more abundant but can occur with less available gas if it can flow for a longer period due to inadequate casing cementing. In the oil & gas industry, better cement formulas have been developed to prevent any gas channeling and better procedures for placing and holding the cement before it sets have also been developed. The use of spaced centralizers to hold cement in place and make it more uniform are important for improved cement bonding. Better assurances of a good cement job are attained by utilizing cement bond logs to determine bonding, any problems from a washed out section of the borehole, and the top of the cement. A bad bond may cause channeling or cracking in the cement. Temperature logs can also be run to check for gas and a cement top. An adequate cement job should stop any methane migration due to drilling from continuing. An inadequate cement job may allow the gas to continue to migrate. The appearance of gas or significant pressure on the “backside” or annulus of the casing of a gas well is a telltale sign of an inadequate cement job and could be indicative of a danger of a serious methane migration event.

Stray Gas Incident Response

There are several good information sources addressing what to do when a stray gas or methane migration “incident” occurs. Marcellus Shale Coalition offers useful guidelines and protocols about how to deal with possible stray gas (methane) migration incidents. The Groundwater Protection Council also offers a very informative white paper developed from a forum on the subject. B.F. Environmental, Inc. offers very useful guidelines for water well owners for incident response and more long-term monitoring of methane levels in the water.

Water Well Construction

Water well construction can also be a factor influencing methane migration incidents. In northeastern PA many water wells penetrate the gas-bearing Catskill Sandstone formation and some penetrate part of the Lock Haven formation. They are often cased or lined shallower than they are drilled in order to be able to produce water from multiple horizons/aquifers. The casing is typically only partially grouted. Such open-hole completions are probably more amenable to both naturally occurring methane migration and incidents induced by drilling gas wells nearby. Since water wells in this area are typically drilled through the methane source it should be noted that the water wells themselves are likely to cause drilling-induced methane migration, possibly migration that is even more continuous and problematic than drilling induced migration from gas wells since the gas wells cement casing to surface after drilling. This could be a significant part of the so-called naturally occurring methane migration, although bubbling streams and flammable faucets have been documented in the area since the 1700’s. 

     

High Pressure Air Used in Drilling

Oil and gas drilling as well as water well and possibly even geothermal drilling may utilize air compressors (“drilling on air”) to push the cuttings away from the drill bit and up through the annulus (the area on the outside of the drill pipe) so that the cuttings return to surface and the hole is kept clear of cuttings. This process can also help move migrating gas, especially in the case of oil and gas drilling where the pressure may be boosted to high pressures and where the shallow part of the hole near the fresh water zones may be of a larger diameter. It has been noted that this may cause groundwater “surging” as high pressure may become trapped in the existing fracture networks of the aquifer. I am unsure if these processes have been altered to make methane migration during drilling less likely but I suspect some companies have addressed it. It has long been known that the high pressure air used in drilling can sometimes cause fluids to move laterally in very shallow zones that are very permeable, and possibly in unconfined aquifers close to discharge areas. A recent case of groundwater contamination occurred in north central PA in Potter County where isopropanol, a surfactant used as drilling soap to free cuttings and in this case to help free a broken drill bit, was illegally dumped down an open hole where freshwater aquifers were uncased. Water wells 800 ft. away detected the foaming agent within 2 days and 3 days later wells 9000 ft. away detected it. Ponds 2.8 miles away also detected it. This shows that these particular aquifers can transmit water quickly. It is not known if air pressure was used to help free the bit which may have helped move the water but it may have been a factor. A few weeks later the soap was found to have dissipated but nearby public water supplies were impacted (recharge areas were shut-in) and this was irresponsible on the part of the well operators. Mudlogging of the shallow vertical air drilling of the gas wells is an important technique for recording shows of gas and water and tying them to specific formations and stratigraphic intervals. Drilling personnel should also try to note any shallow gas or water shows in absence of a mud log.

Possibility of Using Pulverized Limestone or Some Other Agent to Seal off the Borehole When Drilling Surface Hole Through Fresh Water Zones and Gas Migration Source Zones to Prevent or Inhibit Gas from Migrating

Note: This idea is speculative. When monitoring gas through mud logging in air-drilled holes and some mud filled holes it has been noted that certain limestone formations will tend to pack off the borehole and suppress gas entry into the hole from above formations that were continuously entering previously at higher rates. This might offer an opportunity to reduce gas migration via drilling by introducing such a limestone component during drilling of the potential gas zones below the aquifers. I am unaware if such a method has ever been tried. The way to test it would be to note changes in background gas coming into the borehole. However, it may not work in the typically large diameter boreholes drilled through the aquifer through which surface casing is set. It also may be pressure-dependent and only occur in deeper higher pressure zones. Even if it only worked partially it could be an inexpensive and non-destructive technique for slowing any drilling induced migration that might occur.

Methane Contamination Studies Early in the Marcellus Play

The issue of gas migration in northeast PA was confounded by an early study by Duke University that tested some water supplies near early Marcellus drilling in 2011. There was not a lot of baseline water testing at the time but there was some. PADEP and Cabot Oil & Gas tested wells throughout Susquehanna County in 2009 and 2010 after methane migration incidents near Dimock. Baseline testing was limited in the Duke study. The study noted more methane in groundwater near where there was Marcellus drilling but also noted that it could be naturally-occurring methane as many knew was present at the time. However, the headlines ran with it and it became an often stated critique against fracking. Later in 2011 a study came out that utilized 1700 baseline water samples to show that methane was ubiquitous in the local groundwater (78% of samples had detectable levels of methane) and was unrelated to isotopic signatures of Marcellus gas so could not have migrated from the Marcellus due to hydraulic fracturing. This study noted that in Dimock Township in Susquehanna County there was no difference found between methane concentrations in gas drilling areas compared to non-gas drilling areas. The authors of the Duke study conjectured that it could be gas that leaked up from the Marcellus shale through the process of hydraulic fracturing. Geologically and technically, this was very unlikely and has since been disproven through gas isotope and compositional analyses. The source of the gas is shallow Upper Devonian and possibly Middle Devonian gas just below the freshwater aquifers. It is likely that gas in the Catskill and Lock Haven formations is self-sourced thermogenic gas from organic lenses within those formations. I would say that it was overly suggestive and perhaps a bit irresponsible of the Duke Study authors to suggest that the Marcellus could be the source of the methane via hydraulic fracturing and a bit suspicious as they were partially funded by anti-drilling interests.  A second Duke study came out in 2013 that also noted more methane in water wells proximal to Marcellus wells than in less drilled areas there in northeast PA. This time some of the findings were very likely due to methane migration caused by Marcellus drilling and increased in some cases by inadequate cementing as the wells were concentrated near wells with known inadequate cement jobs.  By this time the mechanisms for methane migration via drilling and inadequate cementing had become well established in the area.

 

The Importance and Necessity of Baseline Water Source Sampling Before Drilling Oil and Gas Wells

Most states that host oil and gas resources have developed regulations for pre-drill baseline water source sampling. The amount of samples required, the radial distances around the well, the timeframe before and sometimes after, the sampling and analysis methods, and the reporting requirements all vary by state. Some states, including Pennsylvania, have a presumption of liability/guilt, so that operators must test in order to prove non-contamination previous to drilling. If a potential incident were to occur, the well operator would be presumed to be guilty of causing the contamination if there was no baseline testing. Different states also require testing for different analytes. The most common water parameters tested are water quality (turbidity, total dissolved solids (TDS), total suspended solids (TSS), pH, etc), ions (cations and anions), inorganics and trace minerals, volatile organic compounds (VOCs), and dissolved gases. Certain VOCs like BTEX (benzene, toluene, ethylbenzene, and xylenes) and metals such as strontium and barium are typical of oil and gas contamination. However, they can also come from other sources than spills or leaks from the oil and gas industry. Barium and strontium are also associated with saline waters. Gas migration is typically just methane but thermogenic gas can contain trace amounts of ethane and propane. Biogenic gas is more likely to be entirely methane. Different states also have different thresholds to require further analysis. For instance, Pennsylvania requires isotopic analysis for any finding of methane above 7mg/liter. 28mg/liter is the saturation point for methane in water at surface pressures. It would be greater at higher pressures associated with greater depths. Above 28mg/liter at surface pressure, methane would begin to bubble out of solution.

Baseline water testing by oil and gas companies has also served to identify serious previously existing water quality issues. Many Pennsylvania water wells were found to have pre-existing problems such as iron bacteria, TDS, and other quality and contaminant issues.

 

Water Contamination Due to the Hydraulic Fracturing Process is Not Documented

Thus far, there are no confirmed cases of water contamination caused by the process of hydraulic fracturing itself and there are not likely to be due to the nature of subsurface fluid migration. Some have argued that high volume hydraulically fracturing more shallow oil and gas reservoirs has a higher chance of fracking into aquifers. There is some concern with this, however, shallower zones with less overburden have a tendency to frack horizontally, in so-called “pancake” fracks, due to prevailing rock stresses at these shallower depths. The danger could be where there are extensive existing fracture networks. Oil and gas companies have done multitudes of studies with microseismic technology in order to map where the induced fractures (and the proppant placed in them) go, so the industry has a pretty good idea about the hydraulic fracturing characteristics of different stratigraphic formations and different regions. A serious gas migration incident involving frothing water wells and one explosion occurred in 2007 in Bainbridge Township, Summit County, Ohio due to an inadequate cement job that was not properly remediated. Cement was lost in a fault and an important saltwater and gas-bearing zone was not properly sealed off. The problem was later resolved by a “squeeze job” where cement was re-introduced through perforated casing to seal off the zone. Some people thought that the gas migration was due to hydraulic fracturing since the well was fracked (with a low volume job) previous to the incident. Subsequent investigation proved conclusively that the problem was caused by the inadequate cement job. Studies of methane migration incidents in Pennsylvania, Ohio, Colorado, and Texas, have all concluded that the sources of methane were shallow reservoirs not impacted by hydraulic fracturing in deeper zones. One of the Duke Study authors, Vengosh, has argued that it is possible that Marcellus brines could have migrated up faults and fractures contributing to the saltwater and methane migrations and that this could lead to migration of gas and possibly frac fluids from the Marcellus through hydraulic fracturing. However, both regional isotopic and compositional gas studies and regional structural geology (Middle Devonian positioned detachment faults typically terminate at the top of the Middle Devonian) tend to refute this hypothesis. It has also been pointed out that the Marcellus is not saturated with brine in this area. Based on the evidence I would say the possibility is remote. The Duke Study authors noted the presence of thermogenic gas due to the detection of ethane and propane and that some water well samples near gas wells showed the possibility of being near Marcellus gas composition but others have attributed them to Upper Devonian and possibly Middle Devonian-sourced composition that was altered microbially during migration. They also offered an explanation that involved both faulty cement jobs and apparently faulty casing for the migration of methane into the annular spaces. While the inadequate cement jobs early in Marcellus development are implicated, the faulty casing scenario is unlikely as that is quite rare. Earlier this year the EPA completed a study that noted that there was no widespread contamination of drinking water due to oil and gas operations utilizing high volume hydraulic fracturing. In March 2015 a study came out headed by Donald I. Siegel of Syracuse University utilizing 11,309 baseline water samples collected by Chesapeake Energy that found no statistical association of higher methane concentrations in water wells more proximal to Marcellus gas wells as the Duke study had found. The author noted that the reason the Duke study found that association was simply that it utilized a small sample set centered in an area with known inadequate cement jobs early in the play. Penn State geologist Terry Engelder noted that the large dataset used in the Syracuse study along with the other studies has led to a much more detailed knowledge of methane migration and of groundwater geology in northeastern Pennsylvania.  A more recent limited study by Yale came to the same conclusion but also noted that the presence of minute quantities (ppb) below safe drinking water limits of VOCs and DROs (diesel range organic compounds) that are probably related to oil and gas spills of diesel and frac chemicals, known to have occurred in the area studied. The more radical anti-fracking environmental groups such as the Sierra Club (once less radical) and the EcoWatch blog have thus far refused to accept the findings of these studies and still contend that fracking contaminates groundwater. Of course, their definition of “fracking” includes accidents associated with oil and gas operations. Of course, no one disputes that these accidents have caused water contamination issues as have accidents associated with other industries.     

Conclusions

Drilling induced methane migration does occur in areas and situations with the right conditions. It is possible and important to take maximum precautions in order to prevent it or minimize it if it were to occur. Knowledge of the local and regional hydrology and shallow gas occurrence, baseline water testing, proper well construction and casing cementing formulas and procedures, possibly altering air pressure, possibly coating the borehole when drilling the zones in danger of migration, and having a clear plan and protocol for stray gas incidents seem to be the keys to preventing and minimizing any problems. It should be noted that groundwater protection in oil and gas operations is only possible if the well operators and contractors act responsibly. All should be knowledgeable of potential impacts and regulations regardless of how things were done in the past.            

References:

Recommended Practices: Responding to Stray Gas Incidents – by Marcellus Shale Coalition, October 16, 2012

Evaluation of Methane Sources in Groundwater in Northeastern Pennsylvania – by Lisa J. Molofsky, John A. Connor, Albert S. Wylie, Tom Wagner, and Shahla K. Farhat, in Groundwater, Vol. 51, No. 3, May-June 2013 (pages 333-349)

Regulations and Guidance for Baseline Sampling of Water Sources in Areas of Shale Oil and Gas Development – by Jenna Kromann, GSI Environmental, presented at RPSEA Onshore Technology Workshop, October 27, 2015

Methane in Pennsylvania Water Wells Unrelated to Marcellus Shale Fracturing – by L. Molofsky, J. Connor, S. Farhat, A. Wylie, and T. Wagner, in Oil & Gas Journal, December 2011, (pages 54-67)

Methane and Other Gases in Drinking Water and Groundwater – by Brian Oram, P.G., B.F. Environmental Consultants, Inc. and  Water Research Center, 2011

A White Paper Summarizing the Stray Gas Incidence and Response Forum – by John Veil of Veil Environmental, derived from the Stray Gas Incidence Response Forum by the Groundwater Protection Council, Cleveland, Ohio, July 24-26, 2012

Increased Stray Gas Abundance in a Subset of Drinking Water Wells Near Marcellus Shale Gas Extraction – by Robert B. Jackson, Avner Vengosh, Thomas H. Darrah, Nathaniel L. Warner, Adrian Down, Robert J. Poreda, Stephen G. Osborn, Kaiguang Zhao, and Jonathan D. Karr, in Proceedings of the National Academy of Sciences, 2013

The Environmental Costs and Benefits of Fracking – by Robert B. Jackson, Avner Vengosh, J.  William Carey, Richard J. Davies, Thomas H. Darrah, Francis O’Sullivan, and Gabriel Petron, in Annual Review of Environment and Resources, Vol 39, pages 327 – 362, October 2014

Methane Concentrations in Water Wells Unrelated to Proximity to Existing Oil and Gas Wells in Northeastern Pennsylvania – by Donald I. Siegel, Nicholas A. Azzolina, Bert J. Smith, A. Elizabeth Perry, and Rikka L. Bothun, in Environmental Science and Technology, Vol. 49, No. 7, pgs 4106-4112, March, 2015

Methane in Drinking Water, Unrelated to Fracking, Study Suggests – news story – by Eric Hand, in Science Insider (Science Magazine), March 30, 2015

Elevated Levels of Diesel Range Organic Compounds in Groundwater Near Marcellus Operations Are Derived From Surface Activities – by Drollette, etal, in Proceedings of the National Academy of Sciences, Vol. 112, No. 43, pgs 13184 – 13189, October 2015

Assessment of the Potential Impacts of Hydraulic Fracturing for Oil and Gas on Drinking Water Resources (External Review Draft) – by U.S. Environmental Protection Agency, EPA 600/R15/047, 2015 (June)

Yale Study Concludes Fracking Does Not Contaminate Drinking Water – news story by Andrew Follett, in The Daily Caller News Foundation, October 14, 2015.

Breaking: Oil & Gas Drilling Impacts Public Drinking Water Supplies in Potter County – news story in Public Herald, September 24, 2015.