These are my notes from a seminar I attended last night. Anyone interested in shale gas and fracking should find them of interest. The event was recorded, and will hopefully arrive online in the near future.
Speaker: Richard Davies, Professor of Energy at Durham University, and Director of the Durham Energy Institute
Title: Risks and realities: shale gas, geology, and public opinion
‘Fracking’ is the process of induced hydraulic fracturing to extract methane gas from sub-surface shales. Recently, this process has been adopted by companies in the UK such as Cuadrilla, but the techniques have been met by public, media, and professional opposition. This is partly due to previous issues such as groundwater contamination in the US, and microseismicity induced by Cuadrilla in Lancashire when conducting a pilot test. What has become apparent is the lack of effective regulatory regimes, and their management, as well as a general paucity in empirical backing to the processes and risks. These issues need to be addressed before further shale gas exploration and extraction is conducted.
Davies chose to focus his talk on addressing two of the major concerns related to shale gas extraction: groundwater contamination, and microseismicity. These have received the most attention in the media, and by the people who oppose shale gas extraction. He also emphasized that ‘fracking’ should stop being used for an all-encompassing term for all methane-based processes; it refers purely to the process of hydraulic fracturing.
In Pennsylvania (USA), the issue of methane-contaminated groundwater is quite prevalent, and has been ascribed to methane released from hydraulic fracturing. This has caused major health issues, but the attribution of the contaminants to shale gases was never actually tested before the media storm that built groundwater contamination into a major weapon for the opposition.
During the fracking process, a secondary well is used to house a seismic detector that monitors the fracture process, and the extent and magnitude of the fractures. What has been found is that microseismicity induced by fracturing is actually confined to shale horizons, and is largely constrained by the overlying and underlying lithology. Considering that wells are drilled to shale horizons that are many thousands of meters underground, faults would have to propagate a substantial distance to penetrate near-surface water aquifers. This is not found by empirical studies; there is a substantial buffer zone between the zones of fracturing and any aquifers, although the distance between is highly case dependant (based on the internal structure of overlying lithologies).
Davies explained a probability-based model that had been designed to predict the maximum extent of fractures, based on sampling the fracture extent from five fracturing regions. Simulations of this model were executed using existing published data from a large number of potential fracking sites, and it was found that even for the largest fracture systems (that were created on a geological time scale, with enormous amounts of water acting as the fracture mechanism) the probability of fractures penetrating aquifer zones was less than 1%. However, this was a model based entirely on wells in the USA, so may not be applicable to sites in the UK, where lithologies and stress regimes may substantially differ.
This does provide pretty solid evidence against fracking-induced methane release as a contaminant in groundwater. However, in Pennsylvania, fracking procedures were often conducted at shallow levels, which may have been close enough to groundwater supplies to contaminate them. Furthermore, there are 183,000 wells that pre-date records in Pennsylvania, any of which could potential act as conduits for natural methane movement. Natural seeping of biogenic and thermogenic methane is commonplace, therefore it is not unreasonable to suggest this as an alternative cause. The issue then becomes that sloppy regulation and management led to poor sealing of the wells (through cementation), which allowed leakage to occur. This has actually led to several companies being sued for poor practice by the Environmental Agency in the US. So the issue is not that fracking and shale gas extraction contaminate groundwater, it’s that poor regulatory regimes and relaxed management have led to poor practices that ultimately have caused damage in Pennsylvania.
However, these are things to learn from, not to assume that they will be mimicked. The empirical analyses suggest caution must be applied, and it’s simply a case of employing this. The take-home message is that the data implies methane contamination was not the result of shale gas extraction processes. To maintain this, and avoid issues such as that in Pennsylvania, an empirical regulatory threshold must be set on the fracture zone to aquifer distance. If properly managed, groundwater contamination is not an issue.
Every fracture created in the Earth’s crust is a microseismic event. Based on the Moment Magnitude Scale (calculated using the seismic moment of earthquakes, as opposed to the Richter Scale which is a measure of the amplitude on seismographs), the biggest fracking-induced event was measured at 0.75, a magnitude that cannot be felt by humans. Seismic events of this magnitude occur naturally, and frequently. Events of larger magnitude occur less frequently (there is a proportional relationship between occurrence frequency and magnitude), but are still natural. Induced seismicity, however, is involved in numerous processes (e.g., hydrocarbon extraction, geothermal energy, mining), and is a well-understood and well-managed aspect of the industry.
The issue surrounding this is based on a recent seismic event, related to a pilot drill program in Lancashire conducted by Cuadrilla. A seismic event measuring of 2.3M occurred at the same time as the fracking operations, and naturally cause and effect was established. However, using a rather rudimentary system of 2D acoustic imaging, it was determined that the borehole did not drill a fault, which raises the question of where did the seismicity come from. When considering the magnitude, there is a power law relationship between the ‘levels’ – the event related to Cuadrilla was in fact approximately 30,000 times more powerful than the majority of fracking-induced earthquakes in the US (maximum 0.75M). So were Cuadrilla just unlucky? The jury is still out on that one, but in the mean-time a traffic-light system has been established, whereby certain protocols must be followed if certain magnitudes are reached by hydraulic fracturing. An initial threshold for ‘stop and shut down all processes’ was set as 1.7M. DECC say that a 0.5M threshold should be used. However, these thresholds are somewhat arbitrary and not based on any empirical analysis. A recent consultation by DECC hopes to provide some clarity to this system.
Davies mentioned that he had suggested several safeguards in his response to the consultation. These included photographing the borehole to detect any faults, and if any were found to immediately cancel all planned fracking. He highlighted that, with respect to the recent consultation document, there are still many critical questions or points that remain unanswered. These must be considered if any effective regulatory framework based on empirical principles is to be established, which would both engender public confidence, and enable fracking processes to be efficiently conducted and managed.
- More data is required on fracture height. The heterogeneous behavior of rocks within and between lithologies means that fracture extent needs to be considered on a case-by-case basis. This data is held by companies such as Schlumberger, and needs to be integrated into future models to provide an empirical basis for fracture prediction.
- Before claims of groundwater contamination as the result of gas extraction are made, background chemical levels must be known. This way, attribution of potential contaminants to fracking-induced methane seepage can be confidently made. A database for all relevant aquifers is needed.
- The reasons behind exceptional seismic events (such as that in Lancashire) are still not understood. Explanations are required.
- Fracking fluids left behind need managing in some way. Methods of transport and disposal are still not established.
- Is there any subsidence related to fracturing, and if so, how much? Potential processes like this need to be modeled and considered.
- What are the methods for ‘flowback’ disposal? About 40% of the fracking fluids return to the surface, which may have low levels of radioactivity and other harmful chemicals. No regulation currently exists in dealing with this, in terms of cleaning or disposal.
Davies ended the talk on a positive note. In the USA, shale gas extraction has led to cheaper energy and the generation of thousands more jobs. The greatest issue now is actually over-production, making shale gas non-economical. In the UK, despite its relative size, there are numerous basins hosting shale-horizons (e.g., the Cleveland Basin, Yorkshire). Constrains on extracting gas from though, apart from the issues mentioned above, include the relative density of urban populations compared to that of the USA. Social impact must be considered before drilling and extraction. However, the low cost of shale gas would be a great economic boost in the UK (most gas currently is imported from a single line from Norway). An issue may be that shale gas prospecting and extraction is moving much faster in several Eastern European countries (Bulgaria, Poland, Romania, and Ukraine) backed by companies such as Chevron and ExxonMobil, due to the relatively low social impacts, and the desire to move dependency away from Russia as primary gas supplier. Davies finished by stating that the public should not get caught up in the current bipolar media frenzy, and let the evidence speak for itself.
Many good questions were asked during the subsequent Q&A session.
What is keeping the frenzy going when the evidence for shale gas extraction is so overwhelming?
Misinformation and politicians playing to the crowd. Uncertainty about issues is enhanced by the media. We need openness, and reassurance about the risks and their management. Companies must be open with their data, otherwise this fosters mistrust, which is something already plaguing the hydrocarbon industry.
How commercially viable is shale gas in the UK?
More data and more wells are needed to assess the shale gas potential in the UK. We don’t have the same land extent or population density as the USA, and also have restrictions based on planning policy.
What are the effects of bounding layer variation on dampening seismic activity?
More data is required for aspects such as Young’s Modulus and Poisson’s Ratio for the total range of lithologies. Pilot testing is needed to look for permeable beds and faults and fractures that have a strong effect on seismic activity.
What will be the social impact in Eastern Europe?
Although things like fugitive emissions can occur, these can be managed, and aspects like road damage can be repaired if needed. Most issues so far have arisen from poor regulation and management, so any success criteria will be based on these.
An event hosted by the Geological Society of London will continue a similar theme of talks, looking purely at the geoscience behind shale gas extraction. This will be publicly webcast, and recorded with videos made available shortly after.