Monday, 25 January 2016

Disposal of produced water during conventional oil extraction, and flowback fluid from fracking

The recent article “UK failing to learn U.S. lessons on fracking waste water” by O’Donnell, Gilfillan and Haszeldine (ODGH hereafter) is misleading, both in the way it describes events associated with fluid re-injection in the United States, and how it characterises the Environment Agency’s position with regards to the practice.  

ODGH describes at length the difference between produced water from conventional operations, and flowback fluid from shale gas hydraulic fracturing, and are correct to do so. However, under the terms of their own descriptions, the statements they then make using these definitions are false.

It is claimed that the increases in seismicity observed in Oklahoma are “linked to the subsurface re-injection of vast quantities of waste water from shale gas operations” (my emphasis added). This is not correct, and the USGS have been very clear in stating that this is a myth, propagated in the main by anti-fracking activists. In reality, the vast preponderance of wastewater being disposed of by deep injection in Oklahoma is produced water from conventional oilfields, and not hydraulic fracturing flowback fluids [1]. 

The same is true in California, where ODGH make claims of “environmental contamination due to leaks caused by poor borehole construction.” Firstly, we note that, as is the case in Oklahoma, the vast majority of such wells are disposing produced water from conventional oilfield operations, not flowback water from hydraulic fracturing of shale rocks. I also note in passing that the concerns raised regarding these wells were of an administrative nature, regarding how licenses were awarded and which formations did or did not qualify as protected aquifers. No actual environmental contamination is thought to have occurred: the State Water Resources Control Board found that “the injection wells have not degraded groundwater quality”[2], while CalEPA stated that: “To date, preliminary water sampling of select, high-risk groundwater supply wells has not detected any contamination from oil production wastewater.”[3] 

More importantly, I believe that through their selective quotation, ODGH have seriously misrepresented the position of the Environment Agency. Their full position on re-injection of flowback water in the document in question is quoted below [4]: 
Flowback fluid that cannot feasibly be re-used, is considered by us to be an extractive waste and may contain a concentration of NORM waste above the out of scope values. It will then require a radioactive substances activity permit for its disposal. You must send this to an appropriate permitted waste facility for treatment or disposal.  
The Environment Agency will generally not permit the re-injection of flowback fluid for disposal into any formation, whether or not it contains a concentration of NORM waste above the out of scope values. The re-injection of flowback fluid for disposal is not necessarily prohibited and may be permissible where, for example, it is injected back into formations from which hydrocarbons have been extracted and will have no impact on the status of water bodies or pose any risk to groundwater.  
The Environment Agency takes a precautionary approach to this activity and we do not consider it has been demonstrated that re-injection in these circumstances is BAT.
At present and in the absence of BAT being demonstrated we have determined that overall the long-term objective of ensuring good status of water bodies takes precedence over arguments in favour of the disposal of flowback fluid to underground formations.   
This is reinforced by our view that there are available and viable alternatives, namely disposal at permitted waste disposal facilities or by using onsite waste water treatment facilities. We consider that these techniques are a better environmental option.  
We will review this position in light of increased evidence from hydraulic fracturing operations and from the monitoring of underground waste facilities. 
The following image is also used to outline the EA position regarding flowback re-injection:


By neglecting to provide the full EA position, the ODGF give the impression that the EA are considering allowing “large volumes” of flowback re-injection in the UK. In fact, such disposal is very unlikely. However, the EA take the eminently sensible position that there may be exceptional circumstances under which disposal by re-injection does really present the optimum solution with respect to minimizing the risk of environmental contamination. I can’t think of any such conditions at present, but I think that it’s a wise position for a regulator to consider every individual case on its merits, and to then take the best available route to ensure environmental protection, rather than imposing an arbitrary blanket ban. Regardless, disposal of flowback at treatment facilities is considered BAT under usual circumstances, and subsurface re-injection is not, and therefore it is very unlikely to be allowed in the majority of cases. 

Importantly, even given the above position, the EA statement suggests that flowback is to be “injected back into formations from which hydrocarbons have been extracted”. ODGH appear to have missed the significance of this statement, because it renders most of their remaining arguments invalid. If re-injection is limited to formations from which hydrocarbons have already been extracted, then formation pressures will have been reduced. While their analogy with a clay-covered balloon is an interesting metaphor, it is not relevant to the situation at hand, if, as stated, flowback disposal were only allowed into formations that had already seen hydrocarbon production and pressure drawdown (if flowback re-injection is ever allowed, which as discussed above, is unlikely).

It is also worth considering the volumes involved. The existing onshore conventional industry in the UK disposes of approximately 12 million cubic metres of produced water every year. The Institute of Directors forecasts a UK shale industry developing, between now and 2030, 100 pads with 40 multilateral wells each, with each pad using 0.5 million cubic metres of water[5]. Assuming that 50% of this fracturing fluid flows back, a total of 27 million cubic metres of flowback will need to be disposed of during this period. In comparison, as stated above, the conventional industry will dispose of as much produced water every two-and-a-bit years. 

ODGH have mischaracterized hazard with risk. Hazard is the potential impact of a pollutant, while risk is the hazard posed multiplied by the probability of exposure. It may be true that flowback fluids have different chemical signatures to conventional produced waters that may make them more hazardous (although in reality the chemical signature of every fluid, whether flowback or conventional, will be specific to its geographic locality and rock formation, so the comparison provided by the authors is likely to be an oversimplification). 

However, no evidence is provided to suggest that groundwater contamination is more probable during flowback re-injection (nor have they provided evidence of such contamination from the USA). In fact, given that the volumes of flowback disposal are likely to be far lower, the likelihood of contamination from conventional produced water might be considered higher simply from a volumetric argument. 

With or without NORM, produced water from conventional oil production is usually hypersaline, and its spillage into freshwater sources would represent, I would expect, a significant environmental incident. The public can and should take reassurance from the fact that the existing onshore industry has been able to dispose of very large volumes of such fluid without causing environmental impacts. The risk of contamination, by either produced water or flowback fluid, would appear therefore to be low.  

I will close by pointing out that the authors of this piece are keen promotors of carbon capture and storage (CCS) for greenhouse gas emissions mitigation. I share their belief in the potential this technology; I believe that without it, the UK will not be able to meet its emissions targets; and I share their dismay at recent government decisions to shelve the proposed demonstration projects. However, many of the issues raised here by ODGH are in fact far more applicable to CO2 storage than they are to flowback disposal for re-injection: the injected fluid is in chemical dis-equilibrium with the in situ geological formation – indeed CO2 will dissolve into formation waters to create a mildly corrosive acid, which is not a problem for flowback or produced water re-injection; the CO2 will likely be stored in saline aquifers that have not been depleted, leading to the same issues with pressure increases and potential seismicity described by ODGH (the clay-covered balloon), only the volumes of CO2 to be stored far exceed the volumes of flowback typically re-injected[6], which will exacerbate the risk of larger seismic events. I would appreciate it if the authors could outline in more detail how it is that they can be fully supportive of CCS development, and yet believe these same issues pose a major issue with respect to the re-injection of other waste fluids.  

[1] Rubinstein et al., 2015. Myths and facts on Wastewater Injection, Hydraulic Fracturing, Enhanced Oil Recovery, and Induced Seismicity. Seis. Res. Letts.. 
[2] State Water Resources Control Board September 23, 2014, Item 13 - Executive Director’s Report. 
[3] CalEPA Memorandum, 2015. CalEPA Review of UIC Program. 
[4] Environment Agency, 2015. Onshore Oil & Gas Sector Guidance Consultation Draft, November 2015. 
[5] Institute of Directors, 2013. Getting shale gas working.
[6] Verdon et al., 2014. Significance for secure CO2 storage of earthquakes induced by fluid injection: Env. Res. Letts..