Marine and Petroleum Geology 43 (2013) 519e521

Contents lists available at SciVerse ScienceDirect

Marine and Petroleum Geology
journal homepage:


Reply: Davies et al. (2012), Hydraulic fractures: How far can they go?
Richard J. Davies a, *, Gillian R. Foulger a, Simon Mathias a, Jennifer Moss b, Steinar Hustoft c, Leo Newport a

Durham Energy Institute, Department of Earth Sciences, Durham University, Science Labs, Durham DH1 3LE, UK 3DLab, School of Earth, Ocean and Planetary Sciences, Main Building, Park Place, Cardiff, University, Cardiff CF10 3YE, UK c University of Troms?, Department of Geology, Dramsveien 201, N-9037 Troms?, Norway

a r t i c l e i n f o
Article history: Received 26 January 2013 Accepted 2 February 2013 Available online 24 February 2013

1. Introduction Davies et al. (2012) measured the heights of stimulated and natural hydraulic fractures caused by high ?uid pressure from eight sedimentary successions from around the world. They found the tallest natural hydraulic fractures to be w1133 m in height and the tallest upward propagating stimulated hydraulic fractures, generated by fracking operations for gas and oil exploitation to be 588 m in height. This provided a rationale for an initial, safe separation distance of 600 m between aquifers and the deeper shale gas and oil reservoirs where hydraulic fractures are being stimulated. Three months after the paper went online, Geiser et al. (2012) published a new method, tomographic fracture imaging, which potentially detects the movement of a ?uid pressure pulse in pre-existing natural fracture systems located close to where stimulated hydraulic fractures are forming. These fracture systems are not necessarily natural hydraulic fractures, but could be joints and faults formed due to folding or faulting. They found the maximum vertical extent of these to be w1000 m. Here we respond to the comment made by Lacazette and Geiser (2013) and consider the implications of the new ?ndings of Geiser et al. (2012) for the conclusions we made (Davies et al., 2012). 2. The hydraulic fracturing controversy Hydraulic fractures are stimulated to increase the rate of ?uid ?ow from low permeability oil and gas reservoirs (e.g. shale). The

aim of Davies et al. (2012) was to test the hypothesis that hydraulic fracturing has caused methane contamination of drinking water in the USA and to provide an evidence base for the safe vertical separation distance between shale reservoirs and aquifers. The contamination hypothesis was explicit in the title of the Osborn et al. (2011) paper ‘Methane contamination of drinking water accompanying gas-well drilling and hydraulic fracturing’ and popularised by the 2010 ?lm ‘Gaslands’. The approach adopted by Davies et al. (2012) was entirely empirical and based upon measuring the heights of natural and stimulated hydraulic fractures. We did not consider the vertical extent of fractures unrelated to pore pressure caused by tectonic stresses exceeding the tensile strength of the rock. Also for the stimulated hydraulic fractures we relied upon the microseismicity measurements of Fisher and Warpinski (2011). From this database of thousands of the tallest hydraulic fracture systems, we derived probability of exceedance plots for hydraulic fracture heights. These provide a range of probabilities of natural and stimulated hydraulic fractures extending vertically beyond speci?c distances. The results indicated that no stimulated hydraulic fractures heights measured using microseismicity and published by Fisher and Warpinski (2011) propagated upwards past 588 m in height and the chances of an arti?cially stimulated hydraulic fracture propagating vertically past 350 m was only 1%.

3. Is a 600 m vertical separation distance safe? Davies et al. (2012) was purely statistical and therefore blind to factors such as local geology and operational variables such as the volume of fracturing ?uid used which would need to be considered for safe operations at a speci?c site. If the geology of a region where

DOI of original article: * Corresponding author. Tel.: ?44 1913342346. E-mail address: (R.J. Davies). 0264-8172/$ e see front matter ? 2013 Elsevier Ltd. All rights reserved.


R.J. Davies et al. / Marine and Petroleum Geology 43 (2013) 519e521

hydraulic fracturing is carried out is characterised by evidence for vertically extensive ?uid ?ow (e.g. mud volcanoes which can extend vertically for >>1 km), then this introduces a signi?cant risk that there are open pathways for ?uid migration. But there may also be natural barriers to fracture propagation, known as ‘frack barriers’, which could limit the extent of fractures so that the tallest fractures are <<600 m. Lacazette and Geiser (2013) in their comment propose that ?uid pressure pulses triggered by hydraulic fracturing move vertical distances of w1 km through pre-existing natural fracture systems, hundreds of metres further than the maximum propagation

distance for stimulated hydraulic fractures (Fisher and Warpinksi, 2011; Davies et al., 2012). This is detected using a new tomographic fracture imaging method (Geiser et al., 2012). The work of Davies et al. (2012) remains valid as a statistical analysis of stimulated hydraulic fracture height measurements derived using microseismicity. But this avoids the important question; does the new tomographic fracture imaging method reveal pre-existing fractures, not necessarily generated by natural hydraulic fracturing, that allow for a far more vertically extensive transmission of fracking or pore ?uid? If so what are the implications? The new method is a passive seismic monitoring technique which may detect energy released as a result of the transmission of ?uid pressure pulses. The method assumes that energy emission is linearly related to the sum of the area of failure over time and that regions of highest crack density have the highest semblance value. They also state that they use a summation method to capture a greater fraction of the acoustic energy generated by fracturing, allowing imaging of very weak activity. In the form the method is presented by Geiser et al. (2012), there are three shortcomings. Our ?rst concern is that perhaps because of proprietorial reasons, the exact work?ow they use to detect this pressure pulse is not described in detail. Secondly, they are unclear on the exact physical process that is potentially being detecting. Geiser et al. (2012) hypothesize that it may be some sort of the Biot ‘slow wave’ (Biot, 1962). Lacazette and Geiser (2013) propose that two processes are potentially operating, the transmission of a ?uid pressure pulse in the fracture due to its direct connection with fracking, and coupling of stress in the rock matrix by in-situ ?uid. Thirdly, although they document some validation of their method (e.g. using boreholes which detect

Figure 1. Approximate maximum vertical extent of ?uid transmission in natural fracture systems. (a) Fractures, faults and hydraulic fractures normally located within the crest of anticlines can allow ?uid ?ow in mud volcano systems (Kopf et al., 2003; Davies and Stewart, 2005; Stewart and Davies, 2006). Fluid ?ow may be in stages to intermediate ?uid reservoirs and the ?uid has been traced to reservoirs >2 km in depth (Kopf et al., 2003); (b) injectites are thought to extend a maximum of up to w1 km, form due to hydraulic fracturing and the remobilisation of sand, driven by overpressure (Hurst et al., 2011); (c) chimneys or pipes are probably clusters of hydraulic fractures imaged with seismic re?ection data (L?seth et al., 2001; Hustoft et al., 2010; Moss and Cartwright, 2010).

Figure 2. Potential maximum vertical extent of ?uid transmission and ?uid pressure pulse transmission related to fracking operations. (a) and (b) ?uid pressure pulses may be transmitted through pre-existing fracture systems of 1 km in vertical extent (Geiser et al., 2012); (c) stimulated hydraulic fractures may extend for w 600 m vertically (Fisher and Warpinski, 2011; Davies et al., 2012). The vertical scale is indicative as the depth of the shale reservoir varies.

R.J. Davies et al. / Marine and Petroleum Geology 43 (2013) 519e521


fractures located in similar positions to those imaged), more validations need to be published before the method is fully substantiated. Despite these issues, the method and results are potentially a very signi?cant addition to existing seismic approaches used to monitor fracking operations. If the method performs well it will extend the ability of passive seismic monitoring to map fractures activated over time-scales longer than the nucleation time of stimulated hydraulic fractures. It may dramatically improve our understanding of the extent of pre-existing fracture systems and ultimately verify whether fractures allow ?uid transmission to shallower levels than previously thought possible, over human time-scales. 4. Implications for the protection of water supplies It has long been know that fracture systems of 1000 m extent occur in sedimentary rocks (L?seth et al., 2001) and Davies et al. (2012) showed that three-dimensional seismic data can image natural hydraulic fractures that extend this far. If we assume fractures (hydraulic or otherwise) are also being imaged by the tomographic fracture imaging approach then the key question is whether they remain open after the fracking operations to enable the ascent of ?uid. Con?ning stresses would cause fractures to close-up when pumping stops and the pressure in the ?uid drops so a system of open fractures to shallow levels is dif?cult conceive. It would require there to be sedimentary strata at the level of the reservoir that are permeable and natural ovepressure that keeps the conduits open and active. But we cannot be certain that there are no permeable routes through pre-existing fracture systems. It is important to state that after thousands of fracking operations, there are no proven examples of contamination of drinking water aquifers due to hydraulic fracturing. But we take the opportunity to incorporate the new measurements of Geiser et al. (2012) in a new summary diagram of the heights of fractures potentially stimulated by the ?uid injection during hydraulic fracturing operations (Fig. 1). We also provide the maximum heights for a range of natural vertical ?uid ?ow pathways, which include hydraulic fractures and other routes, such as joints and faults (Fig. 2).

The new work of Geiser et al. (2012) highlighted in the Lacazette and Geiser (2013) comment shows that consideration of local geology and speci?cally the existance of vertically extensive fracture systems are important parts of risk assessments prior to fracking operations. In this reply we have taken the opportunity to include the potential existance of other types of vertically extensive fracture system evidenced by active mud volanoes and seeps which were not described in our earlier paper (Davies et al., 2012). Such phenomena are easily recognised and their locations are well known so they can readily be avoided.

Biot, M.A., 1962. Mechanics of deformation and acoustic propagation in porous media. Journal of Applied Physics 33, 1482e1498. Davies, R.J., Stewart, S.A., 2005. Emplacement of giant mud volcanoes in the South Caspian Basin: three-dimensional seismic re?ection imaging of root zones. Journal of the Geological Society 162, 1e4. Davies, R.J., Mathias, S.A., Moss, J., Hustof, S., Newport, L., 2012. Hydraulic fractures: how far can they go? Marine and Petroleum Geology 37, 1e6. Fisher, K., Warpinski, N., 2011. Hydraulic Fracture-Height Growth: Real Data, SPE 145949. Geiser, P., Lacazette, A., Vermilye, J., 2012. Beyond ‘dots in a box’: an empirical view of reservoir permeability with tomographic fracture imaging. The Leading Edge 30, 63e69. Hurst, A., Scott, A., Vigorito, M., 2011. Physical characteristics of sand injectites. Earth-Science Reviews 106, 215e246. Hustoft, S., Bünz, S., Mienert, M., 2010. Three-dimensional seismic analysis of the morphology and spatial distribution of chimneys beneath the Nyegga pockmark ?eld, offshore mid-Norway. Basin Research 22, 465e480. Kopf, A., Deyhle, A., Lavrushin, V.Y., Polyak, B.G., Gieskes, J.M., Buachidze, G.I., Wallmann, K., Eisenhauer, A., 2003. Isotopic evidence (He, B, C) for deep ?uid and mud mobilization from mud volcanoes in the Caucasus continental collision zone. International Journal of Earth Sciences 92, 407e425. Lacazette, A., Geiser, P., 2013. Comment on Davies et al., 2012 e Hydraulic fractures: how far can they go? Marine and Petroleum Geology 43, 517e519. L?seth, H., Wensaas, L., Arntsen, B., Hanken, N., Basire, C., Graue, K., 2001. 1000 m Long Gas Blow Out Chimneys. 63rd EAGE Conference and Exhibition, Extended Abstracts, 524 pp. Moss, J.L., Cartwright, J., 2010. 3D seismic expression of km-scale ?uid escape chimneys from offshore Namibia. Basin Research 22, 481e501. Osborn, S.G., Vengosh, A., Warner, N.R., Jackson, R.B., 2011. Methane contamination of drinking water accompanying gas-well drilling and hydraulic fracturing. Proceedings of the National Academy of Sciences 108, 8172e8176. Stewart, S.A., Davies, R.J., 2006. Structure and emplacement of mud volcano systems in the South Caspian Basin. AAPG Bulletin 90, 753e770.


Computer simulation of hydraulic fractures
Multiple Hydraulic Fractures in Horizontal Wells
IPTC-12303-MS-P Application of Horizontal Wells with Multiple Hydraulic Fractures
The Influence of Natural Fractures on Hydraulic Fracture Propagation
Discrete Fracture Network Models of Natural and Hydraulic Fractures
Morphology and Propagation of Hydraulic Fractures for CBM Wells
Evaluation of a downhole tiltmeter array for monitoring hydraulic fractures