OR/15/066 Shale variability

Cuss, R J, Wiseall, A C, Hennissen, J A I, Waters, C N, Kemp, S J, Ougier-Simonin, A, Holyoake, S, and Haslam, R B. 2015. Hydraulic fracturing: a review of theory and field experience. British Geological Survey Internal Report, OR/15/066.

In this chapter we discuss the variability that is inherent in shale formations. Shale is a fine-grained sedimentary rock that constitutes approximately half the geological column (Spears, 1980) and is the most abundant geological rock type present in sedimentary basins worldwide (Meissner, 1986). Few geological rock types encompass such variability and as a result shale successions will have considerable differences in sedimentology, organic content, gas content, and strength properties within individual facies. Thus, shale has been used as a group name for all fine-grained sediments (Spears, 1980). The Dictionary of Geological Terms published by the American Geological Institute (Bates & Jackson, 1984) defines shale as:

“A fine-grained detrital sedimentary rock, formed by the compaction of clay, silt, or mud. It has a finely laminated structure, which gives it a fissility along which the rock splits readily, especially on weathered surfaces. Shale is well indurated, but not as hard as argillite or slate. It may be red, brown, black, or gray.”

Even this simplistic definition hints at considerable variation based on visual appearance. In this chapter we will highlight that this variation occurs not only over the geographical extent of a basin and between basins, but also on small distances within the geological succession.

It is outside of the scope of this report to review all of the European potential shale gas basins and to compare these with North American equivalents. Instead, we highlight the variability seen within two boreholes and two field outcrops from the United Kingdom, highlighting variability that will be significant for hydraulic fracturing.

Carsington dam reconstruction C4 borehole, UK

The Carsington Dam Reconstruction C4 borehole, in Derbyshire (UK), was studied extensively by Könitzer (Könitzer, 2014; Könitzer et al., 2014) and is part of on-going research at the British Geological Survey. This shallow borehole (55.25 m deep) was drilled as part of engineering works at Carsington Dam. The borehole intersects lithofacies of organic-rich lower Namurian (Serpukhovian) mudstones from the Widmerpool Gulf, one of several confined early Carboniferous basins in the Pennine Province of the UK. A cored section of 40 metres of Arnsbergian sediments was studied in detail.

Figure 1    Sedimentary log of the Carsington Dam Reconstruction C4 borehole, Derbyshire (UK).

Figure 1 shows the sedimentary log of the Carsington Dam Reconstruction C4 borehole. There is considerable variation in lithology over small vertical distances. In a broad sense, the borehole has clay-dominated mudstone, silty mudstone, siltstone, intercalated siltstone & sandstone, sandstone, limestone and coalified layers. Könitzer (Könitzer, 2014; Könitzer et al., 2014) details 7 facies in the sequence; 1) thin-bedded carbonate-bearing clay-rich mudstones; 2) calcareous mudstones; 3) lenticular clay-dominated mudstones; 4) thin-bedded silt-bearing clay-rich mudstones; 5) thick-bedded graded silt-bearing mudstones; 6) sand-bearing silt-rich mudstones; and 7) plant-debris and sand- bearing mudstones. Facies 4 was sub-divided into; a) lenticular thin-bedded silt-bearing mudstones; b) homogeneous thin-bedded silt-bearing mudstones; and c) organic-rich thin-bedded silt-bearing mudstones.

Within the individually identified facies, considerable variation in total organic carbon (TOC) was observed. For facies 1 to 7 the TOC was 2.4–6.6, 0.3, 1.9–4.5, 0.9–4.1, 0.9–4.1, 0.4–2.8, and 7.1–9.7% respectively. This shows that facies 2 (calcareous mudstone) has a very low TOC, whilst facies 7 (plant-debris and sand-bearing mudstone) has the highest TOC.

The main observation from the Carsington C4 Borehole that is relevant to the current study is the considerable variability seen vertically within a 40 metre sequence of shale. This sequence includes siltstone, mudstone, sandstone, limestone, and coal. The variation is seen on the centimetre and sub-centimetre scale. It should also be noted that Könitzer et al. (2014) report considerable variation in lithology thicknesses between the Carsington Dam Reconstruction boreholes C3 and C4, which were separated by less than 50 metres. This shows that not only does shale vary vertically with depth, but does so laterally.

An important consideration is that seismic resolution is often estimated to be 10–20 m in ideal conditions. Therefore in twice seismic resolution (i.e. 40 m), 7 clear facies and multiple layers of geological variation can occur.

Roosecote-1 borehole, UK

The Roosecote-1 Borehole has been studied for variations in physical, mineralogical, and chemical properties at the British Geological Survey. The 800.88 metre deep (TD) borehole is located approximately 3 km to the south-east of Barrow-in-Furness, Cumbria (UK). It was drilled in 1970–71 as an Institute of Geological Sciences stratigraphic borehole. The borehole proved the succession from the Quaternary and bottomed in Lower Carboniferous limestones, and importantly is a defined stratotype section for the Bowland Shale Formation (Dean et al., 2011). The borehole was drilled in the Lancaster Fells Basin, a small basin located in the northern part of the main Craven Basin that is defined by the Lake District Block to the north, and the Bowland High (separating it from the Bowland Basin) to the south. The borehole was fully cored through the Bowland Shale succession, although much of the core was disposed of following palyntological analysis, leaving short core samples typically 5–20 cm long spaced at metre intervals throughout.

Table 2    Sedimentary log of the Roosecote-1 borehole, Cumbria (UK)

	Description	Thickness (m)	Depth (m)
Namurian	Siltstone to coarsely silty mudstone, dark grey, micaceous, and sandstone pale grey, fine- to medium-grained; interlaminated and interbedded in five major upward fining cycles based at depths of 199.00, 278.68, 326.71, 382.34 and 491.68 metres respectively. Sandstone beds usually predominate in the lower parts of each cycle and often show graded bedding and sharp bases with directional or organic sole structures. A few mudflake conglomerates and chaotically laminated slumped beds are also present. Macrofossils are restricted to rare fish scales and bivalves in the finest lithologies but finely comminuted plant debris is generally abundant. Traces of gaseous oil from 465 m to 487 m	333.55	491.68
	Mudstone, dark grey, silty, with a few siltstone laminae and ferruginous bands	29.80	521.48
	Mudstone, dark grey, slightly calcareous; goniatite/bivalve fauna representing the Cravenoceras malhamense Marine Band	7.59	529.07
	Mudstone, dark grey, silty, with ferruginous bands	23.93	553.00
	Mudstone, dark grey, silty, slightly calcareous; indeterminate marine faunas	2.02	555.02
	Mudstone, dark grey, Silty with fish debris	28.98	584.00
	Mudstone, dark grey, slightly calcareous; marine fauna representing the Eumorphoceras pseudobilingue Marine Band	3.66	578.66
	Mudstone, dark grey, sporadically calcareous, poorly fossiliferous	14.84	608.30
	Mudstone, dark grey, calcareous; goniatite/bivalve fauna representing the Cravenoceras leion Marine Band	5.80	608.30
Visean	Mudstone, dark grey, very silty, micaceous	5.01	613.31
	Limestone, dark grey, very finely granular, bituminous, interbedded with dark grey mudstone; dispersed fine crinoidal debris; 2 mm green mudstone band at 615.59, apparently eroded limestone bedding surface at 616.20; 17 cm bed of conglomeratic mudstone at base	6.38	619.69
	Limestone, coarsely granular, pyritic matrix	0.37	620.06
	Limestone, dark grey, well bedded, finely granular, with dark grey or black mudstone partings every 10–50 cm; bands and nodules of black chert common; thin bands of greey pyritous mudstone at 692.30, 695.24, 704.05 and 704.91 m respectively; very poorly fossililferous except for a 2.27 m bioclastic sequence at 682.94 m with a few indeterminate brachiopod shells and Zaphrentoid corals of probable P2 age	97.94	718.00

The preliminary sedimentary log (Table 2) produced at the time of drilling describes the detail for the Carboniferous (Namurian and upper part of Visean) part of the succession. The sedimentary log shows that a range of geological lithologies were observed within the shale formation; including siltstone, sandstone, mudstone, limestone, and nodules of chert.

Figure 2    Results of XRD and TOC (Rock-Eval) analysis on samples taken from Roosecote-1; a) Whole rock analysis; b) <2 µm fractions.

Figure 2 shows the results of X-ray diffraction (XRD; Cave et al., 2013; Kemp et al., in prep) and total organic carbon (TOC) as determined by Rock-Eval pyrolysis (Hough et al., 2014) from 14 samples taken along the Roosecote-1 borehole. As can be seen, considerable variability in mineralogy is seen for the bulk-rock along the sequence studied from 472–661 metres depth. Quartz content, for instance, varies significantly between 2.6 and 70.8%, while siderite is very low or below detection limits in all but one sample where it accounted for 72.9%. Considerable variation is also observed in the clay content, with illite/smectite ranging from 29–86%. Variability is observed in TOC in the organic rich shale units, which ranged between 1.76 and 3.72%, with a mean of 2.6%. Certain intervals had very low TOC readings.

The Roosecote-1 borehole shows that within a 190 metre sequence of shale a range of lithologies are observed including siltstone, sandstone, mudstone, limestone, and nodules of chert. This is reflected in the mineralogy measured using XRD on the bulk rock, and also on the clay content obserevd on fractions of less than two microns. Variation is also seen in the TOC, showing that certain facies will not be propspective.

Mam Tor and Edale outcrops, UK

In this section we describe variations seen in shale at outcrops in the UK. Figure 3 shows an exposure that clearly shows variation in lithology in the dipping shale sequence at Mam Tor, Derbyshire (UK). The dark-grey shales include harder beds, these are turbidite sandstones and include some ironstones. This photo clearly shows variation over a sequence of about 4 metres. Note also that a close-spaced joint development is present within the harder lithologies.

A finer-scale variation in shale is shown in Figure 4 and Table 3. This example was observed at Edale in Derbyshire (UK) and represents 8.4 metres of the shale succession from the Bowland Shale Formation. The pale layers seen in Figure 4 are much harder ironstone bands and lenses. The sedimentary log (Table 3) shows a range of lithologies, including mudstone, ironstone, and claystone. Some of these facies were as thin as 5 cm, with the thickest being less than 2 metres. It should also be noted that Figure 4 hows a fault running through the sequence with clear offset of beds.

Figure 3    Photo of the shale formation at Mam Tor, Derbyshire (UK).

Figure 4    Photo of the shale formation at Edale, Derbyshire (UK).

Table 3    Thickness of beds observed at Edale, Derbyshire (UK)

Facies	Thickness (m)
Mudstone, dark grey, very thinly bedded, fissile, harder bands are non-calcareous, sharp base	1.8
Ironstone	0.06
Ironstone, thin and interbedded mudstone, dark grey, fissile	0.95
Ironstone	0.05
Mudstone, lighter grey in weathered section, thin bedded, nodular, non-calcareous with very thin ironstones	0.8
Mudstone, dark grey, very thinly bedded, fissile	0.7
Ironstone band	0.08
Mudstone, dark grey, fissile, becoming less calcareous upwards, thin interbedded ironstone bands in upper part	0.5
Gap, vegetated but probably the same unit as below	0.5
Claystone, dark grey, fissile, very thinly bedded with very thin lenticles of wispy paler calcareous mudstone	0.9
Mudstone, dark grey, very thin bedded, lenticular calcareous zones, small goniatite seen; with fairly sharp base	0.35
Claystone, dark grey, fissile, no mica, very homogeneous, gradational base	1.1
Mudstone, dark grey, fissile with large calcareous bullions	0.6
TOTAL	8.39

These field exposures clearly show variations in physical properties over short distances in shale sequences; differences can clearly be seen in weathering rates. Individual beds have been observed to have as little as 5 cm thickness, with the thickest beds of the order of 2 metres thick.

Variations in physical properties

The examples listed above show that ‘shale’ formations can include mudstone, claystone, ironstone, sandstone, limestone, coal measures and chert nodules. This variability is likely to be evident in differences in physical properties.

Table 4    Typical physical properties of lithologies seen within shale formations. From Waltham (1994) and Hobbs (1964)

Rock type	Dry density (g/cc)	Porosity (%)	Dry UCS range (MPa)	Dry UCS mean (MPa)	Young’s modulus (GPa)	Tensile strength (MPa)	Shear strength (MPa)
Greywacke	2.6	3	100–200	180	60	15	30
Sandstone (Carboniferous)	2.2	12	40–100	70	30	5	15
Limestone (Carboniferous)	2.6	3	50–150	100	60	10	30
Mudstone(Carboniferous)	2.3	10	10–50	40	10	1
Shale (Carboniferous)	2.3	15	5-30	20	2	0.5
Clay (Cretaceous)	1.8	30	1–4	2	0.2	2	0.7
Coal	1.4	10	2–100	30	10	2
Ironstone#				190		44
TOTAL RANGE	1.4–2.6	3–30	1–200	2–190	0.2–60	0.5–44	0.7–30
MEAN	2.2	12		79	25	10	19

Table 4 shows typical physical properties for the lithologies listed above. The uniaxial compressive strength (UCS) is often used as a comparative measure of strength. A UCS range of 2 to 190 MPa represents a rock classification from weak to strong rock (Waltham, 1994). A weak rock can be viewed as one that crumbles under a pick blow, whilst a strong rock can be broken by a hammer in the hand. The average UCS of 79 MPa represents a moderately strong rock; one which can be dented with a hammer pick. The tensile strength is of direct relevance to hydraulic fracturing. The range of lithologies have tensile strengths of between 0.5 and 44 MPa, with an average of 10 MPa. This clearly shows that certain beds will be much easier to hydraulic fracture than others. It should, however, be noted that the simplistic data represented in Table 4 does not capture the full range in physical properties seen within highly variable shale sequences.

Knowledge gaps and recommendations

This chapter has introduced the variability seen within shale sequences. The following statements on our current knowledge, knowledge gaps and recommendations can be made:

• The term ‘shale’ includes complex sequences of geological beds that include siltstone, mudstone, sandstone, limestone, ironstone, coal, and chert. These vary over the centimetre scale vertically and vary in thickness and extent laterally. This variation may occur over the 10’s centimetre to 100 meter scale. Making accurate predictions of the full sedimentary sequence is thus very difficult. A better understanding of geological sequence stratigraphy is needed in order to understand the control this variability has on hydraulic fracturing.
• The mineralogy seen within geological sequences varies considerably and this is also evident in total organic carbon (TOC). Hydraulic stimulation may be more successful in certain beds and these might not necessarily be high in TOC. Therefore recoverability will be dependent on both TOC and ease of hydraulic fracturing. A better understanding is needed of both of these properties so that hydraulic fracturing does not just occur where high TOC occurs.
• Bedding thickness is variable, ranging upwards from thinly laminated (less than 6 mm) but typically less than very thickly bedded (2 m). This range in bed thickness is much less than the seismic resolution of 20 metres and therefore the full variability of shale sequences cannot be achieved by seismic techniques alone. The significance of such small beds needs to be understood and the risk of failing to determine the full geological sequence from geophysical methods needs to be assessed.
• The strength properties of litholigies found within shale formations has a considerable range. A better understanding of the variability in physical properties relevant to hydraulic fracturing is required. The interplay between mineralogy and strength also requires more research.
• This chapter has given examples from the United Kingdom. A better understanding of the variability of shale within Europe is required. Similarities are likely, as are differences that are specific to individual basins or geological domains.