Towards Prediction of Saturated-Zone Pollutant Movement in Groundwaters in Fractured Permeable-Matrix Aquifers: the Case of the UK Permo-Triassic Sandstones

Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

Towards prediction of saturated-zone pollutant movement in groundwaters in fractured permeable-matrix aquifers: the case of the UK Permo-Triassic sandstones

JOHN H. TELLAM & RONALD D. BARKER Hydrogeology Research Group, Earth Sciences, School of Geography, Earth and Environmental Sciences, Birmingham University, Birmingham B15 2TT, UK (e-maik J.H. [email protected]. uk; R.D. Barker@bham. ac. uk)

Abstract: The UK on-shore Permo-Triassic sandstones are fluvial and aeolian red beds showing a nested cyclic architecture on scales from millimetres to 100s of metres. They are typical of many continental sandstone sequences throughout the world. Groundwater flows through both matrix and fractures, with natural flow rates generally of less than 200 m year-1. At less than 30 m horizontal distances, below important minimum representative volumes for both matrix and fracture network permeability, breakthroughs are likely to be multimodal, especially close to wells, with proportionately large apparent dispersivities. 'Antifractures' - discontinuities with permeability much less than that of the host rock - may have a dominating effect. Where present, low-permeability matrix (e.g. mudstones) will significantly affect vertical flow, but will rarely prevent eventual breakthrough. Quan- titative prediction of breakthrough is associated with large uncertainty. At scales of 30 to a few 100s of metres, multimodal breakthroughs from a single source become less common, although very rapid fracture flow has been recorded. At distances of hundreds of metres to a few kilometres, there is evidence that breakthroughs are unimodal, and may be more immediately amenable to quantitative prediction, even in some cases for reacting solutes. At this and greater scales, regional fault structures (both slip surfaces and granulation seams) can have major effects on sub-horizontal solute movement, and mudstones and cemented units will discourage vertical penetration. The aquifer has limited oxidizing capacity despite the almost ubiquitous presence of oxides, limited reductive capacity and limited organic sorption capacity. It has a moderate cation-exchange capacity, and frequently contains carbonate. Mn oxides are important for sorption and oxidation, but are present in limited quantity. Relationships between hydraulic and chemical properties are largely unknown. 'Hard' evidence for the solute transport conceptual model presented above is relatively limited. To be able to predict to a reasonably estimated degree of uncertainty requires knowledge of: the geological, and thence the hydraulic and geo-chemical, structure of the complex sandstone architecture (including the correlations between these properties); the development of suitable investigation techniques (especially geophysical) for mapping the structures; and the development of modelling tools incorporating matrix, fractures, 'antimatrix' and antifracture elements, each with associated hydraulic and possibly geochemical properties. In common with solute movement studies in most aquifer types, much more geological characterization needs to be undertaken. Although new investigation and modelling tools are being developed specifically for (shallow) hydrogeological applications with some considerable success, much greater advantage could be taken of importing techniques from other disciplines, and in particular from oil exploration and development.

The development of a quantitative understand- continental USA. They are also well known in ing of groundwater solute movement is an Gondwanaland, especially South America, important goal for all aquifer types. This paper Africa, Antarctica and Australia. Although the is concerned with evaluating the progress character of the sandstones varies with location, towards this goal in an example fractured in broad terms the UK sandstones are typical permeable matrix aquifer - the UK Permo- fractured continental red-bed sequences with Triassic fluvial-aeolian sandstone sequence. well-marked sedimentary structures. After the Permian and Triassic continental sandstones (Cretaceous) chalk, they constitute the most were widely distributed prior to the break-up of heavily used set of aquifers in the UK; they are Pangaea. They now occur in NW Europe, East also extensively used elsewhere for water Greenland and in many locations across the supply, and form hydrocarbon reservoirs

From: BARKER,R. D. & TELLAM,J. H. (eds) 2006. Fluid Flow and Solute Movement in Sandstones: The Onshore UK Permo-Triassic Red Bed Sequence. Geological Society, London, Special Publications, 263,1-48. 0305-8719/06/$15 9The Geological Society of London 2006. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

2 J.H. TELLAM & R. D. BARKER particularly in the shelf areas surrounding the of overlying deposits will not be directly UK. considered: although these have a major effect The development of the attributes that affect on flow patterns and water chemistry, they are reacting solute movement in present-day not intrinsic to the phenomenon of solute groundwater systems is determined by the inter- movement in the sandstone. In addition, the action of three main geological factors: deposi- unsaturated zone and non-aqueous phase liquid tional environments; palaeo-groundwater-rock movement will not be covered, but papers on interactions; and stress (Fig. 1). In most ground- these topics are included within the volume water systems, solute movement, as observed, (Binley et al. 2006; Gooddy & Bloomfield 2006; depends strongly on the scale(s) of heterogen- Privett 2006; Rees 2006; Taylor & Barker 2006). eity of the sequence, the scale of measurement Evidence is drawn from the very considerable and the scale of interest (e.g. Dagan 1989), and literature, 'white', 'grey' and unpublished: it is well recognized that this needs to be because of the difficulty of obtaining much of considered when evaluating evidence. Accord- the latter two, the selection is unavoidably ingly, the structure of this paper (and this biased. Wherever relevant, an attempt has been Special Publication) reflects these issues: first, made to include reference to the very signifi- the geological setting is briefly described, then cant petroleum geology literature, a resource flow and, finally, solute movement. In the latter that often is not fully exploited in hydrogeology: two sections, the general approach is to consider it is, however, important to note that observa- evidence in order of increasing scale of investi- tions from deep systems with a different geo- gation, and in the final discussion scale is a major logical history are not necessarily directly consideration. Aquifer geometry and presence transferable.

:De~iti~a~ I Geological Environments I ff [Gw Interactions]dxdt ! Development i I

/ ' gu~fi~ltx % 1< > ' + Uipli~tit!x) ~ fi [Stress]dxd! I Lithology(t, x) I ~ _ N) J~ : ' ~ ctures(t,x) ~' -- --~ I-Fig -- II |

Km/4,/ mll I,:/+/.,(t.x) z5 Flow l .l tx, II Reactivity(t,x~~ K /d~ /'r (t,x) ~

I Dispersivity(t'x) I

Breakthrough SoluteMovement dislributions(t,x)

Fig. 1. Geologicalfactors and their relationships with solute movement (ff[]dxdt = integrated effect over space and time; K is hydraulic conductivity; 0 is porosity; x is tortuosity; subscripts f, m, and rm are fracture, matrix and rock mass; double line boxes indicate properties that may be amenable to geophysical measurement). Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 3

Geological considerations example, Wills (1970) recognized 'microcycles' at the scale of a few beds, 'miocycles' compris- Introduction ing groups of microcycles up to approximately 100 m thick at most and 'magnacycles' compris- An outline stratigraphy is given in Table 1, and ing groups of miocycles (usually equivalent to the distribution of outcrop is indicated in Figure formations). Each microcycle represents a wet 2 (for comprehensive details see, for example, to dry transition. Thus, an ideal microcycle Warrington et al. 1980 and Benton et al. 2002). sequence according to Thompson (1970a) and In general, the sequence, up to over 1 km in Wills (1970) might be: thickness at its maximum development, is underlain by deposits of low permeability, top Aeolian sandstones sometimes Permian in age but often Carbonif- Mudstone, with desiccation cracks erous or older. Usually it is immediately Finer-grained plane-laminated sandstone overlain by either the Triassic Mercia Mudstone Medium-grained cross-laminated sand- Group (mudstones and evaporates with stone occasional thin sandstones) or by Quaternary Coarse (pebbly) channel deposit deposits (tills to outwash gravels). The Permian base Erosion surface. and Triassic sandstone sequences are broadly Miocycles were defined by Wills (1970), similar in origin and lithology, and, for the rather subjectively, using indicators of maximum purposes of this review, will be considered water velocity with high-energy conditions together. passing upwards into lower energy conditions. At the largest scale (magnacycles), the main Depositional environments formations in any one locality may be paired into a cycle: for example, in the Triassic During the Permo-Triassic, the area that is now Sherwood Sandstone Group of the Midlands the UK migrated from approximately 10 ~ to (Table 1) the first magnacycle would comprise 30 ~ N of the equator. The climate was semi-arid the Kidderminster Formation (coarser, pebbly) to arid, with flash flood episodes bringing debris and the Wildmoor Sandstone Formation (finer, from the eroding Variscan mountains in the non-pebbly): the overlying Bromsgrove Sand- south (present continental Europe and SW stone Formation (coarser, pebbly) would form England). Northward-flowing braided river the first part of the upper magnacycle. systems deposited increasingly finer-grained In recent years the significance of the cycles sediments in tectonically active, often half- described by Wills (1970) has become evident graben, subsiding basins. Locally this north- and they can be related to orbital forcing mech- wards trend of decreasing grain size is reversed, anisms (Clemmensen et al. 1994). Descriptions and conglomerates are deposited as a result of have become rather more sophisticated, and steep slopes (e.g. in southern Scotland: Akhurst similar conceptualizations are now widely et al. 2006). The main depositional environments applied to Triassic stratigraphy and sedimentol- are summarized in Table 2. ogy on a global basis (Szurlies et al. 2003). In The wide range of fluvial and aeolian environ- the UK Triassic the distinctive wetting-drying ments produced a variety of sedimentary cycles (Mountney & Thompson 2002) are most structures, including plane lamination, cross- conveniently explained using orbital forcing lamination, trough and planar tabular cross- mechanisms. stratification, water escape structures, imbricate In strong contrast to the practice in petroleum gravels, debris flows, palaeosols and desiccation geology, such geologically oriented conceptual- cracks (e.g. Thompson 1970a; Steel & Thompson izations have seldom been applied in water 1983) (Fig. 3). Bed size varies from less than a resources and non-nuclear pollution studies in few centimetres to at least several metres. the UK Permo-Triassic sandstones (but From the 1960s, vertical depositional 'cycles', compare Bloomfield et al. 2006 and Bouch et al. or sequences, have been recognized at various 2006). scales in the fluvial-dominated parts of the Lithologically, the sandstones range from succession, with cycle thicknesses from less lithic arkoses to quartz arenites. Detrital clays than 1 to more than 100 m (Fitch et al. 1966; and mica are commonly present, although a few Thompson 1970a; Wills 1970, 1976): such sequences are effectively 'clean' (e.g. the conceptualizations are now embedded within Penrith Sandstone, Cumbria: Lovell et al. 2006). sequence stratigraphy (e.g. Vail et al. 1991; Organic carbon contents are typically less than Jensen et al. 1996; Howell & Mountney 1997; 0.1% (Steventon-Barnes 2001; Shepherd et al. Mountney & Thompson 2002). Thus, for 2006). The generally less than 1 m-thick Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

4 J.H. TELLAM & R. D. BARKER

.,.4, t-~ ~xl:l o

,,...-, ..~, ~ II ~L3

4= o O o ~r ~ ~o ~

"o o o o o

o o ~r ~ o'~ ~o ~

4= o o o .~ ~o~

4= o o ~r ~ ~o ~ b E

..1~ N

~=.~ o 4=

~.~

~.~ ~o~

o o ~'~

"~ ~ ~ r . ,..,~ . ,..,, Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 5

Fig. 2. The main outcrops of Permo-Triassic sandstone in the UK, and locations of aquifers mentioned in the text.

mudstones within the sequence have a similar single deformation bands, zones of deformation mineral composition (clays, fine quartz, feldspar, bands, and slip surfaces (e.g. Fisher & Knipe mica and haematite): illite, kaolinite, chlorite 1998; Manzocchi et al. 1998). Deformation and smectite have all been recognized (Burley bands (or granulation seams: Aydan 1978; Aydin 1984; Hough et al. 2001). & Johnson 1978; Fowles & Burley 1994) are common, often as swarms (Fig. 4a). Slip surfaces Stress (burial and uplift) now exist as open, clay-filled or cataclastite- filled features (Fig. 4b). The basins in which the Permo-Triassic Jointing occurs within the sandstone sequence was deposited were often active sequence to varying degrees. Only a few sets of tectonically during deposition (Poole & quantitative data are available (Barnes et al. Whiteman 1955, 1966; Audley-Charles 1970; 1998; Gutmanis et al. 1998; Wealthall et aL 2001; Plant et al. 1999). Typically they were asymmet- Jeffcoat 2002), and some of these are discussed ric grabens, controlled by reactivation of below. Fracturing becomes less frequent with basement faults: in the Midlands and NW depth (see below). England this results in the longer axes of the Dewatering structures are not uncommon, basins being aligned roughly N-S. Maximum but large-scale sedimentary dykes appear to be burial depths vary from basin to basin, but are rarely noted. usually no more than a few kilometres, and sometimes much less (e.g. Burley 1984; Plant et al. 1999). Components of uplift/inversion Palaeo-groundwater interactions/diagenesis occurred at various times from the Jurassic onwards. Generally, in all the UK on-shore Although locally very variable, in outline the Permo-Triassic basins, bedding dips are typically diagenesis of the sandstones has included the less than 10 ~ following phases (e.g. Burley 1984; Metcalf et al. As with many porous sandstones, faults 1994; Strong et al. 1994; Milodowski & Gillespie display a range of morphologies, including 1997; Milodowski et al. 1999): Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

6 J.H. TELLAM & R. D. BARKER

Table 2. Fluvial and aeolian depositional environments seen in the UK Permo-Triassic sandstone sequence (after Benton et al. 2002)

Environment Lithologies Structures

Fluvial Fan Poorly sorted gravels Sheet floods Poorly stratified or imbricate Upwards fining Debris flows Matrix supported Chaotic internal organization High-energy braided Texturally mature gravels Conglomerates river with interbedded sands Crude stratification Large foresets Sandstones Cross-stratified Lower energy braided Sands with subordinate Sandstones river silts and muds Upper-phase plain stratification Planar tabular stratification Trough cross-stratification Upwards-fining cycles Mudstones Usually <1 m thick, and often much less Meandering river Upwards-fining cycles channel and overbank Sands with 30-60% silts Sandstones with palaeosols Lateral accretion cross-stratification floodplain Variety of smaller scale cross- stratified, rippled and plane-bedded sandstones Thin overbank sands Aeolian Dunes Texturally mature sands Large-scale cross-stratification, asymptotic foresets up to 30 ~ inclination, sets separated by sharp, generally planar bounding surfaces Sheets Sand Bi-modally laminated on mm-scale scours, granule lags, thin cross-laminated units Damp interdune Silty sandstones Irregular wavy or lenticular lamination, slump structures, with adhesion ripples dominant

Early diagenesis. It would appear that in Post-inversion diagenesis. Following inver- many areas, early diagenesis included: sion, dissolution of carbonate and sulphate

- dissolution of unstable silicates; and occurred, with weathering of some of the

- precipitation of feldspars to clay. This process continues, 9 clays (e.g. smectite/illite), but at rates that are unlikely to be hydro- 9 K-feldspar, geologically significant except in very rare

9 non-ferroan carbonate, cases (e.g. Bath et al. 1987). 9 haematite and 9 gypsum. In some cases, the mineralogy allows the past Burial diagenesis. Subsequent burial, typi- presence of saline groundwaters to be inferred cally to less than a few kilometres, (Milodowski et al. 1998). Saline groundwaters compacted the sandstone particularly are common in the deeper parts of many of the where cements were not pervasive and sandstone basins, the majority being derived converted at least some of the smectite to from evaporite dissolution (Edmunds 1986; illite. As many of the sandstones have Barker 1990; Tellam 1995a; Bottrell et al. 2006), present porosities of approximately 0.2--0.3, Especially in central England, mineralization is cementation probably frequently occurred sometimes present (e.g. baryte with Cu, and, less prior to burial. During this phase, some abundantly, Pb, Co, Zn, Mn, V, Ni and As in the precipitation of ferroan dolomite and Cheshire Basin: Ixer & Vaughan 1982; Naylor ankerite occurred. et al. 1989; Plant et al. 1999. This may be the Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 7

(a) result of flow of reducing brines from the underlying Carboniferous up fault planes into oxidized (evaporite-derived) brines in the sandstone during uplift in the Tertiary (Naylor et al. 1989; Tellam 1995a; Milodowski et al. 1998). Although localized, these mineral deposits may play a wider role as sources for dissolved metals (Kinniburgh et al. 2006). The distribution of the mineralization and the often-associated bleaching (removal of the ferric oxides) strongly suggest it to be associated with the faulting, and to be affected by the presence of low- permeability beds (Rowe & Burley 1997). In a few places, in otherwise unaffected aquifers, natural hydrocarbons have been observed to issue from the sandstones (Rowe & Burley 1997; Plant et al. 1999). (b) The main pattern in cementation appears to be: local dissolution of carbonates and feldspars (especially from the upper 50-100 m of the sandstone sequence) (Travis & Greenwood 1911; Burley 1984; Kinniburgh et al. 2006); the presence of remnant gypsum cements in some low-flow systems (e.g. Jackson & Lloyd 1983); and the absence of carbonates in some formations (e.g. Edmunds & Morgan-Jones 1976; Walton 1981).

The present-day sandstone sequence The present-day sequence therefore has the following principal properties: 9 at most a moderately well-developed cementation; (c) 9 often well-developed sedimentary structures; 9 layering, often with a lithological cyclicity at several scales; 9 occasional mudstones, usually less than 1 m thick; 9 usually shallow bedding dips; 9 fracturing, especially at shallower depths; 9 frequent occurrence of faults of a range of morphologies: - granulation seams, single and swarmed; - slip surfaces (well cemented, clay gouge-filled, breccia-filled or empty); 9 veining (see below); 9 carbonate cementation (although sometimes removed at shallow depths); 9 very low organic carbon content; 9 clay as a common constituent; 9 ferric and often manganese oxides/hydroxides. Fig. 3. Example outcrop sections of the UK Permo-Triassic sandstone sequence. (a) Dawlish, SW England; (b) Torbay, SW England; and (c) Runcorn, NW England. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

8 J. H. TELLAM & R. D. BARKER

(a)

(b)

Fig. 4. Fault outcrops (Wirral Peninsula, NW England). (a) A granulation seam 'swarm'. (b) A slip surface, with bleached zones (notebook is about 8 cm wide). Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 9

Groundwater flow some cases, found (Seif el-Dein 1983; Digges la Touche 1998; Bloomfield et Matrix flow al. 2001; Lovell et al. 2006; Pokar et al. 2006), but scaling may be an issue Laboratory studies of hydraulic properties. Very (Worthington 2004); many studies have examined the porosity and mudstone properties have been much permeability of laboratory samples of the sand- less frequently studied. Limited work stones, from Roberts (1869) to various papers in in the West Midlands and NW England this volume (Bloomfield et al. 2006; Lovell et al. (Curi~o unpublished data 2002) 2006; Pokar et al. 2006). Allen et al. (1997) confirms that intergranular perme- summarize much of this work. In general the abilities are very low (c. 104 m day -1) following observations can be made: (Fig. 5), although none of the samples 9 Porosity tested contained infilled desiccation

- distributions appear to be Gaussian, cracks, a common sedimentary feature. with values in Allen et al.'s (1997) data- base ranging between 0.02 and 0.35, Sandstone permeability structure and flow. In with a median of 0.26 (although it is general, estimation of a large-scale hydraulic likely that the data set is slightly biased, conductivity value for the sandstone has usually with too few results from friable involved taking a weighted mean of the samples with higher than average hydraulic conductivity data from small-scale porosity (and permeability)). tests (e.g. Pokar et al. 2006) or finding an effec- 9 Permeability tive hydraulic conductivity distribution by

- values are log-normally distributed; calibration of a numerical model against field

- geometric means are often in the range data. These deterministic approaches are often 0.1-10 m day-l; satisfactory for flow assessment, but usually are

- there is some variation with formation, not adequate for prediction of the flow at least locally (e.g. Campbell 1982) components of solute movement. Historical and between basins; data are often inadequate for estimation of the

- permeability parallel to laminations in effect of the local-scale hydraulic conductivity the sandstones can be up to at least 50 variations on the transport and fate of solutes. times that perpendicular to lamina- For the latter, stochastic approaches are avail- tions, but the ratio is usually much able, although application of these techniques is closer to 1-3 (e.g. Barker & Worthing- much rarer in practice in hydrogeology than in ton 1973a, b; Lovelock 1977; Campbell petroleum geology. They include the use of 1982); geostatistical, process-imitating and structure- - except possibly locally, there appears to imitating models (e.g. Guadagnini & Winter be no systematic variation in matrix 2004; de Marsily et al. 2005). permeability with depth recorded in Publications on geostatistical modelling of the the literature, despite the fairly sandstone hydraulic conductivity distributions frequent dissolution of carbonates appear to be lacking in the accessible hydro- from shallow depths; geological literature. In petroleum geology, the - systematic investigation of three- most significant early application of geostatistics dimensional permeability structures to UK sandstones was the work of Matheron are rare (cf. Prosser & Maskall 1993; et al. (1987) on the coastal Jurassic sandstone Pickup et al. 1994; Ringrose & Corbett outcrop in Yorkshire. This application used a 1994), as are studies on the relationship truncated Gaussian method to simulate the between bed form and permeability/ distribution of rock types within the sandstone porosity; based on the division of a continuous variable

- the few studies available suggest that into a discrete set of indicators. More recently, the permeability of some sandstones is the markedly structured nature of many sensitive to change in water chemistry sequences, for example the UK Permo-Triassic (Braney et al. 2001; Mitchener 2003); sandstone, has lead to suggestions that discon-

- the logarithm of permeability is often tinuous, facies-based approaches are more linearly correlated with porosity, albeit appropriate (e.g. Aigner et al. 1999). Such with considerable scatter and with only approaches have been developed and applied in locally validity (e.g. Campbell 1982); the oil industry. They require detailed geological

- correlations between permeability and characterization, making heavy use of core, pore size have been sought and, in geophysical and outcrop analogue data, the Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

10 J.H. TELLAM & R. D. BARKER

Mudstone Bed Thickness (m)

Sandstone Thickness Without a Mudstone Bed (m)

Log[K (m d-l)]

Fig. 5. Mudstone thickness and frequency (Lower Mersey Basin), and permeability (Cheshire Basin and Birmingham) (Curi~o unpublished data 2002). Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 11 latter increasingly collected with geophysical permeability initially present through deposi- tools (e.g. Aigner et al. 1995, 1999; Bryant et al. tional processes may well affect the develop- 2000; Pringle et al. 2004). There are many prac- ment of cementation (and thence also to tical problems with this approach: for example, fracturing - Fig. 1). quantifying uncertainty, dealing with qualitative A conceptual framework for modelling data, appropriate definition of facies and cali- hydraulic conductivity heterogeneity is supplied bration of resulting models (e.g. Wong et al. by the nested cyclic sedimentation-sequence 1997; Vaughan et al. 1999; Biver et al. 2002; Liu stratigraphy model described earlier (cf. Jensen & Oliver 2005; Martinius & Naess 2005). Never- et al. 1996). However, this model possesses form theless, codes that include explicit accounts of and noise (randomness) that needs to be placed bedforms and facies are starting to be used not on a much sounder statistical foundation: this in only in research but also in practical appli- turn requires much improved geological data cations (e.g. Smith &Moller 2003; Liwanag collection. At present, the required geological 2005). information is largely lacking from the existing At the heart of the facies approach is the water well logs, as Thompson pointed out as characterization of the properties and distri- long ago as 1970 (Thompson 1970a). However, bution of facies and lithotypes. There are it may be possible to take advantage of the very numerous petroleum industry examples of this, considerable number of accessible water wells but a good hydrogeological example, incorpor- by relogging with geophysical tools (e.g. Turner ating both hydraulic and geochemical data is et al. 2001; Houston 2004), including acoustic the study of Allen-King et al. (1998) and Moysey and optical televiewer logging and resistivity et al. (2003). In the UK, although comparisons imaging or spectral IP (induced polarization) of hydraulic properties and sandstone lithol- logging, to gain enough geological detail to ogies have long been attempted (e.g. Lovelock quantify the structure and variability. Many 1977), detailed work of this type is uncommon. possibilities exist for searching for structure An exception is the study of Bloomfield et al. using such geophysical approaches (e.g. Yang & (2006) who show that various lithofacies of the Baumfalk 1997). The application of geophysics Wilmslow Standstone and Helsby Sandstone to hydraulic property estimation is discussed in formations (Table 1) have distinctive hydraulic more detail below. properties. (See also Bloomfield et al. 2001.) Although the development of geologically Bouch et al. (2006) similarly present detailed informed stochastic approaches have developed lithofacies data for a predominantly fluvial largely to study transport problems, there is sequence. potentially substantial benefit to be gained from A particular difficulty in practice is the deter- their application to flow-only studies, notably mination of the lateral extents of each facies through a greater awareness of predictive ('micro'- and 'mio' scales in particular). Generic uncertainty. data are available (e.g. Mial11990), but they may not be appropriate for a given system. One The effect of mudstone permeability on flow. relatively recent development is the use of Little quantitative information is available on sediment deposition models to help constrain the distribution and hydraulic properties of the sedimentary structure architecture. These mudstone beds, but some example data models often use process-based physics models obtained from analysis of cores and geophysical and/or empirical rules to develop sediment logs are shown in Figure 5: comparison of the distributions associated with a particular geophysical and core data indicate that usually environment (e.g. Kolterman & Gorelick 1996). only mudstones thicker than approximately Teles et al. (2004) have used this approach in 10 cm can be resolved by geophysical logs run modelling an alluvial aquifer in France. Bunch under standard conditions, and that sometimes et al. (2004) presented a sedimentation model distinguishing mudstone from mud-pellet- designed to apply to the UK Permo-Triassic containing beds using natural gamma-ray and sandstone sequence, and current work is resistivity logs alone is difficult. Lateral continu- concentrating on the SW England outcrop area ity is generally uncertain. Some mudstones can (Fig. 6) (Bunch 2006). This approach should be traced for at least about 1 km (Thompson provide insights into the vertical and horizontal 1970b), but often it is found that mudstones in components of facies architecture, but a diffi- boreholes separated by less than a few tens of culty for translation of these to yield hydraulic metres cannot be correlated, although whether conductivity distributions that is not usually this results from limited extent, local erosion or important in studies of more recent sediments is fault displacement is usually uncertain (cf. the presence of cements and fractures. The Stanistreet & Stollhofen 2002). Palaeosols may Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

12 J.H. TELLAM & R. D. BARKER

Fig. 6. Modelled alluvial-fan deposition showing grain-size distribution architecture (based on Permian sequence, Torbay, SW England, viewing towards the NW, area approximately 5 • 5 km) (Bunch (2006); see also Bunch et al. (2004)).

also have significantly lower permeability than flow will go around them, but if they are the majority of the sequence (Bouch et al. 2006; hundreds of metres in length, most flow will pass Newell 2006). through them, albeit still imparting a directional Rushton & Salmon (1993) demonstrated the influence on flow (see also Alexander et al. 1987 importance of low-permeability beds, especially and the section on Unreacting solute movement mudstones, in controlling vertical head gradi- later). ents in the sandstones. Vertical head gradients of a few per cent are often seen in boreholes Geophysical estimation of hydraulic properties. (Brassington 1992; Taylor et al. 2003), and As mentioned above, geophysics is a potentially usually most head change is seen across powerful tool for mapping geological structures, mudstone beds (Segar 1993). An extreme but it also has potential for more directly example is seen at a location in the southern mapping the matrix hydraulic property architec- Cheshire Basin where head differences of ture of the sandstones, provided appropriately approximately 65 m have been recorded in wells sensitive methods can be found. Early attempts of different depths at effectively the same to estimate hydraulic properties used equations location (Voyce pets. comm. 2002). It is now developed for oil reservoir studies. Application common practice to model regional flow in the of the Archie (1942, 1950) formation factor sandstones using layered models. Although the equations (P~ = F, where 90 is bulk resistivity, mudstone matrix permeabilities are very low, Pw flow will pass across the mudstone beds to some P,, is pore fluid resistivity and F is the formation extent: in general, simple estimates suggest that factor) to the Permo-Triassic sandstones (e.g. if the mudstones are of limited lateral extent, Worthington 1973) quickly showed that typical Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 13

UK Permo-Triassic sandstone did not behave as Dbscontinuities a 'clean' sandstone as it contained electrically conductive material, mainly clay, disseminated Introduction. The main discontinuities in the throughout the rock (cf. the Penrith Sandstone UK Permo-Triassic sandstone sequence are Formation described by Lovell et al. 2006). In joints, faults/granulation seams and veins. Joints this situation, the Archie equations are only and some faults appear to have greater perme- valid for high-salinity pore waters. abilities than the host sandstone: some faults In order to determine the porosity (~)) from including granulation seams appear to act as the formation factor (F) using Wyllie & barriers, and presumably this is the case also for veins. The following sections discuss fractures Gregory's (1953) equation, F = a-E-~m ' corrections (defined here as discontinuities with permeabil- must be made for the excess conductivity intro- ity apparently greater than the matrix) and duced by the clay, and many methods to do this antifractures (discontinuities with permeability have been suggested (e.g. Patnode & Wyllie less than the matrix). 1950; Winsauer et al. 1952). In order to employ these equations successfully, values of the excess Fractures. Occurrence of fractures. Joints and conductivity together with the constants a and bedding-plane joints are seen in many outcrops rn have to be determined for the local aquifer. of the sandstones, and are often detected in These relationships are further complicated borehole logs (although apparent frequency can when the sandstone is partially saturated vary significantly with technique used: Jeffcoat (Taylor & Barker 2002, 2006; cf. Dalla et al. 2002) (Fig. 7). There appears to be no system- 2004). So although the relationships between atic study of joint distributions and properties electrical properties and porosity are estab- across the UK, but some local data sets are lished, their practical application is not trivial available (Barnes et al. 1998; Gutmanis et aL and can be expensive. In addition, the import- 1998; Wealthall et al. 2001; Jeffcoat 2002). Hitch- ant precise indirect determination of permeabil- mough et al. (submitted) recognize six fracture ity has hitherto proved elusive. sets for the Cheshire Basin (Table 3). In an attempt to overcome these shortcom- The distribution of bedding-plane joints may ings, other geophysical techniques have been be partly controlled by lithology according to used, particularly induced polarization. Some Allen et al. (1998), occurring especially in lami- partial success was achieved for the Permo- nated sandstones (cf. suggestions by Gabrielsen Triassic sandstone of the Fylde (Fig. 2) when et aL 1998). Allen et al. (1998) cite the example Collar & Griffiths (1976) presented an empiri- of the well-cemented Helsby Sandstone cal relation between induced polarization and Formation (Table 1) for which the spacing is permeability. Similar results were reported from approximately 0.1-10 m, and the underlying the Permo-Triassic sandstone of the Lower more friable Wilmslow Sandstone Formation Mersey Basin by Olorunfemi & Griffiths (1985), with a spacing of 25-100 m. Bouch et aL (2006) who also used induced polarization field surveys present detailed fracture data from the to determine groundwater salinity and excess Wildmoor Sandstone Formation of the English conductivity. Midlands, a similar sandstone to the Wilmslow Most recently, it has been shown that there Sandstone. Allen et al. (1998) suggest that are strong relationships between the spectral bedding-plane fractures are relatively short resistivity parameters and the pore-size distri- (less than tens of metres) in channel sandstones, bution obtained using mercury injection capil- but longer where developed at the junctions lary pressure instruments (Scott & Barker 2003, with fine-grained sheet deposits (up to a few 2005, 2006). This opens up the possibility of hundreds of metres) (cf. Table 3). estimation of permeability, using both surface- Borehole geophysical and other data suggest based and borehole-based resistivity measure- that bedding-plane fractures become less ments. frequent below about 200 m in the Lower Cross-borehole surveys may also be used to Mersey Basin (Fig. 2) (University of Birming- examine the properties of sandstone over ham/NWWA 1981; Campbell 1986) and below volumes of several tens of cubic metres approximately 120 m depth in Cumbria (Allen (Winship et al. 2006). Once techniques have et al. 1998): however, there appears to be no been demonstrated on this scale, it is a rela- frequency-depth correlation above this depth in tively small step to their application in large- the Cheshire Basin (Jeffcoat 2002). scale field surveys. Allen et al. (1998) suggest that bedding-plane joints have apertures of up to 10 ram, although it is often difficult to estimate apertures in Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

14 J.H. TELLAM & R. D. BARKER

(a) (b)

(c)

Fig. 7. (a) Example fractures (with granulation seams), southern Cheshire Basin; (b) example fractures, northern Cheshire Basin; (e) fracture trace length distributions, Cheshire Basin (after Hitchmough et aL submitted; unshaded, length censored by edge of outcrop).

Table 3. Summary discontinuity properties from the Cheshire Basin (Hitchmough et al. submitted). K14 is horizontal permeability; Kv is vertical permeability; MRV is the minimum representative volume

Set Dip Strike Length % Description Potential importance (assuming all granulation 'flowing' fractures have the same, seams isotropic, transmissivity)

1 Subhorizontal - 10s of m 0 Bedding planes Dominates KH; increases Kv; reduces tortuosity 2 Subvertical N-S <10 m 43 Joints, faults, Minor effect on KH/Kv ratio granulation seams 3 Subvertical E-W <10 m 26 Joints, faults Little 4* Various Various <10 m 23 Joints, granulation Increases Kv. Increases tortuosity seams, faults 5 Subhorizontal - <10 m 0 Stratification planes Important in absence of Set i for KIa and (slightly less so) for K v 6 Subvertical N-S >10 m 90 Granulation seams; If fractures present, significantly rarely faults, joints changes K tensor in vertical and N-S directions, increases MRV above 35 • 35 x 35 m, marginally increases N-S tortuosity

* All fractures not included in the definitions of the other sets. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 15 eroded and, sometimes, stress-relieved outcrop short matrix 'bridges'). These points are con- (cf. Yeo et al. 1998). There is little evidence of sidered below. the shape of the fractures. Allen et al. (1998) suggest that joint apertures are much less 9 It would be difficult to argue that fracture variable than bedding-plane joint apertures (i.e. permeability is not increased by well in the range of tens of micrometres to milli- development, but to demonstrate that metres). Fractures with unfilled apertures of natural fill material, as well as drilling millmetre size would have a considerable affect debris, is removed during development is on local flows. not straightforward. Outcrops and well In general, fracture properties would be faces are prone to erosion, and are hence expected to vary with bed thickness and lithol- not ideal sites to search for evidence for fill material. To avoid this problem, Wealthall ogy (Narr & Suppe 1991; Allen et al. 1998; et al. Gabrielsen et al. 1998) (Fig. 1), and if this is (2001) examined outcrops in hand-dug confirmed for the sandstones, then fracture tunnels in Runcorn, Lower Mersey Basin. properties as well as matrix properties may be They found fracture apertures up to several correlated to lithofacies and a cyclic nested centimetres wide, filled in every case with distribution may be present. However, frac- sand and/or clay (cf. Hewitt 1898), and tures will also be considerably influenced by speculate that this is the normal situation nearby faulting. within the rock. To test this idea, Hough et In common with many sequences, the UK al. (2001) examined quarry and other Permo-Triassic sandstones are often extensively outcrops of sandstones at approximately 20 faulted on all scales as outcrop mapping and sites from southern Scotland to SW borehole evidence shows (e.g. Knott 1994; England, but found very few such fractures Chadwick 1997) (hence, both fracture and (an example being in the Dumfries Basin in matrix systems show some degree of scale- southern Scotland: Kemp et al. 2003; see invariance, at least informally). also Akhurst et al. 2006). The absence of In some areas of the sandstones (e.g. southern large infilled fractures does not mean that Lancashire in NW England) longwall mining in infill in narrow aperture fractures is not the underlying Coal Measures has probably important, and indeed Jeffcoat (2002) enhanced fracturing, but no detailed quantita- reports that around 20% of fractures in her tive evidence is available. The importance of outcrop surveys had obvious fill. The same fractures may be less where weathering is study found evidence from packer and extreme because of both greater intergranular geophysical log data of flow in less than permeabilities and less ability to maintain open 10% of fractures (Hitchmough et al. apertures, but again little information is avail- submitted). Packer testing of fractures in able. the Wildmoor Sandstone Formation of the English Midlands in boreholes that were purposely only very gently developed failed Evidence of flow in fractures. Fractures have to show permeabilities greater than matrix long been assumed to be important for flow values; yet, less than 100 m distant in the (Tellam 2004), and Table 4 summarizes the same sequence, normally developed bore- evidence now available from geological, holes contained clearly identifiable geophysical, aquifer testing and modelling fractures that were seen to be further studies. Much more detail is provided by the developed during production testing reviews of Allen et aL (1997, 1998). (unpublished data; site described by Bouch In a few areas, comparisons of matrix and et al. 2006). Sand yields from wells may also field-scale permeability estimates show that provide evidence of fracture development, fracture flow must occur to a significant but even in the rare cases where sand yield extent (e.g. in the St Bees Sandstone Forma- is recorded, it is often difficult to determine tion). However, often the evidence is not whether the particles are released from unequivocal, as indicated in Table 4. within the formation or from the borehole Considering only that evidence obtained from walls as suggested by Bouch et al. (2006). borehole studies, there are two main uncertain- Overall, it would appear that borehole data ties: (i) the zone around a borehole may be overestimate the importance of fracture atypical of the rest of the aquifer because it has flow, at least in the sandstones with higher been developed or clogged during and after matrix permeabilities. drilling; and (ii) the fractures detected may not Using outcrop scan-line survey, borehole be connected on a regional scale (even via geophysics and packer test data from the Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

16 J.H. TELLAM & R. D. BARKER

,..-x ~.F ~= ~.~ ~" == ~ ~o~ ~

~-~ o ~N~ag ~N

9--~ e.i ~ ,4

~.~ o~ . ,...i ~

K ~'~ ~. ~ ~ ~'~ ~ .~ ~"~ "a '~ 0 -o ~ ; .~ ~ .~ ~.~

.. ~ = ~ ._~ ~ "~ ~ .~ o ~'~ ~ ~. .... =o ~

m ~ o ~o o ~ ~'~ ~ o ~-n

"a -~ ~~o~ ~'~-~ .~~._.~ ~ ~ ~ ~

o ~ ~ o ~ < ,--; e,i om ~ = z_~< ~ ~.~~ ~ ~ "-" ~ o .9. ~.

~B-~ ~ .. ~= .o~ ..~

9; Z o. ~8o

9n ~ ~'~

.~-~ 0 *--~ 0 0 r ~ ~.~ E~v'~ ~9 ~ ~'~ eO ~'~ o ~~o o..~l ~.-;.~ ~ = "~ 4 E

o~ .~ oo~ o~a

0 P~ ,J= ~ ~

8 l t~ K P. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 17

Cheshire Basin, Hitchmough et al. Davies & Handschy 2003), and there are many (submitted) investigated the properties of studies of low-permeability fault structures the fracture network using the code including in the off-shore UK Permo-Triassic NAPSAC (Herbert 1992; Hartley 1998) sandstone reservoirs (Edwards et al. 1993, (Table 3). It was estimated that about 9% of includes on-shore analogues; Leveille et al. the fractures carried significant flow. If all 1997). the fracture sets had the same permeability Experience in the on-shore sequences indi- distribution, the fracture network would be cates that many faults here also have restricted very anisotropic, being approximately 23 permeability perpendicular to the slip plane times more permeable in the horizontal (e.g. Allen et al. 1997), examples including major direction than the vertical. With this faults in the Birmingham and Nottingham assumption, the permeability tensor is much aquifers (Knipe et al. 1993; Yang et al. 1999; cf. more sensitive to the presence and proper- Trowsdale & Lerner 2003), and the 'bucket' at ties of some of the fracture sets than it is to Nurton, in central England (Fletcher 1989; others. One joint set (Set 6,Table 3) contains Hunter-Williams 1995). The most comprehen- rare but very long through-going subvertical sive demonstration of the importance of faults fractures, which, when present, will radically in regional flows is provided by Seymour et al. modify the flow system. The bedding-plane (2006), who detail a set of case studies across joints (Set 1,Table 3) were found to facilitate NW England involving flow modelling vertical flow as well as horizontal flow, as supported by water-level measurement, they connect up the shorter vertical frac- geophysics and chemical evidence. In most cases tures (cf. Allen et al. 1998). The minimum there is no information on the nature of the representative volume for the 'flowing' fault, and the low permeability could be due to fracture network permeability, ignoring the clay infill, fault rock or granulation seam rare through-going subvertical fractures, 'swarms' (see below). was estimated to have a characteristic As yet, predictions of likely fault properties length of 35 m. In the present context, the are rarely attempted, despite the availability of most significant finding was that the system oil industry techniques (e.g. Gibson & Bentham appears to be connected regionally. Some 2003) and potentially useful fault property fractures may, of course, be filled, but in correlations (Huntoon 1986; Knott 1994). For other cases matrix flow will allow good estimating the probability of 'sealing', it would connection between close but not inter- be useful to discover what proportion of secting fractures (e.g. Odling & Roden pumping tests show barrier-boundary 1997). responses. Although there is convincing evidence of Overall, it seems likely that fracture flow faults acting as barriers, there is also convincing does occur in the sandstones, but in most cases evidence, even from the earliest hydrogeological where intergranular permeability is relatively records (Tellam 2004), of faults acting as high it is not the dominant mode of flow, except high-permeability pathways (Allen et al. 1997). locally around pumping wells. However, there Clearly, some faults may have low permeability are parts of the sequence, especially in northern where mudstone beds have been smeared along England and southern Scotland, where matrix their planes or where some form of cataclastite permeability is low and fracture flow is has formed, and other faults may be effectively dominant. Where it occurs, it will disrupt the open fractures. However, it is also possible, as a effects of the matrix permeability structures, number of authors have pointed out (e.g. but any correlations between the distributions Gutmanis et al. 1998), that a fault may have a of permeable fractures and lithology are largely high permeability in a direction parallel with the unknown. Application of the 'generalized flow fault plane, and a low permeability at right model' pumping test interpretation technique angles to it: faults are not simple single features, suggested by Barker (1988) would be interest- and often may comprise a complex of planes of ing at a site with detailed fracture and matrix a wide range in permeability (e.g. Edwards et al. characterization. 1993; Fowles & Burley 1994; Allen et al. 1998). As with any fracture, the permeability along the Anti fractures. discontinuity may be far from uniform, and flow may be channelled (Yeo et al. 1998; Steele & Faults. In hydrocarbon development, faults are Lerner 2001). Combinations of faults will result often assumed to be barriers in porous sand- in changes in effective large-scale permeability stone reservoirs until proved otherwise (e.g. (e.g. Manzocchi et al. 1998). Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

18 J.H. TELLAM & R. D. BARKER

Granulation seams. Granulation seams (or Bouch et al. 2006). Their permeability is likely deformation bands: e.g. Main et al. 2001; or cata- to be very low. clastic slip bands: e.g. Rowe & Burley 1997) are a type of fault expression common in the UK Geophysical detection or fractures, antifractures Permo-Triassic sandstone sequences (Underhill and fractureflow. Often accurate delineation of & Woodcock 1987; Fowles & Burley 1994; Main even major structures is difficult using outcrop et al. 2001), and in many other high-porosity and borehole evidence alone, and, locally, where sandstones throughout the world (e.g. Aydin & lithological contrasts are present, geophysical Johnson 1978, 1983; Pittman 1981; Jamison & surveys have been used in determining faulted Stearns 1982; Antonellini & Aydin 1994; structure. For example, Collar (1974) used Antonellini et al. 1994; Gibson 1998). In the UK gravity and resistivity surveys to determine the Permo-Triassic sandstones, these deformation structure of the Clwyd Basin (Fig. 2), Barker bands typically comprise approximately 1 mm- (1974) used resistivity and induced polarization wide crushed, almost zero porosity, zones, often in the Fylde, and resistivity surveys have been lacking haematite. They often occur in swarms, widely used in many other areas of the UK and occasionally dominating the host rock and Europe. In recent times, electrical imaging has forming a wall-like feature 1 m or so wide often been increasingly employed (Fig. 9): its greater associated with adjacent main slip surfaces (Fig. resolution can improve fault and other structure 4a). Not surprisingly, granulation-seam orienta- recognition in the sandstone. Bunch (2000) was tions may parallel main slip-surface orientations unable to detect known sandstone-sandstone (e.g. Bouch et al. 2006) (Table 3). Reactivation faulting at shallow depth using resistivity is common, and complex structures can form imaging at one site, but more work on this needs (Edwards et al. 1993). to be done. Reflection seismic data have been Laboratory measurements indicate that the decisive in mapping the larger structures in NW seams typically have permeabilities of 3-6 England especially (Seymour et al. 2006), but orders of magnitude lower than the host sand- often data do not exist, features are too small stone (Fowles & Burley 1994; Ballard 2000; for conventional interpretation or depths are Main et al. 2001; Bouch et al. 2006). Fowles & too shallow. Burley (1994) also found that the permeability It is possible that in future, surface geophysics of the sandstone immediately surrounding the may also be able to provide indirect evidence of seams is increased (by 3-4 times in the case fracture flow, although this is likely to be diffi- described). Gutmanis et al. (1998) found a cult. Griffiths et al. (1981) demonstrated that negative correlation between 'potential flowing induced polarization could be used in tracing zones' and deformation band presence in pollution through fractures in the Permo- Cumbria. The effect on flow of the seams, and Triassic sandstone in the East Midlands. particularly where they occur in swarms, is likely However, no direct link to permeability was to be substantial. Fractures appear to cut found. New technology is almost certainly through the seams, but the latter often restrict required. It is possible that developing tech- the aperture, at least in outcrop (Fig. 7a). It is niques such as spectral resistivity (Scott & therefore possible that flow is funnelled towards Barker 2006) and magnetic resonance sounding the fractures leaving zones of limited flow. More (Lubczynski & Roy 2005) may hold the key to information is required on the probability of future large-scale indirect field characterization encountering granulation seams in different of sandstone properties. lithofacies (not necessarily an easy task as Gabrielsen et al. 1998 warn). Flow through the rock mass Veins. Veins of iron-rich rock up to a few centi- Flow in the rock mass occurs through both metres thick snake through the sandstone in matrix and fractures. Some partially filled or some places (Moore 1896; Walton 1981) (Fig. 8). filled fractures may have properties intermedi- Walton (1981) suggests their formation is due to ate between these extremes. Both fractures and pH rise following low pH groundwater flow into matrix have their low-permeability equivalents

carbonate-bearing units (Budleigh Salterton - some faults/granulation seams and mudstone Pebble Beds to Otter Sandstone Formation, or unusually well-cemented beds. Geologically, in his case; Table 1). Their permeability is the development of these features will not have unknown, but from visual inspection is likely to been independent, and there may be correla- be extremely low. tions in spatial distributions. On the other hand, Carbonate veins are also common in some effects of one permeability type may be moder- parts of the sequence (Milodowski et aL 1998; ated by the presence of the second. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 19

Fig. 8. Veining (looped structures to left of fault), Exeter, SW England.

Fig. 9. Resistivity imaging across the Birmingham Fault. Resistivity more than c. 60 f~m (i.e. everything to the left of 320 m, and below the line between 320 and 390 m) indicates Permo-Triassic sandstone; lower resistivities indicate mudstones.

At the regional scale, flow systems have been system is forgiving "when flow calculations are modelled apparently successfully on many involved, and that flow at large scales at least occasions using equivalent porous medium can be fairly readily represented by an equiva- approaches. Many models include some account lent porous medium. Only major hetero- of specific types of heterogeneity: for example, geneities affect the observation network decrease of fracturing with depth (NW England; sufficiently to require inclusion, and in some University of Birmingham/NWWA 1981), aquifers, where mudstones (Rushton & Salmon presence of mudstones (Midlands; Rushton & 1993) or cementation are not well developed, Salmon 1993) and regional faulting (Fylde area, even layered models are unnecessary - trans- NW England; Seymour et al. 2006). Sometimes missivity models still reproduce field data within account is taken of more than one feature (e.g. their uncertainty limits. Overall, with a few Furlong 2002), but it is rare that models, even of exceptions where the sandstone is much better pumping tests, include all the features described cemented, the apparent effective permeabilities in the previous sections in explicit form (cf. at the regional scale are often relatively close to Edwards et al. 1993). It thus appears that the laboratory measured values. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

20 J.H. TELLAM & R. D. BARKER

At the local (site/well) scale, and possibly at (Table 1). For flow parallel to laminations, they slightly larger scales, rock mass flow, however, found significant tailing of the breakthrough often appears to be strongly affected by frac- curves to occur as the tracer swept through each tures and matrix heterogeneities (Table 4). It is lamination at a different velocity (Fig. 11), and unclear over what distances local scale becomes positron emission projection images of larger regional scale: the method of Streetly et al. blocks also indicated differential movement (2006) offers a possible valuable method for through sets of laminae. The breakthrough can testing at intermediate scales (see also Streetly be described explicitly by modelling advective et al. 2000). This question is considered again movement along laminae with diffusive after solute transport has been discussed. exchange: dual region first-order transfer analytical models also can reproduce the tailing (Fig. lld), but the parameters are scale- Unreacting solute transport dependent (Bashar 1997). For solute movement perpendicular to laminations, untailed, Ogata & Introduction Banks (1961)-type breakthrough occurs, and this is also the case for unlaminated sandstones. Much less work has been completed on solute The limited laboratory data available appear movement than on flow, and the evidence is to suggest that, for the latter cases, kinematic rather fragmentary. Average linear velocities porosity is very similar to total porosity as expected from bulk permeabilities and a range measured by water saturation methods. of head gradients are expected to be slow Measurements of diffusion through the sand- (Fig. 10). stones appear to be rare (Bashar 1997; Braney et al. 2001). Bashar (1997) gives measurements Laboratory scale investigations for Br and amino-G-acid for 10 samples, with Di (i.e. free water diffusion coefficient • porosity/ Investigations of the movement of non-reacting tortuosity) values ranging between 9.9 • 10-11 solutes through laboratory samples have been and 2.4 • 10 -1~ m 2 s -1 for CaBr2 and 3.5 • reported by White (1986), Read et aL (1993), 10-11-9.6 • 10-11 m 2 s-1 for amino-G-acid, with Bashar (1997), Braney et al. (2001) and Bashar implied tortuosities (in this case, path length/ & Tellam (2006). Lovell et al. (2006) show how straight-line path length) of less than 1.5. Unfor- geophysical methods can provide detailed infor- tunately, there appears to be no diffusion mation on sedimentary structure at this scale, coefficient data for mudstones, and their role as and Greswell et al. (1998) imaged flow using a diffusive sink for solutes is uncertain. positron emission projection techniques. Bashar & Tellam (2006) investigated break- Borehole testing and pore-water sampling through of tracers in relatively low-permeability sandstone cores and slabs from the Wildmoor In their review of tracer tests in UK aquifers, Sandstone Formation of the West Midlands Ward et al. (1998) discovered only three

16oo- E K/~) = 40 "-" 1400- om ~ 1200- mO ~ 1000-

~ 800- K/r = 20 ,mC -J 600-

~ 400- . K/~) = 10 - K/~=8 ~ 200- - K/~=4 0-4 K/~=I 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 Hydraulic Gradient Fig. 10. Average linear velocity ranges for typical ratios of hydraulic conductivity (K) to porosity (~) and typical hydraulic gradients: K values in m day-1. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 21

(d) Relative Conc. C/Co

1 ...... --~ Experiment 1-3

~ Amino-G-acid 0.75 TRM

0.5 ORM

0.25

0 2 4 6 8 10 12 14

Pore volumes

Fig. 11. Laboratory-scale breakthrough of tracers (Bashar 1997; Bashar & TeUam 2006). (a) & (b) Permeability distributions through a slab and a small column of sandstone showing different lamination-scale permeability structures (Kay is the mean K for the sample assuming constant porosity). (e) Rhodamine movement through a core, flow parallel to laminations. The experiment was stopped before complete breakthrough and the core sectioned to observe sorbed tracer. (d) Non-reacting tracer breakthrough showing tailing, and one-region (ORM) and two-region (TRM) first-order exchange model fits. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

22 J.H. TELLAM & R. D. BARKER locations in the Permo-Triassic sandstones Similar tests have been carried out recently at where such tests had been completed, and at this two sites on the Birmingham University time lack of test data seems to have been typical campus, and four examples are shown in Figure worldwide for fractured permeable-matrix 12a-d. Although quantitative interpretations aquifers (Stafford et al. 1998). are not yet available, again, over the 7-10 m One test listed by Ward et al. (1998) was at distances involved, breakthrough curves are Heath House, near Hodnet, Shropshire, in the multimodal, with the earliest being so fast as to NW Midlands [national grid reference suggest strongly the presence of fracture flow. In SJ 6028 2459] (Coleby 1996). In the test, amino- one test, no breakthrough was achieved after G-acid was injected into an observation several pore volumes had been pumped, and borehole 20 m from an approximately 120 m- this may suggest that there was a barrier (granu- deep well that had been abstracting at 3.83 MI lation seam or vein) between the injection and day -1 for over a week and was assumed to be recovery boreholes in this interval (see Bouch et close to steady state. Breakthrough commenced al. 2006). More typical recoveries of tracer are at approximately 50 h after injection started, 60-80% at this site. Vertical tracer tests were and peak concentrations occurred at 105 h. The also carried out at the first test site, and break- breakthrough curve showed a single peak with through was again recorded despite the prominent tailing with 93% recovery of tracer. presence of intervening mudstone beds (Fig. Although detailed analysis was not attempted, 12e) (Sauer in prep.). the timing of the peak suggests an effective Qualitatively, these results appear to agree porosity of 0.14. with the conclusions of the two-dimensional The second location was at Haskayne, a (2D) numerical modelling work of Odling & research site north of Liverpool in NW England Roden (1997). These workers considered three (also called Plex Moss: see also Jones & Lerner systems, each with a heterogeneous isotropic 1995) [SJ 357 089]. Here three tests were matrix permeability, containing: (A) four attempted with injection of a slug of tracer from unconnected en echelon fractures; (B) a poorly one 0.4 m-long interval in a piezometer nest connected fracture system based on that approximately 5 m from an open borehole observed in a shale; and (C) a connected sealed by a packer at the elevation of the base fracture network based on that observed in a of the injection interval (Streetly et al. 2002). sandstone. The fracture/matrix permeability Fluorescein and amino-G-acid at the two ratio was set at 105 , probably rather higher than successful test intervals showed a multimodal seen in most UK Permo-Triassic sandstones. It breakthrough. As the first peaks appeared much was found that the fracturing increased the flow earlier than expected from calculations heterogeneity, even where fractures were assuming piston flow through the matrix, they unconnected, with maximum-minimum flow were interpreted as fracture flows, although it is rate ratios being 10 in the matrix-only system possible that they represent fast intergranular but more than 106 in systems B and C. Enhanced pathways. Unlike the earlier pulses, which had flow occurs in the matrix where there are dispersivities of less than 10 cm, the last pulse in 'bottlenecks' in the fracture flow systems, and, both tests was consistent with dispersivities of where the fracture flow predominates, flow in around 2 m, very large given that the travel the matrix is reduced resulting in zones of very distance was only 5 m. An apparent kinematic slow flow. This produces very heterogeneous porosity of approximately 0.2 was determined distributions of contaminants throughout the for these pulses: this is very close to the average modelled regions, even for the simple case A, porosity for the profile at this site (Segar 1993). with the downflow ends of fractures acting as The fastest pathways are 10-40 times faster than solute plume sources. At the downflow constant if the systems were homogeneous. Streetly et al. head boundary, most contaminant was (2002) estimated that there would have been discharged via fractures, despite most mass 30-45% recovery had the tests been continued being held within the matrix at distance from to apparent completion. During the drilling of the boundary. The breakthrough curves were one of the test boreholes, breakthrough was multimodal, the matrix breakthrough only being observed at all 1 m-long intake zones within a recognized as a distinctly identifiable entity in piezometer nest 15m away: calculations case A where only four fractures were present. assuming homogeneous piston flow suggest that Rubin et al. (1996) show that the complex flushing to only 1 m radius from the borehole breakthrough at small distances in fractured should have taken place, and again this strongly permeable media becomes similar to that suggests the presence of transport in (frequent) obtained in an ideal non-fractured permeable fractures. medium at longer distances, as might be Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 23

(a) (b) Concentration (ppb) Concentration (ppb) 40- 40-

20- 20-

0 i 0 5 10 15 / m Time (days) 0 i / I o Io 20 0 40 Time (days) (c) 2.0E-03] I -0- Fluorescein I: 9 Rhodamine WT ~ 1.0E-O3J

t- O 5.0E-04

O.OE+OON~P- 0 0.2 0.4 0.6 0.8 I 1.2 1.4

Time (hours) (d) (e) 1.0E-04- ~-- 1.5E-O~ C

i 5.0E-05

8 O.OE+O~ , , , -- 0 5 10 15 O.OE+OO] , , , Time (hours) o 24 48 72 Time (hours)

Fig. 12. Example tracer tests at Birmingham University Campus sites. (a) & (b) Breakthrough over 7 m between packered, apparently unfractured, intervals in the boreholes described by Bouch et al. (2006) (Joyce et al. 2006). (e) & (d) Breakthrough in fractured and developed intervals over similar distances between pairs of boreholes approximately 100 m distant from those of (a) & (b) (Sauer in prep). Note that pumping rates/interval thickness ratios for the tests vary considerably. (e) Breakthrough at the site of (c) & (d) resulting from forcing the tracer to flow vertically through the profile across mudstone beds (Sauer in prep.). expected. Modelling work by Stafford et aL the unusually large spreading of plumes in a (1998) shows that a small number of higher fractured permeable-matrix aquifer. permeability truncated fractures in an aquifer The third tracer test listed by Ward et al. with a low-permeability matrix could explain (1998) was undertaken by Barker et al. (1998) Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

24 J.H. TELLAM & R. D. BARKER by injecting fluorescein into an observation profiles from piezometer and core pore-water borehole and recording the breakthrough at studies often show variations on metre or even inflows along a tunnel in Liverpool (Fig. 2). submetre scale (e.g. Tellam 1995a; Kinniburgh et Breakthroughs were seen in seven of the eight al. 2006), method error estimates are difficult to inflow points into the tunnel, the fastest being at quantify, and are rarely available. a straight-line rate of approximately 140 m day-1 However, pore-water and piezometer profiles over about 280m. This is far faster than through plumes do confirm that transverse expected from average groundwater flow rates dispersion is often limited (<1-2 m) for the in this system, again very suggestive of fracture distance-time scales of even large pollution flow. Note that this test differs from conven- plumes (Lerner et al. 2000; Thornton et al. tional tests in that the observation point for 2001b; Taylor et al. submitted). On larger space breakthrough was not defined before the test and timescales, considerable lateral dispersion began, and hence the solute pathway was much can develop: for example, the dispersion profile less constrained compared with the other tests associated with a fresh-saline groundwater described here: although the head gradients in interface in an aquifer in NW England is many this test were much greater than in most non- tens of metres thick, despite the fact that most test situations, the fact that the pathway was of these profiles are in poorly fractured, deeper only poorly constrained may be more typical of (100-300 m) parts of the aquifer (Tellam et al. many systems. The pathway may correspond 1986; Brassington et al. 1992; Tellam 1995a; with either the large subvertical N-S fractures, Tellam & Lloyd 1997). In this case, the role of or the large bedding-plane fractures identified mechanical dispersion and diffusion warrants by Hitchmough et al. (submitted) (Table 3). further investigation (cf. Rubin & Buddemeier Taken together, the tracer test results 1996). Deep pore-water profiles provide an summarized here confirm the importance of insight into depth of 'active' flow, but such data fracture flow near wells, and indicate that some are very site specific: for example, in the case of fracture flow can occur over distances of up to the profiles in NW England, the interface at least hundreds of metres, but that often a between the fresher and more saline water significant percentage of tracer is not recovered occurs from less than 50 m to approximately or only very slowly recovered. 250m below ground level depending on Given the effect of layering on breakthrough location within the regional flow system (Tellam seen in the laboratory, a similar phenomenon et al. 1986; Brassington et al. 1992). Example might be expected in the field, at least over short studies of pollution penetration depths include longitudinal distances. However, diffusional Stagg & Tellam (1998), Cronin et al. (2003) and effects will be less marked as transverse Taylor et al. (submitted). distances are significant: characteristic diffusion There are some interesting insights emerging times for a range of typical bed thicknesses are from work on particulates that have relevance given in Figure 13. Although detailed chemical for dissolved solute movement. Viable human

14000 ~ 400 12000 ,_'-u 600t J .mE ~ 200 10000 ~i= 0,-- , , 0 u 8000 o 1 2 Distance (m) j a .~ 6000

4000

~= 2000 U o I i i | I i 0 2 4 6 8 10 12 Distance (m)

Fig. 13. Characteristic diffusion times [=distance2/(2 x diffusion coefficient)] over distances equivalent to bed thickness ranges. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 25 enteric viruses have been found in pumped well Geophysics waters in Birmingham and Nottingham by Powell et al. (2000, 2001), and subsequently Geophysical techniques may be used to charac- recovered in piezometers to at least 50 m below terize the extent of groundwater contamination, ground level (Cronin et al. 2003; Powell et al. with the application of rapid EM (electro- 2003). This is significant given that survival times magnetic) mapping techniques being popular in for such viruses are usually considered to be in cases where the plume concentrations are terms of months. Some piezometer samples significantly greater than the background even appear to show concentration seasonality groundwater. Vertical detail may be provided by in phase with that in the human gut (Powell et resistivity imaging (e.g. Jones et al. 1998), but al. 2003; Taylor et aL 2004), although this needs heterogeneity of sandstone properties can confirmation. All field detections are at concen- sometimes obscure some of the finer detail. trations up to at least eight orders of magnitude Time-lapse imaging offers a potentially below probable source concentrations - virus powerful tool for monitoring movement of con- detection is possible over very considerable tamination, particularly where concentrations concentration ranges. It is unclear whether the are expected to change quickly (Barker & presence of the virus particles suggests an Moore 1998). In this case, features that are extremely small proportion of fast pathways too small to be directly detected can some- through the sandstones or greater survival times times be detected because they affect flow than previously thought for viable viruses; significantly (cf. Hatzichristodulu et al. 2002), confirmation of seasonality is therefore very and the examination of resistivity changes important. Recent bacteriophage tracer tests rather than absolute values can help circum- suggest breakthrough pathways are indeed rare, vent the problem of sandstone property even over lateral distances of 7 m (Joyce et al. heterogeneity. 2006; same site as Bouch et al. 2006, and Fig. 12a, b). Regional-scale investigations There is other evidence for fast pathways between ground surface and pumped wells. Using groundwater sampling survey data, a few Modelling studies by Tellam & Thomas (2002) studies have found significant changes in and Butcher et al. (2006) indicate that chemical chemistry across faults, confirming their effects steady state following uniform addition of a on solute transport at the regional scale: for pollutant at ground surface is only achieved for example in Birmingham, central England typical sandstone wells after a few hundred (Jackson & Lloyd 1983) and Merseyside, NW years; both studies were able to show that time England (Barker et al. 1998; Seymour et al. to breakthrough could explain the observed 2006). wide range of pumped well pollutant concentra- Few appropriate, long-term, detailed tions. However, both studies also found that chemical data sets exist that show the break- predicted first breakthrough times were some- through of a significant change in water chem- times too long. Butcher et al. (2006) found that istry. An exception is the data set from estuary even finite-length fractures extending horizon- water intrusion in Widnes, NW England (Fig. tally from a well could reduce first breakthrough 15) (Carlyle et al. 2004). Here an apparently times considerably. well-behaved breakthrough occurred over a Furlong (2002) takes the argument further. scale of c. 2 km/40 years that can be self- He applies time-variant 3D regional-scale consistently modelled using an equivalent modelling to determine ages of groundwaters at porous medium approach with observed head the time of sampling, and is able to demonstrate gradients, realistic matrix permeability/porosity that pumped well waters can show very great ratios and dispersivities within a normally age differentials not only with depth but also acceptable range for the distances travelled with the direction of inflow into the well (Fig. (10-100 m, depending on assumptions). This is 14). Although his analysis does not take into interesting given the model's gross oversimpli- account the local heterogeneities around the fication of a system that is undoubtedly frac- sampled wells, the implied potential for mixing tured, faulted and 3D in geometry. However, is qualitatively consistent with the mixed the data show some scatter and, as most wells isotope ages seen in many pumped well samples, do not approach complete breakthrough, some including those he modelled (Tellam 1994; see detail in the breakthrough curves may have also Bath et al. 1987). been missed. Regional patterns are often discernable in pumped water sample chemical concentrations. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

26 J.H. TELLAM & R. D. BARKER

Fig. 14. Ages of water entering example pumping wells in the Lower Mersey Basin aquifer at the time of sampling in 1980, and locations of recharge, as predicted by 3D transient modelling (Furlong 2002). Area marked as sandstone is bordered to the north by underlying low-permeability Carboniferous, and bordered to the south by the overlying Mercia Mudstone Group: Quaternary deposits (sands, clays and peat) cover most of the sandstone area.

The overlying sequence usually exerts the major (Tellam 1996). However, as indicated in the control by its effect on flow distributions (and previous section and implied in well-catchment ages), be it the Triassic mudstones (e.g. models (e.g. Papatolios & Lerner 1993; Evers & Edmunds et al. 1982; Jackson & Lloyd 1983; Lerner 1998; Trowsdale & Lerner 2003), water Rivett et al. 1990; Ford & Tellam 1994) or the arriving at a well may be from a wide range of Quaternary deposits (e.g. Sage & Lloyd 1978; flow paths (Fig. 14), and the regional patterning Parker et aL 1985; Tellam 1994). There has even should be considered as patterning of well- been some success at using simple rule-based water chemistry rather than groundwater systems for describing regional distributions chemistry. Even quite simple descriptions of 3D Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 27

1.5 - Mersey Basin indicates a very complex pattern of ages reflecting recharge zones, hydraulic 'traps' associated with discharge zones and low- o 1- permeability faulting, and the regional effects of 0 abstraction. o 0.5- There is much chemical and isotopic evidence that very old water is present in some of the sandstone aquifers (Bath et al. 1979; Edmunds _ et al. 1996, 1997; Darling et al. 1997, 2003). In some cases these Devensian or older 'palaeowa- 1900 1950 2000 ters' occur because discharge is slowed by the presence of the overlying Mercia Mudstone Year Group, for example in the case of the East Midlands aquifer (Andrews & Lee 1979; 1.2 Edmunds et al. 1982; Andrews & Kay 1983; Bath et al. 1987) or the Birmingham/West Midlands 1 aquifer (Lloyd 1976; Jackson & Lloyd 1983). The saline waters at the base of the sandstone 0.8 in NW England have 180, 2H and U series 0 isotopic signatures indicating significant dilution o 0.6 I0 o by Devesian (?subglacier) (Tellam 1995a) and 0.4 post-glacial/pre-industrial (Ivanovich et al. 1992) recharge: although these waters are not 0.2 confined by the Mercia Mudstone Group, they have been prevented from being flushed during 0 low glacial base levels by low-permeability faults (see Seymour et al. 2006) and these faults 1900 1950 2000 also contribute to the present relatively ineffi- Year cient flushing by regional flow (cf. Furlong 2002). Sparse data suggest the deepest saline water has a heavier isotopic signature. 1.2

Summary 0.8 The evidence reviewed above suggests that: 0.6 9 in laboratory column experiments, solute 0.4 movement is multiregion, with diffusion playing an important role; 0.2 9 in approximately 10 m borehole-to- 0 borehole forced-gradient tracer tests, fast pathways are important, with overall 1900 1950 2000 dispersion being large and breakthroughs often multimodal (and incomplete); this is Year consistent with the experience of borehole hydraulic tests and with the heavily skewed Fig. 15. Example breakthrough curves for estuary distribution of fracture trace lengths (Fig. water intrusion in response to a regional head 7c), but the presence also of fast inter- gradient, Widnes, Lower Mersey Basin (Carlyle et al. granular pathways cannot be ruled out; 2004; data from Widnes Water Users Association). 9 fast pathways appear to exist from ground One-dimensional Ogata & Banks (1961)-style surface to pumped borehole and unpumped breakthrough curves also indicated. piezometer intakes at up to at least 50 m depth; 9 in one forced-gradient borehole-to-tunnel transient flow systems imply a considerable tracer test over about 200 m, single frac- complexity in the distribution of groundwater tures (possibly members of Set 1 or Set 6 in ages, as the work of Toth (1963) onwards has Table 3) were important; indicated. Furlong's (2002) model for the Lower 9 at the regional scale, fracture flow becomes Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

28 J.H. TELLAM & R. D. BARKER

less obvious, and in the only long-term data are even available, for dissolution if not for set of which we are aware apparent disper- precipitation. sivity appears to be much as expected. However, because such minerals are often sparsely distributed, and although locally the in Note that it may be possible for different situ groundwater may be saturated, ground- results to occur at the same scale when different water samples, being averaged over a larger methods are used (e.g. comparing tests where scale, are often not at saturation (Atteia et al. the sampling points are defined with those 2005). For example, high [SO4] but gypsum- where they are not). undersaturated groundwaters from the confined zone of the Birmingham (Fig. 2) aquifer are Movement of reacting solutes interpreted by Jackson & Lloyd (1983) and Hughes et al. (1999) to derive from dissolution Introduction of patchy gypsum cement. Edmunds et al. (1982), Smedley & Brewerton (1997), Shand et Most of the available chemical studies attempt al. (1997) and Edmunds & Smedley (2000) also to explain rather than predict reacting solute deduce dissolution sources for Li, Cs, K and Sr movement, an indication of the inherent in the East Midlands and Cumbrian aquifers difficulties. Four reaction classes - dissolution- (Fig. 2): with distance down flow gradient, the precipitation, reduction-oxidation, acid-base concentrations increase, again suggesting a and sorption-desorption are considered in turn source limitation. below. Discussion will be limited to sandstone- In principle, for a particular scale of measure- groundwater interactions, despite the import- ment, such apparently kinetic dissolution should ance Quaternary deposits can have on recharge be quantifiable, given adequate data on mineral water chemistry (e.g. Spears & Reeves 1975; distributions and their relationships with Sage & Lloyd 1978). hydraulic properties. Alternatively, the observed chemical data could be used to investigate the Dissolution-precipitation reactions latter. Considerable amounts of data exist on mineral distributions as a function of depth and Of the minerals commonly found in the sand- sometimes of lithology (e.g. Jones et al. 1999), stones (Table 5), a few will dissolve in a simple but such chemostratigraphic data appear not to and quantitatively predictable way (gypsum, have been examined for this purpose (and the barite and fluorite): in some cases kinetic data role of potentially protective oxide coatings is

Table 5. Main mineral components of the sandstones, and suggested main water-rock interactions Mineral Main origin Role

Quartz Detrital/framework (authigenic) Source of Si K-Feldspar Detrital/ffamework (authigenic) Source of Si, trace determinands Plagioclase Detrital/framework Source of Si, trace determinands Lithics Detrital/framework Source of Si, trace. ?removal of 02 in long term by Fell release Mica Detrital ?removal of 02 in long term by Fe II release. Source of trace determinands Clays Detrital, authigenic Sorption Calcite Authigenic (/detrital) Source of Ca, CO3, minor Sr, Mg. pH control. ?removal of 02 in long term by Fe II release. Possibly some sorption Dolomite Autliigenic (/detrital) Source of Ca, Mg, CO3, minor Sr. pH control. ?removal of 02 in long term by Fe II release Gypsum Authigenic Source of Ca, SOn. Effect on pH via Ca-carbonate equilibria Fe ox/hyd Authigenic pH; cation (/anion) sorption. Oxidant (?small %). ?catalysis of redox reactions, pH control in absence of carbonates Mn oxides Authigenic Sorption of cations (anions). Oxidant. ?catalysis of redox reactions Organics Detrital Sorption of organics Pyrite Authigenic Source of Fe, SO4. Reductant Flourite Authigenic Source of F Heavy Detrital Source of trace Barite Authigenic Source of Ba Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 29 unclear). At a site adjacent to that described by studied to any great extent in the specific Bouch et al. (2006), Mitchener (2003) could context of the sandstones, although dissolved establish no correlations between mineral organic carbon concentrations are low (see composition and permeability or porosity, with below). Colloidal transport can also increase the exception, curiously, of a positive correlation apparent solubility. Even in pumped well between permeability and feldspar content. samples, total colloid concentrations appear to Except at interbasin scale, knowledge of mineral be less than a few mg 1-1 (c. 1011 particles 1-1) distributions in horizontal directions is less well (Stagg et al. 1997, 1998; Stagg & Tellam 1998; documented. Clearly, this distribution issue is Stagg 2000): most of the colloids are dominantly common to any groundwater reaction involving composed of silicates, with significant but a solid phase. secondary amounts of Fe- and, at shallow As in many aquifers, especially deep ones, depths, organic-rich particles (Stagg et al. apparent equilibrium states in pumped water 1998a). Metal-rich particles have been identified samples are affected by mixing within a well in polluted urban groundwaters, but not at profile. Examples are shown in Figure 16: here concentrations that approach water quality groundwater is saturated throughout the depth standards. However, the number of sites that profile with barite, but shallower waters have have been examined is limited, and little is lower [Ba]/[SO4] ratios; due to the 'algebraic' known of situations such as landfill leachate effect (Plummer 1975), the pumped water is plume migration or recharge through peat oversaturated. In theory, with different compo- deposits where much higher concentrations of sitions, undersaturation also can similarly occur. particulate matter might be expected. Such issues obscure the geochemical interpretation of pumped samples. Reduction-oxidation reactions Complexation by dissolved organic matter may increase the apparent solubility (especially The reduction capability of the aquifer, and for those species where acid-base and/or reduc- redox sequences. Classical redox sequences (e.g. tion-oxidation are involved in the dissolution Champ et al. 1979) have been mapped within process): such effects appear not to have been the sandstones at a regional scale: Edmunds

Number of 5O Number of Samples 60 Samples

1o 20 A | V V i -4 -3 -2 -1 o 1 2 | Calcite Saturation Index -4 -2 0 2 Barite Saturation Index

500 .' NH4 and Fe (mg 1-1) 9 Fe 400 9 9 300 JII 9 NH4 200 m 9 | 9 100

0 2 4 6 8 10 12 02 (rag I-I) Fig. 16. Disequilibrium in pumped well samples from the Birmingham aquifer: calcite and barite saturation indices, and NH4 and 02 concentrations. Barite colloid particle (<1 pm) also shown. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

30 J.H. TELLAM & R. D. BARKER et al. (1982, 1984), for example, describe the possible that high Fe concentrations suggest in sequential removal of 02, NO3 and SO4 with general a source of reductant outside the sand- distance downflow in the East Midlands aquifer stone: often this will be either from Quaternary as the sandstone dips below and continues deposits or from pollutant sources (e.g. Tellam under the overlying Triassic mudstones. This 1996; Torstensson et al. 1998). pattern is also recognized in aquifers with more Another sink for dissolved Fe II is reduction complex, Quaternary deposit-influenced, flow of Mn oxides (Thornton et al. 2000b). Mn oxides systems in northern England being particularly are likely also to play a significant role in a noticeable in the distributions of Fe II, NO3 and number of redox transformation processes O2 (e.g. Parker et al. 1985; Tellam 1994). In involving metals (e.g. Cr) and possibly N general, detailed bacterial or H2 studies have compounds (e.g. Guha et al. 2001), although not yet been undertaken except in research on studies specific to the sandstones appear as yet ?one contaminant plume (Thornton et al. to be lacking. 2001b). Nitrate reduction (sensu lato) is suspected to Given that the sandstone is usually coated in occur in many of the sandstone aquifers, but ferric minerals, what are the electron donors? In evidence is often equivocal. However, in some some cases the overlying Quaternary deposits cases denitrification has been positively identi- supply the reductants, usually in the form of fied using N2/Ar and/or isotope data and/or organic matter or sulphides. Where Quaternary bacterial data (Wilson et al. 1994, in only two of deposits are missing, some dissolved and par- 23 samples; Parker et al. 1988; Spence et al. ticulate organic matter will enter with recharge. 2001a, b; Williams et al. 2001; Fukada et aL 2004; Buss et al. (2005) give a sandstone average Buss et al. 2005). In some cases - especially dissolved organic carbon (DOC) concentration where the nitrate source is also organic-rich - of approximately 2.8 mg 1-1, but two discrete there will be enough organic matter to facilitate depth studies on sites with limited Quaternary NO3 reduction, but elsewhere there may be cover (Edmunds et al. 1992:0-5 m depth; Stagg other, as yet uninvestigated, processes occurring et al. 1998: 0-50m) suggest that in this case involving Fe II or Mn, for example (e.g. Staub et DOC concentrations may be lower, falling al. 1996; Luther et al. 1997). Records of human rapidly with depth to less than c. 1 mg 1-1. Some pathogenic bacteria at depth (e.g. Barrett et al. of this decrease may result from 02 reduction, 1999; Cronin et al. 2003; Taylor et al. 2004), the but some may also result from reduction of Fe presence of NO3 and SO4 reducers (e.g. Pickup III and Mn (III) IV oxides and hydroxides (or et al. 2001), and the positive identification of possibly from filtration if some of the load is not NO3 and SO4 reduction (see above) suggest that truly dissolved) (e.g. Banwart 1999; Banerjee & the mobility of bacteria within the sandstones is Nesbitt 1999; Klewicki & Morgan 1999). not limiting. Sandstone median pore diameters Edmunds et aL (1982) suggest that a long-term are typically tens of microns (Bloomfield et al. (approximately several thousands of years: 2001), although mudstone units may provide Smedley & Edmunds 2002) sink for 02 may be more substantial barriers: the dimensions of Fe II release as ferroan carbonate is dissolved micro-redox-environments, if they exist as such, (and ferro-magnesian minerals, continuing the are unknown. Overall the capacity for NO3 diagenetic process of red-bed production?). reduction is almost certainly limited (Smedley Petrographic and redox titration (Jones 2001) & Brewerton 1997; Buss et al. 2005). data indicate that small amounts of reactive Sulphate reduction has been detected in a reduced mineral phases, other than those number of aquifers (e.g. Edmunds et al. 1982; mentioned above, do sometimes occur even Tellam 1994), but sulphides are almost always at within the non-bleached red sandstones. low concentrations, and even 34S results are Once 02 has been removed Fe concentrations sometimes equivocal (Barker 1996). Sulphide usually rise, although the mechanism again has sinks include aqueous Fe II-solid phase Fe III not in general been quantified: in particular the (cf. Mayer et al. 2001). Apart from cases of electron donor is usually uncertain. Smedley & pollution (including by drilling fluid polymers: Edmunds (2002) suggest that for the East Barker 1996), there are few examples where Midlands aquifer, where 02 removal appears SO4 has been substantially removed within the to have been predominantly from water- aquifer by reduction, the obvious limitation sandstone interaction over long periods, the rise again being the lack of organic matter. Where in Fe II concentrations is modest because either reduction is occurring, the mechanisms in the solubility of the bulk of the Fe oxides is general have not received detailed study. limited, these being predominantly crystalline Methane has been detected in a number of haematite, or because of lack of reductant. It is locations (Tellam 1994; Gooddy & Darling Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 31

2005). Possible allochthanous sources include but may be greater near the surface where overlying Quaternary peat and underlying coal- authigenesis in the unsaturated zone is bearing sequences. The role of CH 4 in sulphate commonly observed. reduction is unknown. When oxides are dissolved, their sorbed loads From H2 and other data from a detailed are also released into solution as seen, for study of a large phenol plume in the West example, by Thornton et al. (2000b) in the Midlands, Thornton et al. (2001b) concluded laboratory and Lewin et al. (1994) in the field: that under some circumstances, in their case there is no information as to whether the oxides high concentrations of phenol inhibiting most prone to reductive dissolution are also biodegradation (Pickup et al. 2001), the those that have the greatest sorption capacity. classical redox sequence may not be followed: at their site, they suggest that phenol oxidation Fossil evidence. 'Fossil flow features' also occur may be possible via NO3, SO4, Fe and Mn in the sandstone sequence where reducing fluids pathways simultaneously. Unfortunately, lack have passed through removing or reducing the of mineralogical, thermodynamic and ATP iron oxides/hydroxides and bleached the rock threshold data prevented fuller quantitative (e.g. Rowe & Burley 1997; cf. Parry et aL 2004), analysis, but a method was suggested. This situ- turning it from red- to grey- or buff-coloured. ation is unlikely to be unique to organic cont- The distribution of bleaching often can be seen aminants. Such analysis can only be attempted to be influenced by changes in permeability in cases where detailed small-scale sampling associated with lamination, mudstones, granula- has been undertaken: most chemical data avail- tion seams or slip surfaces (Fig. 17). Sometimes able are from pumping wells, and as such bleaching can be traced laterally for over 100 m, mixing will always reduce the resolution, often often running parallel with bedding, despite producing disequilibrium effects not actually being only a few centimetres in thickness, present in the groundwater (Fig. 16). suggesting large volumes of flow and limited transverse dispersion. In some locations, the The oxidation capability of the aquifer. The bleaching affects almost all the rock, and in presence of a few per cent Fe III in the sand- other places complex patches suggest tortuous stones suggests considerable oxidation capacity. 3D flow paths. The presence of occasional oil However, as the previous section implies, this seeps in the sandstones suggests that even apparent potential capacity is often not fully flushing, albeit natural, over millions of years realized (see also Thornton et al. 1995, 2000a, b, does not necessarily remove all 'pollution' in all 2005). In the detailed field and laboratory cases. investigation of the phenol plume mentioned above, mass balance and microcosm studies Acid-base reactions strongly suggested that only a very small amount, less than 0.15%, of the apparent Fe III If it is present, dissolution of carbonate will and Mn IV capacity of the aquifer was used in usually buffer the pH to around neutral in oxidizing the organic load (Harrison et al. 2001; recently recharged groundwaters. Incongruent Thornton et al. 2001a, b). Smedley & Edmunds dissolution, especially of dolomite, can subse- (2002) report oxalate-extractable Fe, i.e. an quently increase the pH (and 13C) considerably indication of 'amorphous' Fe, of <500 mg kg -1, (e.g. Edmunds et al. 1982; Bath et al. 1987), as which is less than 2% total Fe. In the case of the can processes that rejuvenate carbonate dis- phenol plume, over 90% of the observed degra- solution, for example Ca removal by ion dation used dissolved 02 and NO3 supplied by exchange (e.g. Tellam 1994). In the case of dispersive processes in the 2 m-wide dispersion polluted recharge water of low pH, buffering by zones at the edges of the plume (Lerner et al. the aquifer is, of course, dependent on the 2000). composition of the intruding solution and its Whether this oxidizing capacity is typical is existing carbonate saturation state (e.g. uncertain, and capacity will vary with reaction Thornton et al. 2000b, 2001a). system. In some systems, oxidative capacity may In many of the outcrop areas, carbonates have be underutilized because of other limitations. been removed from the rock by previous dis- Of the two, MnO2 is more labile than Fe III solution, and pHs of less than 5 are common oxides/hydroxides in the aquifer. Accessible (e.g. Greenwood & Travis 1915; Edmunds & MnO2 is finite, perhaps 50-100 mg kg -1 (not Kinniburgh 1986; Edmunds et al. 1992; Moss & much less than oxalate-extractable concentra- Edmunds 1992; Tellam 1995b; Shand et al. 1997). tions determined by Smedley & Edmunds 2002) In other cases, certain formations are apparently (Thornton et al. 1995; Kinniburgh et al. 2006), carbonate-free, or nearly so (Edmunds & Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

32 J.H. TELLAM & R. D. BARKER

addition, if the pH is being buffered at a low value, silicate dissolution will occur to a significant extent; although not as quick as most sorption reactions, saturation with simpler silicates appears to occur within hydrogeologically small travel distances (e.g. Haines & Lloyd 1985; Edmunds et al. 1992). Buss (1999) has presented a quantitative model of the interaction of H § with samples of the sandstone over short timescales, but, in common with similar aquifers elsewhere, a generic useable quantitative description of the complex array of processes involved has yet to be developed.

Sorption-desorption reactions Cation-exchange capacities (CECs) for the sandstone are often in the range 1-20 meq/100 g dry mass (Gillespie et al. 2001; Mitchener 2003; Carlyle et al. 2004). Although most frequently at the lower end of this range, and therefore modest, there is abundant laboratory (e.g. E1- Ghonemy 1997,1998; Thornton et al. 2000b) and field evidence (e.g. Tellam & Lloyd 1986; Lucey 1987; TeUam 1994) that exchange can modify concentrations significantly. The main exchange phase appears to be provided by clay minerals, but Mitchener (2003) found that for the sequence described by Bouch et al. (2006), at pHs of 6.5-7.5, approximately 50% was supplied by oxides, in particular Mn oxides. Mitchener (2003) also found a complex relationship between permeability, porosity, CEC and MnO2, implying that when MnO2 is present it increases the CEC significantly which then decreases the significance of the (negative) permeability/CEC correlation: however, a (negative) POrosity/ CEC correlation remains significant whatever the MnO2 content. Taylor & Barker (2006) show that cation- exchange capacity is correlated in the sand- Fig. 17. Three examples of bleached zone outcrops, NW England, showing respectively: effects of stones with matrix conduction, opening up a faulting; flow controlled by cross-lamination; and 3D possible approach for assessing CEC. Scott & sinuous pathways. Barker (2005) found that surface area is an important influence on the quadrature conductivity in spectral induced polarization measure- Morgan-Jones 1976; Walton 1981). In these ments. systems, there is often still a considerable At a regional scale, in common with many buffering capacity due to sorption-exchange sand aquifers, there is evidence that the Gaines- and dissolution-precipitation processes. The Thomas convention provides an acceptable processes are as often suggested for acid soils approximation for predicting major cation- (e.g. McBride 1994), including A13+ hydrolysis, exchange reactions (Carlyle et al. 2004): estuary base cation/H+/A13+/Fe/Mn exchange, and dis- water intrusion over a period of about 40 years solution (Moss & Edmunds 1992; Thornton et al. was predicted successfully using laboratory 2000b). If the intruding water is reducing (e.g. a determined exchange parameters. However, landfill leachate, as for Thornton et al. 2000a, b), further detailed laboratory investigations further complications arise through reductive suggest that an even better description may be dissolution, even if limited, of the oxides. In afforded by the Rothmund-Kornfeld power Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 33 function relationship (EI-Ghonemy 1997, 1998; occasionally pollutant concentrations rise to Tellam et al. 2002), an empirical approach which tens or, even, hundreds of mg 1-1 (Ford & Tellam appears to produce constants independent of 1994). sorbed-site composition. Parker et al. (2000) and Again, little is known of anion sorption. Parker (2005) show that exchange on reference Limited work has been done on F- (Gresswell samples of the dominant clays present in the 2005), but very little apparently on phosphate: sandstone, and mixtures thereof, can also be neither is commonly at high concentration in described well using the power function the sandstone groundwaters, although there is approach: similar consideration of the other evidence that F- distributions are pH-depen- possible exchange phases has yet to be under- dent as might be expected. taken. Preliminary work suggests that linear Sorption of organic pollutants to the sand- up-scaling may be appropriate in cases where stone has been the subject of a few studies, the range of property variation is modest including those of Williamson (1994), Barrett (Parker 2005). Furlong (2002) suggests that ion (1995) and Shepherd (2003). The fraction of exchange may allow groundwater age (up to inorganic carbon within the sandstones is about 15 pore volumes) to be estimated in limited, being between 0.001 and 0.15% certain circumstances in the sandstones using (Williamson 1994; Barrett 1995; and Thornton et such quantification. al. 2000a; Steventon-Barnes 2001; Shepherd The laboratory experiments of Thornton et al. 2003). The role of organic coatings, if they exist, (1995, 2000b, 2005) suggest that NH4 § exchange on the rock surface is unknown, as is the distri- behaves very similarly, as might be expected, to bution of the detrital organic matter. Mitchener K § exchange, and this seems to be the case in (2003) could find no significant relationship the field too (Lewin et al. 1994). Usually a Kd between foc (fraction of organic carbon) and approach is applied when predicting NH4 migra- permeability or porosity for a core from the tion (Erskine 2000; Gillespie et al. 2000; Jones Wildmoor Sandstone Formation (Bouch et aL 2001; Adey 2004; Buss et al. 2004), and the 2006). Farris (1999) found for sandstones from experiments of Thornton et al. (2000b) show the Midlands and the Central North Sea that that breakthroughs for NH4 § (and K +) can organic matter was correlated with the finer indeed be described by retardation factors: deposits, and completely missing from more geochemical modelling work generally supports than 90% of the sequence. Interestingly, as was this conclusion, but also suggests that the found for MnO2 contents as mentioned above, appropriate Kd value is very difficult to predict/ there is a correlation between organic matter measure without extensive knowledge of the content and colour. system (Clarke 2005). Sorption of chlorinated aliphatics, in particu- Sorption of metals/metalloids to the sand- lar tetrachloroethene, has been investigated by stone has not been investigated systematically. Shepherd (2003). He found approximately The results of some empirical studies are avail- linear isotherms with slopes that vary with lithol- able (e.g. Mimides & Lloyd 1987; A1-Hosni ogy; for the more hydrophobic compounds, he 2001; Thornton et al. 2000b), but few attempts concluded that sorption to inorganic phases was using recent advances in surface modelling not important, despite the low fractions of appear to have been published (Read et al. organic carbon in the sandstones. Thornton et al. 1993). Field studies usually conclude that pH- (2000a), who investigated migration in landfill dependent sorption on the oxide fraction is leachates, found less partitioning to the sand- particularly important (e.g. Spears 1986; stone than Shepherd (2003). Barrett (1995) Edmunds et al. 1989; Ford et al. 1992; Ford & found that co-solutes increased the sorption of Tellam 1994; Smedley & Edmunds 2002; xylene. Kinniburgh et al. 2006; Shepherd et al. 2006). Much has yet to be learnt of the sorption behav- Synthesis iour of the oxide/hydroxide fractions, and how they correspond with those fractions also The following attempted synthesis is broad and actively involved with redox reactions. Where based directly on the evidence summarized MnO2 reductive dissolution occurs, sorption above: it is hence biased towards the shallow capacity will be decreased at the same time, but English succession, for which most evidence is does this also occur with some of the Fe III available. Table 6 provides more detail. Scale is oxides? The sorption capacity is far from domi- stressed. However, it should be noted that the nating: even natural metal concentrations can scales of heterogeneity, measurement and exceed water quality standards (Smedley & interest are not necessarily independent of each Edmunds 2002; Kinniburgh et al. 2006), and other: for example, measurement devices often Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

34 J.H. TELLAM & R. D. BARKER

Table 6. Summary of evidence and conclusions (?, limited evidence; ??, very limited evidence; OLL,longitudinal dispersivity; ST, transverse dispersivity; +k/+y., = ratio of kinematic porosity to total porosity; B TCs = breakthrough curves) Feature Scale <30 m 30--few 100s of Few 100s of More than a few km metres metres-few km

Breakthrough Multimodal and/or Usually unimodal? Unimodal? Unimodal?? highly dispersive Lateral fracture Often obvious, Rare None?? None? pathways becoming less so with greater distance Antifractures Least likely to More likely to May encounter Likely to encounter, encounter, but, when encounter at least in some present, can be regions dominant Vertical Small amount of rapid 10s of metres? Many 10s of metres? Penetration to at least penetration penetration from 250 m in places ground surface Matrix barriers Major effect, but does Lesser effect Lesser effect Overall seen as not completely stop anisotropy solute migration Large given the scale Moderate given Moderate given Moderate given ~L (but not a good measure scale? scale? scale?? given multimodal BTCs) ~x High overall, but limited Can be high overall, ?? Can result in mixing at edges? but limited at plume over many lOs of edges? metres? (~k/(~ ~ Lowest ---) 1? ---) 1? ~ 1?? Lithofacies 'Microcycles' significant 'Microcycles' Microscale much 'Microcycles' /lithofacies for flow and chemistry becoming less less significant; insignificant; association significant 'mio/macro'- macroscale importance sequences potentially significant increasingly significant Chemical Greater chance of low Greater chance of controls pH, high 02 if small high pH, anaerobic lateral distances are conditions associated with shallow depths Predictability Low Moderate Moderate? Moderate?

are different when monitoring at the regional effect to the same degree, thus resulting in an scale than when monitoring at the site scale, and anisotropic dispersive behaviour. The lower the evidence summarized in the preceding permeability units will, however, tend to encour- sections is not directly comparable at all scales age layer-parallel flow: if flow passes around the of interest. lower permeability layers, dispersion will The sandstone sequence is layered at scales increase further. Anisotropy will also be present from 1 mm to over 100 m, and this layering within the plane of the bedform (i.e. in the often shows some cyclicity. With some notable approximately horizontal plane) as a result of exceptions, these nested cyclic structures have cross-stratification (Fig. 3). Given that the rock not been mapped. fabrics are present on various scales, there are Over small distances, when flow is parallel likely to be several minimum representative with layering, solute transport is multiperme- volumes (MRVs) for hydraulic properties for ability in character, with diffusive exchange matrix flow, and these will be directionally- between higher and lower permeability layers dependent: values have yet to be established. (Fig. 11 for small-scale example; Fig. 13). Flow Analyses of idealized versions of such systems perpendicular to layers does not experience this have been available for a long time (e.g. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 35

Mercado 1967; Gelhar et al. 1979; Dagan stone where flow is restricted (Fig. 17). Fault 1989): it is found that dispersivity eventually structures will often have a predominant approaches a constant value dependent on the regional alignment, and compartments may statistical properties of the sequence, but up to sometimes not be completely closed, or they this point non-Fickian behaviour is seen. There may be partly closed to flows in one direction is little evidence of systematic change in matrix only. Compartment size will vary from semi- permeability with depth in most aquifers, regional to subhand specimen size. Often frac- although dissolution would be expected to tures cut through granulation seams, and some increase permeability near the surface in origi- degree of funnelling therefore will occur (Fig. nally carbonate-containing units. 7a). However, kinematic porosity for large-scale The layering will often result in multimodal flows appears to be similar to total porosity. breakthrough up to distances equal to the Relationships between the distributions of lateral continuity of the layer (Fig. 12). As the fractures, antifractures, matrix permeability and lateral distance travelled increases beyond this, lithofacies are poorly known. However, it is the effects of layering become less obvious as likely that such relationships exist (cf. Fig. 1). the migrating solution encounters repeated The presence of each of these features will changes in lithofacies or cementation. The modify the effects of the others. distance over which this happens is uncertain, Groundwater pH is often buffered at neutral but possibly up to the order of tens of metres to slightly alkaline when carbonates are present, (30 m in Table 6). Given the log-normal perme- but at 4-6 when absent (Table 5). Highest pH ability distributions, unusually high-permeability values are normally found when significant pathways may occasionally connect up, forming incongruent dissolution or ion exchange has fast solute-conducting pathways (cf. Gutmanis occurred, and can be up to at least 9. Carbon- et al. 1998). ates are absent in some formations, and are Superimposed on the lithological variations often removed from the shallower parts of other are structural elements - discontinuities (Figs 4, formations. The reducing capacity of the aquifer 7 & 8; Table 3). A small proportion (<10% ?) will is limited, but over extended periods dissolved be of high permeability as a result of permeable oxygen is completely removed, possibly in part fill and/or wide apertures. Exponential fracture through Fe II release during incongruent length distributions mean that over short carbonate and silicate dissolution. The oxidizing distances fracture flow is more common than capacity is also limited. Mn oxides are relatively over longer distances (Fig. 7c). However, when redox-reactive, but are present in finite amounts. present, permeable long fractures (Table 3) will In contrast, Fe oxides are abundant but only a have a disproportionate effect on flows. Because very small proportion is bioavailable in the of their size and frequency, bedding-plane short term. Sorption capacity is moderate, with fractures can impart a strong anisotropy to the cation-exchange capacities of 1-20 meq/100 g fracture network system; they are also import- dry wt, mostly supplied by clays and Mn oxides, ant in connecting fracture systems vertically. If the latter being vulnerable to reductive dissolu- the rare long subvertical factures are ignored, tion. pH-dependent sorption on Fe oxides is the permeability MRV of the fracture network also important for metal and some anion is of the order of a few tens of metres in each sorption, foc values are typically <0.1%, but direction. Rapid, fracture-flow tracer break- nevertheless appear to control the sorption of at through can occur over at least 100 m, but as least the more hydrophobic organic pollutants. distance increases the chances of this occurring In regions of where palaeoflows have been diminish rapidly. At the kilometre scale, break- limited, some soluble minerals (e.g. gypsum) are through is effectively single-modal as far as can still present. Distributions of reactive mineral be discerned (Fig. 15). A relationship between components within the sandstones, and especi- lithology and fracture properties is suspected ally in relation to the hydraulic properties, are but not proved. Fractures reduce in frequency poorly known. Limited evidence suggests that with depth, especially below 100-200 m. the organic carbon is more associated with the Discontinuities can also be of lower perme- finer grained beds, and that the CEC is inversely ability than the host sandstone. Such 'anti- related to the matrix permeability, albeit weakly fractures' include faults/granulation seams and and only when Mn oxides contents are low. veins (Figs 4 & 8). Some faults have enhanced Chemical properties of the fracture fills are permeability parallel to strike, and reduced unknown, except in a few cases, and thus the permeability in directions perpendicular to effects on solute compositions of flow across or strike. When intersecting mudstones, antifrac- along filled fractures are unknown. Up-scaling tures may form compartments within the sand- procedures for using laboratory geochemical Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

36 J.H. TELLAM & R. D. BARKER parameter determinations for field prediction apparent solute movement at this scale is cannot be justified without collection and possible in some sense. analysis of more mineralogical and hydraulic The point at which 'local' becomes 'sub- data on the same sequences: there is an import- regional' is uncertain, and will be fuzzy in both ant role here for the use of borehole geophysics the sense of being different at different in hydraulic and cation-exchange capacity locations and being gradual at any one location. determination at least. However, it has been argued that a possible How much of the often fairly limited natural transition range may be 30-100 m or so. This is attenuation capacity of the sandstones is avail- an important scale range for many point-source able to a migrating solute plume and how pollution problems, including remediation. To readily it is retained during flushing depends on improve prediction in this distance range especi- the flow system, the layering and the presence ally, it will be necessary to develop appropriate of potential flow barriers (mudstones, antifrac- stochastic descriptions. As is often attempted in tures) (Figs 4 & 5). Examination of bleached oil reservoir engineering contexts, these should zones and consideration of the results of numer- be conditioned by the probable lithofacies- ical modelling suggests that pockets of sand- controlled structure of the matrix permeability stone are sometimes left unaffected (Fig. 17). distribution: lithofacies is also suspected to However, flow around compartments may affect fracture/antifracture development and actually increase the attenuating capacity by geochemical distributions. The local- to extending flow paths and dividing plumes, thus regional-scale transition is very much deter- increasing transverse dispersion. The tortuosity mined by the measurement device used, and of the matrix and the macroscopic fracture direct modelling of measurement devices may network is limited (actual/straight-line path be needed. length is <2). To confirm and subsequently use these suspected relationships between geological Concluding comments properties and the required hydraulic/chemical properties, it will be necessary to improve the The UK Permo-Triassic sandstone is a continen- characterization procedures significantly, tal red-bed sequence unremarkable in many despite the wealth of data already available. ways in a worldwide context, displaying hydro- Much could be achieved from adapting tech- geological characteristics which are seen in niques that are more commonly used in other many other fractured permeable-matrix geological disciplines, and in particular petro- aquifers. leum geology, especially in geological character- As expected, on a local scale solute ization and modelling, as advocated strongly by movement rates are extremely variable, affected de Marsily et al. (2005) in a wider context. as they are by matrix, fracture and geochemical Existing, largely geophysical, methods could be heterogeneity. Breakthroughs in tracer tests are used to retrieve geological data using the generally multimodal, with extreme cases existing extensive borehole network, but new ranging from almost karst-like to no break- developments in, for example, resistivity/ through at all depending on local fracture and spectral IP and magnetic resonance sounding 'antifracture' geometry. Precise prediction is could enable hydraulic and even some difficult. Low tracer test recoveries in general geochemical properties to be evaluated more suggest remediation may often be difficult to directly. Potentially, geophysics has a very complete. significant role to play, but, in addition, direct At the subregional scale of perhaps a few rock characterization of cores and outcrop is kilometres, breakthrough appears to be much also needed. less complex because the scale of measurement Perhaps the main fundamental uncertainties is much greater than the minimum representa- in the geochemical behaviour of the sandstone tive volumes for both the hydraulic and are: (i) the role of the Fe and Mn oxides and chemical systems. At the regional scale, ground- hydroxides in both redox and sorption water age distributions can be complicated due reactions; and (ii) the distributions of the to the effects of varying and sometimes geo- geochemical components. Distributions may be metrically intricate boundary conditions, and related to lithofacies, and over short distances chemical variations are further obscured by the the effects of correlations between hydraulic usual practice of using pumping wells for moni- and geochemical parameters may give rise to toring purposes. Nevertheless, there is structure further (interesting) complexity. in regional distributions of pumped well-sample chemistry, implying that quantitative analysis of Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 37

This article is in part a review, drawing heavily on the water from the Bunter Sandstone of England as work of many people. We would like to acknowledge indicators of age and palaeoclimatic trends. the input of all, whether they are cited or not. We Journal of Hydrology, 41, 233-252. would also like to thank funders with whom we have ANDREWS, J.N., EDMUNDS, W.M., SMEDLEY, P.L., been involved, especially the UK Natural Environ- FONTES, J.-C.H., FIELD, L.K. & ALLAN, G.L. 1994. ment and Engineering and Physical Sciences Chlorine-36 in groundwater as a palaeoclimatic Research Councils, and the Environment Agency. We indicator: the East Midlands Triassic aquifer (UK). are very grateful to M. Riley, P. Turner and R. Mackay Earth and Planetary Science Letters, 122, 159-172. for commenting on parts of the drafts, and to P. Turner ANTONELLINI, M. & AYDIN,A. 1994. Effect of faulting for permission to use his photographs. on fluid flow in porous sandstones. AAPG Bulletin, 78, 355-377. References ANTONELLINI, M., AYDIN, A. & POLLARD, D.D. 1994. Microstructure of deformation bands in porous ADEV, J. 2004. Development of the Partition Coefficient sandstones at Arches National Park, Utah. Journal (Kd) Test Method for Use in Environmental Risk of Structural Geology, 16, 941-959. Assessments. (UK) Environment Agency R&D ARCHIE, G.E. 1942. The electrical resistivity log as an Technical Report, P1-50014t'I'R. aid in determining some reservoir characteristics. AIGNER, W., HEINZ, J., HORNUNG, J. & ASPRION, U. Transactions of the American Institute of Mining 1999. A hierarchical process-approach to reservoir and Metallurgical Engineers, 146, 54-62. heterogeneity: Examples from outcrop analogues. ARCHIE, G.E. 1950. Introduction to petrophysics of Bulletin du Centre de Recherches Elf Exploration reservoir rocks. AAPG Bulletin, 34, 943-961. Production, 22, 1-11. ATI'EIA, O., ANDRE, L., DuPuY., A. & FRANCESCHI, M. AIGNER, T., SCHAUER, M., JUNGHANS, W.D. & RE1N- 2005. Contributions of diffusion, dissolution, ion HARDT, L. 1995. Outcrop gamma-ray logging and exchange, and leakage from low-permeability its applications: Examples from the German layers to confined aquifers. Water Resources Triassic. Sedimentary Geology, 100, 47-61. Research, 41, WO9412, doi: 10.1029/2003WR00 AKHURST, M.C., BALL, D.E ET AL. 2006. Towards 2593/. understanding the Dumfries Basin aquifer, SW AUDLEY-CHARLES,M.G. 1970. Stratigraphical correla- Scotland. In: BARKER, R.D. & TELLAM, J.H. (eds) tion of the Triassic rocks of the British Isles. Quar- Fluid Flow and Solute Movement in Sandstones: terly Journal of the Geological Society, London, The Offshore UK Permo- Triassic Red Bed 126, 19-47. Sequence. Geological Society, London, Special AYDIN, A. 1978. Small faults formed as deformation Publications, 263, 187-198. bands in sandstones. Pure and Applied Geo- ALEXANDER, J., BLACK, J.H. & BRIGHTMAN, M.A. physics, 116, 913-930. 1987. The role of low-permability rocks in regional AYD1N, A. & JOHNSON, A.M. 1978. Development of flow. In: GOFF, J.C. • WILLIAMS, B.J.P. (eds) Fhdd faults as zones of deformation bands and as slip Flow in Sedimentary Basins and Aquifers. Geo- surfaces in sandstone. Pure and Applied logical Society, London, Special Publications, 34, Geophysics, 116, 931-942. 173-183. AYDIN,A. & JOHNSON,A.M. 1983. Analysis of faulting AL-HOSNI, T. 2001. Attenuation of cadmium by UK in porous sandstones. Journal of Structural Triassic sandstone aquifer materials. MSc Project Geology, 5, 19-31. Report, Earth Sciences, University of Birmingham. BALLARD, M. 2000. The role of granulation seams in ALLEN, D.J., BREWERTON, L.J. ET AL. 1997. The pollutant transport in the UK. Triassic Sandstone. Physical Properties of Major Aquifers in England MSc Project Report, University of Birmingham. and Wales. British Geological Survey Technical BANERJEE, D. & NESBITT, H.W. 1999. XPS study of Report, WDI97134. Environment Agency R&D reductive dissolution of birnessite by oxalate: Publication, 8. Rates and mechanistic aspects of dissolution and ALLEN, D.J., BLOOMFIELD, J.P., G1BBS, B.R. & redox processes. Geochimica et Cosmochimica WAGSTAFF, S.J. 1998. Fracturing and the Hydro- Acta, 63, 3025-3038. geology of the Permo-Triassic Sandstones in BANWART, S.A. 1999. Reduction of iron (III) minerals England and Wales. British Geological Survey by natural organic matter in groundwater. Technical Report WD/98/1, and Environment Geochimica et Cosmochimica Acta, 63, 2919-2928. Agency R&D Technical Report W164. BARKER, A.P. 1996. Isotopic studies of groundwater. ALLEN-KING, R.M., HALKET, R.M., GAYLORD,D.R. & PhD thesis, University of Leeds. ROBIN, M.J.L. 1998. Characterizing the hetero- BARKER, A.R, NEWTON, R.J., BOTTRELL, S.H. & geneity and correlation of perchloroethene TELLAM, J.H. 1998. Processes affecting groundwater sorption and hydraulic conductivity using a facies- chemistry in a zone of saline intrusion into an urban based approach. Water Resources Research, 34, aquifer. Applied Geochemistry, 6,735-750. 385-396. BARKER, J.A. 1988. A generalized flow model for ANDREWS, J.N. & KAY, R.L.F. 1983. The U contents hydrualic tests in fractured rock. Water Resources and 234U/238U activity ratios of dissolved Research, 24, 1796-1804. uranium in groundwaters from some Triassic sand- BARKER, R.D. 1974. Interpretation of induced-polar- stones in England. Isotope Geoscience, 1, 101-117. ization sounding curves in the frequency domain. ANDREWS,J.N. & LEE, D.J. 1979. Inert gases in ground- Geophysical Prospecting, 22, 61(~626. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

38 J.H. TELLAM & R. D. BARKER

BARKER, R.D. 1990. Investigation of groundwater BIVER, P., HAAS, A. & BACQUET, C. 2002. Uncertain- salinity by geophysical methods. In: WARD, S.H. ties in facies proportion estimation II: Application (eds) Geotechnical and Environmental Geo- to geostatistical simulation of facies and assess- physics. Volume 2: Environmental and Ground- ment of volumetric uncertainties. Mathematical water. Society of Exploration Geophysics, Geology, 34, 703-714. Houston, Texas, USA, 201-212. BLOOMFIELD, J., GOODDY, D.C., BRIGHT, M.I. & BARKER, R.D. & MOORE, J. 1998. The application of WILLIAMS, RJ. 2001. Pore-throat size distributions time-lapse electrical tomography in groundwater in Permo-Triassic sandstones from the United studies. The Leading Edge, 17, 1454-1458. Kingdom and some implications for contaminant BARKER, R.D. & WORTHINGTON, EE 1973a. The hydrogeology. Hydrogeology Journal, 9, 219-230. hydrological and electrical anisotropy of the BLOOMFIELD, J., MOREAU, M.E & NEWELL,A.J. 2006. Bunter Sandstone of Northwest Lancashire. Characterization of permeability distributions in Quarterly Journal of Engineering Geology and six lithofacies from the Helsby and Wilmslow Hydrogeology, 6, 169-175. Sandstone formations of the Cheshire Basin, UK. BARKER, R.D. & WORTHINGTON, EE 1973b. Some In: BARKER, R.D. & TELLAM,J.H. (eds) Fluid Flow hydrogeophysical properties of the Bunter Sand- and Solute Movement in Sandstones: The Offshore stone of northwest England. Geoexploration, 11, UK Permo-Triassic Red Bed Sequence. Geological 151-170. Society, London, Special Publications, 263, 83-101. BARNES, R.R, DEE, S.J., SANDERSON, D.J. & BOWDEN, BOTTRELL, S.H., WEST, L.J. • YOSHIDA, K. 2006. R.A. 1998. Interpretation of structural domains in Combined isotopic and modelling approach to discontinuity data from Nirex deep boreholes at determine the source of saline groundwaters in the Sellafield. Proceedings of the Yorkshire Geological Selby Triassic Sandstone aquifer, UK. In" BARKER, Society, 52, 177-187. R.D. & TELLAM, J.H. (eds) Fluid Flow and Solute BARRETF, M.H. 1995. The transport of mixtures of Movement in Sandstones: The Offshore UK Permo- organic contaminants in groundwater. PhD thesis, Triassic Red Bed Sequence. Geological Society, University of Birmingham, UK. London, Special Publications, 263, 325-338. BARRETL M.H., HISCOCK, K.M., PEDLEY, S.J., LERNER, BOUCH, J.E., HOUGH, E., KEMP, S.J., MCKERVEY, J.A., D.N., TELLAM, J.H. & FRENCH, M.J. 1999. Marker WILLIAMS, G.M. & GRESWELL, R.B. 2006. Sedi- species for identifying urban groundwater mentary and diagenetic environments of the recharge sources; the Nottingham case study. Wildmoor Sandstone Formation (UK): impli- Water Research, 33, 3083-3097. cations for groundwater and contaminant trans- BASHAR, K. 1997. Developing a conceptual model of port, and sand production. In: BARKER, R.D. & intergranular conservative solute transport TELLAM, J.H. (eds) Fluid Flow and Solute processes for water flow through laboratory scale Movement in Sandstones: The Offshore UK samples of the U.K. Triassic Sandstones. PhD Permo-Triassic Red Bed Sequence. Geological thesis, University of Birmingham, UK. Society, London, Special Publications, 263, BASHAR, K. & TELLAM,J.H. 2006. Non-reactive solute 129-153. movement through saturated laboratory samples BRANEY, M.C., GILLING,D., JEFFERIES, N.L., LINEHAM, of undisturbed stratified sandstone. In: BARKER, T.R. & STONE, N.S. 2001. Laboratory Measure- R.D. & TELLAM, J.H. (eds) Fluid Flow and Solute ments of the Mass Transfer Properties of Samples Movement in Sandstones: The Offshore UK of Permo- Triassic Sedimentary Strata From Permo-Triassic Red Bed Sequence. Geological Cumbria. UK Nirex Ltd, Report, NSS/R278. Society, London, Special Publications, 263, 233-251. BRASSINGTON, E 1992. Measurements of head vari- BATH, A.H., EDMUNDS, W.M. & ANDREWS, J.N. 1979. ations within observation boreholes and their Palaeoclimatic trends deduced from the hydro- implications for groundwater monitoring. Journal chemistry of a Triassic sandstone aquifer. Inter- of the Institute of Water and Environmental national Symposium on Isotope Hydrology, Management, 6, 91-100. Vienna, II, IAEA-SM-228/27, 545-568. BRASSINGTON, EC. & WALTHALL, S. 1985. Field tech- BATH, A.H., MCCARa-NEV, R.A., RICHARDS, H.G., niques using borehole packers in hydrogeological METCALF, R. & CRAWFORD, M.B. 1996. Ground- investiations. Quarterly Journal of Engineering water chemistry in the Sellafield area: a prelimi- Geology, 18, 181-193. nary interpretation. Quarterly Journal of BRASSINGTON,EC., LUCEY,EA. & PEACOCK,A.J. 1992. Engineering Geology and Hydrogeology, 29, The use of down-hole focussed electric logs to Supplement 1, $39-$58. investigate saline groundwaters. Quarterly Journal BATH, A.H., MILOWDOWSKI, A.E. & STRONG, G.E. of Engineering Geology, 25, 343-349. 1987. Fluid flow and diagenesis in the East BRERETON, N.R. & SKINNER,A.C. 1974. Groundwater Midlands Triassic sandstone aquifer. In: GovF, flow characteristics in the Triassic Sandstone in the J.C.W. & WILLIAMS, B.J.E (eds) Fluid Flow in Sedi- Fylde area of Lancashire. Water Services, 78, mentary Basins and Aquifers. Geological Society, 275-279. London, Special Publications, 34, 127-140. BROHOLM, M.M., JONES, I., TORSTENSSON, O. & ARVIN, BENTON, M.J., COOK, E. & TURNER, P. 2002. Permian E. 1998. Groundwater contamination from a coal and Triassic Red Beds and the Penarth Group of carbonization plant. In: LERNER, D.N. & WALTON, Great Britain. Joint Nature Conservation N.R.G. (eds) Contaminated Land and Ground- Committee, Peterborough. water: Future Directions. Geological Society, Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 39

London, Engineering Geology, Special Publi- attenuation of ammonium in aquifers? MSc Project cations, 14, 159-165. Report, University of Birmingham, School of BRYANT, I., CARR, D., CIRILLI, P., DRINKWATER, N., Geography, Earth and Environmental Sciences. MCCORMICK, D., TILKE, P. & THURMOND, J. 2000. CLEMMENSEN, L.B., ~XNEVAD, I.E.I. & DE BOER, EL. Use of 3D digital analogues as templates in reser- 1994. Climatic controls on ancient desert sedimen- voir modelling. Petroleum Geoscience, 6, 195-201. tation: some late Palaeozoic and Mesozoic BUNCH, M.A. 2000. Spectral IP signatures of granula- examples from NW Europe and the Western tion seams in Triassic sandstone. MSc Project Interior of the USA. In: DE BOER, P.L. & D.G.S. Report, University of Birmingham. (eds) Orbital Forcing and Cyclic Sequences. Inter- BUNCH, M.A. 2006. A model for simulating dryland national Association of Sedimentologists, Special alluvial fan development. PhD thesis, University of Publications, 19, 439-457. Birmingham, UK. COLEBY, L.M. 1996. Heath House Tracer Test Data BUNCH, M.A., MACKAY,R., TELLAM, J.H. & TURNER, Report. British Geological Survey Technical E 2004. A model for simulating the deposition of Report, WD/96/74. water-lain sediments in dryland environments. COLLAR, EA., 1974. A geophysical interpretation of Hydrology and Earth Systems Science, 8, 122-134. the structure of the Vale of Clywd, North Wales. BURLEY, S.D. 1984. Patterns of diagenesis in the Geological Journal, 9, 65-76. Sherwood Sandstone Group (Triassic), United COLLAR, EA. & GRIFFITHS, D.H. 1976. A laboratory Kingdom. Clay Minerals, 19, 403-440. study of the relationships between induced polar- Buss, S.R. 1999. Attenuation of strong acids in the ization, permeability and matrix electrical con- Birmingham sandstone aquifer. PhD thesis, ductivity in Bunter Sandstones. Quarterly Journal University of Birmingham. of Engineering Geology and Hydrogeology, 9, Buss, S.R., HERBERT, A.W., MORGAN, P., THORNTON, 57-63. S.E & SMITH, J.W.N. 2004. A review of ammonium CRONIN, A., TAYLOR, R.G., POWELL, K.L., BARRETT, attenuation in soil and groundwater. Quarterly M.H., TROWSDALE, S.A. & LERNER, D.N. 2003. Journal of Engineering Geology and Hydro- Temporal variations in the depth-specific hydrogeology, 37, 347-359. chemistry and sewage-related microbiology of an Buss, S.R., RIVETr, M.O., MORGAN, P. & BEMMENT, urban sandstone aquifer, Nottingham, United C.D. 2005. Attenuation of Nitrate in the Sub-surface Kingdom. Hydrogeology Journal, 11, 205-216. Environment, (UK) Environment Agency Science DAGAN, G., 1989. Flow and Transport in Porous Report SC030155/SR2. Formations. Springer, Berlin, 465. BUTCHER, A., LAWRENCE,A., JACKSON, C., CULLIS, E., DALLA, E., CASSIANI, G., BROVELLI, A. & PITEA, D., CUNNINGHAM, J., HASAN, K. & INGRAM, J. 2006. 2004. Electrical conductivity of unsaturated Investigating rising nitrate concentrations in porous media: pore-scale model and comparison groundwater in the Permo-Triassic aquifer, Eden with laboratory data. Geophysical Research Valley, Cumbria, UK. In: BARKER, R.D. & Letters, 31, Article No. L05609. TELLAM, J.H. (eds) Fluid Flow and Solute DARLING, W.G., EDMUNDS, W.M. & SMEDLEY, EL. Movement in Sandstones: The Offshore UK 1997. Isotopic evidence for palaeowaters in the Permo-Triassic Red Bed Sequence. Geological British Isles. Applied Geochemistry, 12, 813-829. Society, London, Special Publications, 263, DARLING, W.G., BATH, A.H. & TALBOT, J.C. 2003. The 285-296. O and H stable isotopic composition of fresh CAM]BELL, J.E. 1982. Permeability characteristicsofthe waters in the British Isles: 2. Surface waters and Permo-Triassic sandstones of the Lower Mersey groundwaters. Hydrology and Earth Systems Basin. MSc thesis, University of Birmingham. Science, 7, 183-195. CAMPBELL, J.E. 1986. West Cheshire Saline Ground- DAVIES, R.K. & HANDSCHY,J.W. 2003. Introduction to water Investigation (Phase 3). North West Water the AAPG Bulletin thematic issue on fault seals. Authority, Warrington, Cheshire, UK. AAPG Bulletin, 87, 377-380. CARLYLE, H.E, TELLAM, J.H. & PARKER, K.E. 2004. DE MARSILY,G., DELAYE, GONCALVES,J., RENARD, P., The use of laboratory-determined ion exchange TELES, V. & VIOLETTE, S. 2005. Dealing with spatial parameters in the prediction of field-scale major heterogeneity. Hydrogeology Journal, 13,161-183. cation migration over a 40 year period. Journal of DIGGES LA TOUCHE, S.V. 1998. Unsaturated flow in the Contaminant Hydrology, 68, 55-81. Triassic Sandstones of the United Kingdom. PhD CHADWICK, R.A. 1997. Fault analysis of the Cheshire thesis, University of Birmingham. Basin, NW England. In: MEADOWS, N.S., EDMUNDS, W.M. 1986. Geochemistry of geothermal TRUEBLOOD, S.P., HARDMAN, M. & COWAN, G. waters in the U.K. In: DOWNING, R.A. & GRAY, (eds) Petroleum Geology of the Irish Sea and D.A. (eds) Geothermal Energy: The Potential in the Adjacent Areas. Geological Society, London, United Kingdom. HMSO, London, 111-122. Special Publications, 124, 297-313. EDMUNDS, W.M. 1996. Indicators of Rapid Environ- CHAMP, D.R., GULENS,J. & JACKSON,R.E. 1979. Oxida- mental Change in the Groundwater Environment. tion reduction sequences in groundwater flow Balkema, Rotterdam, 135-150. systems. Canadian Journal of Earth Sciences, 16, EDMUNDS, W.M. & KINNIBURGH, D.G. 1986. The 12-23. susceptibility of UK groundwaters to acidic depo- CLARKE, L.E. 2005. Under what circumstances can the sition. Journal of Geological Society, London, 143, Kd approach (i.e. linear sorption) be used to predict 707-720. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

40 J.H. TELLAM & R. D. BARKER

EDMUNDS, W.M. & MORGAN-JONES, M. 1976. FISHER, Q.L. & KNIPE, R.J. 1998. Fault sealing Geochemistry of groundwaters in British Triassic processes in in siliclastic sediments. In: JONES, G. Sandstones: the Wolverhampton-East Shropshire ET AL. (eds) Faulting and Fault Sealing in Hydro- area. Quarterly Journal of Engineering Geology carbon Reservoirs. Geological Society, London, and Hydrogeology, 9, 73-101. Special Publications, 147, 117-134. EDMUNDS, W.M. & SMEDLEY, EL. 2000. Residence FITCH, EJ., MILLER, J.A. & THOMPSON, D.B. 1966. The Time Indicators in Groundwater: The East palaeogeographic significance of isotopic age Midlands Triassic Sandstone Aquifer. Applied determination on detrital micas from the Triassic Geochemistry, 15, 737-752. of the Stockport-Macclesfield district, Cheshire. EDMUNDS, W.M., BATH, A.H. & MILES, D.L. 1982. Palaeogeography, Palaeoclimatology, Palaeo- Hydrochemical evolution of the East Midlands ecology, 2, 281-312. Triassic sandstone aquifer, England. Geochimica et FLETCHER, S.W. 1989. Groundwater Resource Cosmochimica Acta, 46, 2069-2081. Development in the Region Between Wolverhamp- EDMUNDS, W.M., BREWERTON, L.J., SHAND, P. & ton and Bridgnorth. Unpublished report, Severn SMEDLEY, EL. 1997. The Natural (Baseline) Trent Water Authority, Birmingham, UK. Quality of Groundwater in England and Wales. FORD, M. & TELLAM, J.H. 1994. Source, type of extent Part 1: A Guide to the Natural (Baseline) Quality of inorganic contamination within the Birming- Study. British Geological Survey Technical Report, ham urban aquifer system, UK. Journal of Hydrol- WD/97/54. ogy, 156, 101-135. EDMUNDS, W.M., BREWERTON, L.J. & SMEDLEY, P.L. FORD, M., TELLAM, J.H. & HUGHES, M. 1992. Pollu- 1996. Natural Quality of Groundwaters in Aquifers tion-related acidification in the urban aquifer, in England and Wales: A Scoping Study. British Birmingham, UK. Journal of Hydrology, 140, Geological Survey Technical Report, WD196146R. 297-312. EDMUNDS, W.M., COOK, J.M., KINNIBURGH, D.G., FOWLES, J. & BURLEY, S. 1994. Textural and perme- MILES, D.L. & TRAFFORD, J.M. 1989. Trace Element ability characteristics of faulted, high porosity sand- Occurrences in British Groundwaters. British stones. Marine and Petroleum Geology, 11, 608-623. Geological Survey, Research Report, SD/89/3. FUKADA, T., HISCOCK, K.M. & DENNIS, EE 2004. A EDMUNDS, W.M., KINNIBURGH, D.G. & Moss, ED. dual isotope approach to the nitrogen hydro- 1992. Trace metals in interstitial waters from sand- chemistry of an uban aquifer. Applied Geo- stones: acidic inputs to shallow groundwaters. chemistry, 19, 709-719. Environmental Pollution, 77, 129-141. FURLONG, B.V. 2002. Regional scale solute transport in EDMUNDS, W.M., MILES, D.L. & COOK, J.M. 1984. A the Permo- Triassic Sandstone aquifer of the Lower comparative study of sequential redox processes in Mersey Basin, north west England. PhD thesis, three British aquifers. In: ERIKSSON, E. (ed.) University of Birmingham. Hydrochemical Balances of Freshwater Systems. GABRIELSEN, R.H., AARLAND, R.-K. & ALSAKER, E. IAHS Publication, 150, 55-70. 1998. Identification and spatial distribution of frac- EDWARDS, H.E., BECKER, A.D. & HOWELL, J.A. 1993. tures in porous, siliclastic sediments. In: COWARD, Compartmentalization of an aeolian sandstone by M.E, DALTABAN,T.S. & JOHNSON, H. (eds) Struc- structural heterogeneities: Permo-Triassic Sand- tural Geology in Reservoir Characterization. stone, Moray Firth, Scotland. In: NORTH, C.E & Geological Society, London, Special Publications, PROSSER, D.J. (eds) Characterization of Fluvial and 127, 49-64. Aeolian Reservoirs. Geological Society, London, GELHAR, L.W., GUTJAHR, A.L. & NAFF, R.L. 1979. Special Publications, 73, 339-365. Stocastic analysis of macrodispersion in a strati- EL-GHONEMY, H. 1997. Laboratory experiments for fied aquifer. Water Resources Research, 15, quantifying and describing cation exchange in UK 1387-1397. Triassic Sandstones. PhD thesis, University of GIBSON, R.G. 1998. Physical character and fluid-flow Birmingham. properties of sandstone-derived fault zones. In: EL-GHONEMY, H. 1998. Estimation of Solid Phase COWARD, M.E, DALTABAN, T.S. & JOHNSON, H. Activity Coefficients for Non-ideal Cation (eds) Structural Geology in Reservoir Characteriz- Exchange Behaviour in Laboratory Samples of UK ation. Geological Society, London, Special Publi- Triassic Sandstones. Geological Society, London, cations, 127, 83-97. Engineering Geology Special Publications, 14, GIBSON, R.G. & BENTHAM,EA. 2003. Use of fault-seal 213-217. analysis in understanding petroleum migration in ERSKINE, A. 2000. Transport of ammonium in aquifers: a complexly faulted anticlinal trap, Columbus retardation and degradation. Quarterly Journal of Basin, offshore Trinidad. AAPG Bulletin, 87, Engineering Geology and Hydrogeology, 33, 465-478. 161-170. GILLESPIE, M.R., KEMP, S.J., VICKERS, B.E, WATERS, C. EVERS, S. & LERNER, D.N. 1998. How uncertain is our & GOWING, C.J. 2001. Cation Exchange Capacity estimate of a wellhead protection zone. Ground (CEC) of Selected Lithologies from England, Wales Water, 36, 49-57. and Scotland. (UK) Environment Agency R&D FARRIS, M. 1999. Palynomorph taphonomy in conti- Technical Report, P2-222/TR. nental red-bed facies, UK West Midlands and GILLESP1E, M.R., LEADER, R.U. ETAL. 2000. CEC and Central North Sea. PhD thesis, University of Keele, K d Determination in Landfill Performance Evaula- UK. tion: A Review of Methodologies and Preparation Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 41

of Standard Materials for Laboratory Analysis. HARTLEY, L.T. 1998. NAPSAC (Release 4.1) Technical (UK) Environment Agency R&D Project Record, Summary Document. AEA Technology Report, P1/254/01. AEA-D&R.0271. GOODDY, D.C. & BLOOMFIELD, J.P. 2006. Controls on HATZICHRISTODULU, V.C., BARKER, R.D. & TELLAM, dense non-aqueous phase liquid transport in J.H. 2002. High Resolution Electrical Monitoring of Permo-Triassic Sandstones, UK. In: BARKER, R.D. Fluid Flow Through the Unsaturated Zone. & TELLAM, J.H. (eds) Fluid Flow and Solute Environment Agency Technical Report, P2-042/ Movement in Sandstones: The Offshore UK Permo- TR1. Triassic Red Bed Sequence. Geological Society, HEATHr J.A., JONES, M.A. & HERBERT,A.W. 1996. London, Special Publications, 263, 253-264. Modelling groundwater flow in the Sellafield area. GOODDY, D.C. & DARLING,W.G. 2005. The potential Quarterly Journal of Engineering Geology and for methane emissions from groundwaters in the Hydrogeology, 29, Supplement 1, $59-$82. UK. Science of the Total Environment, 339, HERBERT, A.W. 1992. Groundwater flow and transport 117-126. in fractured rock. PhD thesis, University of Bath. GOUZE, P., TENCH1NE, S. ET AL. 2003. A study of the HEWITT, W. 1898. Notes on some sections exposed by dispersion in a single thin fracture using Positron excavations on the site of the new technical Emission Projected Imaging. In: KRASN't, J., schools, Byrom Street, Liverpool. Proceedings of HRKAL, Z. & BRUTHANS, J. (eds) Groundwater in the Liverpool Geological Society Session the Fractured Rock, IHP-VI. Series on Groundwater, Thirty-ninth, 8, 268-273. 7, 407-408. HITCHMOUGH, A.M., RILEY, M.S., HERBERT, A.W. & GREENWOOD, H.W. & TRAVIS, C.B. 1915. The miner- TELLAM, J.H. Submitted. Estimating the hydraulic alogical and chemical constitution of the Triassic properties of the fracture network in a sandstone rocks of Wirral, Part II. Proceedings of the Liver- aquifer. Journal of Contaminant Hydrology. pool Geological Society, 12, 161-188. HOUGH, E., PEARCE, J.M., KEMP, S.J., MCKERVEY,J. & GRESSWELL, R.A. 2005. Quantitative characterisation DATES, D. 2001. Regional Survey and Sampling of of fluoride sorption in the UK Triassic sandstone Sediment Filled Fractures and Evaluation of and the development of a fluoride reactive transport Prospective Research Sites: A Report Based on model. MSc Project Report, University of Fieldwork Undertaken in 2000-2001. British Birmingham. Geological Survey Technical Report, CR/01/14. GRESWELL, R.B., TELLAM, J.n., LLOYD, J.W. & HOUSTON, J. 2004. High-resolution sequence PARKER, D. 1998. Positron emission tomography stratigraphy as a tool in hydrogeological explo- studies of solute movement through rock. In: ration in the Atacama Desert. Quarterly Journal of BRAHANA, J., ECKSTEIN, Y., ONGLEY, L.K., Engineering Geology and Hydrogeology, 37, 7-17. SCHNEIDER, R. & MOORE, J.E. (eds) Gambling HOWARD, K.W.E 1988. Beneficial aspects of sea-water with Groundwater, Proceedings of the IAH Confer- intrusion. Ground Water, 17, 250-257. ence, Las Vegas, 561-566. HOWELL, J. & MOUNTNEY, N. 1997. Climate cyclicity GRIFFITHS, D.H., BARKER, R.D. & FINCH, J.W. 1981. and accommodation space in arid to semi-arid Recent applications of the electrical resistivity and depositional systems: an example from the induced-polarisation methods to hydrogeological Roliegend Group of the UK southern North Sea. problems. In: A Survey of British Hydrogeology In: ZIEGLER, K., TURNER, P. & DAINES, S.R. (eds) 1980. The Royal Society, London. Petroleum Geology of the Southern North Sea: GUADAGNINI,A. & WINTER, C.L. 2004. Special Issue: Future Potential. Geological Society, London, Stochastic models of flow and transport in Special Publications, 123, 63-86. multiple-scale heterogeneous porous media. HUGHES, A.J., TELLAM, J.H., BOTFRELL, S.H., LLOYD, Journal of Hydrology, 294, 1-212. J.W., BARKER,A.R, STAGG, K.A. & BARRETT, M.H. GUHA, H., SAIERS, J.E., BROOKS, S., JARDINE, P. & 1999. Sulphate Isotope signatures in borehole KRISHNASWAMYJAYACHANDRAN. 2001. Chromium waters from three urban Triassic Sandstone transport, oxidation, and adsorption in aquifers, UK. In: ELLIS, J.B. (ed.) Impacts of Urban manganese-coated sand. Journal of Contaminant Growth on Surface Water and Groundwater Hydrology, 49, 311-334. Quality. IAHS Publication, 259, 143-149. GUTMANIS, J.C., LANYON,G.W., WYNN,T.J. & WATSON, HUNTER-WmLIAMS, N. 1995. Investigation of the C.R. 1998. Fluid flow in faults: a study of fault response of the Triassic Sandstone aquifer Nurton hydrogeology in Triassic sandstone and Ordo- to extended abstraction and of the controls on vician volcaniclastic rocks at Sellafield, north-west groundwater flow. MSc Project Report, University England. Proceedings of the Yorkshire Geological of Birmingham. Society, 52, 159-175. HUNTOON, P.W. 1986. Incredible tale of Texasgulf Well HAINES, T.S. & LLOYD, J.W. 1985. Controls on silica in 7 and fracture permeability, Paradox Basin, Utah. groundwater environments in the United Ground Water, 24, 643--653. Kingdom. Journal of Hydrology, 81, 277-295. IVANOVICH, M., TELLAM, J.H., LONGWORTH, G. & HARRISON, I., WILLIAMS,G.M., HIGGO, J.J.W., LEADER, MONAGHAN, J.J. 1992. Rock/water interaction R.U., KIM, A.W. & NOY, D.J. 2001. Microcosm timescales involving U and Th isotopes in a Permo- studies of microbial degradation in a coal tar distil- Triassic sandstone. Radiochimica Acta, 58/59, late plume. Journal of Contaminant Hydrology, 53, 423-432. 31%340. IXER, R.A. & VAUGHAN,D.J. 1982. The primary ore Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

42 J.H. TELLAM & R. D. BARKER

mineralogy of the Alderley Edge deposit, ethylenediaminetetratacetic acid, and citrate. Cheshire. Mineralogical Magazine, 46, 485-492. Geochimica et Cosmochimica Acta, 63, 3017-3024. JACKSON, D. & LLOYD,J.W. 1983. Groundwater chem- KNIPE, C., LLOYD, J.W., LERNER, D.N. & GRESWELL, istry of the Birmingham Triassic sandstone aquifer R.B. 1993. Rising Groundwater Levels in Birming- and its relation to structure. Quarterly Journal of ham, and Their Engineering Significance. ed.; Engineering Geology, 16, 135-142. CIRIA (Construction Industry Research and JAMISON,W.R. & STEARNS,D.W. 1982. Tectonic defor- Information Association), London, Special Publi- mation of the Wingate sandstone, Colorado cations, 92, 114. National Monument. AAPG Bulletin, 66, KNOTt, S.D. 1994. Fault thickness versus displacement 2584-2608. in the Permo-Triassic sandstones of NW England. JEFFEOAT, A.M. 2002. Exploring the hydraulic proper- Journal of the Geological Society, London, 151, ties of discontinuity geometry in the UK Triassic 17-25. Sandstones. PhD thesis, University of Birmingham. KOLTERMAN, C.E. & GORELICK, S.M. 1996. Hetero- JENSEN, J.L., CORBETI', P.W.M., PICKUP, G.E. & geneity in sedimentary deposits: a review of struc- RINGROSE, P.S. 1996. Permeability semivariograms, ture-imitating, process-imitating, and descriptive geological structure and flow performance. Mathe- approaches. Water Resources Research, 32, matical Geology, 28, 419-435. 2617-2658. JONES, D.G., MORTON, A.C., LENG, M.J., HASLAM, LERNER, D.N., THORNTON,S.E ETAL. 2000. Ineffective H.W., MILOWDOWSKI,A.E., STRONG, G.E. & KEMP, natural attenuation of degradable organic S.J. 1999. Provenance of basin fill material. In: compounds in a phenol-contaminated aquifer. PLANT, J.A., JONES, D.G. & HASLAM, H.W. (eds) Ground Water, 38, 922-928. The Cheshire Basin: Basin Evolution, Fluid LEVEILLE, G.E, KNIPE, R. ET AL. 1997. Compartmen- Movement and Mineral Resources in a Permo- talization of Rotliegendes gas reservois by sealing Triassic Rift Setting. British Geological Survey, faults, Jupiter Fields area, southern North Sea. In: Keyworth, 90-124. ZIEGLER, K., TURNER, P. & DAINES, S.R. (eds) JONES, I. 2001. Processes controlling the transport and Petroleum Geology of the Southern North Sea: attenuation of contamination from a coking plant Future Potential. Geological Society, London, in a sandstone aquifer. PhD thesis, University of Special Publications, 123, 87-104. Bradford. LEWlN, K., YOUNG, C.P., BRADSHAW,K., FLEET, M. & JONES, I., DAVISON, R.M. & LERNER, D.N. 1998. The BLAKEY, N.C. 1994. Landfill Monitoring Investi- importance of understanding groundwater flow gations at Burntstump Landfill. Sherwood Sand- history in assessing present-day groundwater stone, Nottingham 1978-1993 (ENV 9003), Final contamination patterns: a case study. In: LERNER, Report of the Department of the Environment D.N. & WALTON,N.R.G. (eds) Contaminated Land CWMD35194: DOE3388/1. and Groundwater: Future Directions. Geological LIU, N. & OLIVER, O.S. 2004. Automatic history Society, London, Engineering Geology, Special matching of geologic facies. Journal of Petroleum Publications, 14, 137-148. Engineers, 9, 429-436. JONES, I. & LEARNER, D.N. 1995. Level-determined LIU, N. & OLIVER, D.S. 2005. Ensemble Kalman filter sampling in an uncased borehole. Journal of for automatic history matching of geologic facies. Hydrology, 171, 291-317. Journal of Petroleum Science and Engineering, 47, JOYCE, E., RUEEDI, J., CRONIN, A.A., PEDLEY, S., 147-161. GRESWELL, R.B. & TELLAM, J.H. 2006. Fate and LIWANG, J. 2005. Introducing geological processes in transport of phage and viruses in UK sandstone reservoir models. Geo ExPro, 2, 28-32. aquifers. Final report on Environment Agency LLOYD, J.W. 1976. Hydrochemistry and groundwater Research Project, NC/02/47. flow patterns in the vicinity of Stratford-upon- KELLER, G.V. 1966. Electrical properties of rocks and Avon. Quarterly Journal of Engineering Geology, minerals. In: CLARK, S.P. (eds) Handbook of 9, 315-326. Physical Constants. Geological Society of America, LOVELL, M.A., JACKSON, ED., HARVEY, EK. & FLINT, Memoir, 97, 553-578. R.C. 2006. High-resolution petrophysical charac- KEMP, S.J., PEARCE, J.M. & OATES, D. 2003. In situ terization of samples from an aeolian sandstone: sampling of sediment-filled fractures. Geotech- the Permian Penrith Sandstone of NW England. nique, 53, 665-668. In: BARKER, R.D. & TELLAM, J.H. (eds) Fluid KINNIBURGH,D.G., NEWELL,A.J., DAVIES,J., SMEDLEY, Flow and Solute Movement in Sandstones: The EL., MILODOWSKI,A.E., INGRAM, J.A. t~; MERRIN, Offshore UK Permo-Triassic Red Bed Sequence. ED. 2006. The arsenic concentration in ground- Geological Society, London, Special Publications, water from the Abbey Arms Wood observation 263, 49-63. borehole, Delamere, Cheshire, UK. In: BARKER, LOVELOCK, EE.R. 1977. Aquifer Properties of the R.D. & TELLAM, J.H. (eds) Fluid Flow and Solute Permo-Triassic Sandstones in the United Movement in Sandstones: The Offshore UK Kingdom. Bulletin of the Geological Survey of Permo-Triassic Red Bed Sequence. Geological Great Britain, 56. Society, London, Special Publications, 263, LUBCZYNSKI, M. & RoY, J. 2005. MRS contribution to 265-284. hydrogeological system parametrization. Near KLEWICKI,J.K. & MORGAN,J.J. 1999. Dissolution of 13- Surface Geophysics, 3, 131-139. MnOOH particles by ligands: Pyrophosphate, LUCEY, EA. 1987. The hydrochemistry of groundwater Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 43

in the West Cheshire aquifer. MPhil, University of METCALF, R. 1998. The petrology and paragenesis Lancaster. of fracture mineralization in the Sellafield area, LUTHER, G.W., SUNDBY,B., LEWIS, B.L., BRENDEL, P.J. west Cumbria. Proceedings of the Yorkshire & SILVERBERG,N. 1997. Interactions of manganese Geological Society, 52, 215-241. with the nitrogen cycle: Alternative pathways to MILODOWSKI, A.E., STRONG, G.E. ET AL. 1999. Dia- denitrogen. Geochimica et Cosmochimica Acta, 61, genesis of the Permo-Triassic rocks. In: PLANT, 4043-4052. J.A., JONES, D.G. & HASLAM, H.W. (eds) The MAIN, I., MAIR, K., OHMYOUNG KWON, ELPHICK, S. 8/: Cheshire Basin. British Geological Survey, NGWENYA, B. 2001. Experimental constraints on Keyworth, 125-175. the mechanical and hydraulic properties of MIMIDES, Z. & LLOYD, J.W, 1987. Toxic metal adsorp- deformation bands in porous sandstones: a review. tion in the Triassic Sandstone aquifer of the In: HOLDSWORTH, R.E., STRACHAN, R.A., English East Midlands. Environmental Geology MAGLOUGHLIN,J.E & KNIFE, R.J. (eds) The Nature and Water Science, 10, 135-140. and Tectonic Significance of Fault Zone Weakening. MITCHENER, R.G.R. 2003. Hydraulic and chemical Geological Society, London, Special Publications, property correlations of the Triassic sandstone of 186, 43-63. Birmingham. PhD thesis, University of Birming- MANZOCCHI, T., RINGROSE, PS. & UNDERHILL, J.R. ham, UK. 1998. Flow through fault systems in high-porosity MOORE, C.C. 1898. The chemical examination of sand- sandstones. In: COWARD, M.E, DALTABAN,Z.S. & stones from Prenton Hill and Bidston Hill. JOHNSON, H. (eds) Structural Geology in Reservoir Proceedings of the Liverpool Geological Society, Characterization. Geological Society, London, Session the thirty-ninth, 8, 241-267. Special Publications, 127, 65-82. Moss, ED. & EDMUNDS,W.M. 1992. Processes control- MARTINIUS, A.W. & NAESS, A. 2005. Uncertainty ling acid attenuation in the unsaturated zone of a analysis of fluvial outcrop data for stochastic reser- Triassic Sandstone aquifer (UK), in the absence of voir modelling. Petroleum Geoscience, 11, 203-214. carbonate minerals. Applied Geochemistry, 7, MATHERON, G., BEUCHER, H., DE FOUQUET,C., GALLI, 573-583. A., GUERILLOT, D. & RAVENNE, C. 1987. MOUNTNEY, N.ET. & THOMPSON, D.B. 2002. Strati- Conditional simulation of the geometry of fluvio- graphic evolution and preservation of aeolian deltaic reservoirs. In: Conditional Simulation of the dune and damp/wet interdune strata: an example Geometry of Huvio-deltaic Reservoirs, SPE Annual from the Triassic Helsby Sandstone Formation, Technical Conference and Exhibition, Society of Cheshire Basin, UK. Sedimentology, 49, 805-833. Petroleum Engineers, Dallas, TX, paper 16753-MS. MOYSEY, S., CAERS, J., KNIGHT, R. & ALLEN-KING, MAYER, K.U., BENNER, S.G., FRIND, E.O., THORNTON, R.M. 2003. Stochastic estimation of facies using S.E & LERNER, D.N. 2001. Reactive transport ground penetrating radar data. Stochastic Environ- modeling of processes controlling the distribution mental Research and Risk Assessment, 17, 306-318. and natural attenuation of phenolic compounds in NARR, W. & SUPPE, J. 1991. Joint spacing in sedimen- a deep sandstone aquifer. Journal of Contaminant tary rocks. Journal of Structural Geology, 13, Hydrology, 53, 341-368. 63-72. MCBRIDE, M.B. 1994. Environmental Chemistry of NAYLOR, H., TURNER, P, VAUGHAN,D.J., BOTCE, A.J. Soils. Oxford University Press, Oxford, 406. & FALLICK,A.E. 1989. Genetic studies of red bed MERCADO, A. 1967. The spreading of injected waters mineralization in the Triassic of the Cheshire in a permeable stratified aquifer. In: Artificial Basin, northwest England. Journal of the Geologi- Recharge and Management of Aquifers. IAHS cal Society, London, 146, 685-699. Publication, 72, 23-36. NEWELL, A.J. 2006. Calcrete as a source of hetero- METCALF, R., ROCHELLE, C.A., SAVAGE,D. t~ HIGGO, geneity in Triassic fluvial sandstone aquifers (Otter J.W. 1994. Huid-rock interactions during contine- Sandstone Formation, SW England). In: BARKER, tal red bed diagenesis: implications for theoretical R.D. & TELLAM, J.H. (eds) Fluid Flow and Solute models of mineralization in sedimentary basins. In: Movement in Sandstones: The Offshore UK PARNELL, J. (ed.) Geofluids: Origin, Migration and Perrno-Triassic Red Bed Sequence. Geological Evolution of Fluids in Sedimentary Basins. Society, London, Special Publications, 263, Geological Society, London, Special Publications, 119-127. 78, 301-324. ODLING, N.E. & RODEN, J.E. 1997. Contaminant trans- MIALL, A.D. 1990. Principles of Sedimentary Basin port in fractured rocks with significant matrix Analysis. Springer, New York. permeability, using natural fracture geometries. MILODOWSKI, A.E. & GILLESPIE, M.R. 1997. The Journal of Contaminant Hydrology, 27, 263-283. Petrology of Fractures and Fracture Mineralisation OGATA, A. & BANKS, R.B. 1961. A Solution of the in Sellafield BH3, British Geological Survey Differential Equation of Longitudinal Dispersion Report CC945/527/CF-I-A. in Porous Media. US Geological Survey, MILODOWSKI,A.E., GILLESPIE, M.R. & KEMP, S.J. 1994. Professional Paper, 411-A, A1-A9. A Review of the Fracture Mineralogy and Mineral OLORUNFEMI, M.O. & GRIFFITHS, D.H. 1985. A labora- Distribution in the Sellafield Boreholes, British tory investigation of the induced polarization of Geological Survey, Technical Report WG-194137. the Triassic Sherwood Sandstone of Lancashire MILODOWSKI, A.E., GILLESPIE, M.R., NADEN, J., and its hydrogeological applications. Geophysical FORTEY, N.J., SHEPHERD, Z.J., PEARCE, J.M. & Prospecting, 33, 110-127. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

44 J.H. TELLAM & R. D. BARKER

PAPATOLIOS, K.T. & LERNER, D.N. 1993. Defining a the Country Around Nantwich and Whitchurch. borehole capture zone in a complex sandstone Memoir of the Geological Survey of Great aquifer: a modelling case study from Shropshire, Britain. UK. Quarterly Journal of Engineering Geology POWELL, K.L., BARRETt, M.H. ET At. 2001. Virus- and Hydrogeology, 26, 193-204. colloid interactions: implications for groundwater PARKER, J., FOSTER, SS..D., SHERRATT, R. & ALDRICK, monitoring. In: SEILER, K.-P. & WOHNLICH, S. (eds) J. 1985. Diffuse Pollution and Groundwater New Approaches Characterizing Groundwater Quality in the Triassic Sandstone Aquifer in Flow, 1, Balkema, Rotterdam, 169-174. Southern Yorkshire. British Geological Survey POWELL, K.L., BARRETt, M.H., PEDLEY, S., TELLAM, Report, 17, No. 5. J.H., STAGG, K.A., GRESWELL, R.B. & RIVETr, M. PARKER, J.M., MASON, J. & KELLY, D.E 1988. A Study 2000. Enteric virus detection in groundwater using of Denitrification in the Triassic Sandstone Aquifer glass wool. In: SILILO, O.T.N. (ed.) Groundwater: in Yorkshire, England. British Geological Survey Past Achievements and Future Challenges. Hydrogeology Report, WD/88/23. Balkema, Rotterdam, 813-816. PARKER, K.E. 2005. Calculating ion exchange POWELL, K.L., TAYLOR, R.G. ET" At.. 2003. Microbial parameters for pure clay assemblages. PhD thesis, contamination of two urban sandstone aquifers in University of Birmingham. the UK. Water Research, 37, 339-352. PARKER, K.E., TELLAM, J.H. & CLIFF., M.I. 2000. PRICE, M.M., MORRIS, B.L. & ROBERTSON, A.S. 1982. Predicting ion exchange parameters for Triassic A study of intergranular and fissure permeability Sandstone aquifers. In: SILILO, O.T.N. (ed.) in Chalk and Permian aquifers, using double Groundwater: Past Achievements and Future packer injection testing. Journal of Hydrology, 54, Challenges. Balkema, Rotterdam, 581-586. 401-423. PARRY, W.T., CHAN, M.A. & BEITLER, B. 2004. PRINGLE, J.K., WESTERMAN, A.R., CLARK, J.D., Chemical bleaching indicates episodes of fluid flow DRINKWATER, N.J. & GARDINER, A.R. 2004. 3D in deformation bands in sandstone. AAPG high resolution digital models of outcrop analogue Bulletin, 88, 175-191. study sites to constrain rservoir model uncertainty: PATNODE, H.W. & WYLLIE, M.R.J. 1950. The presence an example from Alport, Derbyshire, UK. Petro- of conductive solids in reservoir rocks as a factor leum Geoscience, 10, 343-352. in electric log interpretation. Transactions of the PRIVET1", K.D. 2006. The capillary characteristic model American Institute of Mining and Metallurgical of petroleum hydrocarbon saturation in the Engineers, 189, 47-52. Permo-Triassic sandstone and its implications for PICKUP, G.E., RINGROSE, ES., JENSON, J.L. & SORBIE, remediation. In: BARKER, R.D. & TELLAM, J.H. K.S. 1994. Permeability tensors for sedimentary (eds) Fluid Flow and Solute Movement in Sand- structures. Mathematical Geology, 26, 227-250. stones: The Offshore UK Permo-Triassic Red Bed PICKUP, R.W., RHODES, G., ALAMILLO, M.L., Sequence. Geological Society, London, Special MALLINSON, H.E.H., THORNTON, S.E & LERNER, Publications, 263, 297-309. D.N. 2001. Microbiological analysis of multi-level PROSSER, D.J. & MASKALL,R. 1993. Permeability vari- borehole samples from a contaminated ground- ation within aeolian sandstone: a case study using water system. Journal of Contaminant Hydrology, core cut sub-parallel to slipface bedding, the Auk 53, 269-284. Field, Central North Sea, UK. In: NORTH, C.E & PITI'MAN, E.D. 1981. Effects of fault-related granula- PROSSER, D.J. (eds) Characterization of Fluvial and tion on porosity and permeability of quartz sand- Aeolian Reservoirs. Geological Society, London, stones. AAPG Bulletin, 65, 2381-2387. Special Publications, 73, 377-397. PLANT, J.A., JONES, D.G. & HASLAM, H.W. 1999. The READ, D., LAWLESS,T.A., SIMS, R.J. & BUTTER, K.R. Cheshire Basin: Basin Evolution, Fluid Movement 1993. Uranium migration through intact sandstone and Mineral Resources in a Permo-Triassic Rift cores. Journal of Contaminant Hydrology, 13, Setting. British Geological Survey: Keyworth, 277-289. 263. REES, S.B. 2006. Investigation and management of a PLUMMER, L.N. 1975. Mixing of sea water with calcium kerosene leakage into a Permo-Triassic sandstone carbonate groundwater. Geological Society of aquifer in the UK. In: BARKER, R.D. & TELLAM, America, Memoir, 42, 219-236. J.H. (eds) Fluid Flow and Solute Movement in POKAR, M., WEST, L.J. & ODLING, N.E. 2006. Petro- Sandstones: The Offshore UK Permo-Triassic Red physical characterization of the Sherwood Sand- Bed Sequence. Geological Society, London, Special stone from East Yorkshire, UK. In: BARKER, R.D. Publications, 263, 311-324. & TELLAM, J.H. (eds) Fluid Flow and Solute REEVES, M.J., SKINNER,A.C. & WILKINSON,W.B. 1975. Movement in Sandstones: The Offshore UK The relevance of aquifer-flow mechanisms to Permo-Triassic Red Bed Sequence. Geological exploration and development of groundwater Society, London, Special Publications, 263, resources. Journal of Hydrology, 25, 1-21. 103-118. RIMSTIDT, J.D. & VAUGHAN, D.J. 2003. Pyrite oxida- POOLE, E.G. & WHITEMAN,A.J. 1955. Variations in the tion: a state-of-the-art assessment of the reaction thickness of the Collyhurst Sandstone in the mechanism. Geochimica et Cosmochimica Acta, Manchester area. Bulletin of the Geological Survey 67, 873-880. of Great Britain, 9, 33-41. RINGROSE, P.S. & CORBETF, EW.M. 1994. Controls on POOLE, E.G. & WHITEMAN,A.J. 1966. The Geology of two-phase fluid flow in heterogeneous sandstones. Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 45

In: PARNELL,J. (eds) Geofluids: Origin, Migration UK Permo-Triassic Red Bed Sequence. Geological and Evolution of Fluids in Sedimentary Basins. Society, London, Special Publications, 263, Geological Society, London, Special Publications, 169-186. 78, 141-150. SHAND, P., HARGREAVES, R. & BREWERTON,L.J. 1997. RIVETr, M.O., LERNER, D.N., LLOYD,J.W. & CLARK, L. The Natural (Baseline) Quality of Groundwaters in 1990. Organic contamination of the Birmingham England and Wales: The Permo-Triassic Sand- aquifer, UK. Journal of Hydrology, 113, 307-323. stones of Cumbria, North-west England. (UK) ROBERTS, I. 1869. On the wells and water of Liverpool. Environment Agency R&D Project Record, Proceedings of the Liverpool Geological Society, 1, W6/i722/3, Birmingham, UK. 84-97. SHEPHERD, K.A. 2003. Contamination and ground- ROWE, J.R. & BURLEY,S.D. 1997. Faulting and porosity water quality in the Birmingham aquifer. PhD modification in the Sherwood Sandstone at thesis, University of Birmingham. Alderley Edge, northern Cheshire: an exhumed SHEPHERD, K.A., ELLIS, P.A. & RIVETf, M.O. 2006. example of fault-related diagenesis. In: MEADOWS, Integrated understanding of urban land, ground- N.S.,TRUEBLOOD, S.P, HARDMAN,M. & COWAN,G. water, baseflow and surface-water quality - The (eds) Petroleum Geology of the Irish Sea and City of Birmingham, UK. Science of the Total Adjacent Areas. Geological Society, London, Environment, 360, 180-195. Special Publications, 124, 325-352. SMEDLEY, P.L. & BREWERTON, t.J. 1997. The Natural RUBIN, H. & BUDDEMEIER, R.W. 1996. Transverse (Baseline) Quality of Groundwaters in England dispersion of contaminants in fractured permable and Wales: The Triassic Sherwood Sandstone of the formations. Journal of Contaminant Hydrology, East Midlands and South Yorkshire. (UK) 176, 133-151. Environment Agency R&D Project Record, RUBIN, H., SOLIMAN,A.M., BIRKHOLZER,J. t~ ROUVE, W6/i722/1. G. 1996. Transport of a tracer slug in a fractured SMEDLEY, P.L. ~ EDMUNDS, W.M. 2002. Redox permeable formation. Journal of Hydrology, 176, patterns and trace-element behavior in the East 133-151. Midlands Triassic Sandstone Aquifer, U.K. RUSHTON, K.R. & SALMON, S. 1993. Significance of Ground Water, 40, 44-58. vertical flow through low-conductivity zones in SMITH, R. & MOLLER, N. 2003. Sedimentology and Bromsgrove Sandstone aquifer. Journal of Hydrol- reservoir modelling of the Ormen Lange field, mid ogy, 152, 131-152. Norway. Marine and Petroleum Geology, 20, SAGE, R.C. & LLOYD, J.W. 1978. Drift deposit influ- 601-613. ences on the Triassic Sandstone aquifer of N.W. SPEARS, D. 1986. Pollution investigation of a Triassic Lancashire as inferred by hydrochemistry. Quar- sandstone aquifer: the role of mineralogy. In: terly Journal of Engineering Geology, 11, 209-218. CRIPPS, J., BELL, EG. & CULSHAW, MG. (eds) SAUER, R.I. In preparation. Solute transport around Groundwater in Engineering Geology. Geological pumping wells in a sandstone aquifer. PhD thesis, Society, London, Engineering Geology, Special University of Birmingham, UK. Publications, 3, 211-218. SCOTT, J.B.T. & BARKER, R.D. 2003. Determining SPEARS, D.A. & REEVES, M.J. 1975. The influence of pore-throat size in Permo-Triassic sandstones from drift deposits on groundwater quality in the Vae of low-frequency electrical spectroscopy. Geophysi- York. Quarterly Journal of Engineering Geology, cal Research Letters, 30, 3-1-3-4. 8, 255-269. ScoTT, J.B.T. & BARKER, R.D. 2005. Characterization SPENCE, M.J., BOTrRELL, S.H., HIGGO, J.J.W., of sandstone by electrical spectroscopy for strati- HARRISON, I. d~ FALLICK,A.E. 2001a. Denitrifica- graphical and hydrogeological investigations. tion and phenol degradation in a contaminated Quarterly Journal of Engineering Geology and aquifer. Journal of Contaminant Hydrology, 53, Hydrogeology, 38, 143-154. 305-318. ScoTT, J.B.T. & BARKER, R.D. 2006. Pore geometry of SPENCE, M.J., BOTFRELL, S.H., HIGGO, J.J.W., Permo-Triassic sandstone from measurements of THORNTON, S.E & LERNER, D.N. 2001b. Isotopic electrical spectroscopy. In: BARKER, R.D. & modelling of the significance of bacterial sulphate TELLAM, J.H. (eds) Fluid Flow and Solute reduction for phenol attenuation in a contamnated Movement in Sandstones: The Offshore UK aquifer. Journal of Contaminant Hydrology, 53, Permo-Triassic Red Bed Sequence. Geological 285-304. Society, London, Special Publications, 263, 65-81. STAFFORD, P,TORAN, L. & MACKAY,L. 1998. Influence SEGAR, D.A. 1993. The effect of open boreholes on of fracture truncation on dispersion: A dual groundwater flows and chemistry. PhD thesis, permeability model. Journal of Contaminant University of Birmingham, UK. Hydrology, 30, 79-100. SElF EL-DEIN, S.A.E.-B. 1983. Diagenesis and aquifer STAGG, K.A. 2000. An investigation into colloids in characteristics of the Sherwood Sandstone Group Triassic Sandstone groundwaters, UK. PhD thesis, in north Nottinghamshire. University of Birmingham. SEYMOUR, K.J., INGRAM, J.A. & GEBBETT, S.J. 2006. STAGG, K.A. & TELLAM, J.H. 1998. The use of image Structural controls on groundwater flow in the analysis in interpreting electron microscope Permo-Triassic sandstones of NW England. In: photographs of groundwater colloids. In: BARKER, R.D. & TELLAM, J.H. (eds) Fluid Flow BRAHANA, J., ECKSTEIN,Y., ONGLEY, L.K., SCHNEI- and Solute Movement in Sandstones: The Offshore DER, R. & MOORE, J.E. (eds) Gambling with Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

46 J.H. TELLAM & R. D. BARKER

Groundwater. Proc IAH Conference, Las Vegas. Central Germany. Earth and Planetary Science IAH, Smyrna, GA, 533-539. Letters, 212, 263-278. STAGG, K.A., KLEINERT, U.O., TELLAM, J.H. & LLOYD, TAYLOR, R.G., CRONIN, A.A., LERNER, D.N., TELLAM, J.W. 1997. Colloidal populations in urban and rural J.H., BOTTRELL, S.H., RUEEDI, J. & BARRETI', M.H. groundwaters, UK. In: CHILTON, EJ. Er AL. (eds) Submitted. Penetration of anthropogenic solutes Urban groundwater. Proc IAH Conference, in sandstone aquifers underlying two mature cities Nottingham, UK. Balkema, Rotterdam, 187-192. in the UK. Hydrogeology Journal. STAGG, K.A., TELLAM,J.H., BARRETt, M.H. & LERNER, TAYLOR, R.G., CRONIN, A.A., PEDLEY, S., ATKINSON, D.N. 1998. Hydrochemical variations with depth T.C. & BARKER, J.A. 2004. The implications of in a major UK aquifer: the fractured, high perme- groundwater velocity variations on microbial ability Triassic Sandstone. In: BRAHANA, J., transport and wellhead protection: review of field ECKSTEIN, Y., ONGLEu L.K., SCHNEIDER, R. & evidence. FEMS Microbiology and Ecology, 49, MOORE, J.E. (eds) Gambling with Groundwater. 17-26. Proc IAH Conference, Las Vegas. IAH, Smyrna, TAYLOR, R.G., CRONIN, A.A., TROWSDALE, S.A., GA, 53-58. BAINES, O.P., BARRETT, M.H. & LERNER, D.N. 2003. STANISTREET, I.G. & STOLLHOFEN, H. 2002. Hoanib Vertical groundwater flow in Permo-Triassic sedi- River flood deposits of Namib Desert interdunes ments underlying two cities in the Trent River as analogues for thin permeability barrier Basin (UK). Journal of Hydrology, 284, 92-113. mudstone layers in aeolianite reservoirs. Sedimen- TAYLOR, S. & BARKER, R.D. 2002. Resistivity of tology, 49, 719-736. partially saturated Triassic Sandstone. Geophysical STAUB, K.L., BENZ, M., SCHINK, B. & WIDDEL, E 1996. Prospecting, 50, 603q513. Anaeorobic, nitrate-dependent microbial oxida- TAYLOR, S.B. & BARKER, R.D. 2006. DC electrical tion of ferrous iron. Applied and Evironmental properties of Permo-Triassic sandstone. In: Microbiology, 62, 1458-1460. BARKER, R.D. & TELLAM, J.H. (eds) Fluid Flow STEEL, R.J. & THOMPSON, D.B. 1983. Structures and and Solute Movement in Sandstones: The Offshore textures in Triassic braided stream conglomerates UK Permo-Triassic Red Bed Sequence. Geological ('Bunter Pebble Beds') in the Sherwood Sand- Society, London, Special Publications, 263, stone Group, north Staffordshire, England. Sedi- 199-217. mentology, 30, 341-367. TELLS, V., DELAY,E & DE MARSILY,G. 2004. Compari- STEELE, A. & LERNER, D.N. 2001. Predictive model- son of genesis and geostatistical methods for char- ling of NAPL injection tests in variable aperture acterizing the heterogeneity of alluvial media: spatially correlated fractures. Journal of Contami- Groundwater flow and transport. Journal of nant Hydrology, 49, 287-310. Hydrology, 294, 103-121. STEVENTON-BARNES, H. 2001. Solid organic carbon in TELLAM, J.H. 1994. The groundwater chemistry of the UK aquifers: its role in sorption of organic con- Lower Mersey Basin Permo-Triassic Sandstone taminants. PhD thesis, University of London. aquifer system, UK: 1980 and pre-industrialisa- STREETLY, H., HAMILTON, C.A.L., BETrS, C., TELLAM, tion/urbanisation. Journal of Hydrology, 161, J.H. & HERBERT, A.W. 2002. Reconnaissance 287-325. tracer tests in the Triassic Sandstone aquifer north TELLAM, J.H. 1995a. Hydrochemistry of the saline of Liverpool, UK. Quarterly Journal of Engineer- groundwaters of the lower Mersey Basin Permo- ing Geology and Hydrogeology, 35, 167-178. Triassic sandstone aquifer, UK. Journal of Hydrol- STREETLY, M., CHAKRABARTY,C. & MCLEOD, R. 2000. ogy, 165, 45-84. Interpretation of pumping tests in the Sherwood TELLAM, J.H. 1995b. Urban groundwater pollution in Sandstone Group, Sellafield, Cumbria, UK. the Birmingham Triassic Sandstone aquifer. In: Quarterly Journal of Engineering Geology and 4th Annual IBC Conference on Groundwater Hydrogeology, 33, 281-299. Pollution. IBC, London, 10. STREETLY, M.J., HEATHCOTE,J.A. & DEGNAN,P.J. 2006. TELLAM, J.H. 1996. Interpreting the borehole water Estimation of vertical diffusivity from seasonal chemistry of the Permo-Triassic sandstone aquifer fluctuations in groundwater pressures in deep of the Liverpool area, UK. Geological Journal, 31, boreholes near Sellafield, NW England. In: 61-87. BARKER, R.D. & TELLAM, J.H. (eds) Fluid Flow TELLAM, J.H. 2004. 19th century studies of the hydro- and Solute Movement in Sandstones: The Offshore geology of the Permo-Triassic Sandstones of the UK Permo-Triassic Red Bed Sequence. Geological northern Cheshire Basin, England. In: MArI-IER, J. Society, London, Special Publications, 263, (ed.) 200 Years of British Hydrogeology. Geologi- 155-167. cal Society, London, Special Publications, 225, STRONG, G.E., MILOWDOWSKI, A.E., PEARCE, J.M., 89-105. KEMP, S.J., PRIOR, S.V. & MORTON, A.C. 1994. The TELLAM, J.H. & LLOYD, J.W. 1986. Problems in the petrology and diagenesis of Permo-Triassic rocks recognition of seawater intrusion by chemical of the Sellafield area, Cumbria. Proceedings of the means: an example of apparent chemical equiva- Yorkshire Geological Society, 50, 77-89. lence. Quarterly Journal of Engineering Geology SZURLIES, M., BACHMANN, G.H., MENNING, M., and Hydrogeology, 19, 389-398. NOWACZYK,N.R. & KADING,K.-C. 2003. Magneto- TELLAM, J.H. & LLOYD, J.W. 1997. The flushing of stratigraphy and high-resolution lithostratigraphy saline groundwater from a Permo-Triassic Sand- of the Permian-Triassic boundary interval in stone aquifer system in NW England. In: FEI JIN Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

SOLUTE TRANSPORT 47

& KROTHE, N.C. (eds) Proceedings of the 30th Experimental and modelling approaches for the International Geological Congress (Beijing, China, assessment of chemical impacts of leachate migra- Aug 1996), 22, VSB, Utrect, 177-184. tion from landfills: A case study of a site on the TELLAM, J.H. & THOMAS, A. 2002. Well water quality Triassic sandstone aquifer in the UK East and pollutant source distributions in an urban Midlands. Geotechnical and Geological Engineer- aquifer. In: HOWARD, K.W.E & ISRAFILOV, R.G. ing, 23, 811-829. (eds) Current Problems in Hydrogeology in Urban THORNTON, S.E, LERNER, D.N. & TELLAM, J.H. 1995. Areas, Urban Agglomerates and Industrial Centres. Laboratory Studies of Landfill Leachate-Triassic NATO Science Series, IV. Earth and Environ- Sandstone Interactions. Wastes Technical Division, mental Sciences, 8, 139-158. Department of the Environment Report, TELLAM, J.H., LLOYD, J.W. & WALTERS, M. 1986. The CWM/035A/94. morphology of a saline groundwater body; its TORSTENSSON, D., THORNTON, S.E, BROHOLM, M.M. & investigation, description and possible expla- LERNER, D.N. 1998. Hydrochemistry of pollutant nation. Journal of Hydrology, 83, 1-21. attenuation in groundwater contaminated by coal TELLAM, J.H., CARLYLE, H.E, EL-GHONEMY, H., tar wastes. In: LERNER, D.N. & WALTON, N.R.G. PARKER, K.E. & MITCHENER, R.G.R. 2002. (eds) Contaminated Land and Groundwater: Towards quantification of ion exchange in a sand- Future Directions. Geological Society, London, stone aquifer., In: THANGARAJAN,M., RAI, S.N. & Engineering Geology, Special Publications, 14, SINGH, V.S. (eds) Towards Quantification of Ion 149-157. Exchange in a Sandstone Aquifer. Sustainable TOTH, J. 1963. A theoretical analysis of groundwater Development and Management of Groundwater flow in a small drainage basin. Journal of Geo- Resources in Semi-arid Regions with Special Refer- physical Research, 68, 4795-4812. ence to Hard Rocks. Oxford & IBH Publishing, TRAVIS, C.B. & GREENWOOD, H.W. 1911. The miner- New Delhi, 205-214. alogical and chemical constitution of the Triassic TELLAM, J.H., THORNTON, S.E, LERNER, D.N., rocks of Wirral, Part I. Proceedings of the Liver- KLEINERT, U.O. & HARRIS, R.C. 1997. Leachate pool Geological Society, 11, 116-139. Migration Through the Triassic Sandstones at TROWSDALE, S.A. & LERNER, D.N. 2003. Implications Burntstump Landfill and Elsewhere. Environment of flow patterns in the sandstone aquifer beneath Agency R&D Report P183, (UK) Environment the mature conurbation of Nottingham (UK) for Agency, Bristol. source protection. Quarterly Journal of Engineer- THOMPSON, D.B. 1970a. Sedimentation of the Triassic ing Geology and Hydrogeology, 36, 197-206. (Scythian) red pebbly sandstones in the Cheshire TURNER, E, PILLING, D., WALKER, D., EXTON, J., Basin and its margins. Geological Journal, 7, BINNIE, J. & SABAOU, N. 2001. Sequence stratig- 183-216. raphy and sedimentology of the late Triassic TAG- THOMPSON, D.B. 1970b. The stratigraphy of the so- I (Blocks 401/402, Berkine Basin, Algeria). Marine called Keuper Sandstone Formation (Scythian- and Petroleum Geology, 18, 959-981. ?Anisian) in the Permo-Triassic Cheshire Basin. UNDERHILL, J.R. • WOODCOCK, N.H. 1987. Faulting Quarterly Journal of the Geological Society, mechanisms in high-porosity sandstones: New Red London, 126, 151-181. Sandstone, Arran, Scotland. In: JONES, M.E. & THORNTON, S.E, TELLAM, J.H. ~; LERNER, D.N. 2000b. PRESTON, R.M.E (eds) Deformation of Sediments Attenuation of landfill leachate by Triassic sand- and Sedimentary Rocks. Geological Society, stone aquifer materials, 1. Fate of inorganic pollu- London, Special Publications, 29, 91-105. tants in laboratory columns. Journal of UNIVERSITY OF BIRMINGHAM/NWWA. 1981. Lower Contaminant Hydrology, 43, 327-354. Mersey Basin Saline Groundwater Study: Final THORNTON, S.E, BRIGHT, M.I., LERNER, D.N. & Summary Report. North West Water Authority, TELLAM, J.H. 2000a. Attenuation of landfill Warrington, Cheshire, UK. leachate by Triassic sandstone aquifer materials, 2. VAIL, RR., AUDEMARD,E, BOWMAN,S.A., EISNER, P.N. Sorption and degradation of organic pollutants in & PEREZ-CRUZ, C. 1991. The stratigraphic signa- laboratory columns. Journal of Contaminant tures of tectonics, eustacy, and sedimentology - an Hydrology, 43, 355-383. overview. In: EINSELE, G., RICKEN, W. & THORNTON, S.F., QUIGLEY, S., SPENCE, M.J., BANWART, SEILACHER, A. (eds) Cycles and Events in Stratig- S.A., BOTTRELL, S. & LERNER, D.N. 2001b. raphy. Springer, Berlin, 617--657. Processes controlling the distribution and natural VAUGHAN,A., HANSEN, T., CARDON, H. & RADCLIFFE, attenuation of dissolved phenolic compounds in a N. 1999. Litho-flow facies prediction in an alluvial deep sandstone aquifer. Journal of Contaminant fan/fluvial system, Central North Sea. Petroleum Hydrology, 53, 233-267. Geoscience, 5, 409-419. THORNTON, S.E, LERNER, D.N. & BANWART, S.A. WALSH, J.J., WATrERSON, J., HEATH, A.E. & CHILDS, C. 2001a. Assessing the natural attenuation of 1998. Representation and scaling of faults in fluid organic contaminants in aquifers using plume- flow models. Petroleum Geoscience, 4, 241-251. scale electron and carbon balances: model WALTHALL, S. & CAMPBELL, J.E. 1986. The measure- development with analysis of uncertainty and ment, interpretation and use of permeability parameter sensitivity. Journal of Contaminant values with specific reference to fissured aquifers. Hydrology, 53, 199-232. In: CRIPPS, J.C., BELL, EG. & CULSHAW,M.G. (eds) THORNTON, S.E, TELLAM, J.H. & LERNER, D.N. 2005. Groundwater in Engineering Geology. Geological Downloaded from http://sp.lyellcollection.org/ by guest on September 24, 2021

48 J.H. TELLAM & R. D. BARKER

Society, London, Engineering Geology, Special WILLIAMS, M. 1952. Resistivity of brine-saturated Publications, 3, 273-278. sands in relation to pore geometry. AAPG WALTHALL, S. & INGRAM,J.A. 1984. The investigation Bulletin, 36, 253-277. of aquifer parameters using multiple piezometers. WINSHIP, E, BINLEY,A. & GOMEZ, D. 2006. Flow and Ground Water, 22, 25-30. transport in the unsaturated Sherwood Sandstone: WALTON, N.R.G. 1981. The origin of high iron concen- characterization using cross-borehole geophysical trations in groundwater from the Trassic aquifer in methods. In: BARKER, R.D. & TELLAM, J.H. (eds) East Devon. Journal of the Institute of Water Engi- Fluid Flow and Solute Movement in Sandstones: neers, 36, 63-77. The Offshore UK Permo-Triassic Red Bed WARD, R.S., WILLIAMS,A.T., BARKER, J.A., BREWER- Sequence. Geological Society, London, Special TON, L.J. & GALE, I.N. 1998. Groundwater Tracer Publications, 263, 219-231. Tests: A Review and Guidelines for Their Use in WONG, P.M.,TAMHANE, D. & WANG, L. 1997. A neural- British Aquifers. British Geological Survey, network approach to knowledge-based well inter- Report, WD/98/19. polation: A case study of a fluvial sandstone WARRINGTON, G., AUDLEY-CHARLES, M.G., ELLIOTr, reservoir. Journal of Petroleum Geology, 20, R.E., IVIMEY-COOK, H.C., ROBINSON, EL., 363-372. SHOTTON, EW. & TAYLOR,EM. 1980. A Correlation WORTHINGTON, EE 2004. The effect of scale on the of the Triassic Rocks in the British Isles. Geologi- petrophysical estimation of intergranular perme- cal Society, London, Special Report, 13, 78. ability. Petrophysics, 45, 59-72. WEALTHALL, G.E, STEELE, A., BLOOMFIELD,J.E, MOSS, WORTHINGTON, EE & BARKER, R.D. 1972. Methods R.H. & LERNER, D.N. 2001. Sediment filled frac- for the calculation of true formation factors in the tures in the Permo-Triassic sandstones of the Bunter Sandstone of northwest England. Engi- Cheshire Basin: observations and implications for neering Geology, 6, 213-228. pollutant transport. Journal of Contaminant WORTHINGTON, EE & GRIFFITHS, D.H. 1975. The Hydrology, 50, 41-51. application of geophysical methods in the explo- WHITE, K.E. 1986. Water tracers and investigatory ration and development of sandstone aquifers. techniques. In: BRANDON,T.W. (ed.) Groundwater." Quarterly Journal of Engineering Geology and Occurrence, Development and Protection. The Hydrogeology, 8, 73-102. Institution of Water Engineers and Scientists, WORTHINGTON, EH. 1973. Estimation of the perme- London, 353-383. ability of a Bunter Sandstone aquifer from labora- WILLIAMS, G.M., PICKUP, R.W., THORNTON, S.E, tory investigations and borehole resistivity LERNER, D.N., MALLINSON, H.E.H., MOORE, Y. & measurements. Water and Water Engineering, 77, WHITE, C. 2001. Biogeochemical characterisation 251-257. of a coal tar distillate plume. Journal of Contami- WYLLIE, M.R.J. & GREGORY, G.H.E 1953. Formation nant Hydrology, 53, 175-197. factors of unconsolidated porous media: influence WILLIAMSON, D.J. 1994. Controls on the transport of of particle shape and effect of cementation. Trans- organic pollutants through Permo-Triassic sand- actions of the American Institute of Mining and stone aquifers: low concentration sorption of Metallurgical and Petroleum Engineers, 198, selected species on mineral substrates. PhD thesis, 103-110. University of Birmingham, UK. YANG, C.S. & BAUMFALK,Y.A. 1997. Application of WILLIS, B.J. & WHITE, C.D. 2000. Quantitative outcrop high-frequency cycle analysis in high-resolution data from flow simulation. Journal of Sediment sequence stratigraphy. In: ZIEGLER, K., TURNER, E Research, 70, 788-802. & DAINES, S.R. (eds) Petroleum Geology of the WILLS, L.J. 1970. The Triassic succession in the central Southern North Sea: Future Potential. Geological Midlands in its regional setting. Quarterly Journal Society, London, Special Publications, 123, of the Geological Society, London, 126, 225-283. 181-203. WILLS, L.J. 1976. The Trias of Worcestershire and YANG, Y., LERNER, D.N., BARRETt, M.H. & TELLAM, Warwickshire. Institute of Geological Sciences, J.H. 1999. Quantification of groundwater recharge Report No. 7612. in the city of Nottingham, UK. Environmental WILSON, G., ANDREWS, J.N. & BATH, A.H. 1994. The Geology, 38, 183-198. nitrogen isotope composition of groundwater YEO, I.W., DE FREITAS, M.H. & ZIMMERMAN, R.W. nitrates from the East Midlands Triassic Sandstone 1998. Effect of shear displacement on the aperture aquifer, England. Journal of Hydrology, 157, and permeability of a rock fracture. International 35-46. Journal of Rock Mechanics and Mining Sciences, WINSAUER, W.O., SHEARIN, H.M., MASSON, RH. & 35, 1051-1070.