Sub-seismic-scale-fracture-pattern-and-in-situ-permeability-

Sub-seismic scale fracture pattern and in situ permeability data in the chalk atop of the Krempe salt ridge at L?gerdorf,NW Germany:Inferences on synfolding stress ?eld evolution and its impact on fracture connectivity

Fabrizio Storti a ,*,Fabrizio Balsamo b ,Federico Cappanera c ,Giambattista Tosi d

a

Dipartimento di Scienze della Terra,Universitàdegli Studi di Parma,Parma,Italy b

Dipartimento di Scienze Geologiche,Università“Roma Tre ”,Rome,Italy c

ENI E.&P.Division,S.Donato Milanese,Italy d

Eni Norge AS,Stavanger,Norway

a r t i c l e i n f o

Article history:

Received 19July 2010Received in revised form 24January 2011

Accepted 26March 2011

Available online 3April 2011Keywords:

Structural geology Joint

Fault zone Diapir

Permeability

a b s t r a c t

Chalk is exposed in the Heidestrasse quarry at L?gerdorf,at the top of the NE-SW trending Krempe salt ridge.Structural data indicate the presence of two joint sets,striking almost parallel and perpendicular to the salt ridge,respectively,and of a set of conjugate extensional faults and fault zones striking NW-SE,i.e.almost perpendicular to the salt ridge.Within the overall NW-SE trend of joints and faults,strike vari-ations occur from the massive chalk exposed in the lower half of the quarry,to the overlying layered chalk.A large variability characterizes the normalized spacing of both joint sets,which does not show any clear trend with layer dip.In situ measurements indicate that the cross-sectional permeability of tight joints increases 1e 2orders of magnitude with respect to the undeformed chalk.We propose that joint and fault azimuthal variability resulted from changes through time of the stress ellipsoid at the top of the salt ridge,while joint spacing variability is associated with the weak mechanical in ?uence of bedding in chalk.Azimuthal variability improves fracture connectivity and,hence,permeability and ?uid ?ow.

ó2011Elsevier Ltd.All rights reserved.

1.Introduction

Chalk is a very important lithology in petroleum geology,particularly in regions with salt diapirs and ridges like the Gulf of Mexico and North Sea.Owing to the lack of signi ?cant cementation and to the very small particle size,undeformed chalk is generally a high porosity,very low permeability biomicrite (e.g.Scholle,1977;Bell et al.,1999).Accordingly,it can act as a seal to hydrocarbons and as an effective pressure barrier to formation water in under-lying high pressure reservoirs (e.g.Mallon and Swarbrick,2002).However,the high effective permeability imparted by tectonic fracturing,makes chalk a well known hydrocarbon reservoir rock (e.g.Watts,1983;Corbett et al.,1987).This is favoured by the evidence that well developed joint patterns can be generated by very small extensional strains (e.g.Hooker et al.,2009;Olson et al.,2009).It is quite common that salt diapirs control the deformation pattern and geometry of hydrocarbon traps in the overlying chalk,which is typically affected by intense extensional fracturing

producing tectonic thinning in the diapir overburden (e.g.Davison et al.,2000).

The orientation,spacing,aperture,persistence,and type of mechanical discontinuities are all important factors controlling both the magnitude and anisotropy of rock permeability (Watts,1983;Tsang,1984;Narr and Suppe,1991;Antonellini and Aydin,1995;Caine et al.,1996;Agarwal et al.,1997;Child et al.,1997;Dholakia et al.,1998;Bell et al.,1999;Odling et al.,1999;Aydin,2000;Eichhubl and Boles,2000;Cello et al.,2001;Doolin and Mauldon,2001;M.Fisher and Knipe,2001;Labaume et al.,2001;Rawling et al.,2001;Bailey et al.,2002;Gale,2002;Laubach,2003;Davatzes et al.,2005;Géraud et al.,2006;Fossen et al.,2007;Sheldon and Micklethwaite,2007;Ortega et al.,2009).Owing to the wide variety of kinematic pathways that are associ-ated with the growth of salt-related structures (e.g.Vendeville and Jackson,1992a,b;Coward and Stewart,1995;Ge et al.,1997;Bonini,2003;Rowan et al.,2003;Stewart,2006),predictions of fracture patterns at the sub-seismic scale (joints and fault zones),and of stylolite attributes,are not straightforward (e.g.Watts,1983;Johnson and Bredeson,1971;Rowan et al.,1999;Davison et al.,2000).Small-scale variations of the local stress ?eld during the rise of diapirs can signi ?cantly in ?uence the resulting deformation

*Corresponding author.

E-mail address:fabrizio.storti@unipr.it (F.

Storti).

Contents lists available at ScienceDirect

Marine and Petroleum Geology

journal h omepage:ww w.elsevi 24ad71a4a45177232e60a256/locate/marp

etgeo

0264-8172/$e see front matter ó2011Elsevier Ltd.All rights reserved.doi:10.1016/j.marpetgeo.2011.03.014

Marine and Petroleum Geology 28(2011)1315e 1332

pattern and,hence,the associated permeability properties.This is particularly true for chalk,given its peculiar stratigraphical and petrophysical properties and their signi?cant variability(e.g. Hardman,1982),which control the mechanical response to stress (Corbett et al.,1987;da Silva et al.,1985;Hermansen et al.,2000; Frykman,2001;Risnes,2001;Fabricius and Borre,2007;Talesnick and Shehadeh,2007;Hjuler and Fabricius,2009).It follows that guidelines obtained by quantitative studies of?eld analogue structures of fractured chalk can provide fundamental support to ?uid?ow predictions in hydrocarbon reservoirs(e.g.Corbett et al., 1987;Koestler and Ehrmann,1991;Koestler and Reksten,1994; Alsop et al.,2000;Rijken and Cooke,2001;Schroeder et al.,2006; Bahat et al.,2007;Richard,2008;Gaviglio et al.,2009).For this purpose,we performed a?eld study of the deformation pattern in the Upper Cretaceous chalk overlying the northeastern termination of the Krempe salt ridge(e.g.Koestler and Ehrmann,1991),near L?gerdorf in Holstein,NW Germany(Fig.1).The variability of both fault and joint azimuth and spacing is interpreted as induced by a variability of the mechanical stratigraphy of the chalk and of the stress?eld during the rise of the salt ridge.The impact of such variability on permeability predictions is discussed.

2.Structural data

Our structural dataset includes fault orientation,kinematics and spacing,and joint orientation,spacing(S)and height(H)in order to compute normalized joint spacing(H/S;e.g.Tavani et al.,2006).We did not collect joint aperture data due to the poor reliability imparted them by the fact that the chalk underwent signi?cant uplift and erosion,and that the studied exposures were produced by recent quarrying.Data were statistically analysed by assuming that they have a Gaussian,or normal,distribution.The Gaussian distri-bution is a widely used probability distribution in statistical analysis and is based on the reasonable assumption that the data have a normal distribution around their mean value(e.g.Swan and Sandilands,1995).In structural geology,Gaussian distribution statistics have been used in lineament swarm analysis(e.g.Wise et al.,1985),cleavage(e.g.Salvini et al.,1999;Tavani et al.,2006), joint(e.g.Balsamo et al.,2008),and subsidiary fault population analysis(e.g.Storti et al.,2006),and also in particle shape and size, and permeability data analysis(e.g.Balsamo et al.,2010).To analyse the data by Gaussian distribution statistics,original data histograms are smoothed to reduce the noise component in the row data and then automatically?tted with Gaussian curves using the approxi-mation of the polymodal probability function in Fraser and Suzuki (1966).Each Gaussian curve is considered in its?1.5s interval to reduce the ambiguity among adjacent Gaussian curves.The poly-modal?tting procedure increases the number of Gaussian curves until the residual is below a given threshold value(10%)of the maximum height of the curves.The statistical procedure outlined above is described in detail in Salvini et al.(1999)and Storti et al. (2006),and implemented in the Daisy3software(Salvini,2011).

2.1.Overview of the quarry site

The large opencast quarries near L?gerdorf provide the oppor-tunity to study the deformation pattern in the shallowly dipping chalk uplifted about1000m by the rise of the Krempe salt ridge (Koestler and Ehrmann,1986,1987,1991)(Fig.1).The salt was deposited near the center of the Southern Permian basin(Ziegler, 1982).Diapirism started in Triassic time and continued during the Tertiary(Jaritz,1973).The present-day top of the Krempe ridge

at

Fig.1.a)Schematic location of the L?gerdorf area,at the northern tip of the NE-SW trending Krempe salt Ridge(after Koestler and Ehrmann,1991).b)Location of the Heidestrasse and Schinkel quarries,near the L?gerdorf village(image from Google Earth).c)Detail of(b)showing the old Heidestrasse quarry and the location of the?eld analysis sites.The white line indicates the trace of the pro?le cross section in Figure2.

F.Storti et al./Marine and Petroleum Geology28(2011)1315e1332

1316

L?gerdorf is inferred to be at a depth of 400e 800m (Jaritz,1973).The quarried chalk is a very homogeneous and almost mono-mineralic autochthonous white biomicrite with calcite content of 94e 98%,mostly provided by skeletal material,particularly cocco-liths (Hermann,1986).Its age spans from Middle Coniacian to Lower Maastrictian (Schulz et al.,1984).The upper few meters of the quarried chalk are affected by heavy fracturing,with also curvi-planar surfaces suggesting frost shattering,which likely occurred during the late Pleistocene (e.g.Bell et al.,1999).

Despite their kilometric size,the quarries provide very limited exposures of the Krempe salt ridge and this prevents exhaustive three-dimensional studies at the diapir scale.Our work focused on a gentle hectometer-scale anticline whose axis strikes almost perpendicular to the overall NE-SW oriented salt ridge (transversal subsidiary anticline;Fig.2a e d).The fold is exposed on the south-eastern wall of the Heidestrasse quarry,which is more than 60m high and cuts shallowly northwestward dipping chalk almost parallel to the strike.Analysis of bed thickness data indicates a bimodal distribution characterized by a large variability (Fig.2e).In particular,in the lower part of the section,the chalk is poorly layered (Fig.2f).Weak indications of bedding are provided by nodular chert layers,by rare and very thin marl beds,and by pyrite impregnations parallel to layering.In the upper part of the section,the sedimentary layering is more evident and bed

thickness

Fig.2.a)Panoramic view of the SE wall of the old Heidestrasse quarry:a cross-sectional view of the subsidiary gentle transversal anticline is well exposed.b)Pro ?le section of the transversal anticline showing the projected location of the ?eld analysis sites and their apparent dip values along the cross section azimuth (see Figure 1for section trace location).c)Rose diagram of bedding strike,showing a preferential clustering at N37 E ?7 .d)Stereographic projection (Schmidt net,lower hemisphere)of poles to bedding,their con-touring,and the corresponding p àdiagram showing the reconstructed average trend of the fold axis;thin circles paralleling the mean circle (thick)indicate the standard deviation associated with the computed mean circle.e)Histogram of bed thickness distribution and its best ?tting by Gaussian curves.The values corresponding to the two major Gaussian peaks are indicated.f)Photograph showing the typical ?eld appearance of poorly layered (massive)chalk.g)Photograph showing the typical ?eld appearance of layered chalk.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321317

decreases to an average value of about352mm(Fig.2g).No evidence of bedding-parallel slip was found.

Data were collected in twenty sites;nineteen of them are located in the old part of the Heidestrasse quarry(Fig.1).Defor-mation structures mostly consist of joints and extensional fault zones that,in some cases,preserve striated slickensides(Figs.3a,b, 4).Extremely thin bed-parallel compaction bands were also found. Joints are the dominant deformation structures in many outcrops (Fig.4).In some cases,both joints and conjugate low-displacement extensional faults are present,particularly in the crestal region of the anticline(Fig.5).No evidence of cement deposits in the joints and faults was found.Cumulative analysis of joint and shear frac-ture attitude in fault damage zones and in unfaulted sites,indicates the presence of a dominant set with an average NW-SE strike, characterized by a large variability and a near vertical dip(Fig.3c).A subordinate set strikes NE-SW and has a near vertical dip.Faults (i.e.single slip surfaces with very small displacement,in the order of few centimeters,and no visible damage zone and fault core) and fault zones have an average WNW-ESE orientation and “Andersonian dip”of about60 (Fig.3d).The angular pergence between the average strike of joints and faults is24 .Fault slick-enlines have an average NE-SW rake and a northeastward plunge (Fig.3e).

In the following,we?rst describe the patter of joints in sites not affected by signi?cant faulting,and then the fault pattern is illus-trated.A speci?c sub-section is devoted to the description of fault zones.

2.2.Joint pattern

Joints in sites not affected by signi?cant faulting are almost perpendicular to bedding and near vertical.Three azimuthal trends can be identi?ed by cumulative contouring of joint poles:NNW-SSE,WNW-ESE,and NE-SW(Fig.6).Owing to the NE strike of the quarry wall,where most of the data were collected,the relative abundances between the three joint sets might be biased.However, data from?eld site18(Fig.4),located in the NW striking northern wall of the new Heidestrasse quarry(Fig.1),con?rm results

from

Fig.3.a)Detail of the layered chalk showing stratabound NW-SE striking joints(adjacent to the hammer)abutting unstratabound NE-SW striking joints.b)Cross-sectional view of a fault zone affecting layered chalk.c)Stereographic projection(Schmidt net,lower hemisphere)showing contoured poles to total fractures:a clustering of fracture attitudes about an average NW-SE strike and near vertical dip is evident.d)Stereographic projection(Schmidt net,lower hemisphere)of poles to faults:their contouring shows a preferential clustering about an NW-SE strike.e)Stereographic projection(Schmidt net,lower hemisphere)of fault slickenlines:their contouring shows typical Andersonian dip values.

F.Storti et al./Marine and Petroleum Geology28(2011)1315e1332

1318

the cumulative statistical analysis of data collected in the old Hei-

destrasse quarry,i.e.most joints strike almost perpendicular to the

trend of the Krempe salt ridge.When only NNW-SSE to WNW-ESE

trending joints are analyzed by Gaussian distribution statistics,

almost 72%of the total data pertain to the ?rst trend (Fig.6c).The second peak indicates that about 31%of total data ?t into the WNW-ESE trend.The latter joint set dominantly occurs in the layered chalk,exposed in the upper part of quarry wall.NNW-SSE joints preferentially but not exclusively occur in the underlying massive chalk (Fig.4).If a single Gaussian curve is used,it provides an azimuthal peak at N38 W (322 ;joint set J1),characterized by a greater variability (T in Fig.6c).Analysis of data pertaining to the NE striking joint set (joint set J2)indicates a well de ?ned Gaussian peak with average strike at N39 E (039 ;Fig.6d).Joints are more abundant in the layered chalk exposed in the upper part of the quarry wall.In many cases,they are stratabound structures;in places,however,joints cut across some beds (Fig.3a).Generally J1joints abut against J2joints (Fig.3a);the opposite occurs in few cases.The spacing of J1joints in the layered chalk (sites 2,6,10,14)ranges from 348mm to 430mm.The standard deviations indicate that the difference between the two means is not statistically signi ?cant (Fig.7).Data from site 18refer to much thinner beds.On the other hand,J1spacing in the massive chalk (sites 12,19,20)shows values ranging from 3288mm to 3448mm (Fig.7).The smaller values from site 12(617mm)can be explained by the location of the analysed quarry wall segment,which is close to a transtensional fault zone that could have in ?uenced

the Fig.4.Stereographic projection (Schmidt net,lower hemisphere)showing bedding (broken circles),joint (thin circles)and extensional fault (thick circles)attitude,and slickenline orientation (white dots)in representative ?eld analysis sites projected on the pro ?le section of the subsidiary transversal

anticline.

Fig.5.Photograph showing the concomitant occurrence of low-displacement,conju-

gate NW-SE striking extensional faults and joints in layered chalk in the crest of the

subsidiary transversal anticline. F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321319

deformation pattern in the surrounding area.Unimodal Gaussian ?tting of cumulative J1joint spacing data provides an average value of about 321mm (Fig.8a).

The ratio between bed thickness (H)and the corresponding joint spacing data (S)provides values ranging from about 0.9to 3.2(Fig.7).Among the eight sites,three have average J1H/S close to 1,and other three have average H/S of about 1.6.Sites characterized by comparable joint spacing have either very similar (c.f.sites 2,10,14)or quite different average H/S values (c.f.sites 6and 14;19and 20).The anomalous situation of site 14,which is characterized by two distinct peaks,relates to the evidence that at this site some joints are stratabound structures with an average H/S value of about 1.7,while others are unstratabound structures that cut across a thickness of about 1900mm and have an average H/S value of about 3.2(Fig.7).Cumulative unimodal Gaussian distribution statistics of J1H/S values provide a peak of about 1.23(Fig.8b).

Analysis of J2joint attributes in sites not affected by signi ?cant faulting provides average spacing values ranging between about 240mm and 615mm (Fig.9).Spacing values at sites 2,6,10,and 14are comparable to J1joint spacing values in the corresponding sites when standard deviation values are considered.On the other hand,the average J2spacing in massive chalk (site 20)is about one ?fth of the corresponding J1joint spacing values.Cumulative unimodal Gaussian distribution statistics of J2joint spacing data provides a peak at about 337mm,i.e.very similar to the corresponding J1joint value (Fig.8c).When J2H/S values are considered,some are comparable with J1values in the corresponding sites (sites 2,10,14),whereas others are signi ?cantly different (sites 6and 20)(Fig.9).Cumulative unimodal Gaussian distribution statistics of J2joint H/S values provides a peak at about 1.6,i.e.higher than the corresponding result from J1joint values (Fig.8d).The standard deviations indicate that the difference between the two means is not statistically signi ?cant.

Analysis of the behavior of average H/S values in sites not affected by signi ?cant faulting does not show any clear relationship to bedding dip for both J1and J2joint sets (Fig.10a,b).A cumulative plot of J1joint spacing versus bed thickness indicates an overall positive correlation characterized by a large variability of spacing values in each bed.A similar behavior persists when the subset including only spacing values pertaining to beds thinner than 1m is considered (Fig.10c,d).Plotting fracture spacing ratios (Gross et al.,1995;Becker and Gross,1996)versus bed thickness indicates a large variability for values associated with comparable bed thicknesses (Fig.10e).The same analyses performed on thinner beds (site 18)con ?rm the overall results from total J1joints data with an improved linear best ?t (Fig.10f).J2joints spacing data show a quite poor positive correlation with bed thickness,charac-terized by large scattering particularly for thicker beds.The corre-sponding fracture spacing ratios show a positive correlation with bed thickness (Fig.10g).When only the subset of J2joints spacing pertaining to beds thinner than 1m is analyzed,the positive correlation has a very poor linear best ?t (Fig.10h).

Azimuthal angular relationships among J1joints and bedding strike show a cluster of high-angle values between 60 and 80 ,and another one of low angle values between 25 and 40 ,which are not affected by variations of bedding azimuth.A similar behavior characterizes the angle between J2joint azimuths and bedding azimuth,which shows more abundant values between 0 and 20 (Fig.11a).Comparable results are obtained when the same data are plotted against bedding dip (Fig.11b).The angle between the azi-muth of J1and J2joint sets varies between about 60 and 90 and does not show any correlation with bedding azimuth an dip (Fig.11c,d).2.3.Fault pattern

Most of the faults exposed in the old Heidestrasse quarry have small displacements,generally not exceeding a few hundred millimeters.They are preferentially located in the crestal region of the studied transversal subsidiary anticline and are commonly organized in conjugate extensional arrays that cut across jointed layers (Figs.5and 12a).On both fold limbs,low-displacement faults have a dominant southwestward dip.Most shear fractures are stratabound faults;many of them have height values roughly cor-responding to those of the quarry step walls (about 10e 15m as a maximum),and ?ve fault zones,having displacement exceeding about 1m,cut across the entire height of the stepped quarry wall.The spacing of the latter fault zones increases with increasing displacement (Fig.12b).

The spacing of low-displacement faults,measured as the distance between near parallel faults in adjacent conjugate

pairs

Fig.6.Stereographic projection (Schmidt net,lower hemisphere)showing contoured poles to joints at sites not affected by signi ?cant faulting.A major cluster about an NW-SE strike and near vertical dip is evident.b)Histogram of the corresponding dip values and its best ?tting by a single Gaussian curve having its peak (Gp)at about 85 .c)Stereographic projection (Schmidt net,lower hemisphere)showing contoured poles to J1joints:they are clustered into two azimuthal maxima,as also indicated by the associated histogram of joint azimuth which can be best ?tted by two Gaussian curves (1and 2).Fitting by a single Gaussian curve (T)is also provided.d)Stereographic projection (Schmidt net,lower hemisphere)showing contoured poles to J2joints,which are clustered about a well de ?ned single maximum,as also indicated by the associated histogram of joint azimuth which can be best ?tted by a single Gaussian curve.Gp means Gaussian peak.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 1332

1320

(Fig.12c),can be best ?tted by three Gaussian curves (Fig.12d).About 72%of the data belong to a curve with a peak at 1935mm of spacing.The second peak is at about 4400mm and involves about 21%of the data.Finally,about 10%of the data can be ?tted by a curve with a peak at 6089mm of 24ad71a4a45177232e60a256ing only two Gaussian curves provides a peak at about 1728mm of spacing,which involves about 60%of the data.The second curve involves about 36%of the data and has a peak at 4330mm of spacing (Fig.12e).Finally using a single Gaussian curve provides a peak value of 2391mm of spacing (Fig.12f).Analysis of the angular relationships between fault azimuth and bedding azimuth,and fault azimuth and J1or J2joints azimuth shows a pattern compa-rable with the corresponding joints (Fig.11).2.4.Fault zone architecture

Fault zone architecture,when displacement is greater than about 1m,is strongly dependent on the mechanical stratigraphy of the deformed chalk.No gouge layers were found in fault cores exposed in the old Heidestrasse quarry.Fault zones in layered chalk have a peculiar internal architecture,dominated by fracturing in damage zones of highly variable cross-sectional width,due to triangular fracture envelopes in both hangingwall and footwall (Fig.13a e c).In most cases,there was no clear evidence for sepa-rating shear fractures from joints in the outcrop.For this reason,we prefer to use the general term fracture to include both shear and tensile deformation structures (e.g.Price and Cosgrove,1990).

Sectors with different fracture frequencies occur in damage zones.Particularly,an inner damage zone is well distinguished from hangingwall and footwall outer damage zones by an abrupt increase of about one order of magnitude of fracture frequency.These frac-tures dominantly consist of fault-parallel shears encompassing fractured domains.Poorly coherent cataclastic fault core rocks occur in thin and discontinuous layers along principal slip surfaces.In many cases,however cataclastic rocks did not develop (Fig.13c).Fracture spacing analysis by Gaussian curve ?tting in a fault zone at site 11,having 1627mm of displacement in the analyzed cross section,illustrates the in ?uence of faulting on fracture frequency.The background fracture spacing at this site is 442mm and the corresponding H/S is 1.8(Fig.13d,e).In the hangingwall outer damage zone (sector A),fracture spacing is ?tted by two Gaussian curves having peaks at about 121mm and 66mm,respectively (Fig.13f).Approaching the inner damage zone (sector B),fracture spacing is again ?tted by two Gaussian curves,with peaks at about 69mm and 33mm,respectively.In the inner damage zone (sector C),fracture spacing is ?tted by a single Gaussian curve with a peak at about 9mm.Finally,fracture spacing in the footwall outer damage zone (sector D)is again ?tted by two Gaussian curves having peaks at about 84mm and 398mm,respectively.These data indicate that fracture frequency in the hangingwall outer damage zone is about 4e 7times higher than the background value and becomes 6e 13times higher approaching the inner damage zone,where it further increases to about 50times the background value.In the footwall outer damage zone,fracture frequency is four times higher and then it comes back to the background value.

The fault zone with the higher displacement value (3456mm in the analyzed cross section)is also exposed at site 11(Fig.14).It is characterized by an anomalously wide damage zone,that

is

Fig.7.Histograms showing spacing data of J1joints at sites not affected by signi ?cant faulting (upper two rows)and their corresponding values normalized by bed thickness (H/S)(lower two rows).Apart from H/S values at site 14,all data can be best ?tted by a single Gaussian curve.Gp means Gaussian peak.See text for details.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321321

bounded to the southwest by a pair of near vertical subsidiary fault zones.Six sectors have been identi ?ed within this fault zone,based on fracture spacing.Sector A is located in the outer footwall damage zone approaching the inner damage zone;fracture spacing

is best ?tted by a Gaussian curve with a peak at about 30mm.In the inner damage zone (sector B),fracture spacing decreases to about 10.5mm and then it increases again in the adjacent sector C (two Gaussian peaks at about 34mm and 75mm,respectively;Fig.14c,d).In the wide outer hangingwall damage zone,sector D has fracture spacing ?tted by two Gaussian peaks at about 23mm and 95mm,respectively.Fracture spacing of sector E is again ?tted by two Gaussian curves having peaks at about 50mm and 91mm,respectively.Finally,two Gaussian peaks at about 31mm and 156mm,respectively,characterize spacing of fractures in between the two near vertical subsidiary faults at the end of the hanging-wall outer damage zone (Fig.14e,f).These spacing values are generally comparable to those collected in the corresponding sectors of the previously described fault zone,and similar infer-ences on fracture frequency patterns can be made.Sectors E and

F

Fig.8.a)Histogram showing cumulative spacing data analysis of J1joints at sites not affected by signi ?cant faulting.The inset shows a subset of spacing data lower than 2000mm:it can be best ?tted by a single Gaussian curve with a peak at about 321mm.b)normalized spacing (H/S)values of J1joints in (a):they can be best ?tted by a single Gaussian curve with a peak at about 1.23.c)Histogram showing cumulative spacing of J2joint data at sites not affected by signi ?cant faulting:they can be best ?tted by a single Gaussian curve with a peak at about 337mm d)normalized spacing (H/S)values of J2joints in (c):they can be best ?tted by a single Gaussian curve with a peak at about 1.61.Gp means Gaussian peak.See text for

details.

Fig.9.Histograms showing spacing data of J2joints at sites not affected by signi ?cant faulting (upper ?ve graphs)and their corresponding values normalized by bed thick-ness (H/S)(lower ?ve graphs).Apart from H/S values at site 20,all data can be best ?tted by a single Gaussian curve.Gp means Gaussian peak.See text for details.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 1332

1322

are characterized by the presence of shallow-dipping stylolites

(Fig.14g)that make an average angle of 64 with the corre-

sponding high-angle faults.

Fault zones in massive chalk have a “conventional ”cross-

sectional structural architecture that consists of an outer damage

zone with a highly variable fracture intensity,an intensely fractured inner damage zone,and a fault core made up by poorly coherent protocataclasites.Typically,the architecture is strongly asymmet-rical,with a well developed damage zone in the hangingwall,commonly con ?ned by conjugate subsidiary faults (Fig.15).Fault-related deformation in the footwall of the main slip surface is limited to few synthetic and antithetic

shears.Fig.10.Scaling relationships of joints at sites not affected by signi ?cant faulting.a)Diagram showing J1joints H/S mean values plotted versus bedding dip.b)Diagram showing J2joints H/S mean values plotted versus bedding dip.c)Diagram showing the distribution of J1joint spacing plotted versus bed thickness.Note the large variability associated with bed thickness of about 600mm,2500mm,and 6000mm,respectively.d)Subset of the previous data for H values lower than 1000mm;open circles indicate data that were not included in the best ?t.e)J1fracture spacing ratio values plotted versus bed thickness.f)Diagram showing the distribution of J1joints spacing data plotted versus bed thickness at site 18;the inset shows the behavior of J1fracture spacing ratios for the same site.g)Diagram showing the distribution of J2joints spacing data plotted versus the corresponding bed thickness;the inset shows the behavior of J2fracture spacing ratios at the same sites.h)Subset of the same data for H values lower than 1000mm.See text for details.F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321323

Cumulative Gaussian distribution statistics of total fracture data associated with the ?rst described fault zone provide a peak,involving almost 71%of total data,at about 39mm (Fig.16a).The same analysis on data from the more complex fault zone exposed in the same site provides a peak at about 26mm when a single Gaussian curve is used (Fig.16b).Finally,the entire fault-related fracture dataset collected in the old Heidestrasse quarry is ?tted by two Gaussian curves:about 60%of total data provide a peak at almost 49mm of fracture spacing,while the second peak is at about 1/7of the former value,i.e.close to 7mm.Fitting the same data by a single Gaussian curve provides a peak at about 24mm of fracture spacing (Fig.16c).

Plotting the total average fault zone width versus the corre-sponding displacement provides a good linear ?t when an outlier point is excluded from the regression line (Fig.16d).On the other hand,the ratios between fault zone width and the corresponding

displacement,plotted against fault displacement,are best ?tted by an exponential curve (Fig.16e).3.Permeability data

In situ permeability was measured by the portable air mini-permeameter Tiny Perm II,manufactured by NERC,which has a nozzle diameter of 5mm and a nominal measure effectiveness in the 10à4to 10Darcy range (e.g.Rotevatn et al.,2008).Permeability data were acquired in pristine chalk,in tightly closed joints,and in fault breccias,respectively (Fig.17).Permeability data in pristine chalk,when ?tted by assuming a Gaussian distribution,provide two peaks at about 0.7mD (57.9%of the data)and 3.6mD (38.2%of the data),respectively.Permeability in hairline joints was measured by centering the nozzle of the minipermeameter across the joint trace and orienting the instrument parallel to the strike of

)

°(g n i d d e b o t e l g n a bedding dip (°))

°(g n i d d e b o t e l g n a bedding azimuth (°)

bedding dip (°))

°(s e r u t c u r t s n o i t a m r o f e d n e e w t e b e l g n a bedding azimuth (°)

J1 joints

J2 joints

faults

J1-J2

J1- faults

J2-faults

)

°(s e r u t c u r t s n o i t a m r o f e d n e e w t e b e l g n a a

b

c

d

Fig.11.Angular relations between bedding and deformation structure azimuths.a)Angle between bedding and deformation structure strike plotted versus bedding azimuth;symbols as in (b).b)Angle between bedding and deformation structure strike plotted versus bedding dip.c)Angle between deformation structure strike,plotted versus bedding azimuth;symbols as in (d).d)Angle between deformation structure strike plotted versus bedding dip.See text for details.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 1332

1324

the joint itself.Data are clustered in two quite different pop-ulations,?tted by Gaussian peaks at about 71mD (74.2%of the data)and 894mD (44.0%of the data),respectively.The higher perme-ability values were recorded from fault breccias,which provided a good ?t by a single Gaussian curve with a peak at about 1.62D (Fig.17).4.Discussion

The timing of fractures and especially joints is often problematic to specify,particularly when only limited quarry exposures are available.We used the clustering of poles to joints and faults,and their distribution relative to the salt ridge and subsidiary fold geometry to support a salt-induced,folding-related origin (Koestler and Ehrmann,1986,1987,1991;Koestler and Reksten,1994).The sharp contrast between the heavy fracturing in the upper few meters of the quarried chalk,and the much lower fracture intensity and almost constant fracture orientation at greater depth,do not support signi ?cant post-diapir near surface fracturing,including quarrying-related deformation.4.1.Joint azimuth

According to the abutting relationships observed in the ?eld,J2can be interpreted as systematic joints and J1as cross joints (e.g.Gross,1993;Gross and Eyal,2007).Cumulative statistical analysis indicates that,overall,J2joints strike parallel to bedding (c.f.Figs.2c and 6d)and J1make an angle of 75 to bedding strike,rather than about 90 as expected for transversal extensional deformational structures in cylindrical folding (e.g.Hancock,1985;Cooper,1992).This deviation can be explained by a change of the stress ?eld during salt ridge evolution,which can be schematically described by two main stages (Fig.18a).In the early onset of ridge growth,gentle parallel folding induced by the onset of salt ridge growth is accompanied by formation of J2joints perpendicularly to the transversal traction T t ,generated by outer arc extension (e.g.Ramsay,1967).With increasing fold amplitude,periclinal folding produces a component of outer arc extension parallel to the fold axis.The perturbation of transversal outer arc extension by longi-tudinal outer arc extension (T l )in the periclinal fold sector creates a resultant traction (T tl )responsible for the development of J1joints perpendicularly to it.The relative magnitude of T l is about 1/4of T t ,as computed from their vectorial summation (Fig.18a).

A more detailed analysis of the azimuthal relationships between joints and bedding reveals some scattering (Fig.11),which can be linked to the occurrence of the transversal subsidiary fold exposed in the old Heidestrasse quarry wall.The lack of ?eld evidence supporting ?exural slip suggests that the anticline developed by tangential e longitudinal strain (e.g.Ramsay,1967).This inference is supported by the negligible mechanical role of the poorly devel-oped and discontinuous bedding in the exposed chalk,and by the abundance of extensional fracturing.The growth of this fold likely perturbed,at the local scale,the stress ?eld associated with the rise of the Krempe salt ridge.Accordingly,joint scattering in the subsidiary anticline provides the opportunity to make inferences on the evolution through time of the stress ?eld heterogeneity by associating the azimuthal variability of J1joints with that of the T tl traction.The evidence that J1joints have different average azi-muths in the layered and massive chalk,respectively,can provide insights on the relative chronology of jointing if we assume that the two rock types have the same tensile strength and that stress channeling (e.g.Mandl,2000)during salt ridge ampli ?cation

thus

Fig.12.a)Pro ?le section of the subsidiary transversal anticline showing the preferential location of extensional faults (thin lines)and fault zones (thick lines)in the crestal region.b)Spacing of fault zones plotted versus their displacement.c)Cartoon showing the computation of fault spacing from conjugate arrays.d)Histogram of fault spacing data,which can be best ?tted by three Gaussian curves.e)The same data ?tted by two Gaussian curves.f)The same dataset ?tted by a single Gaussian curve.Gp means Gaussian peak.See text for details.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321325

favored initial jointing in the massive chalk.Accordingly,J1joints at sites 6,12,and 19may have formed before J1joints at sites 2,10and 14(Fig.4).The direction perpendicular to the average strike of J1joints in the massive chalk thus provides the azimuth of the trac-tion T tl1during initial “cylindrical ”folding.This azimuth is 46.8 E,very similar to the direction perpendicular to the average strike of cumulative J1joints in Figure 6c (51.7 E),which can be associated with T tl in Figure 18a.With the same reasoning,the azimuth of the ridge-parallel traction in the late deformation stages (T tl2)is 12.7 E.The counterclockwise rotation of 34.1 of T tl can be associated with the formation of the subsidiary across-ridge fold exposed in the old Heidestrasse quarry wall (Fig.18b).In particular,such rotation can be produced by the interplay between a fold-parallel traction induced by the along strike northwestward propagation of the subsidiary anticline (T f ),and T tl1.In particular,T f adds to T t and changes the ratio between T t and T l (Fig.18b).Finally,the virtual parallelism (c.f.Storti et al.,2006)between faults at site 16and J1joints at sites 2,10and 14,may suggest that the former developed in the late stages of deformation,during the growth of the

subsidiary across-ridge fold.Consequently,they may have post-dated faults at sites 10,11,12,13and 19.4.2.Joint spacing

Joint spacing normalized by the corresponding layer thickness is commonly thought to allow for the meaningful comparison of deformation intensity among beds with different thicknesses (e.g.Huang and Angelier,1989;Narr and Suppe,1991;Gross et al.,1995;Bai and Pollard,2000).Different normalization parameters have been proposed,including the fracture spacing index (FSI;Narr and Suppe,1991),the fracture spacing ratio (FSR;Gross et al.,1995),C/B ?ssility (Durney and Kisch,1994),and the H/S ratio (Tavani et al.,2006),which are based on slightly different statis-tical analyses.

Our data indicate a large variability of H/S for both J1and J2joint sets,which is not signi ?cantly in ?uenced by bedding dip.These are apparently unexpected results because invariance with bedding attitude would indicate fracture saturation and,

consequently,

Fig.13.a)Photograph showing the typical architecture of a fault zone exposed in the Heidestrasse quarry:A and B are two sectors in the hangingwall outer damage zone,characterized by increasing fracture frequency;C indicates the inner damage zone;D indicates the footwall outer damage zone.b)Stereographic projection (Schmidt net,lower hemisphere)showing the attitude of the principal slip zone (thick circle)and of subsidiary faults (thin circles),and of fault-related fractures.c)Detail of (a)showing the fracture architecture in the damage zone.d)Background fracture spacing at sit 11.e)Background fracture spacing normalized by bed thickness (H/S).f)Fracture spacing in the four sectors of the damage zone indicated in (a).Gp means Gaussian peak.See text for details.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 1332

1326

comparable H/S values would be expected in all sites (Narr and Suppe,1991).Fracture saturation implies well developed joint sets,that are expected to provide consistent results when the line method is used to collect spacing data (Wu and Pollard,1995).A possible explanation for reconciling this ?eld evidence relies on the mechanical role of layering in the chalk exposed in the Heidestrasse quarry (e.g.Di Naccio et al.,2005;Cooke et al.,2006;Laubach et al.,2009;Lezin et al.,2009;Morris et al.,2009).When not highlighted by chert layers,bedding consists of subtle discontinuities without marly or clay interlayers,which are not expected to produce effective changes in rock mechanical properties across them.These conditions are quite far from those adopted in elastic models of joint in ?lling (e.g.Hobbs,1967;Narr and Suppe,1991)and,consequently,bedding may not systematically provide arrest surfaces to joint propagation (e.g.Underwood et al.,2003;Lezin et al.,2009).Depending on the mechanical coupling between adjacent beds,joints can either arrest or propagate across them (e.g.Rijken and Cooke,2001).This ?uctuation of mechanical layering thickness can explain the large scattering of H/S values in adjacent ?eld sites,particularly when dealing with thick mechan-ical layers,which are likely to deviate from a linear behavior with fracture spacing (e.g.Mandal et al.,1994).4.3.Fault pattern

Previous work on the deformation structures exposed in the chalk quarries at L?gerdorf reported polygonal extensional fault

patterns (Koestler and Ehrmann,1987,1991).Our data from the old Heidestrasse quarry do not support this.Instead,they illustrate the presence of truly Andersonian conjugate extensional faults and fault zones characterized by azimuthal variability at different sites that can be explained in terms of variations of mechanical stratig-raphy and stress ?eld.The reasons of these contrasting interpre-tations can be found in the different statistical approaches.In fact,also in our case,cumulative statistics of fault data (Koestler and Ehrmann,1987,1991)provides an azimuthal variability that can exceed 60 (Fig.3c),thus apparently supporting extensional polygonal faulting (e.g.Reches,1983).Different inferences on fault attitudes and faulting mechanisms in ?eld analogues signi ?cantly affect fracture patterns implemented in reservoir production models and,hence,permeability predictions.We support a multi-step procedure for analyzing the geometric and kinematic congruency of fracture attribute data,from single ?eld site,to neighboring ?eld site clusters,up to cumulative data statistics.This should contribute to reducing the uncertainty associated with the application of ?eld-derived fracture patterns to reservoir perme-ability models.

4.4.Fault zone evolution

The evidence that abundant extensional faults exposed in the upper part of the studied quarry cut across the layered chalk without in ?uencing the frequency of J1joints (Fig.5),and that damage zones only occur in the ?ve major fault zones located

in

Fig.14.Structural architecture of the fault zone with the highest displacement among those exposed in the Heidestrasse quarry.a)Overview of the fault zone.b)Stereographic projection (Schmidt net,lower hemisphere)showing the attitude of the principal slip zone (thick circle)and of fault-related fractures.c)Detail of the fault zone showing the location of sector A,B,and C in the damage zone.The position of the photo is indicated in (a).d)Spacing values of fault-related fractures in sectors A,B,and C.e)View of sectors D,E,and F in the hangingwall outer damage zone.The position of the photo is indicated in (a).f)Spacing values of fault-related fractures in sectors D,E,and F.g)Detail of sector E showing the attitude and spacing of moderately-dipping stylolites adjacent to a near vertical fault.The position of the photo is indicated in (e).See text for details.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321327

Fig.15.a)Overview of a typical fault zone architecture exposed in massive chalk.Fault-related fracturing is concentrated near the principal slip zone of the conjugate pair.b)Stereographic projection (Schmidt net,lower hemisphere)showing the attitude of the principal slip zone (thick circle)and of fault-related fractures.c)Detail of the fault zone showing the cataclastic fault core and the very intense fracturing in the inner damage zone.The position of the photo is indicated in (a).d)Fracture spacing in the inner damage zone.Gp means Gaussian peak.See text for

details.

Fig.16.a)Cumulative fracture spacing data pertaining to the fault zone illustrated in Figure 13.b)Cumulative fracture spacing data pertaining to the fault zone illustrated in Figure 14.c)Cumulative spacing of fault-related fracture data collected in the Heidestrasse quarry.In (b)and (c)1and 2indicate the peaks associated with best ?tting by two Gaussian curves;T indicates the peak associated with ?tting by a single Gaussian curve.d)Diagram showing fault zone width plotted versus the corresponding displacement values.e)Diagram showing the ratio of fault zone width versus the corresponding displacement,plotted versus fault displacement.See text for details.F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 1332

1328

the central part of the subsidiary anticline (Fig.12a),indicates that fracture in ?lling (e.g.Hobbs,1967;Narr and Suppe,1991;Wu and Pollard,1995)and the consequent development of fault damage zones were caused by the contribution of the kinematically-induced steeply-dipping compressional traction produced by simple shear along the principal slip surfaces (e.g.Harding,1985;Sylvester,1988).The exponentially decreasing correlation curve of the ratio between fault zone width and displacement versus fault displacement (Fig.16e)indicates that fracture in ?lling in damage zones occurred in the early stages of faulting because the ?nal width of fault zones was acquired when fault slip was lower than about 1m.

In our preferred evolutionary pathway,extensional faulting followed across-ridge jointing (J1)and started by development of widespread conjugate fault pairs with negligible displacement,in a progressive deformation localization scenario.Most of these faults were rapidly abandoned and further extension was accom-modated by the activity of few fault zones,likely favoured by an increased across-fold dilational component in the subsidiary anti-clinal crest (Fig.19).This inference is supported by the angle greater than 45 between shallow-dipping stylolites and their parent faults (Fig.14).The very steep attitude of the latter,resembling compound fractures (Gross and Eyal,2007),provides additional evidence of

a dilational component,which contributed to increase fault dip from truly Andersonian to near vertical.4.5.Permeability implications

The counterclockwise rotation of deformation structures upward from massive to layered chalk bears important implica-tions for both subsurface ?uid ?ow and recovery.The formation of late thoroughgoing fault zones striking oblique to both early J1and J2joint sets ensures a much more effective fracture connectivity compared with that associated with fault zones striking parallel to J1joints (Fig.20a),particularly when the hydraulic role of damage zones is considered.Optimization

of

Fig.17.Permeability data collected in the Heidestrasse quarry.a)Data collected in undeformed chalk.b)Data collected across tight joints.c)Data collected in fault

breccia.

Fig.18.Proposed relationships between jointing and salt ridge growth.a)Two-step cartoon showing in map view the development of J2joints by transversal outer arc extension (thick black arrows;T t )during early folding of the ridge overburden (1);strike and dip of bedding are schematically indicated.With increasing salt ridge rise (2),periclinal folding at the ridge tip occurs under concomitant longitudinal (T l and white thick arrows)and transversal outer arc extension.J1joints form perpendicular to the resultant traction T tl .b)Map view showing the proposed rotation of traction T tl during salt ridge growth.The initial azimuth of T tl (T tl1),as obtained by the perpen-dicular to J1joints in massive chalk (J1a),is rotated counterclockwise by 34.1 due to the interaction with the new traction T f produced by the growth of the across-ridge subsidiary anticline.The resulting new azimuth of T tl (T tl2)is responsible of J1joint orientation in the layered chalk (J1b).

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 13321329

fracture interception capability for hydrocarbon recovery can be signi ?cantly in ?uenced by the intraformational variability of the strike of deformation structures.The hectometre-scale and gentle limb dip of subsidiary transversal folds developed ahead of salt diapirs may lead to the overlooking of these features during re ?ection seismic data.This,in turn,may lead to neglect of signi ?cant azimuthal rotations of joints and faults,which affects the appropriate planning of deviated deep production wells in reservoirs.

Scan lines along quarry wall steps approximate horizontal borehole in the crestal region of salt ridges or diapirs.Our results indicate that a high variability of fracture interceptions can occur along bed-parallel wells when bedding provides loose mechanical discontinuities.Being associated with the mechanical stratigraphy of chalk,this fracture frequency variability does not depend on,and cannot be predictively modeled by,methods based on bedding dip or curvature.Systematic collection of fracture spacing versus layer thickness datasets in well exposed natural analogues should be routinely carried out to provide robust statistical constraints to synthetic fracture populations in reservoir dynamic models.

The evolution of fracturing inferred in the studied exposures of the Krempe salt ridge indicates an increase of fracture perme-ability through time.Jointing and initial diffuse Andersonian faulting with very low-displacement strongly increase fracture connectivity.Conjugate crossing faults localize deformation in

fault intersection regions,which can produce permeability anisotropy along the fault strike (e.g.Ferrill et al.,2000).Subse-quent deformation localization,enhanced by a dilational compo-nent,creates fracture corridors that likely provide very effective drainage and ?ow pathways for ?uids stored in the background fracture network and primary pores (Fig.20b).This is further enhanced by the development of anomalously wide fault zones (Fig.14)that are outliers in the routinely used fault scaling rela-tions (e.g.Schlische et al.,1996).Provided that an effective over-lying sealing formation occurs,this permeability enhancing evolutionary pathway on deformation structures well below the resolution of re ?ection seismic data can signi ?cantly contribute to the explanation of hydrocarbon accumulations in the chalk deformed by salt diapirs (e.g.Watts,1983).5.Conclusions

Fracture pattern and evolution in space and time,inferred from the studied exposures of chalk in the crest of the Krempe salt ridge,imply an increase through time of fracture-related permeability.The following six conclusive points can be drawn from our results:1)two joint sets overprinted by a set of conjugate extensional faults and fault zones developed in the salt ridge crest:one joint set strikes almost parallel to the ridge,while the other one and most faults and fault zones strike nearly perpendicular to the ridge trend;

2)The frequency of joints,normalized by the thickness of the corresponding mechanical layers,is characterized by a

high

Fig.19.The three main evolutionary steps of deformation progression:initial wide-spread jointing (1)is followed by development of diffuse low-displacement conjugate extensional faulting (2);further extension is accommodated by few fault zones while most previous faults are rapidly abandoned;damage zones formed only along fault zones

(3).

Fig.20.Relationships between faulting and permeability.a)Schematic map view showing the improved connectivity potential of late faults and fault zones striking oblique to both J1and J2joint sets,compared with that of early J1-parallel faults.b)Intense fracturing in fault damage zones without developing sealing fault cores strongly enhances secondary porosity and permeability along fault zones,which behave as preferential storage sites and effective conduits for ?uid ?ow.

F.Storti et al./Marine and Petroleum Geology 28(2011)1315e 1332

1330

variability that likely relates to the weak mechanical in?uence of bedding in the studied chalk,rather than to the very gentle fold curvature;

3)In situ measurements indicate that permeability along tight

joints increases1e2orders of magnitude with respect to undeformed chalk;

4)Azimuthal rotations of faults and joints with depth,likely

induced by shape and orientation variations of the stress ellipsoid through time,signi?cantly in?uence fracture connectivity;

5)The widespread occurrence of extensional conjugate fault pairs

with negligible displacement,and of few fault zones with metre-scale displacements,and heavily fractured damage zones,further improves joint-related secondary porosity and provides effective?uid?ow pathways at the top of the salt ridge that are expected to strongly favour hydrocarbon accu-mulation and recovery;

6)All these deformation structures are well below the resolution

of re?ection seismic volumes and,consequently,?eld work on natural analogues is fundamental for their successful imple-mentation into reservoir permeability numerical predictive tools.

Acknowledgements

We are extremely grateful to the Associate Editor J.W.F.Waldron and to I.Alsop and an anonymous reviewer for their careful reviews that allowed us to signi?cantly improve the originally submitted manuscript.This paper derives from a research project on frac-turing in chalk,funded by ENI E.&P.Division and ENI Norge As.We gratefully acknowledge both companies for their support and for releasing material for publication.We are also grateful to Holcim Cement Industries for kindly providing us access to the Heides-trasse quarry,and to T.Kusche for his help during?eldwork.Data statistics and stereographic projections were made by using the Daisy software(Salvini,2011).This paper is dedicated to the memory of Elisabetta Costa.

References

Agarwal,B.,Allen,L.R.,Farrell,H.E.,1997.Eko?sk?eld reservoir characterization: mapping permeability through facies and fracture intensity.Society of Petro-leum Engineers Formation Evaluation12,227e233.

Alsop,G.I.,Brown,J.P.,Davison,I.,Gibling,M.R.,2000.The Geometry of Drag Zones Adjacent to Salt Diapirs.Journal of the Geological Society,London.157, 1019e1029.

Antonellini,M.,Aydin,A.,1995.Effect of faulting on?uid?ow in porous sandstones: geometry and spatial distribution.American Association of Petroleum Geolo-gists Bulletin79,642e671.

Aydin,A.,2000.Fractures,faults,and hydrocarbon entrapment,migration,and?ow.

Marine and Petroleum Geology17,797e814.

Bahat,D.,Rabinovich,A.,Frid,V.,Bankwitz,P.,24ad71a4a45177232e60a256parative fracture of joints that strike parallel to structures in sedimentary rocks e characterization and implications.Israelian Journal of Earth Sciences55,113e127.

Bai,T.,Pollard,D.D.,2000.Fracture spacing in layered rocks:a new explanation based on the stress transition.Journal of Structural Geology22,43e57. Bailey,W.R.,Manzocchi,T.,Walsh,J.J.,Keogh,K.,Hodgetts,D.,Rippon,J.,Nell,P.A.R., Flint,S.,Strand,J.A.,2002.The effect of faults on the3D connectivity of reservoir bodies:a case study from the East Pennine coal?eld,U.K.Petroleum Geoscience 8,263e277.

Balsamo, F.,Storti, F.,Piovano, B.,Salvini, F.,Cifelli, F.,Lima, C.C.,2008.Time dependent structural architecture of subsidiary fracturing and stress pattern in the tip region of an extensional growth fault system,Tarquinia basin,Italy.

Tectonophysics454,54e69.

Balsamo,F.,Storti,F.,Salvini,F.,Silva,A.T.,Lima,C.C.,2010.Structural and petro-physical evolution of extensional fault zones in low-porosity,poorly lithi?ed sandstones of the Barreiras formation,NE Brazil.Journal of Structural Geology32, 1806e1826.

Becker,A.,Gross,M.R.,1996.Mechanism for joint saturation in mechanically layered rocks:an example from southern Israel.Tectonophysics257,223e237.

Bell, F.G.,Culshaw,M.G.,Cripps,J.C.,1999.A review of selected engineering geological characteristics of English chalk.Engineering Geology54,237e269.Bonini,M.,2003.Detachment folding,fold ampli?cation,and diapirism in thrust wedge experiments.Tectonics22.doi:10.1029/2002TC001458.

Caine,J.S.,Evans,J.P.,Forster,C.B.,1996.Fault zone architecture and permeability structure.Geology24,1025e1028.

Cello,G.,Invernizzi,C.,Mazzoli,S.,Tondi,E.,2001.Fault properties and?uid?ow patterns from Quaternary faults in the Apennines,Italy.Tectonophysics336, 63e78.

Child,C.,Walsh,J.J.,Watterson,J.,24ad71a4a45177232e60a256plexity in fault zone structure and implications for fault seal predictions.In:M?ller-Pedersen,P.,Koestler,A.G.

(Eds.),Hydrocarbon Seals:Importance for Exploration and Production.

Norwegian Petroleum Society Special.publication7,Singapore,pp.61e72. Cooke,M.L.,Simo,J.A.,Underwood,C.A.,Rjiken,P.,2006.Mechanical stratigraphic controls on fracture patterns within carbonates and implications for ground-water?ow.Sedimentary Geology184,225e239.

Cooper,M.,1992.The analysis of fracture systems in subsurface thrust structures from the Foothills of the Canadian Rockies.In:McClay,K.R.(Ed.),Thrust Tectonics.Chapman and Hall,London,pp.391e405.

Corbett,K.,Friedman,M.,Spang,J.,1987.Fracture development and mechanical stratigraphy of Austin chalk,Texas.American Association of Petroleum Geologists Bulletin71,17e28.

Coward,M.P.,Stewart,S.,1995.Salt-in?uenced structures in the Mesozoic-Tertiary cover of the southern North Sea,UK.In:Jackson,M.P.,Roberts,D.G.,Snelson,S.

(Eds.),Salt Tectonics:A Global Perspective.American Association of Petroleum Geologists Memoir65,Tulsa,pp.229e250.

da Silva,F.,Monjoie,A.,Debande,G.,Schroeder,C.,Poot,B.,Detiege,C.,Halleux,L., 1985.Mechanical Behaviour of Chalks.Second North Sea Chalk Symposium Stavanger,Norway1e10.

Davatzes,N.C.,Eichhubl,P.,Aydin,A.,2005.Structural evolution of fault zones in sandstone by multiple deformation mechanisms:Moab fault,southeast Utah.

Geological Society of America Bulletin117,135e148.

Davison,I.,Alsop,G.I.,Evans,N.G.,Safaricz,M.,2000.Overburden deformation patternsand mechanisms of salt diapir penetration in the Central Graben,North Sea.Marine and Petroleum Geology17,601e618.

Dholakia,S.K.,Aydin,A.,Pollard,D.D.,Zoback,M.D.,1998.Fault-controlled hydro-carbon pathways in the Monterey Formation,California.American Association of Petroleum Geologists Bulletin82,1551e1574.

Di Naccio, D.,Boncio,P.,Cirilli,S.,Casaglia, F.,Morettini, E.,Lavecchia,G., Brozzetti,F.,2005.Role of the mechanical stratigraphy on fracture development in carbonate reservoirs:insights from outcropping shallow water carbonates in the Umbria-Marche Apennines,Italy.Journal of Volcanology and Geothermal Research148,98e115.

Doolin, D.M.,Mauldon,M.,2001.Fracture permeability normal to bedding in layered rock masses.International Journal of Rock Mechanics&Mining Sciences 38,199e210.

Durney,D.W.,Kisch,H.,1994.A?eld classi?cation and intensity scale for?rst-generation cleavages.AGSO Journal of Australian Geology and Geophysics15, 257e295.

Eichhubl,P.,Boles,J.R.,2000.Focused?uid?ow along faults in the Monterey Forma-tion,coastal California.Geological Society of America Bulletin112,1667e1679. Fabricius,I.L.,Borre,M.,2007.Stylolites,porosity,depositional texture,and silicates in chalk facies sediments.Ontong Java Plateau e Gorm and Tyra?elds,North Sea.Sedimentology54,183e205.

Ferrill, D.A.,Morris, A.P.,Sims, D.W.,Stamatakos,J.A.,Sims, D.,2000.Crossing conjugate normal faults.American Association of Petroleum Geologists Bulletin 84,1543e1559.

Fisher,Q.J.,Knipe,R.J.,2001.The permeability of faults within siliciclastic petroleum reservoirs of the North Sea and Norwegian Continental Shelf.Marine and Petroleum Geology18,1063e1081.

Fossen,H.,Schultz,R.A.,Shipton,Z.K.,Mair,K.,2007.Deformation bands in sand-stones:a review.Journal of the Geological Society164,755e769.

Fraser,R.D.B.,Suzuki,E.,1966.Resolution of overlapping absorption bands by least squares procedures.Analytical Chemistry38,1770e1773.

Frykman,P.,2001.Spatial variability in petrophysical properties in Upper Maas-trichtian chalk outcrops at Stevns Klint,Denmark.Marine and Petroleum Geology18,1041e1062.

Gale,J.F.W.,2002.Specifying lengths of horizontal wells in fractured reservoirs.

Society of Petroleum Engineers Reservoir Evaluation and Engineering5, 266e272.

Gaviglio,P.,Bekri,S.,Vandycke,S.,Adler,P.M.,Schroeder, C.,Bergerat, F., Darquennes,A.,Coulon,M.,2009.Faulting and deformation in chalk.Journal of Structural Geology31,194e207.

Ge,H.,Jackson,M.P.A.,Vendeville, B.C.,1997.Kinematics and dynamics of salt tectonics driven by progradation.American Association of Petroleum Geologists Bulletin81,398e423.

Géraud,Y.,Diraison,M.,Orellana,N.,2006.Fault zone geometry of a mature active normal fault:a potential high permeability channel(Pirgaki fault,Corinth rift, Greece).Tectonophysics426,61e76.

Gross,M.R.,1993.The origin and spacing of cross joints:examples from the Mon-terey formation,Santa Barbara Coastline,California.Journal of Structural Geology15,737e751.

Gross,M.R.,Eyal,Y.,2007.Throughgoing fractures in layered carbonate rocks.

Geological Society of America Bulletin119,1387e1404.

Gross,M.R.,Fisher,M.P.,Engelder,T.,Green?eld,R.J.,1995.Factors controlling joint spacing in interbedded sedimentary rocks:integrating numerical models with ?eld observations from the Monterey Formation,USA.In:Ameen,M.S.(Ed.),

F.Storti et al./Marine and Petroleum Geology28(2011)1315e13321331

Fractography:Fracture Topography as a Tool in Fracture Mechanics and Stress Analysis.Geological Society Special Publication,pp.215e233.92.

Harding,T.P.,1985.Seismic characteristics and identi?cation of negative?ower structures,positive?ower structures,and positive structural inversion.Amer-ican Association of Petroleum Geologists Bulletin69,582e600.

Hardman,R.F.P.,1982.Chalk reservoirs in the North Sea.Bulletin of the Geological Society of Denmark30,119e137.

Hancock,P.L.,1985.Brittle microtectonics:principles and practice.Journal of Structural Geology7,437e457.

Hermann,W.U.,1986.Zum sedimenteintrag in das zentrale nordwesteurop?ische Oberkreidemeer.Geologisches Jahrbuch A97,3e139.

Hermansen,H.,Landa,G.H.,Sylte,J.E.,Thomas,L.K.,2000.Experiences after10years of water?ooding the Eko?sk?eld,Norway.Journal of Petroleum Science and Engineering26,11e18.

Hjuler,M.L.,Fabricius,I.L.,2009.Engineering properties of chalk related to diage-netic variations of Upper Cretaceous onshore and offshore chalk in the North Sea area.Journal of Petroleum Science and Engineering68,151e170.

Hobbs, D.W.,1967.The formation of tension joints in sedimentary rocks:an explanation.Geological Magazine104,550e556.

Hooker,J.N.,Gale,J.F.C.,Gomez,L.A.,Laubach,S.E.,Marrett,R.,Reed,R.M.,2009.

Aperture-size scaling variations in a low-strain opening-mode fracture set, Cozzette Sandstones,Colorado.Journal of Structural Geology31,707e718. Huang,Q.,Angelier,J.,1989.Fracture spacing and its relation to bed thickness.

Geological Magazine126,355e362.

Jaritz,W.,1973.Zur entstehung der salzstrukturen nordwestdeutschland.Geo-logisches Jahrbuck A10,1e77.

Johnson,H.A.,Bredeson,D.H.,1971.Structural development of some shallow salt domes in Luisiana Miocene productive belt.American Association of Petroleum Geologists Bulletin55,204e226.

Koestler,A.G.,Ehrmann,W.U.,1986.Fault patterns in the calcareous overburden of

a salt diapir:Laegerdorf,NW Germany.Neues Jahrbuch fur Geologie und

Palaeontologie9,555e569.

Koestler,A.G.,Ehrmann,W.U.,1987.Fractured chalk overburden of a salt diapir, Laegerdorf,NW Germany e exposed example of a possible hydrocarbon reservoir.In:Lerche,I.,O’Brien,J.J.(Eds.),Dynamical Geology of Salt and Related Structures.Academic Press,New York,pp.457e477.

Koestler,A.G.,Ehrmann,W.U.,1991.Description of brittle extensional features in chalk on the crest of a salt ridge(NW Germany).In:Roberts,A.M.,Yelding,G., Freeman,B.(Eds.),The Geometry of Normal Faults.Geological Society special publication,pp.113e123.56.

Koestler,A.G.,Reksten,K.,1994.3D characterization of the fracture network in

a deformed chalk reservoir analogue e the L?gerdorf case.SPE Paper28728,

573e583.

Labaume,P.,Sheppard,S.M.F.,Moretti,I.,2001.Fluid?ow in cataclastic thrust fault zones in sandstones,Sub-Andean Zone,southern Bolivia.Tectonophysics340, 141e172.

Laubach,S.E.,2003.Practical approaches to identifying sealed and open fractures.

American Association of Petroleum Geologists Bulletin87,561e579. Laubach,S.E.,Olson,J.,Gross,M.R.,2009.Mechanical and fracture stratigraphy.

American Association of Petroleum Geologists Bulletin93,1413e1426. Lezin,C.,Odonne,F.,Massonnat,G.J.,Escadeillas,G.,2009.Dependence of joint spacing on rock properties in carbonate strata.American Association of Petro-leum Geologists Bulletin93,271e290.

Mallon,A.J.,Swarbrick,R.E.,2002.A compaction trend for non-reservoir North Sea Chalk.Marine and Petroleum Geology19,527e539.

Mandal,N.,Deb,S.K.,Khan,D.,1994.Evidence for a non-linear relationship between fracture spacing and layer thickness.Journal of Structural Geology16, 1275e1281.

Mandl,G.,2000.Faulting in Brittle Rocks,an Introduction to the Mechanics of Tectonic Faults.Springer,Berlin.434p.

Morris,A.P.,Ferrill,D.A.,McGinnis,R.N.,2009.Mechanical stratigraphy and faulting in Cretaceous carbonates.American Association of Petroleum Geologists Bulletin93,1459e1470.

Narr,W.,Suppe,J.,1991.Joint spacing in sedimentary rocks.Journal of Structural Geology13,1037e1048.

Odling,N.E.,Gillespie,P.,Bourgine,B.,Castaing,C.,Chiles,J.-P.,Christensen,N.P., Fillion, E.,Genter, A.,Olsen, C.,Thrane,L.,Trice,R.,Aarseth, E.,Walsh,J.J., Watterson,J.,1999.Variations in fracture system geometry and their implica-tions for?uid?ow in fractured hydrocarbon reservoirs.Petroleum Geoscience 5,373e384.

Olson,J.E.,Laubach,S.E.,Lander,R.H.,2009.Natural fracture characterization in tight gas sandstones:integrating mechanics and diagenesis.American Associ-ation of Petroleum Geologists Bulletin93,1535e1549.

Ortega,O.J.,Marrett,R.A.,Laubach,S.E.,2009.A scale-independent approach to fracture intensity and average spacing measurement.American Association of Petroleum Geologists Bulletin90,193e208.

Price,N.J.,Cosgrove,J.W.,1990.Analysis of Geological Structures.Cambridge,New York,502pp.

Ramsay,J.G.,1967.Folding and Fracturing of Rocks.McGraw-Hill,New York.568p.Rawling,G.C.,Goodwin,L.B.,Wilson,J.L.,2001.Internal architecture,permeability structure,and hydrologic signi?cance of contrasting fault-zone types.Geology 29,43e46.

Reches,Z.,1983.Faulting of rocks in three-dimensional strain?elds.II.Theoretical analysis.Tectonophysics95,133e156.

Richard,J.,2008.Mass transfer and volume changes related to pressure solution-fracturing interactions in carbonate rocks:example of the Oligocene deformation of the Coniacian chalk from the Omey area(Paris Basin,France).

Journal of Structural Geology30,1300e1313.

Rijken,P.,Cooke,M.,2001.Role of shale thickness on vertical connectivity of fractures:application of crack-bridging theory to the Austin chalk,Texas.

Tectonophysics337,117e133.

Risnes,R.,2001.Deformation and yield in high porosity outcrop chalk.Physics and Chemistry of the Earth(A)26,53e57.

Rotevatn,A.,Torabi,A.,Fossen,H.,Braathen,A.,2008.Slipped deformation bands:

a new type of cataclastic deformation bands in Western Sinai,Suez rift,Egypt.

Journal of Structural Geology30,1317e1331.

Rowan,M.G.,Jackson,M.P.A.,Trudgill,B.D.,1999.Salt-related fault families and fault welds in the northern Gulf of Mexico.American Association of Petroleum Geologists Bulletin83,1454e1484.

Rowan,M.G.,Lawton,T.F.,Giles,K.A.,Ratliff,R.A.,2003.Near-salt deformation in La Popa basin,Mexico,and the northern Gulf of Mexico:a general model for passive diapirism.American Association of Petroleum Geologists Bulletin87,733e756. Salvini, F.,Billi, A.,Wise, D.U.,1999.Strike-slip fault-propagation cleavage in carbonate rocks:the Mattinata fault zone,Southern Apennines,Italy.Journal of Structural Geology21,1731e1749.

Salvini, F.,2011.Daisy3version 4.8.12The Structural Data Integrated System Analyzer.Available for free download at.http://host.uniroma3.it/progetti/fralab. Scholle,P.A.,1977.Chalk diagenesis and its relation to petroleum exploration:oil from chalk,a modern miracle?American Association of Petroleum Geologists Bulletin61,982e1009.

Schroeder,C.,Gaviglio,P.,Bergerat,F.,Vandycke,S.,Coulon,M.,2006.Faults and matrix deformations in chalk:contribution of porosity and sonic wave measurements.Bulletin of the Geological Society of France177,203e213. Schulz,M.G.,Ernst,G.,Ernst,H.,Schmid,F.,1984.Coniacian to Maastrictian stage boundaries in the standard section for the upper Cretaceous white chalk of NW Germany(Laegerdorf-Kronsmoor-Hemmoor):de?nitions and proposals.

Bulletin of the Geological Society of Denmark33,203e215.

Schlische,R.W.,Young,S.S.,Ackermann,R.V.,Gupta,A.,1996.Geometry and scaling relations of a population of very small rift-related normal faults.Geology23, 683e686.

Sheldon,H.A.,Micklethwaite,S.,2007.Damage and permeability around faults: implications for mineralization.Geology35,903e906.

Stewart,S.A.,2006.Implications of passive salt diapir kinematics for reservoir segmentation by radial and concentric faults.Marine and Petroleum Geology 23,843e853.

Storti,F.,Rossetti,F.,Laüfer,A.,Salvini,F.,2006.Consistent kinematic architecture in the damage zones of intraplate strike-slip fault systems in North Victoria Land, Antarctica and implications for fault zone evolution.Journal of Structural Geology28,50e63.

Swan, A.R.H.,Sandilands,M.,1995.Introduction to Geological Data Analysis.

Blackwell,Oxford.446p.

Sylvester,A.G.,1988.Strike-slip faults.Geological Society of America Bulletin100, 1666e1703.

Talesnick,M.,Shehadeh,S.,2007.The effect of water content on the mechanical response of a high-porosity chalk.International Journal of Rock Mechanics and Mining Sciences44,584e600.

Tavani,S.,Storti,F.,Fernandez,O.,Mu?oz,J.A.,Salvini,F.,2006.3-D deformation pattern analysis and evolution of the A?isclo anticline,southern Pyrenees.

Journal of Structural Geology28,695e712.

Tsang,Y.W.,1984.The effect of tortuosity on?uid?ow through a single fracture.

Water Resources Research20,1209e1215.

Underwood, C.A.,Cooke,M.L.,Simo,J.A.,Muldoon,M.A.,2003.Stratigraphic controls on vertical fracture patterns in Silurian dolomite,northeastern Wis-consis.American Association of Petroleum Geologists Bulletin87,121e142. Vendeville, B.,Jackson,M.P.A.,1992a.The rise of diapirs during thin-skinned extension.Marine and Petroleum Geology9,331e353.

Vendeville, B.,Jackson,M.P.A.,1992b.The fall of diapirs during thin-skinned extension.Marine and Petroleum Geology9,354e371.

Watts,N.L.,1983.Microfractures in chalks of Albuskjell Field,Norwegian sector, North Sea:possible origin and distribution.American Association of Petroleum Geologists Bulletin67,201e234.

Wise, D.U.,Funiciello,R.,Parotto,M.,Salvini, F.,1985.Topographic lineament swarms:clues to their origin from domains analysis of Italy.Geological Society of America Bulletin96,952e967.

Wu,H.,Pollard,D.D.,1995.An experimental study of the relationship between joint spacing and layer thickness.Journal of Structural Geology17,887e905. Ziegler,P.A.,1982.Triassic rifts and facies patterns in Western and Central Europe.

Geologische Rundschau71,747e772.

F.Storti et al./Marine and Petroleum Geology28(2011)1315e1332 1332

本文来源：https://www.bwwdw.com/article/olnq.html