Near-salt deformation in La Popa basin, Mexico, and the northern Gulf of Mexico A general model

AUTHO RS

Mark G.Rowan Rowan Consulting,Inc.,8508th Street,Boulder,Colorado 80302;mgrowan@https://www.360docs.net/doc/718012954.html,

Mark Rowan has degrees from Caltech,Berkeley,and the University of Colorado.He was an exploration geologist with Sohio,a consultant with Geo-Logic Systems and Alastair Beach Associates,and a research professor at the University of Colorado.He is now an independent consultant specializing in structural geology and salt tectonics and is the instructor for AAPG’s salt tectonics school.Timothy https://www.360docs.net/doc/718012954.html,wton Institute for Tectonic Studies,New Mexico State University,Las Cruces,New Mexico 88003

Tim Lawton received degrees from the University of California,Santa Cruz (1973),Stanford (1980),and the University of Arizona (1983).He worked as a development geologist and research sedimentologist for Sohio Petro-leum and is currently a professor in the Department of Geological Sciences at New Mexico State University.He specializes in clastic depositional systems,stratigraphy,and tectonic analysis.

Katherine A.Giles Institute for Tectonic Studies,New Mexico State University,Las Cruces,New Mexico 88003

Kate Giles graduated from the University of Wisconsin,Madison,the University of Iowa,and the University of Arizona.She worked at Exxon Production Research and is currently an associate professor at New Mexico State University and director of the Institute of Tectonic Studies.She specializes in carbonate depositional systems,sequence stratigraphy,and sedimentation as it relates to tectonics.Robert A.Ratliff Geo-Logic Systems,LLC,Boulder,Colorado

Bob Ratliff received his Ph.D.from the

University of Colorado.He worked for Geo-Logic Systems (GLS)and CogniSeis Develop-ment,where he was responsible for the

technical content and marketing of the GeoSec and GeoStrat software packages.He rejoined GLS and is currently director of Product Development and active in research on the geometry,kinematics,and interpretation of rock deformation.

Near-salt deformation in La Popa basin,Mexico,and the northern Gulf of Mexico:A general model for passive diapirism

Mark G.Rowan,Timothy https://www.360docs.net/doc/718012954.html,wton,Katherine A.Giles,and Robert A.Ratliff

ABSTRACT

Strata adjacent to exposed diapirs in La Popa basin,northeastern Mexico,comprise stacked halokinetic sequences consisting of un-conformity-bounded packages of thinned and rotated strata cut by radial faults.Deformation results from shallow drape folding over the flanks of the rising diapirs and not from deep drag folding in diapir-peripheral shear zones.Subsurface analogs from the Gulf of Mexico have diapir-flanking geometries ranging from similar,wide zones of upturned and thinned strata to undeformed,constant-thickness strata.Subhorizontal salt tongues display little subsalt deformation and thinning.

We propose a general model for passive diapirism and flank deformation that includes (1)gradually varying salt-flow rates,(2)superposed episodic sedimentation that results in changing bathy-metric relief,(3)rotation of near-surface strata as salt inflates rel-ative to the adjacent basin,(4)failure and erosion of strata in the steepening bathymetric halo,and (5)bedding-parallel slip surfaces that converge on unconformities and onlap surfaces.A primary factor influencing flank geometries is the width of the bathymetric high extending beyond the diapir edge.This is largely dependent on the thickness of the halokinetic sequence onlapping the diapir,which in turn is controlled mostly by the interplay between salt inflation/deflation rates and sedimentation rates.Other factors include the amount of concurrent shortening,which produces a wider but less intense zone of deformation,and the position on the scarp of salt breakout and extrusion.

Our model is important for exploration and production in diapir-flank and subsalt settings because of its implications for trap size and geometry,reservoir distribution,trap compartmentalization

Copyright #2003.The American Association of Petroleum Geologists.All rights reserved.

Manuscript received August 16,2002;provisional acceptance September 30,2002;revised manuscript received December 11,2002;final acceptance January 15,2003.DOI:10.1306/01150302012

AAPG Bulletin,v.87,no.5(May 2003),pp.733–756733

734Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of Mexico

and pressure seals,and hydrocarbon charge.It can help in explaining complex and enigmatic well data and in better assessing risk in areas of poor seismic imaging.

INTRODUCTION

Passive diapirism is the syndepositional growth of a salt body near the sea floor or land surface (Nelson,1989;Jackson and Talbot,1991;Vendeville and Jackson,1992a).It is commonly considered equivalent to downbuilding (Barton,1933),in which surrounding minibasins sink into the salt-source layer,thereby driving salt flow into the growing salt body.However,salt can grow passively in the absence of minibasin subsidence if the salt is inflated,for example,by shortening.In any case,the salt body grows as a vertical stock when salt-flow rates equal sediment-aggradation rates,or it ex-pands as a subhorizontal salt tongue if salt flow greatly outpaces sedimentation (Vendeville and Jackson,1991;McGuinness and Hossack,1993;Talbot,1993).

Diapirism may be initiated in several ways.For example,it may be triggered by extension and reactive diapirism (Vendeville and Jackson,1992a),it may start by the breaching of contractional detachment folds (Coward and Stewart,1995),or it may be driven by progradational loading (Ge et al.,1997).In each case,there is a brief episode of active diapirism (Vendeville and Jackson,1992a;Schultz-Ela et al.,1993),during which the salt body pierces a thin,weak overburden to reach the depositional surface,before passive diapirism takes over.Passive diapirism dominates the history of most diapiric salt bodies and ceases only when salt flow cannot keep up with sedimentation,usually when the source salt layer is depleted,the feeder closes off,or shortening ceases.

Diapirism is commonly accompanied by near-salt deformation in which adjacent or underlying strata are folded and faulted.Faulting may be perpendicular to the diapir boundary (radial),parallel to the diapir (concentric),or more complex (e.g.,Johnson and Bredeson,1971;Rowan et al.,1999;Davison et al.,2000b).Folding may range from negligible,with little thinning of flanking strata,to substantial,with vertical or even overturned beds,local unconformities,and significant stratal thinning (e.g.,Bornhauser,1969;Johnson and Bredeson,1971;Giles and Lawton,2002).

The area of folded strata around a diapir is typically called the ‘‘drag zone,’’because near-diapir folding is commonly attributed to shearing (or drag)of the rocks surrounding the diapir as the salt rises and the sediments sink (e.g.,Bornhauser,1969;Brown,1998;Alsop et al.,2000;Davison et al.,2000a).Such folding may be a product of either active piercement or passive rise.Experimental models using a variety of salt and sediment analogs,in which diapir-flanking folds have been generated by shear,have been used to support this interpretation (e.g.,Parker and McDowell,1955;Jackson et al.,1988;Davison et al.,1993;Podladchikov et al.,1993;Alsop,1996).

ACKNOWLEDGEMENTS

We thank the sponsors of the La Popa basin Joint Industry Consortium for funding:Amerada Hess,Anadarko,BP,ChevronTexaco,Conoco,Devon,ENI-Agip,Enterprise,ExxonMobil,Phil-lips,and Shell.Discussions with their staff as well as B.Vendeville,F.Vega-Vera,I.Davison,and D.Schultz-Ela were very helpful,and M.Jackson,D.Schultz-Ela,B.Trudgill,and K.Bowker suggested revisions that helped clarify our message.This research could not have been completed without the hard work of graduate students at New Mexico State University:J.Aschoff,S.Furgal,K.Graff,K.Hon,L.Hunnicutt,D.Mercer,D.Shelley,and A.Weislogel.F.Diegel (Shell)and R.Hobbs and L.Liro (Veritas)assisted with permission to show seismic data,and WesternGeco and Veritas kindly allowed the data to be shown.Finally,we thank Geo-Logic Systems,LLC for use of the LithoTect restoration software.

There are several fundamental problems with the drag-fold model.First,it assumes that the salt and adjacent strata have roughly equal strengths and simi-lar mechanical behavior.In fact,salt is a viscous ma-terial having negligible effective strength,whereas even shallowly buried sediments are plastic materials having mechanical strength and anisotropy.Second,drag fold-ing is incompatible with the observation that many di-apirs and subhorizontal salt tongues have little,if any, adjacent deformation or thinning.Third,the drag-fold model cannot explain beds that are folded beyond the dip of the diapir edge(e.g.,overturned beds on a ver-tical diapir).Moreover,it does not account for local unconformities and locally variable degrees of upturn.

Johnson and Bredeson(1971)recognized that dom-ing of a thin overburden above a passive diapir is a more plausible model than drag folding for observed stratal geometries.Other workers have speculated that passive diapirs grow by cycles of true passive diapirism,burial, and active piercement(Jackson et al.,1988;Schultz-Ela et al.,1993).However,it was only when exposed diapirs in La Popa basin in Mexico were studied that such a pro-cess was first documented(Lawton and Giles,1997;Giles and Lawton,2002).A model was proposed in which near-diapir folding is not a product of drag,but takes place at shallow levels as bathymetric relief increases and decreases because of the interplay between gradually varying salt-rise rates and high-frequency fluctuations in sedimentation rate(Rowan et al.,2000a,b).Finite-element modeling(Schultz-Ela,2003,this volume)has substantiated this model in that these simulations fail to produce significant folding by drag but do generate re-alistic fold geometries by near-surface rotation of thin, wedge-shaped flaps of overburden.

In this paper,we use field data from two exposed diapirs in La Popa basin and seismic data over analo-gous subsurface structures to derive a general model for passive diapirism and the associated deformation zone.We first build on the description and model for halokinetic sequences at El Papalote diapir(Giles and Lawton,2002),but focus on the structural geometry, kinematics,and evolution of folding.After comparing the deformation and stratigraphy at El Papalote and El Gordo diapirs,we use seismic images of currently growing salt bodies in the northern Gulf of Mexico to identify the factors that influence bathymetry above diapirs or allochthonous salt tongues.Ultimately,a complex relationship between the geometry of the salt body,salt inflation/deflation rates,sedimentation rate, and the resultant local bathymetric relief dictates whether there will be substantial folding and over-turning of beds,only moderate folding,or effectively no upturn and thinning of flanking or subsalt strata.

The significance of this work is its applicability to diapir-flank traps and subsalt truncation traps that con-tain large volumes of hydrocarbons in salt basins around the world.These traps are typically areas of poor seis-mic image quality because of steep dips or overlying salt. Better appreciation of possible structural geometries and deformation processes in these environments will help explorationists interpret data,understand reser-voir compartmentalization,and predict sand-thinning patterns adjacent to salt.

LA POPA BASIN

Regional Setting

La Popa basin is a small depocenter located in the foreland of the Sierra Madre Oriental foldbelt in north-eastern Mexico(Figure1).The basin contains several elliptical outcrops of gypsum representing eroded salt stocks and a25-km-long faultlike structure having inter-mittent remnant gypsum that marks an exposed subver-tical salt weld(Giles and Lawton,1999).The presence of subsurface salt in the basin is indicated by a Pemex well in the core of an adjacent detachment fold that spudded in gypsum but penetrated more than3000m of halite at depth(Lopez-Ramos,1982;Lawton et al.,2001).

The tectonostratigraphic history of La Popa basin is only summarized here;for further details and refer-ences,see Lawton et al.(2001)and Giles and Lawton (2002).The basin formed during Triassic–Jurassic rift-ing,probably as a pull-apart basin on a strike-slip mar-gin associated with the separation of North and South America and the opening of the Gulf of Mexico.Post-rift subsidence resulted in widespread evaporite dep-osition during the Oxfordian(and possibly earlier)in La Popa basin and adjacent areas.This was followed by a period of shallow-to deep-water carbonate deposition from the Late Jurassic into the Late Cretaceous.The ba-sin is dominated by a siliciclastic successsion of Upper Cretaceous to middle Eocene shales,siltstones,and sand-stones deposited in a foreland basin that developed east and north of the Sierra Madre orogenic belt.Depositional environments ranged from prodeltaic to fluvial.Lentic-ular carbonate strata(‘‘lentils’’of McBride et al.,1974) are interbedded with deeper-water facies of the silici-clastic succession only adjacent to the diapirs and weld.

Observed deformation in La Popa basin is a result of combined salt withdrawal/diapirism and regional

Rowan et al.735

contraction (Figures 1,2).Diapirism as old as the Ap-tian can be documented from surface exposures (Law-ton et al.,2001),although it likely started even earlier,shortly after deposition of the evaporites,as suggested by latest Jurassic cooling ages for metaigneous blocks carried up in the diapirs (Garrison and McMillan,1999).Shortening occurred during the latest Creta-ceous to Eocene Hidalgoan orogeny (Guzma ′n and de Cserna,1963),while diapirism was ongoing,and was accommodated primarily by salt-cored detachment folding and lateral squeezing of the salt body now represented by the weld (Rowan et al.,2001).

El Papalote Diapir

Geometry

El Papalote diapir is located on a northeast-dipping fold limb between the El Gordo anticline and the syncline adjacent to the weld (Figures 2,3).A small flexure on

736

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of

Mexico

Figure https://www.360docs.net/doc/718012954.html,ndsat image of La Popa basin in northeastern Mexico (inset).El Papalote and El Gordo diapirs and a 25-km-long vertical weld are in blue,and the axial traces of plunging,salt-cored detachment folds of the Hidalgoan orogeny are in red.

the limb runs through the diapir,having dips first flattening from 30–35j to about 20j ,then steepening to about 45j before returning to regional dip values (Figure 2).This flexure accommodates 200m of short-ening 2km away from the diapir.

A concentric zone of folded strata is present around El Papalote diapir some 500–1000m in width (Figure 3),in which strikes gradually approach paral-lelism with the edge of the stock and dips increase to vertical or overturned.The geometry is best seen be-neath the flaring southeastern flank of the diapir (Fig-ure 4),where there is a series of unconformity-bound halokinetic sequences (Giles and Lawton,2002)com-prising both siliciclastic strata and carbonate lentils adjacent to the diapir.The oldest exposed sequence consists of a wedge of overturned strata (including len-til 1)that dips moderately westward.Dips shallow to subhorizontal (but overturned)about 200m from the diapir,where the sequence is truncated by a vertical unconformity at the base of lentil 2.This unconformity bounds the next higher sequence,which has a similar overall geometry and is likewise truncated beneath the stratigraphically higher unconformity,a pattern repeat-ed through younger sequences.The angular unconform-ities grade laterally to corrrelative conformities within about 250m of the diapir.

There is a systematic stratigraphy in the haloki-netic sequences (Giles and Lawton,2002).Each package has a carbonate lentil above the lower unconformity.The base of the lentil is a debris flow with dominant

carbonate clasts and subordinate metaigneous clasts derived from blocks entrained in the diapir (Garrison and McMillan,1999).The main body of the lentil grades from thick,proximal facies adjacent to the diapir to thin,more distal calciturbidites encased in black shales.Overlying siliciclastic strata onlap the lentil and thin toward the diapir,near which the entire sequence is truncated by the upper unconformity,which forms the base of the overlying sequence.

The exposed edge of the diapir parallels the base of overturned lentil 1,where it dips about 35j toward the diapir (Figure 4).Although gypsum is not present farther up the hill because of erosion,we infer that the original diapir-sediment interface continued along the base of lentil 1to its intersection with lentil 2.Other-wise,there would be a local pod of older strata,and the diapir edge would have cut back downsection at a higher structural level,an unlikely scenario.If our interpretation is correct,the diapir edge had a local cusp at the unconformity,where it changes abruptly from subhorizontal at the base of lentil 1to vertical at the base of lentil 2.We infer that a similar cusp was present at the contact of each sequence pair,with the cusp’s shape dictated by the angle of truncation at the unconformity.

Observed small-scale deformation in the sedimen-tary section surrounding the diapir is relatively in-significant.Shales contain a steep,northwest-striking cleavage that is axial-planar to the regional folds.Sand-stones and carbonates have minor fractures and veins,

Rowan et al.

737

Figure 2.Cross sections in La Popa basin showing (a)regional structures,including El Gordo anticline,the teepee structure of the vertical weld,and the intervening syncline,and (b)local structures,including El Gordo diapir,along El Gordo anticline,and El Papalote diapir,along a minor flexure on the northeast-dipping fold limb.Short red lines are measured bedding attitudes.Strata include Kac =carbonate lentils equivalent to the Aptian Cupido and Albian Aurora formations;Ki =basinal carbonates of the Cenomanian to Santonian Indidura Formation;Kp =Campanian Parras shale;Km =Maastrichtian Muerto sandstone;Kpsl,Kpml,Kpsm,TKd,Tpmu,and Tpss =Lower Siltstone,Lower Mudstone,Middle Siltstone,Delgado Sandstone,Upper Mudstone,and Upper Sandstone members,respectively,of the Maastrichtian to Paleocene Potrerillos Formation;Ta,Tv,and Tc =Eocene Adjuntas,Viento,and Carroza formations,respectively;Klg,Kmg,Ksj,1,and 5/6=carbonate lentils in mudstone members of the Potrerillos Formation.Section locations are indicated on Figures 1and 3.

most of which fit into three categories:(1)a set with northeast strikes and steep dips that represents axial-parallel extension along the regional folds;(2)a set perpendicular to local folded beds,but with similar strikes,that formed because of outer-arc extension;(3)a set oriented at high angles to the diapir boundary that accommodated tangential extension around the diapir (Figure 5a).The latter deformation is also expressed by radial,steep faults with as much as 20m of offset that extend as much as 1km away from the diapir (Figure 3).Displacement on the radial faults is greatest near the diapir and decreases outward to the fault tips.In addi-tion,radial ‘‘dikes’’of gypsum cut lentil 1adjacent to the diapir (Figure 5a).

The only shear zones or breccias oriented parallel to the diapir edge are present in the outer 100m of the diapir itself.No diapir-parallel high-strain zones are present in adjacent rocks.Instead,shear deformation in diapir-flanking strata is concentrated along the uncon-formities.These are brittle shear zones as much as 20cm wide consisting of brecciated subunconformity li-thologies in a matrix of fine-grained crushed rock and

738

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of

Mexico

Figure 3.Geologic map of the two diapirs.Red lines indicate the axial traces of folds:El Gordo anticline,which passes through El Gordo diapir,a minor flexure expressed as an anticline-syncline pair at El Papalote diapir,and an orthogonal withdrawal syncline to the southwest of El Gordo diapir.Green dashed lines indicate the limits of the deformation zones around the diapirs,where bedding is rotated to steeper dips.The amount of rotation from horizontal (in degrees)is indicated by numbers in red (90is vertical,180is horizontal but upside-down).Siliciclastic strata are in pastel colors (see Figure 2)and carbonate lentils are in brighter shades (Kpgl =Lower Gordo lentil;Kpgm =Middle Gordo lentil;Tpgu =Upper Gordo lentil).

Rowan et al.

739

Figure 4.Cross section of the eastern flank of El Papalote diapir (location shown on Figure 3).Siliciclastics that thin toward the diapir are in pastel colors (see Figure 2),and carbonate lentils (1to 5)that pinch out away from the diapir are in brighter shades.The salt flares to the east and is flanked by stacked halokinetic sequences bounded by unconformities at the bases of the lentils.Unit boundaries are solid where certain,dashed where approxi-mate,and dotted where inferred.See text for

description.

Figure 5.(a)Photograph taken from just inside El Papalote diapir,looking out at the vertical base of lentil 1,which is cut by a swarm of radial veins and a gypsum dike that record concentric extension during drape folding of the overburden.(b)Photograph of the unconformity at the vertical base of lentil 2,which truncates horizontal,overturned sandstones of the Potrerillos Formation;the unconformity is a 10–20-cm-wide brittle shear zone with sense-of-shear indicators showing older beds moved away from the diapir.

secondary carbonate cement(Figure5b).Sense-of-shear indicators show that the underlying,truncated beds moved away from the diapir.

Evolution

To understand the growth of El Papalote diapir and the progressive development and deformation of halo-kinetic sequences,we apply2-D structural restoration to the cross section of Figure4.Numerous algorithms are available and commonly applied in restoration(e.g., Davison,1986;White et al.,1986;Williams and Vann, 1987;Rowan and Kligfield,1989).Simple flexural-slip restoration maintains the length of the horizon being restored and imposes curvilinear sliding surfaces parallel to this horizon.Simple vertical/oblique slip applies lin-ear sliding surfaces at some chosen orientation,common-ly parallel to antithetic faults or axial surfaces of restored folds.Slip-line restoration invokes curvilinear slip parallel to a fault or,in this case,the edge of the diapir,thereby modeling a halo of simple shear around the diapir.

In the case of diapirs like El Papalote,having over-turned beds,convergent and onlapping strata,and angu-lar unconformities,each of these restoration methods requires slip surfaces locally oblique to bedding.Depend-ing on the algorithm and the local geometry,slip surfaces range from bedding-parallel to bedding-perpendicular. Yet there is no field evidence for deformation zones cutting through competent beds or for the consequent dip-dependent thinning.The implication is that the dom-inant deformation mechanism is slip along bedding, which preserves bed lengths and thicknesses.Conse-quently,we employ a new algorithm termed complex flexural slip,which models slip along nonparallel bed-ding surfaces(Ratliff et al.,1999).The lengths and thicknesses of individual beds are maintained,although slip between beds results in thinning or thickening of larger stratal packages.Slip surfaces converge at onlap surfaces and unconformities,consistent with the signifi-cant slip seen on unconformities at El Papalote(Figure 5b).Nevertheless,complex flexural slip still simplifies the true deformation,in that incompetent beds probably experienced complicated small-scale shear and flow that resulted in bed thinning.

The restoration of flanking strata at El Papalote is shown in Figure6.However,it is easier to understand the progressive deformation by examining the restora-tion of modeled halokinetic sequences(Figure7).In both figures,the restoration template(upper surface) is horizontal because there are no constraints on the pa-leobathymetry of each horizon.Variable bathymetric relief is an important component of the deformation (Giles and Lawton,2002)that will be addressed below in the general model for passive diapirism.

Both restorations(Figures6,7)show that de-formation is not a product of shear around the diapir, i.e.,the folds are not drag folds.Instead,they are best termed drape folds because they form by rotation of wedge-shaped flaps of overburden(halokinetic se-quences)as the diapir inflates and flanking strata sink. Synrotational sedimentation produces the wedge shape of each halokinetic sequence.The deformation occurs in the shallow subsurface and folding ceases with increas-ing burial(Figure6),although there may be continued relative salt rise.Similar results in models by Schultz-Ela(2003,this volume)show that this idea is not only kinematically but also dynamically feasible.

During bed rotation,strata directly beneath a diapir-flank unconformity translate along the uncon-formity,thereby increasing the original erosional angle of truncation and creating a cusp along the diapir edge (Figures7,8).These cusps are not primary salt wings caused by lateral extrusion of salt(Yeilding and Travis, 1997),but instead are secondary by-products of the deformation,with the mobile salt filling in the space created by the increasingly angular unconformities.

The drape-fold model has an important conse-quence for three-dimensional deformation around di-apirs with elliptical plan shapes or any local convex-outward geometry.As the overburden rotates,it must experience tangential extension and split much as the petals of a flower separate when they open.This pro-duces the radial faults,veins,and gypsum dikes seen adjacent to El Papalote diapir.The intensity of this deformation,as well as slip along the unconformities, decreases rapidly away from the diapir.

Comparison of El Gordo and El Papalote Diapirs

El Gordo diapir has many of the same stratigraphic and structural features seen at El Papalote diapir.It has a concentric deformation zone having halokinetic sequences,beds that are rotated to vertical and be-yond,local unconformities,and radial faults(Figure 3).It also has exposed cusps at the gypsum-sediment interface where unconformities intersect the diapir (Figure9).However,there are key differences that we examine in the following sections.

Structure

Whereas El Papalote diapir is on the limb of a regional Hidalgoan anticline,El Gordo diapir is located along the hinge of the fold(Figures2,3).This anticline

740Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of Mexico

accommodates about2km of shortening just1.7km away from El Gordo,whereas the small flexure pass-ing through El Papalote records one-tenth as much shortening.As both diapirs predated the folding,one might expect that both would have been squeezed similarly,because the presence of weak salt localizes contractional strain and controls fold development (e.g.,Vendeville and Nilsen,1995;Rowan et al.,2000c). However,the two diapirs and the weld are too close together for all three to form major folds because the thickness of folded strata(>6400m,Lawton et al., 2001)results in a fold wavelength of at least10km (Figure1).Thus,major anticlines developed along El Gordo and the weld,whereas only a minor flexure de-veloped at El Papalote.

Deformation at El Gordo diapir covers a broader area but is not as severe as that at El Papalote diapir. The deformed zone around El Gordo diapir is up to 2km wide,more than the800m at El Papalote dia-pir(Figure3).Correspondingly,the unconformities bounding halokinetic sequences extend farther away from El Gordo diapir(as much as800m)than from El Papalote diapir( 200m).Despite this broader zone of deformation at El Gordo diapir,the degree of folding is less than it is at El Papalote diapir,with beds only rotating a maximum of34j beyond vertical as opposed to being completely overturned at El Papalote diapir(Figure3).Because of this,the max-imum angular discordance observed on El Gordo un-conformities is about75j,compared to90j on those at El Papalote diapir.

Stratigraphy

All siliciclastic units thin as they approach the diapirs. Measured sections of the Delgado Sandstone docu-ment that thinning takes place within400m of El Papa-lote but within1200m of El Gordo(J.Aschoff,2002, personal communication).Carbonate lentils also show

Rowan et al.741 Figure 6.Sequential restorations of the cross section in Figure4constructed in LithoTect using complex flexural slip (slip along nonparallel beds;Ratliff et al.,1999),a horizontal restoration template,and a pin line at the right edge of the section.The first restoration to the top of Tpmu(a)removes the effects of Hidalgoan shortening;although this is a strike line with respect to Hidalgoan structures,a sharp lateral strain gradient caused by squeezing of the weak diapir is manifested as a steep local fold plunge along strike to the southeast.Thus, the Tpmu restoration(a)shows the deformation associated with passive diapirism,which is sequentially restored in(b)to (d).Note the increasing angular discordance of the unconform-ities during progressive rotation and the creation of cusps at the salt edge because of slip along the unconformities. Bathymetric relief is not reconstructed,but would have an important effect;for example,the Kpsm monocline in(b)and (c)was probably a simple,progressive upturn of beds.

key differences between the two diapirs (Figure 10).As compared to El Papalote diapir,those at El Gordo diapir are generally thicker and extend farther away from the diapir,and they typically have shallower water facies (Hunnicutt,1998;Mercer,2002).More-over,there are variations around El Gordo diapir itself,in that lentils reach greater distances from the diapir on the northeast side.

In summary,the stratigraphic trends are consist-ent with the structural data.Both the lentils and the siliciclastics indicate broader and higher bathymetric relief at El Gordo diapir.El Gordo diapir is located on a major anticline and has a wider,but less severe,surrounding zone of deformation.

GENERAL MODEL FOR PASSIVE DIAPIRISM Near-diapir folding associated with passive diapirism is found in salt basins throughout the world,although

742

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of Mexico

Figure 7.Model halokinetic sequences HS1and HS2(a)and restoration to the top of HS1(b).Restoration was carried out in LithoTect using complex flexural slip,a vertical pin line outside the defor-mation halo,and a horizontal restoration template.Slip along convergent bedding planes in HS1is transferred to the over-lying unconformity as the diapir rises and the stratal wedge is folded and onlapped by HS2.The result is that the tip of the wedge (x)draws back along the uncon-formity to point y,thereby creating a cusp in the edge of

salt.

Figure 8.Sequential stages in the model restoration of Figure 7overlapped and slightly offset to show the progressive deformation as the diapir rises and the flanking minibasin subsides.Gradual onlap and rotation of HS2during drape folding is accompanied by further folding of HS1and slip along the intervening unconformity,resulting in an increasing angle of

truncation.

the degree of deformation varies(e.g.,the northern Gulf of Mexico[Johnson and Bredeson,1971],onshore Nova Scotia[Alsop et al.,2000],the southern North Sea [Davison et al.,2000a],onshore Yemen[Davison et al., 1996a],southern and central Iran[Kent,1979;Jackson et al.,1990],and the Adelaide geosyncline of Australia [Dyson,1996]).A particularly well-imaged seismic ex-ample of stacked halokinetic sequences from the Pre-caspian basin of Kazakhstan is illustrated in Figure11.

A single comprehensive model can explain the ob-served variations in salt-flank deformation.Our mod-el for the growth and emplacement of passive diapirs and allochthonous salt tongues is based on several key elements.

Key Elements of Model

Salt-Rise Rate

We assume that the rate of diapir rise relative to as-sociated subsidence of the surrounding strata varies only gradually through time.It is not incremental in-flux of sediment that drives salt movement;instead,the primary factor is the differential load between preexisting sediment in the subsiding minibasin and the thin diapir overburden(Vendeville et al.,1993;Prather et al.,1998;Rowan and Weimer,1998).The rate of diapir rise driven by this differential load steadily in-creases as the minibasin grows in thickness,slows be-cause of viscous drag forces in the thinning source salt layer,and eventually ceases when the layer is com-pletely evacuated and a weld forms(Vendeville et al., 1993).Any pulse of increased sedimentation,such as during a lowstand on the slope,increases the differ-ential load,but this is a second-order effect super-imposed on the primary drive of the existing load.

Diapir-rise rates are also influenced by lateral forces acting on the diapir walls.Extension and con-traction both preferentially affect diapirs because they are the weakest parts of the rock volume.Extension has the effect of slowing the rise of passive diapirs because the diapir widens with time;in some cases, the diapir may even collapse instead of grow(Ven-deville and Jackson,1992b).Conversely,shortening narrows the diapir and accelerates upward salt flow (Vendeville and Nilsen,1995).

Rowan et al.743

Figure9.Geologic map of the north-

western corner of El Gordo diapir showing

a prominent cusp where the halokinetic

unconformity at the base of Middle Gordo

lentil(Kpgm)intersects the diapir.The

map is oriented(see North arrow)so

that it shows a down-plunge view and

thus approximates a cross section.

Sedimentation Rate

In contrast to the gradual changes in subsidence/dia-pirism rates,sedimentation rates show high-frequency variations.These are largely caused by eustatic changes in sea level but also may be caused by lateral shifting of deltas and/or turbidite channel systems and fans.Thus,sedimentation rates sometimes exceed sub-sidence rates and sometimes are slower.The result is that bathymetric relief around the diapir fluctuates:rapid sedimentation decreases the relief as the mini-basin is filled,and slow sedimentation increases the re-lief as the basin subsides and the diapir inflates (Ro-wan and Weimer,1998;Prather et al.,1998;Giles and Lawton,2002).Origin of Folding and Faulting

Salt bodies that grow passively are typically covered by a thin condensed section of finer grained material depos-ited on the bathymetric high (e.g.,Worrall and Snel-son,1989;Fletcher et al.,1995;Moore et al.,1995).As bathymetric relief increases,this cover is gradually draped over the edge of the rising diapir,forming a scarp at the sea floor.It is this draping,not viscous drag,that rotates beds and creates diapir-flank folds.The three-dimensional effects of draping over curved salt boundaries also lead to radial faults adjacent to the diapir.For example,radial faults are most abundant at the curved terminations of elongate salt diapirs in the southern North Sea (Davison et al.,2000b).

744

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of Mexico

Figure 10.Facies,thicknesses,and lengths of carbonate lentils at El Gordo and El Papalote dia-pirs (see Figures 2–4for lentil nomenclature).Range of widths at El Gordo diapir are for the southwestern (shorter)and northeastern (longer)flanks.El Gordo lentils are generally thick-er,wider,and shallower than those at El Papalote diapir,but there are more Paleocene lentils at El Papalote

diapir.

Deformation Mechanism

Folding is accommodated primarily by slip along con-vergent bedding surfaces,slip that is concentrated at angular unconformities and onlap surfaces (complex flexural slip;Figures 7,8).Although near-surface stra-ta are relatively unconsolidated,bedding provides an anisotropy that controls deformation.This is a sim-plification,of course,that neglects a likely component of bed lengthening and thinning,especially in con-densed muds above the diapir.Bed-parallel slip is docu-mented around several diapirs in the southern North Sea (Davison et al.,2000a),although it is unclear whether this is related to local diapiric rise or regional contractional deformation.

Origin of Unconformities

Our model stipulates that when the bathymetric scarp becomes sufficiently steep,it fails gravitationally,as depicted for salt sheets by McGuinness and Hossack (1993).This slumped material may form debris flows,as at El Papalote diapir (Giles and Lawton,2002)and in the southern North Sea (Davison et al.,2000a),or it may be represented by the structurally complex ‘‘gumbo zones’’beneath some salt tongues (Fletcher

et al.,1995).In the slope environment,the slumped fine-grained material of the condensed section may be winnowed away by currents or entrained in later tur-bidity flows and thus be undetectable.Turbidity flows may even contribute to the erosion as they increase in velocity while passing over and down the steep scarp.In any case,the erosion of the draped cover creates a local angular unconformity between the rotated,trun-cated beds beneath the failure surface and subsequent strata that onlap the surface (Giles and Lawton,2002).These unconformities are local to the bathymetric scarp and pass quickly into correlative conformities away from the diapir.Different Model Scenarios

We have already stated that observed near-diapir geom-etries vary widely,from severe overturn and thinning of beds to essentially undeformed,constant-thickness strata.We present below five different scenarios,all based on the common key elements of our model,which produce very different results.The scenarios vary in the relationship between salt-body geometry and the associated bathymetric expression.

Rowan et al.

745

Figure 11.Uninterpreted (a)and interpreted (b)time-migrated 2-D seismic profile from the Precaspian basin in Kazakhstan showing stacked halokinetic sequences bounded by angular unconformities (u)that intersect the diapir at cusps (c)and pass into correlative conformities away from the diapir.The basal unit of each sequence is parallel to the unconformity and then the diapir edge and is onlapped by younger strata,equivalent to the lentils in La Popa basin.Scale not available.

Diapir with Wide Halo of Bathymetric Relief

Many diapirs have areas of bathymetric relief that ex-tend for some distance beyond the diapir edges (Fig-ure 12).When modeled using complex flexural slip,this scenario creates a wide zone of deformation with thick halokinetic sequences,significant onlap and thin-ning,overturned beds,and prominent,highly angular unconformities (Figure 13).Discordance at unconform-ities increases with time as underlying beds translate away from the diapir during folding to create cusps in the salt edge.Bed rotation gradually wanes with depth,so that the stratal geometry remains unchanged after sufficient burial.

Diapir with Narrow Halo of Bathymetric Relief

Other diapirs have narrower zones of bathymetric re-lief that extend only a short distance beyond the diapir.A model of this scenario yields very different diapir-flank geometries than those produced in the previous example.The halokinetic sequences are thinner and more frequent,the width of the deformation halo is nar-rower,the amount of upturn and thinning is markedly less,and the unconformities are subtle and less angular

(Figure 14).An example is seen on the right side of the diapir in Figure 15b.

Diapir with No Halo of Bathymetric Relief

In extreme cases,there is little,if any,thinning and deformation of flanking strata (Figure 15a).This can be seen in a modern example,where the width of the bathymetric high is approximately the same as that of the diapir (left side of Figure 15b).Consequently,there is almost no adjacent deformation,and no radial faults are expected because there is no flap of strata being folded.

Shortened Diapir

Both diapirs in Figure 16have thrust faults and as-sociated contractional folds extending laterally away from them in map view (Rowan,1995).The diapirs have been subjected to shortening at the basinward margin of an allochthonous salt canopy in response to gravity spreading and updip extension.Bathymetric relief is higher and broader than in diapirs with no shortening:in these examples,there is as much as 400m of relief extending as much as 3km from the diapir

746

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of

Mexico

Figure 12.Time-migrated 3-D seismic profile from the northern Gulf of Mexico showing a diapir with a bathymetric halo extending approximately 650m away from the diapir edges.The underlying stratal wedge is strongly rotated,with beds approaching vertical (scale is roughly 1:1).This is the same diapir as in figure 11of Rowan (1995).Data courtesy of WesternGeco.

edge.In contrast,the diapir in Figure12has a bathy-metric halo200m high and650m wide,and that in Figure15b has only about100m of relief immediately above the diapir.

The subsurface deformation is correspondingly dif-ferent,in that there is a much broader zone of thinning and upturn of beds around the salt(Figure16).Fur-thermore,the erosional unconformity on the left flank of the right-hand diapir becomes conformable some1.5 km away from the diapir.In contrast,any unconformi-ties around the diapir in Figure12would have a max-imum extent of roughly500m because the zone of drape folding is narrower.

We have not modeled shortened diapirs because the flanking stratal geometries are largely unconstrained. Folding is a response not only to passive diapirism,but also to regional contraction.The fold geometry is thus influenced by the depth to the detachment,the amount of shortening,the presence and dip of any adjacent reverse faults,and the diapir shape.

Subhorizontal Salt Tongues/Canopies

The previous scenarios have all involved roughly ver-tical salt diapirs.The surrounding deformation is ex-pected to be similar in the case of diapirs that flare outward.In other words,the degree of upturn and thinning in adjacent strata will depend largely on the height and areal extent of the bathymetric relief,which in turn are dependent upon the interplay between salt-flow rates and sedimentation rates.

However,at some point,the similarity between vertical growth and more lateral extrusion of salt fails. Subhorizontal salt bodies typically have subsalt strata that are truncated by the base salt with little thinning or deformation(Figure17).We do not refer here to the complex geometries sometimes found in so-called ‘‘gumbo zones,’’where the section just below salt may have overturned beds,repeated sections,and generally complex deformation and biostratigraphy(Harrison and Patton,1995).These zones probably represent material slumped off the bathymetric scarp above the toe of salt and subsequently overriden as the salt extruded basinward(McGuinness and Hossack,1993; Fletcher et al.,1995).Instead,we refer to the larger scale geometry of subsalt strata,which are typically relatively undisturbed.

The explanation lies again in the relationship be-tween the salt geometry and the bathymetric relief. In the case of a vertical diapir,the drape folding caused by diapiric growth and basin subsidence occurs adja-cent to the rising diapir(Figure13).The breakout of salt occurs relatively high on the bathymetric scarp, and folded strata are preserved around the diapir.In the case of subhorizontal salt,the tip of salt is near the base of the scarp(Figure17a,b),and the salt breaks through at this position.Thus,the rotated strata mak-ing up the scarp are those that are eroded away,so that only undeformed strata are preserved beneath the salt after it extrudes basinward.A contributing factor is that subhorizontal salt tongues/canopies generally serve as detachments for gravitational failure and basinward translation of the overburden.Thus,the tip of the salt commonly lies on a thrust fault that cuts through to the base of the bathymetric scarp.As the salt and its over-burden advance basinward,only the undeformed foot-wall is preserved below the salt.

Discussion

We have shown that the relationship between salt geometry and associated bathymetric relief dictates the nature and degree of near-diapir deformation dur-ing passive diapirism.Because the rotation of beds in the bathymetric scarp produces the folding,the width of the scarp beyond the point of salt breakthrough determines the width of the deformation zone pre-served in the subsurface.

However,what controls the width of the bathy-metric halo?In the case of a vertical or flaring diapir without shortening,it is the thickness of the over-burden wedge(halokinetic sequence)and its mechan-ical strength,as first suggested by Parker and McDowell (1955).A thick wedge of strata produces a wider zone of deformation than a thin wedge folded to the same degree of rotation.Thus,relatively thick overburden produces a broad area of bathymetric relief(Figure12), thinner overburden generates a narrow halo of relief, and an absence of overburden results in bathymetric relief that is restricted to the area directly above the diapir(Figure15b).

In turn,the interplay between salt-flow rates and variations in sedimentation rates during episodic deposi-tion controls the thickness of the halokinetic sequence. When sedimentation rates exceed salt inflation/deflation rates(for example,during lowstands on the slope),bathy-metric relief decreases as a thickening wedge of strata onlaps and covers the diapir(Figure13).When sedimen-tation slows(during transgressive and highstand peri-ods),bathymetric relief builds again to the point of failure,and the result is a broad deformation halo.In contrast,if maximum sedimentation rates are rela-tively slow,onlap of the diapir will be minor,and the

Rowan et al.747

Figure13.Kinematic forward

model of halokinetic sequence

deposition and deformation

during drape folding over the

flank of a vertical diapir having

a broad bathymetric halo and

overall rapid sedimentation.

Each stage(a–i)represents an

equal time interval and amount

of relative inflation of the salt or

subsidence of the basin.Bathy-

metric relief during each of the

two cycles first decreases and

then increases as sedimentation

rates are rapid and then slow.

The top of the diapir and

flanking beds are removed

when the steepness of the scarp

reaches an arbitrary angle(de-

noted by‘‘failure’’),thereby

producing the bounding uncon-

formities of the two halokinetic

sequences.The angularity of the

unconformity increases from(e)

to(g)and creates a cusp in the

diapir edge.The modeling was

carried out in LithoTect using

complex flexural slip(slip along

nonparallel beds)and a pin line

at the right edge.Bed lengths

are maintained except over the

top of the diapir,where there is

lengthening.Not to scale.

748Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of Mexico

overburden wedge will be thin,leading to a narrower zone of deformation(Figure14).

This explanation is compatible with our field observations.The thick,broad halokinetic sequence between lentils1and2at El Papalote diapir formed during deposition of the Middle Siltstone Member of the Potrerillos Formation,whereas the thinner,nar-rower halokinetic sequences between lentils2and6 formed during slower deposition of the black shales of the Upper Mudstone Member(Figures4,6).Recent finite-element modeling is also in agreement:the greatest amount of diapir-flank folding is generated when there is episodic rapid deposition of thick units (Schultz-Ela,2003,this volume).

The proposed link between deformation-zone width and the interplay between diapir rise and sedi-mentation rates apparently fails in the case of diapirs having minimal,if any,flank folding and thinning(Fig-ure15).This geometry typically occurs on the upper slope when both sedimentation and salt-rise rates are very rapid.For example,the average aggradation rate around the diapir in Figure15over the last half-million years was about0.4cm/yr;lowstand sedimentation rates would have been significantly higher.In our model,this scenario should lead to wide zones of bathymetric relief and folding.We offer two possible explanations for the apparent lack of deformation. First,it may be that absolute rates,not just relative rates,of diapir rise and sedimentation are important.If both are very rapid,the high strain rates may cause the diapir edge to behave like a brittle fault,so that there is little folding(just as salt can fault instead of flow under high strain rates,e.g.,Davison et al.,1996b).Alter-natively,bathymetric relief built during a highstand (slow sedimentation rate)may be stripped off by high-velocity,delta-front turbidity currents early in the subsequent lowstand.The lack of preserved bathy-metric relief would result in little diapir-flanking deformation.

We have suggested that the width of the deforma-tion halo depends in part on the thickness of the folded wedge.This thickness also influences the degree of folding,because for a given bed-convergence angle in a wedge,the bottom bed of a thick wedge has a steeper dip adjacent to the salt than the lowest bed of a thinner wedge having the same width.In the former case,there-fore,scarp failure,erosion,and subsequent onlap pro-duces more angular discordance at the unconformity. Further deformation as the overlying halokinetic se-quence is deposited and rotated increases the angular discordance as subjacent beds translate along the un-conformity.

Rowan et al.749

Figure14.Kinematic

forward model similar

to that in Figure13,but

having a narrower ba-

thymetric halo and slower

overall sedimentation

rates.The result is that

bathymetric relief con-

tinuously builds up to

the point of failure,pro-

ducing thinner halokinetic

sequences,less upturn

of beds,only subtle un-

conformities,and no

cusps.The time interval

reflected by the red beds

is double that of the yel-

low beds.Modeling car-

ried out in LithoTect;not

to scale.

Other factors affect the degree to which folding develops.One is the amount of time between deposition and erosion.For a given rate of rotation (which depends on the inflation/subsidence rate),a scarp that fails after a relatively long time will have steeper dips adjacent to the diapir.Beds that are truncated sooner,either through

750

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of

Mexico

Figure 15.(a)Well-log cross section through the overhanging edge of Cote Blanche diapir,onshore Louisiana,showing essentially no thinning or upturn of flanking strata (from Johnson and Bredeson,1971,reprinted by permission of the AAPG whose permission is required for further use).(b)Time-migrated 3-D seismic profile from the northern Gulf of Mexico showing a diapir having no bathymetric halo and correspondingly little thinning and deformation of flanking strata on the left side,and a narrow bathymetric halo and corresponding deformation zone on the right side (this is the same diapir as in figure 12of Rowan,1995;data courtesy of WesternGeco).

gravitational failure or erosion by loop currents or tur-bidity currents,will have less rotation.This may explain the difference between the sharp folding and 90j unconformities of lentils 2and 3at El Papalote diapir (Figures 4,6)and the more subtle folding and truncation in the modeled thin wedge (Figure 14).

Another factor is suggested by finite-element models of Schultz-Ela (2003,this volume),in which the very top of the diapir tends to spread because the thin overburden and free face of the bathymetric scarp offer little lateral resistance.This enhances rotation of the onlapping wedge and,although it cannot be doc-umented,may help explain the large amount of over-turn at El Papalote diapir.

The discussion to this point has focused on vertical or flaring diapirs growing purely passively.Any coeval shortening has a significant impact.If the shortening is relatively minor and thus confined to the diapir itself,it simply increases the rate of upward salt flow as the diapir is squeezed.This may initially increase the de-gree of folding,but the width of the bathymetric halo actually decreases if sedimentation can no longer keep up with and onlap the more rapidly growing bathy-metric relief.

With adequate shortening,an elongate,salt-cored detachment fold develops extending away from the diapir.Beds uplifted and rotated on this fold form a broad,high area of bathymetric relief (Figure 16)and a correspondingly wide zone of deformation.However,the deformation halo is characterized by a combination of growth-fold geometries,with progressive rotation of gradually thinning beds over a wide area,and more local halokinetic geometries.Therefore,the degree of bed upturn in halokinetic sequences tends to be less in diapirs along folds (e.g.,El Gordo)than in diapirs with only minor shortening (e.g.,El Papalote).Unconform-ities are correspondingly less angular,although they may extend farther from the diapir because of the wider bathymetric scarp.

The degree of near-diapir deformation commonly varies around a salt body because asymmetric diapir geometries,subsidence,and/or sedimentation generate asymmetric bathymetric relief.For example,although the left-hand diapir in Figure 16has roughly symmetric bathymetric slopes,the other has a gentle downdip flank and a narrower,steeper updip flank.Both diapirs are being shortened,but there is also a component of vertical subsidence into salt underlying the updip minibasin (right edge of the section).The results of such asymmetric subsidence can be seen at El Gordo diapir,where a linear withdrawal syncline intersects the southwestern diapir flank (Figure 3).Greater bed rotation (Figure 3)and less-extensive lentils (Figure 10)occur along this flank than along the northeastern

Rowan et al.

751

Figure 16.Time-migrated 3-D seismic profile from the northern Gulf of Mexico showing two shortened diapirs (same as in figure 19of Rowan,1995).The bathymetric halos are as much as 3km wide,underlying strata are thinned and rotated over a similar distance,and an angular unconformity extends 1.5km away from the diapir edge.The steeper,narrower bathymetric scarp on the right is caused by subsidence of the flanking minibasin into underlying salt.Data courtesy of WesternGeco.

flank,as would be expected for a steeper and narrower bathymetric halo than one dominated by shortening (Figure 18).

El Papalote diapir provides another example of asymmetry commonly seen on diapirs on the slope.Clastic source areas lay to the west of La Popa basin (Lawton et al.,2001),and bryozoan reefs on the east-ern flank of the diapir indicate a sediment-shadow ef-fect (Hunnicutt,1998).The basinward margin of the diapir would have had a slightly higher,narrower,and steeper bathymetric scarp (Figure 18),which is consist-ent with the greater amount of bed overturn observed on the eastern flank (Figure 3).The higher bathymetric scarp would also favor outward spreading of the salt and consequent folding,as in the simulations of Schultz-Ela (2003,this volume).

Because field data show that La Popa diapirs were episodically eroded,salt emergence is built into our forward models (Figures 13,14).However,it is un-likely that all parts of a diapir are uncovered except in the case of narrow stocks.Since salt breakout occurs mostly along the flanks because of slumping,it is prob-able that salt bodies with larger horizontal extent have significant areas that remain covered by sediment.This is why condensed but incomplete sections of older strata are commonly found on the tops of many diapirs (e.g.,Johnson and Bredeson,1971;Moore et al.,1995).In fact,diapir exposure is not even required in passive diapirism and may not occur in many cases or over prolonged periods of evolution of some diapirs.It is enough that sufficient material is regularly removed or thinned so that the overburden remains thin enough for the diapir to continue rising.There is also no need for salt breakthrough in the case of subhorizontal salt tongues (Figure 17),if the draped overburden remains thin through erosion and keeps moving forward with the advancing salt.

We have proposed that the interplay between salt-rise rates and sedimentation rates largely controls near-diapir deformation.Yet the same relationship is used to explain the dip of the salt-sediment interface,with equal rates allowing vertical diapirism and high ratios of salt-flow rate to sedimentation rate responsible for lateral salt extrusion (Vendeville and Jackson,1991;McGuinness and Hossack,1993;Talbot,1993).There is no incompatibility between these proposals because

752

Near-Salt Deformation in La Popa Basin,Mexico,and the Northern Gulf of

Mexico

Figure 17.(a)and (b)Time-migrated,3-D seismic profile and 1:1depth section,respectively,across a subhorizontal salt tongue and its bathymetric scarp;the tip of the salt is at the toe of the scarp,so that only undeformed strata will underlie the salt after it advances.Data courtesy of Veritas Marine Surveys.(c)Prestack depth-migrated portion of another line showing a lack of subsalt thinning and folding where the base salt is subhorizontal and apparent flank folding where the base salt ramps up (modified from Liro et al.,2001,reprinted by permission of the Gulf Coast Section SEPM Foundation).