-
Journal of Sedimentary Research, 2005, v. 75, 798819
DOI: 10.2110/jsr.2005.064
THE TRANSITION BETWEEN SHEET-LIKE LOBE AND BASIN-PLAIN
TURBIDITES IN THEHECHO BASIN (SOUTH-CENTRAL PYRENEES, SPAIN)
EDUARD REMACHA,1 LUIS P. FERNANDEZ,2 AND EUDALD
MAESTRO11Departament de Geologia, Universitat Auto`noma de
Barcelona, 08193, Bellaterra, Spain
2Departamento de Geologa, Universidad de Oviedo, J. Arias de
Velasco, s/n 33005 Oviedo, Spain
e-mail: [email protected]
ABSTRACT: Genetic facies analysis based on bed-by-bed
correlations fromsheet-like lobes to the basin plain in some Hecho
turbidite systemsdemonstrates that at least 50% of the flows
building the sheet-like lobeskept moving downcurrent to the closely
related basin plain and underwentflow transformations, interpreted
to have resulted from interaction withtopography at the basin
margin(s), that gave rise to specific facies. Thesefacies form a
new facies tract that replaces the fine-grained group ofturbidite
facies (very fine sand to mud) and characterizes the
basin-plainbeds.
Beds in the sheet-like lobes evolve downcurrent in a way that
ispredictable by the existing turbidite facies tract models,
whereas 36% ofbasin-plain beds, which account for 78% of the
basin-plain volume, do not.The latter have deposits from
high-density turbidity currents at their basesand typically
complete basin-floor coverage. The new facies tract developedwhen
the flows obliquely encountered the southern foreland-margin
ramp.At the ramp, the lower, sand-laden and high-density part of
the larger flowswas deflected, evolving downcurrent along the ramp
trend. The upper part ofthe flow, more dilute and thicker, was
reflected from the foreland margin asa train of declining undular
bores (moving hydraulic jumps). Subsequentreflections generated
against the flanking margins in the closed basin led toponding,
which resulted in an overall sheet-like stacking pattern
acrosssheet-like lobes and the basin plain and is the diagnostic
feature of the distalelement in the lower, sand-rich stages of the
turbidite systems. Calcilutiteson top of beds, interpreted until
now as hemipelagites, show field evidence ofhaving a turbiditic
origin (hemiturbidites), thus forming the facies cappingthe new
facies tract.
INTRODUCTION
Diagnostic criteria currently used to recognize and interpret
basin-plainvs. lobe facies associations in outcrop and core stem
largely from theturbidite fan model of Mutti and Ricci Lucchi
(1972, 1975), Mutti (1977),and Mutti and Johns (1979). Mutti and
Johns (1979) proposed a twofoldorigin for basin-plain beds (we will
use the term bed in the sense ofCampbell 1967): (1) thin beds
(thickness , 10 cm) would be most likelyassociated with the final
stages of waning flows that, following the Bouma(1962) sequence,
would have deposited their coarse load mainlyupcurrent; and (2)
thick beds, coarser and more common than might beexpected, would be
probably related to exceptional flows reaching thebasin plain after
bypass of a lobe region. As a consequence, the detailedspatial
relationships between lobes and basin-plain elements (we use
theterm element in the sense of Mutti and Normark 1991) are assumed
to becomplex (Pilkey 1987). In the Hecho basin (south-central
Pyrenees, Spain;Fig. 1A, B), these relationships and the facies
change from lobes to basinplain can be analyzed by means of
high-resolution correlations (seeRemacha and Fernandez 2003) and a
genetic (process-oriented) approachin the sense of Mutti
(1992).
In the sandy (lower) stages of growth of the Hecho turbidite
systems(types I and II turbidites of Mutti 1985), the inner
depositional elements,i.e., channel-lobe transition merging into
sheet-like lobes (we use the termsheet-like lobe following Normark
et al. 1993), form a longitudinalcontinuum (Mutti et al. 1972;
Mutti 1985; Mutti et al. 1985; Mutti et al.1988) and have a
remarkable sheet-like character (Remacha et al. 1998b;Mutti et al.
1999). Additionally, high-resolution correlations (seeRemacha and
Fernandez 2003) show that (1) sheet-like lobes mergedowncurrent
into the basin plain without significant facies breaks,
(2)basin-plain beds are deposited from the flows building the
upcurrentlobes, and (3) the sheet-like character is maintained
across both elementsas the general basinward wedging out of beds is
balanced by thethickening of some beds.
In the Hecho basin, turbidite systems lap out against basin
margins,suggesting that flows are topographically controlled. A
growing body ofevidence from laboratory experiments has shown that
flows change theirproperties after encountering topographic
obstacles (Pantin and Leeder1987; Simpson 1987; Kneller et al.
1991; Edwards 1993; Edwards et al. 1994;Haughton 1994; Kneller
1995; Kneller et al. 1997; and references therein). Infield
studies, such changes have been interpreted from the occurrence
ofmultiple paleocurrent directions commonly associated with abrupt
reversalsin grading within single beds (Ricci Lucchi and Valmori
1980; Marjanac1985; Pickering and Hiscott 1985; Hiscott et al.
1986; Marjanac 1990;Kneller et al. 1991; Edwards et al. 1994;
Haughton 1994; Kneller andMcCaffrey 1999; amongst others). In the
Hecho basin, little attentionhas been paid to the topographic
control on facies (Rupke 1976; Remachaet al. 1998a; Remacha et al.
1998b; Remacha and Fernandez 2003).
In this paper, we review the significance of the basin-plain
element inthe Hecho basin. To do this, we first determined the
downcurrent extentof the beds constituting the sheet-like lobes
and, hence, the dependence ofbasin-plain facies on the flows that
constructed sheet-like lobes (seeRemacha and Fernandez 2003).
Second, we reviewed and geneticallyclassified the basin-plain
facies. Third, we developed interpretations offlow behavior that
may have led to the change between the two faciesassociations.
Finally, we developed a hypothesis for the genesis of
thebasin-plain element on the basis of our detailed work and the
regionalgeological setting.
To do the study, we selected a representative outcrop belt (Fig.
1) fromthe lower part of the Banaston-2 composite sequence
(BanastonAllogroup; Remacha et al. 1998b). The study area extends
for nearly30 km down depositional dip, from what have been
interpreted as puresheet-like lobe facies association in the east
(Aragon valley, north ofJaca), to pure basin-plain facies
association in the west (Veral valley,south of Anso), respectively
(see Mutti et al. 1972). The results presentedhere are consistent
with observations for other Hecho basin turbiditesystems.
Data and interpretations presented here help the understanding
offacies and sand-thickness distribution over large portions (tens
of
Copyright E 2005, SEPM (Society for Sedimentary Geology)
1527-1404/05/075-798/$03.00
-
FIG. 1.A) Highly simplified geological map of the Pyrenees
showing the outcrop belt of the Eocene strata of the south-central
Pyrenees and the main structuralelements discussed in text: Bi5
Binies thrust, Bo5 Boltana anticline (lateral ramp), Co5 Cotiella
thrust, Ga5Gavarnie thrust, Lk5 Lakora thrust, La5 Larra thrust,OB
5 OrozBetelu thrust. B) Simplified geological map of the
south-central Pyrenees showing the outcrop belt of the lowermiddle
Eocene strata and the BanastonAllogroup turbidites, and also the
location of studied area. OB 5 OrozBetelu Massif. C) Simplified
geological map of the studied area showing the stratigraphy of
theBanaston Allogroup and the location of sections. Notice that
sections are aligned roughly parallel to the paleocurrent
directions of sole marks.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 799J S
R
-
kilometers) of the depositional reaches (sheet-like lobe and
basin-plainelements) of turbidite systems in confined basins, and
provide anexplanation for these features in terms of sediment
dispersal anddepositional mechanisms.
GEOLOGICAL SETTING AND SYNDEPOSITIONAL TECTONICS
Mutti et al. (1988) and Remacha et al. (1998b) identified six
majorEocene turbidite wedges (allogroups), which stack to form the
southward-migrating basin fill of the south-central Pyrenees
foreland basin (HechoGroup after Mutti et al. 1972). Turbidite
wedges are largely eroded in thenorth and onlap onto the foreland
margin in the south, where they arepartially overlain by deltaic
and terrestrial sediments of the Pyreneanmolasse stage (Bartonian
to Miocene; Fig. 1A, B). Most of the Hechoturbidite systems fit
into the model of Mutti (1985) and consist of twomain stages of
growth (Fig. 2A). The lower stage of growth accounts forthe bulk of
turbidite sand deposition all over the basin in front of
large-scale submarine erosional features (types I and II turbidites
of Mutti 1985).The upper stage comprises muddy slope-delta wedges
(type III turbidites ofMutti 1985) and subordinate foreland
carbonate-ramp toes. The upperstages are fully developed within the
canyons in the southeast and thin tothe west within the foredeep,
where they may be replaced by majorcarbonate megaturbidites. Eight
major basin-wide megaturbidites arefound (Labaume 1983; Labaume et
al. 1987; Teixell 1992) extending alongmost of the depositional
zone (from channel-lobe transition to the end ofthe basin plain),
although they are best developed in the basin-plain area,where they
are thicker (up to 250 m) and display a complete developmentof the
five sedimentary divisions defined by Labaume et al. (1987).
The Banaston Allogroup formed between 47.8 and 41.8 Ma
(middleLutetian; magnetostratigraphic and biostratigraphic dating
by O. Oms,personal communication). It consists of four
unconformity-boundedturbidite systems, named, from base to top,
Banaston-1, -2, -3, and -4(Fig. 1C; Remacha et al. 1998b). The
second system, or Banaston-2composite depositional sequence, is the
principal subject of study in thiswork. It unconformably overlies
the Mt-5 megaturbidite (also known asRoncal-Fiscal megaturbidite;
nomenclature of megaturbidites afterLabaume 1983; Labaume et al.
1987; Teixell 1992) or the foreland-margin carbonates and is capped
by the Mt-6 megaturbidite. The fourBanaston systems have very
similar facies, spatial distribution of turbiditeelements, and
stacking patterns, being fed through a single,
structurallycontrolled funnel in the southeast of the south-central
Pyrenees (Fig. 2B).
The structural framework of the Hecho basin during Banaston
timescan be related to three thrust systems (Labaume 1983; Teixell
1992, 1996),named LakoraCotiella, LarraBoltana, and GavarnieBinies
(Figs. 1A,B, 2B). Once the LakoraCotiella unit had been emplaced,
the onset ofthrusting of the LarraBoltana cover system and, to a
lesser extent, theinitial thrusting of the Binies cover thrust (an
extension of the basement-involved Gavarnie thrust; Teixell 1996)
led to a reorganization of theforedeep as summarized in Figure 2B.
Synsedimentary thrusting of theLarraBoltana and Binies thrusts is
demonstrated by the progressiveunconformities locally exposed in
the Boltana and Binies ramp anticlines(see Remacha et al. 1998b,
their figs. 7 and 8). At the western end of theBanaston turbidites
outcrop belt, beyond the study area, mapping of theBanaston
megaturbidites and the Mt-8 megaturbidite (Payros et al.
1994;Payros et al. 1999; see also Faci-Paricio et al. 1997 and
references therein)has revealed an onlap pattern onto the
OrozBetelu Massif (Figs. 1A, B,2B), showing that this massif
constituted a submarine high forming thewestern closure of the
deep-water Banaston basin. The OrozBetelubasement-involved thrust
seems to represent the western prolongationof the LarraBoltana
system (Camara and Klimowitz 1985; see alsoFaci-Paricio et al.
1997), connecting the poorly developed submarineorogen in the north
(the Lakora and LarraBoltana thrusts) with thesouthern foreland
margin (Fig. 2B). Therefore, the Banaston basin was
a structurally confined and relatively small basin, extending
longitudi-nally for some 150 km. The basin width (northsouth) is
more difficult tocalculate because of generalized erosion in the
north but can be estimatedto be about 2545 km west of the Boltana
anticline.
CORRELATION FRAMEWORK
We have followed the genetic facies approach of Mutti (1992),
which isbased on high-resolution correlation patterns. In the
depositionalelements of some turbidite systems, the highest
(bed-by-bed) correlationlevel can be reached (see examples in
Pickering et al. 1995). This can beattained only through a
hierarchical procedure by splitting time-equivalent packages of
beds bounded by precise time lines (markerturbidite beds, each
deposited by a single flow) into smaller groups. Weenvisage three
main correlation orders, termed first-, second- and third-order
correlations (see Remacha and Fernandez 2003 for a
detaileddiscussion).
First-order correlation is based on the two major megaturbidites
(Mt-5and Mt-6) that bound the Banaston-2 sandy stage. These
distinctive andmappable megaturbidites (Fig. 1C), which can each be
up to 250 m thick,and which extend for almost the complete foredeep
(see above), are themost outstanding markers for stratigraphic
correlations on both regionaland local scales (Fig. 3). Selected
sections between the megaturbiditeswere measured at maximum detail
(bed by bed, separating all thedivisions forming each bed), forming
the basis for second- and third-ordercorrelations. Second-order
correlation relies on matching outstandinglythick beds between
sections. Such beds, considered by some geologists asminor
megaturbidites, also have a distinctive suite of facies (see
below).The second-order correlation of the interval between Mt-5
and Mt-6(Figs. 3, 4) allows discrete packages of time-equivalent
beds to becorrelated at the meter scale. Third-order correlation
was done bycomparing the number, vertical arrangement, and facies
features of thesingle beds within second-order packages and,
further, on testing thecoherence of the downcurrent evolution of
facies (textural and structuraldivisions) in single beds, thus
providing the bed-by-bed correlationframework. Problematic
correlations due to the presence of coveredintervals, small-scale
faults, and local wedging out of beds were not takeninto account
for calculations. A complete bed-by-bed correlation for
thebest-exposed interval of the Banaston-2 resulted in 64% of the
lobe bedscorrelating across the four studied sections, whereas
17.6% of bedscorrelated partially and 18.4% of beds did not
correlate, and is the basisfor our analysis (see the simplified
cross section in Fig. 4). This crosssection is roughly parallel to
the paleoflow direction as determined fromsole-mark paleocurrents
(see Fig. 1C) and summarizes the downcurrentevolution of beds
between sheet-like lobes to the east (Jaca section) andthe basin
plain to the west (Anso section) through transitional
interveningsections (Estarrun and Aragues sections).
LITHOLOGICAL FEATURES OF BEDS IN SHEET-LIKE LOBE AND
BASIN-PLAIN
ELEMENTS
The turbidite beds in the selected interval of Figure 4 can
consist of upto five basic lithological divisions. From base to top
of a bed, these are:
N Graded sandstones and/or graded coarse siltstones
(clean-sandstonedivision).
N Medium to fine siltstones (clean-siltstone division).N Poorly
sorted (muddy), graded, very fine sandstones to siltstones
(dirty sandstonesiltstone division).
N Homogeneous shales (shale division).N Marlstones or limestones
(calcilutite division).
On the basis of Markov chain analysis, individual beds in the
lobesand the basin plain show variable vertical trends in these
lithologies
800 E. REMACHA ET AL. J S R
-
(Fig. 5). In summary, beds in sheet-like lobes are mainly
bipartite,comprising a clean-sandstone division overlain by a shale
division.Clean siltstone and dirty sandstonesiltstone divisions are
scarce, beingpresent only in the thinnest beds and in some of the
thickest beds,respectively. Finally, calcilutite divisions rarely
appear on top of thelobe beds. Toward the basin plain, clean
siltstone, dirty sandstonesiltstone, and calcilutite divisions
become more common. Basin-plainbeds typically consist of a
clean-sandstone division, a dirty-sandstonesiltstone division, a
shale division and, finally, a calcilutite division.
The thinnest basin-plain beds tend to lack a
dirty-sandstonesiltstone division and contain a clean-siltstone
division, either overlyinga basal clean-sandstone division or
forming the base of the bed. Also,calcilutite divisions can form
single beds or directly overlie a basalclean sandstone or siltstone
division. The relative importance ofseveral lithological divisions
varies from lobes to the basin plain (Fig. 6;Table 1).
Beds at outcrop scale are always tabular. Their bases are flat,
although,in some of the thickest lobe beds, tabular scours (Mutti
and Normark
FIG. 2.A) Highly idealized sketch showing the arrangement of the
Hecho turbidite systems according to Muttis (1985) model. To
broadly illustrate its application tothe Banaston systems, the
distribution of transfer-zone, lobe, and basin-plain elements as
well the approximate position of the Boltana anticline have been
added(compare with Figs. 2B and 3). B) Paleogeographic sketch map
of the Banaston-2 turbidite basin (compare with Fig. 1A, B). The
configuration of the basin for the othersystems of the Banaston
Allogroup is very similar to this one. Structures older than
Banaston times: Co 5 Cotiella thrust, Lk 5 Lakora thrust. Active
structures duringBanaston times: Bi 5 Binies thrust, LaBo 5
LarraBoltana thrust, O-B 5 OrozBetelu thrust.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 801J S
R
-
1987) or, more rarely, irregular scours are present. Contacts
betweenlithological divisions are flat (although tops of
clean-sandstone divisionscan be rippled, especially in the
basin-plain beds), and sharp orgradational. Sharp contacts appear
where clean sandstone is overlain
by shale or calcilutite. In these cases, significant intra-bed
bypass ofthe intermediate grain-size populations, which are found
fartherdowncurrent, took place. Also, where clean sandstone or
clean siltstoneis overlain by dirty sandstonesiltstone, an
intra-bed depositional
FIG. 3.A) Cross section of the Banaston Allogroup between the
Boltana anticline (Janovas section) and the Anso area showing the
Banaston-1, -2, -3, and -4composite depositional sequences (compare
with Fig. 2A). B) Stratigraphic cross section of the Banaston-2
composite depositional sequence in the study area from sheet-like
lobes (Jaca section, right) to basin plain (Anso section, left)
showing the first-order and second-order correlation framework
(reprinted from Remacha and Fernandez2003; with permission from
Elsevier). Black bars at both sides mark the interval detailed in
Figure 4.
802 E. REMACHA ET AL. J S R
-
break, which may be erosional but without a marked grain-size
break, ispresent.
Beds range in thickness between 0.5 and 520 cm (Table 1). Mean
bedthickness increases from the lobe to the basin-plain element
(Table 1;Fig. 7), although the mean thickness of the
clean-sandstone divisiondisplays the opposite trend. Moreover, the
ratio of thick (. 10 cm) tothin (, 10 cm) beds increases from the
lobe (1:3.4) to the basin plain(1:1.9), but clean-sandstone
thickness per meter of measured sectiondecreases from 0.67 to 0.27
along the same trend. This is because thickbeds of the basin-plain
element range widely in terms of clean-sandstone
thickness and typically are mainly built of a thick interval of
dirty sandsiltstone plus shale (Table 1; Fig. 6).
GENETIC FACIES ANALYSIS
The third-order correlation framework enables us to perform a
geneticfacies analysis, i.e., to understand how sediment gravity
flows evolved inboth space and time from the sheet-like lobe to the
basin-plain elements.Following Mutti (1992, p. 4953), the facies
tract in a bed records thedownstream facies changes produced by
transformations of a flow during
FIG. 4.Simplified version of a detailed bed-by-bed correlation
of the selected interval of the Banaston-2 composite depositional
sequence from sheet-like lobes (Jacasection, right) to basin plain
(Anso section, left). Note the overall thickening of beds toward
basin plain. See Figures 1C and 3B for location of sections and
forstratigraphic location of the selected interval, respectively.
Bars at both sides mark the stretch detailed in Figure 12. Bed
marked with asterisk is depicted in Figure 11C.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 803J S
R
-
its motion. The lateral and vertical changes in depositional
divisionswithin beds record the flow evolution in space and
time.
On the basis of the textural and structural features of, and the
vertical andlateral relationships among, the lithological divisions
that form the turbiditebeds of the studied interval, two main
groups of facies have been distin-guished in the transition from
sheet-like lobes to basin plain. The featuresof these facies are
based on field observations. The equivalence
between these facies and the lithological divisions above is
described inTable 2.
GROUP 1: FACIES FROM PRIMARY (DEFLECTED) FLOWS
Table 3 summarizes the main features of Group 1 facies. Although
thetable is largely self-explanatory, some comments help to
understand these
FIG. 5. Lithological structuring of thin(, 10 cm) and thick (.
10 cm) beds fromsheet-like lobes (Jaca section) to basin plain(Anso
section). Markov chain analysis hasbeen done following the method
of Powers andEasterling (1982). Numbers denote residuals[(observed
transitions fitted transitions)/sqrfitted transitions], of which
the positive onesare shown. x2 testing is statistically
significantat the 99.9986% level, well beyond the 95%needed to
reject the null hypothesis ofrandomness (n 5 number of beds
involved incalculations). The commonest bed types aregiven by the
preferred transitions (highestresiduals shown by bold lines);
compare withthe real examples of beds displayed inFigure 12.
804 E. REMACHA ET AL. J S R
-
facies. They form part of the facies tract of Mutti (1992),
refined by Muttiet al. (1999), and therefore are designated
following his nomenclature.From the sheet-like lobes to basin
plain, only the coarse-grained and thefine-grained facies groups
are present (Fig. 8A, B). The very coarse-grained facies (boulders,
cobbles, and pebbles) and the first deposits ofthe coarse facies
(small pebbles to coarse sand) are restricted to thetransfer and
channel-lobe transition elements (see Remacha et al. 1998b)and
therefore are not discussed here. Facies of Group 1 are the
onlyconstituents of all but the thickest lobe beds, and of the
thinnest basin-plain beds. In other cases, they form the lower part
of beds (see below;Fig. 8). Also, these facies are arranged forming
simple fining trends bothupwards and downcurrent.
F5 consists of poorly sorted to well-sorted, graded sandstones.
Coarse-tail grading (Middleton 1967) may be clear or subtle. In the
study area, F5always forms the basal division of beds. The base of
F5 deposits is usuallyflat, although in some beds it locally
displays impact features (Mutti andNormark 1987) and the deposit
contains rip-up clasts. F5 is interpreted torecord deposition from
high-concentration flows (see Mutti 1992, Knellerand Branney 1995,
and references therein; cf. S3 from Lowe 1982; see alsodiscussion
in Kneller and Buckee 2000).
F7 deposits are well-sorted, fine- to medium-grained sandstones
withparallel-stratified, millimeter-thick, inversely graded or
ungraded layers,which thin and fine upward through the deposit. F7
deposits can overliethe base of the bed or evolve from an
underlying F5. Upward, the layerscommonly become indistinct and
pass into F8 deposits (see below) or areoverlain abruptly by
secondary-flow facies. These layers have beeninterpreted as
traction carpets (Lowe 1982) or as longitudinallysegregated coarse
material (Hiscott 1994) possibly due to reworking ofnewly deposited
sediment upcurrent (Mutti et al. 1999). Data from thisproject do
not allow selecting amongst these interpretations or
evenconsideration of others (see Leeder 1999, p. 223).
F8 consists of well-sorted, graded sandstones, apparently
withdistribution grading and, according to Mutti (1992), is the
strictequivalent of Boumas Ta division. F8 deposits can gradually
overlieF7 or form the bases of beds. Mutti (1992) interpreted this
facies as a latesuspension-sedimentation stage from a
gravitationally reconcentrated(Fisher 1983) sandy high-density
turbidity current.
F9 encompasses Boumas Tb through Te divisions (Mutti 1992)
butalso includes calcilutites (formerly referred to as
hemipelagites; seebelow). It is the most common facies in the
transition between lobes andbasin plain. F9 deposits were deposited
by traction-plus-fallout processesfrom low-density turbidity
currents. This facies has been discussedextensively elsewhere (see
review in Pickering et al. 1989; see also Komar
FIG. 6. Lateral trend of the bulk lithological composition of
sections fromsheet-like lobes (Jaca section, right) to basin plain
(Anso section, left). Symbols asin Figure 5.
TABLE 1.Main lithological parameters from sheet-like lobes to
basin plain.
Section Jaca Estarrun Aragues Anso
No. of turbidite beds (*) 367 (5901) 431 287 277No. of thick (.
10 cm) beds 83 123 92 96No. of thin (, 10 cm) beds 284 308 195
181
Bed thickness range (cm) 0.5310 1520 0.5301 0.5353
Mean bed thickness (cm) All beds 8.387 10.788 13.609 16.887thick
(. 10 cm) beds 22.276 25.105 32.140 38.973thin (, 10 cm) beds 4.328
5.071 4.922 5.172
Mean clean-sandstone thickness (cm) All beds 5.585 5.184 5.046
4.563thick (. 10 cm) beds 15.131 13.669 13.431 11.895thin (, 10 cm)
beds 2.795 1.795 1.021 0.674
Mean clean-siltstone thickness (cm) All beds 0.05 0.192 0.265
0.536thick (. 10 cm) beds 0.054 0.284 0.282 0.610thin (, 10 cm)
beds 0.049 0.155 0.255 0.496
Mean dirty sand+siltstone thickness (cm) All beds 0.142 0.492
1.870 3.837thick (. 10 cm) beds 0.627 1.711 5.726 10.607thin (, 10
cm) beds , 0 0.005 0.031 0.075
Mean shale thickness (cm) All beds 2.552 4.690 5.491 6.455thick
(. 10 cm) beds 6.371 8.767 11.061 13.729thin (, 10 cm) beds 1.436
3.062 2.834 2.769
Mean calcilutite thickness (cm) All beds 0.046 0.032 0.839
1.474thick (. 10 cm) beds 0.093 0.027 0.984 2.609thin (, 10 cm)
beds 0.033 0.034 0.781 1.159Beds/meter 11.923 9.269 7.348 5.922Sand
ratio 0.67 0.48 0.37 0.27
(1) In Jaca section there is a fairly thick covered interval,
hence fewer turbidite beds (367). Total number of beds in Jaca
section (590) has been extrapolated froma smaller interval for
which 73% of Jaca beds are in the Estarrun section.
(*) Calcilutite intervals with a transitional base have been
incorporated into the respective underlying beds.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 805J S
R
-
1985) and further discussion is outside the scope of this work.
However,some comments about the Tc divisions are in order.
Tc divisions consist of cross-laminated very fine sandstone to
coarsesiltstone, commonly showing convolute bedding, and lack
mudstonedrapes that separate sets of ripples. Tc ripples are small
(wavelength is lessthan 10 cm and usually less than 7 cm),
asymmetric, 2D to 3D (linguoid)bedforms that migrated in the same
direction as paleocurrents indicatedby sole marks (Figs. 4, 9).
Also, Tc divisions form a part of a strict Boumasequence, and
unless overlain by Group 2 facies (see below) they
underlierelatively thin (usually less than 9.0 cm) Td plus Te
divisions.
Interpretation of the Group 1 Facies
Paleocurrent directions indicated by sole marks (Fig. 1C) and
bothdowncurrent and vertical facies evolution suggest that facies
of this groupwere deposited from axially evolving, simple waning
flows. Also, the sole-mark paleocurrent pattern following the trend
of the southern forelandmargin (Fig. 1C) suggests that these
axially evolving flows were deflectedby the basin topography (cf.
Kneller et al. 1991; Kneller and McCaffrey1999). Because no other
major modifications of the flows depositing thesefacies seem to
have taken place, facies of Group 1 are interpreted as thedeposits
from primary (deflected) flows.
GROUP 2: FACIES FROM SECONDARY (REFLECTED) FLOWS
We define a new group of fine-grained facies that do not fit
Muttis(1992) general facies tract (Fig. 8). The new group involves
the samegrain-size populations as F9 and consists of four main
facies: Fm-1 toFm-4. The volume represented by these facies
increases from 18% in thelobe, where they are only present in , 2%
of beds, which are the thickest,to up to 78% in the basin plain,
where they become widespread,appearing in at least 36% of beds (81%
of thick beds) in the Anso sectionand being a diagnostic feature.
Other key points of this group of facies,compared to the group of
primary (deflected) flow facies, are: (1)paleocurrent directions
diverge from those of the underlying primary(deflected) flow
facies, and also among the facies of this group in the
sameturbidite bed (Figs. 4, 9); and (2) facies of this group in the
same bed forma complex, overall graded sequence with abrupt grading
reversals(sawtooth grading of McCaffrey and Kneller 2001) that
evolves intoa thick mudstone cap (Fig. 8A, C), distinctively
thicker than that of F9facies (Td plus Te; Table 3). The main
features of these facies aresummarized in Table 3. Additional
comments are given below to bettercharacterize them.
Fm-1
Fm-1 refers to rippled very fine sandstone to coarse siltstone.
Bedformsare slightly asymmetrical to symmetrical (cf. Marjanac
1990; McCaffreyand Kneller 2001). In a few cases (some 10 beds in
the studied interval),the plan shape of the bedforms is visible and
shows well-defined laterallycontinuous, sinuous crests that may
bifurcate (Fig. 10). Ripplewavelength, averaging 4060 cm (Fig. 10)
and reaching a maximum of2 m, is the most diagnostic feature of
this facies when compared withprimary ripples (Boumas Tc). Ripple
height ranges between 1 and 12 cm,and crest-sinuosity wavelength is
up to 2 m (Fig. 10A). Rare examples ofrounded, isolated,
hummock-like bedforms have also been detected.Cross lamination is
rarely affected by convolution and displays climbingpatterns that
evolve upward from stoss-erosional to stoss-depositionaland finally
sinusoidal, showing silty drapes that thicken into the troughs.Very
thin laminae of mud can be interleaved toward the top of the
silty
FIG. 7. Lateral trend of lithologicalparameters from sheet-like
lobes (Jaca section,left) to basin plain (Anso section,
right)expressed as fractional variation. See Table 1for the
corresponding numerical values. Rep-rinted from Remacha and
Fernandez (2003),with permission from Elsevier.
TABLE 2.Equivalence between lithological divisions and
facies.
Lithological divisions Facies
Clean sandstone F5, F7, F8, F9lam (Tb), F9rip (Tc), Fm-1,
Fm-2Clean siltstone F9shear (Td), F9grad (coarser part of Te),
Fm-3Dirty sandstone/siltstone Homogeneized intervals consisting of
Fm-2+Fm-
3+Fm-4Shale F9shale (finer part of Te), Fm-4Calcilutite
Hemiturbidite
806 E. REMACHA ET AL. J S R
-
drapes within troughs. This facies commonly displays an
erosional baseand evolves upward, either sharply or transitionally,
into a thin, relativelypoorly sorted mudstone division (Fm-4
facies; see below) forming Fm-1Fm-4 couplets (Fig. 8C).
Fm-1 intervals overlie a primary facies and may consist of up to
two orthree Fm-1Fm-4 stacked couplets, although usually there is
only one.When several couplets are stacked, the mudstone (Fm-4)
caps of the lowerones are preserved only locally, leading to
amalgamation of Fm-1
TABLE 3.Main facies features.
Facies (no. of bedscontaining each facies)
Mean Thickness (cm):(Std. Dev.); Range
Mean MaximumGrain Size (Range) Texture
PrimarySedimentaryStructures
OtherFeatures
InterpretedSedimentary Processes
Group 1: facies from primary (deflected) flows
F5 (13) 19.4 (22.0); 1.070.0 M/C (Granule toM/F)
Poorly sorted.Ungraded tocoarse-tail grading
None Common impactfeatures and rip-upmudstone clasts.Frequent
fluid-escape structures
En massesedimentation.Hindered settling
F7 (22) 6.1 (4.5); 2.220.5 M (C to F) Inversely gradedlayers
forming anoverall fining-upward division
mm-thickhorizontal layers
Commonlyindistinct layers
Frictional freezing oftraction carpets
(?).Longitudinallysegregated grains (?)
F8 (Boumas Ta division)(239)
10.3 (8.5); , 157.0 F/VF (M to VF) Well
sorted.Distributiongrading
None Tabular scours andrip-up mudstoneclasts. Rare fluid-escape
structures
Grain-by-grainsuspensionsedimentation
F9lam (Boumas Tbdivision) (26)
3 (1.8); 1.07.0 VF Rarely F Well sorted Submillimetricparallel
laminae
Rare fluid-escapestructures
Upper-regime tractionplus fallout
F9rip (Boumas Tc division)(1066)
3.7 (3.5); , 129.5 VF to siltExceptionally F
Well sorted Ripple-drift cross-lamination
Commonconvolution.Linguoid, cm-spaced ripples
Lower-regime tractionplus fallout
F9shear (Boumas Tddivision) (241)
0.8 (0.7); , 15.0 Silt and mud Well sorted silt- andclay
laminae.Overall fining-upward trend
Parallel lamination Shear sorting in theviscous sublayer
F9grad (Coarser part ofBoumas Te division)(144)
0.9 (0.7); , 13.5 Silt to mud Well sorted. Normalgrading
None Fallout
F9shale (Finer part ofBoumas Te division)(1039)
3.7 (2.6); , 121.0 Mud to clay Homogeneous None Common
upwardenrichment incarbonate
Fallout
Group 2: facies from secondary (reflected) flows
Fm-1 (38) * 3.6 (2.8); , 111.0 VF to silt Well sorted.
Overallfining-upwardtrend
Ripple-drift crosslamination tosinusoidallamination
Long-wavelength(up to 2 m) ripples.Sinuous crestswhich
sometimesbifurcate
Traction plus fallout.Strongest undularbores
Fm-2 (14) * 1.7 (1.5); , 110.0 VF/silt to silt Well sorted.
Normalgrading
None Shearingdeformationalfeatures
Fallout. Intermediateundular bores
Fm-3 (14) * **; , 13.0 Silt and mud Well sorted silt-and clay
laminae.Overall fining-upward trend
Parallel lamination Shearingdeformationalfeatures
Fallout; shear sortingin the viscoussublayer(?). Weakundular
bores
Fm-4 (125) * 20 (36.5) *; , 1300(?)* Mud and clay Normal
gradingto homogeneous.Poor sorting ofbasal divisions
None Shearingdeformationalfeatures. Commonupward enrichmentin
carbonate
Collapse of dampedprimary flow. Falloutfrom the weakestundular
bores
Hemiturbidite (Finest partof Te/Fm-4 divisions)(257)
2.7 (2.1); , 110 Mud and lime mud.Planktonic organisms
Homogeneous,sometimes w/a graded or graded-laminated
lowerpart
None Common downwardenrichment in clay.Moderatelybioturbated
Fallout ofhydraulically sortedcarbonate particles
* Referred to entire intervals of this facies which may be
composed of several stacked divisions (see text).** Not calculated
because of inaccurate measures of thickness.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 807J S
R
-
808 E. REMACHA ET AL. J S R
-
divisions (Fig. 8C). Upward through the deposit, Fm-1 units thin
andfine, whereas Fm-4 units thicken slightly. Ripple laminae
directly belowthe erosional bases of couplets can show drag
features (folded or evenoverturned laminae).
Paleocurrents from Fm-1 diverge from those of the underlying
primaryripples and sole marks at angles up to 120u (Figs. 4, 9).
Fm-1 paleocurrentvalues from the same section are not constant,
spreading over about 210u.The degree of spread is, at least
partially, attributable to the long-wavelength sinuosity of bedform
crests not fully exposed in the smalleroutcrops.
All these features allow distinction between Fm-1 and Tc
(primaryflow) ripples (see above). However, discriminating between
Fm-1 ripplesand Tc ripples must be undertaken with caution because
their respectivecharacteristics overlap to some extent, and also
because the Fm-1paleocurrents are not constant. Consequently, we
have taken a conserva-tive attitude and classified Fm-1 ripples as
only those bedforms clearlydisplaying the features described above.
Doubtful cases have beenincluded in the primary-flow-ripples (Tc)
category.
Bedforms in Fm-1 seem to differ from those developed in very
fine sandunder purely unidirectional flows (see Yalin 1972, Baas
1994, andreferences therein) and are more similar to bedforms
generated bycombined flows (cf. Arnott and Southard 1990; Banerjee
1996). However,an increase in both ripple spacing and symmetry has
been reportedexperimentally as concentration of fine-grained
sediment increases in thedriving unidirectional flow (Kuenen 1967;
Bradley 1986). For very finesand, Kuenen reported the formation of
low, nearly symmetrical ripplesspaced as much as 25 cm and
consisting of alternating laminae of siltysand and silty clay.
These bedforms resemble the Fm-1 ripples, but thelatter show a
tendency to bifurcate and sometimes have rounded forms,and their
spacing is much greater. In addition, they lack clay-rich
laminae(except in their upper part; see above). On the other hand,
the formationof internal waves on the upper surfaces of turbidity
currents has beendocumented in both natural and experimental flows
(Wright et al. 1988;Kneller et al. 1997). Moreover, Kneller et al.
(1997) have demonstratedthe ability of internal waves to interact
with the unidirectional flow at thebed (see below) with the
resulting combined flow being, thus, able to formcombined-flow
bedforms.
Fm-2, Fm-3, and Fm-4
In most cases, the Fm-1 facies interval or a primary flow facies
unit isoverlain by a thick (up to 4.5 m, averaging 34.6 cm)
interval that can bedescribed in terms of two units (Figs. 8D,
11A). They are: (1) a gradedmuddy sandstonesiltstone overlain by
(2) a graded to homogeneousmudstone cap.
The graded muddy sandstonesiltstone unit is formed of poorly
sorted,muddy sandstonesiltstone grading upward into muddy siltstone
witha dappled appearance produced by darker intraclasts (Fig. 11B).
Thelatter consist of (a) small pseudonodules (usually up to 12 cm)
composedof relatively clean, very fine sandstone or siltstone
(Fm-2; see below) or ofsilt-laminated mudstones (Fm-3; see below),
and (b) pieces of recumb-ently folded silt-laminated mudstones
(Fm-3) displaying clearly vergentfolds. Intraclasts become smaller
and scarcer upward until they disappearat the top of the lower
unit. The maximum grain size of this unit is veryfine sand, always
finer than that of the underlying facies in the same bed.
The graded to homogeneous mudstone cap is thicker than Boumas
Te(F9) primary subfacies (Table 3). It may consist of one to
several gradeddivisions separated by subtle but sharp grain-size
breaks. Commonly, thisunit is in gradational contact with the
underlying one, but in some cases,mainly above an Fm-1 or primary
flow facies unit, its base is sharp.Toward the top it almost always
grades into a calcilutite division.
In some beds, these two units, especially the lower one, pass
laterallyand vertically into an organized deposit formed of three
facies, namelyFm-2, Fm-3, and Fm-4. The transition always takes
place throughincreasing deformation (see below) of the Fm-2 to Fm-4
intervals, whichprogressively become dismembered and homogenized
(Fig. 11B, C), untilonly the intraclasts bear witness of the
original fabric (Fig. 11B).
On the basis of these exceptionally preserved examples, the
originalfeatures of the Fm-2 to Fm-4 facies can be recognized and
theirrelationships reconstructed (Fig. 8A, C).
Fm-2 consists of a thin (less than 10 cm thick) division of
relativelywell-sorted, very fine sandstonecoarse siltstone, grading
upward intomedium siltstone (Fig. 11C). Distribution grading seems
to be present onthe basis of field observations. Fm-3 is a very
thin (up to a few centimetersthick), graded, parallel-laminated
division consisting of one to severalcouplets of silt laminae and
mud laminae (Fig. 11C) and closely resemblesthe Bouma Td division.
Fm-4 is a poorly sorted, silty mudstone,sometimes with scattered
very fine sand grains, grading upward intohomogeneous claystone
that, in some cases, forms the whole division. Nolaboratory
analyses have been made to detect the internal fabric. Thethickness
of this facies ranges from less than a centimeter to 3 m,although,
locally, some of the thicker examples are actually composed
ofseveral stacked Fm-4 divisions (Fig. 11D). The thickness and
compositionof this facies varies depending on position within the
deposit. As a rule,the higher this facies lies, the thicker and
finer-grained it is.
These three facies stack to form an elemental unit, a
fining-upwardsequence composed of up to three divisions from Fm-2
to Fm-4 (Fig. 8C).In turn, several elemental units stack to form an
overall graded interval.The lower elemental units are sandier and
dominated by Fm-2 and Fm-3divisions (see also Fig. 11C). Usually no
more than two stacked elementalunits of this type are found. In
successively higher elemental units, (a)Fm-2 fines and thins (Fig.
11B) until it disappears, (b) Fm-3 decreases inthe number of
siltstonemudstone couplets, and (c) Fm-4 quicklythickens and fines
by losing the silt- and sand-size particles. As a result,the
intermediate elemental units consist of Fm-3 and -4 divisions, and
theupper ones form a stack of Fm-4 divisions that typically are
hard tosubdivide (Fig. 11D). Finally, the uppermost division of
homogeneousFm-4 is always the thickest and merges upward into a
calcilutite division,which represents the end of the reconstructed
sequence.
The basal division of an elemental unit, especially where it is
Fm-2 orFm-3, may have a loaded and sheared base where it overlies
Fm-4 in thetop of the previous elemental unit. There is a complete
spectrum of loadstructures from load casts attached to the parental
layer to detachedpseudonodules in the thicker, underlying Fm-4
divisions (Fig. 11B).Shearing affects both the load structures and
the host sediment toa variable extent, fading downwards and
exhibiting both folding andthrusting of the laminated Fm-3 division
below the Fm-4 division of theelemental unit (Fig. 11C, D). Shear
direction may not agree withpaleocurrents from sole marks (i.e.,
primary-flow direction) and/or fromFm-1 and are even reversely
directed (to the south; Figs. 4, 9, 11C).
r
FIG. 8.A) Facies tracts of both primary flows (cf. Mutti 1992,
his Figs. 26, 27, and 32) and reflected flows. B) Markov chain
analysis of vertical facies transitions.Analysis has been done
following the method of Powers and Easterling (1982). Numbers
denote residuals [(observed transitions fitted transitions)/sqr
fitted transitions]of which the positive ones are shown. x2 testing
is statistically significant at the 99.9986% level. Calculations
have been performed on the whole set of beds from the fourstudied
sections (Jaca, Estarrun, Aragues, and Anso). C) Idealized sequence
of facies in a basin-plain bed deposited from a large-volume flow.
D) Usual appearance of thesequence of facies in a basin-plain bed
deposited from a large-volume flow (see text for details; compare
with Fig. 11A).
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 809J S
R
-
FIG. 9. Simplified map of the Banaston outcrop belt between Jaca
and Anso showing the pattern of paleocurrents from flutes, primary,
and reflected ripples(Boumas Tc and Fm-1, respectively) and from
shear directions (see also Figs. 1C and 4; see text for details).
Paleocurrents have been taken in the four main sectionsdiscussed in
the text plus an additional outcrop located very close to the
wedging out of the Banaston-2 against the foreland margin, and
correspond to the intervaldepicted in Figure 3B.
FIG. 10.Secondary (reflected flow) ripples, Fm-1. A) Field
aspect of bedding surface. Notice the large spacing of bedforms and
both sinuosity and bifurcation (arrow)of crests. Hammer for scale
is 33 cm long. Santaliestra Allogroup at Urdues (see Fig. 1C for
location). B) Close-up view of nearly symmetrical secondary
ripples.Wavelength is 5060 cm (each segment of the meter stick is
20 cm long). Anso section.
810 E. REMACHA ET AL. J S R
-
Moreover, shear directions may diverge amongst groups of
elementalunits.
Fm-2 formed by direct suspension sedimentation from a
turbulentsuspension of relatively high sediment concentration.
Absence of tractionsuggests lower flow power or higher
suspended-load fallout rates than forFm-1. Under weaker,
low-density, turbulent suspensions, Fm-3 mayform, analogously to
Bouma Td divisions, by shear sorting of silt and clayparticles by
burst-and-sweep cycles at the viscous sublayer (Hesse andChough
1980), although other interpretations have been postulated (Stowand
Bowen 1980). In contrast, the rather poorly sorted, gradedmudstones
and/or the apparently homogeneous mudstones of Fm-4 arerelated to
collapse of a muddy flow with a sufficiently high
suspended-sediment concentration to damp turbulence severely (cf.
McCave andJones 1988). However, the uppermost divisions of this
facies within a bed(Fig. 11D) are, wholly or partly, more likely
related to deposition froma progressively more dilute remnant
flow.
Interpretation of the Group 2 Facies
Any interpretation of this group of facies must account for the
featuresdescribed above: (1) the paleocurrent divergence with
respect to theprimary facies and within this group of facies (Figs.
9, 11C); (2) thevertical arrangement in a complex, overall graded
sequence with abruptgrading reversals; and (3) the occurrence of
these facies forming the upperpart of the basin-plain beds, whereas
lobe beds almost always consist onlyof primary flow facies. In
other words, simple waning flows depositing inthe lobe evolved
downcurrent into flows that first deposited as simplewaning flows
(lower part of the basin-plain beds) and then as complexwaning
flows with energy pulses (upper part of the basin-plain beds).
Fm-1Fm-4 Couplets.Fm-1 bedforms, as discussed above,
probablyresult from traction plus fallout under combined-flow
conditions. Theformation of internal waves on the upper surfaces of
flood-derived,sustained turbidity currents has been documented
(Wright et al. 1988), ordeduced from facies features (Mutti et al.
1999), in shallow-water settings,in front of delta or fan-delta
systems, a scenario quite different from thedeep-water environment
discussed here (see below). Also, this interpre-tation fails to
explain why the inferred combined-flow conditions mostlydevelop, or
are at least detected, in the basin-plain element and notupcurrent,
in the lobe region. Indeed, strikingly, the complex waningbehavior
of the flow revealed by the whole vertical sequence of the faciesof
Group 2 is first evident in the basin-plain beds. If this behavior
hadbeen inherent to the flows, it would be expected to be present
also, andmainly, in the lobe beds (see Kneller and McCaffrey 2003).
The complexvertical sequence (sawtooth grading) has been attributed
to flowreflection from basin margins (McCaffrey and Kneller 2001),
a possibilitysupported by the paleocurrent divergence, which is
considered the bestindicator of flow reflection from basin margins
(Kneller et al. 1991). Allof that suggests that Fm-1 ripples could
be the result of reflected flows.Ripple formation by transverse
internal waves generated by reflectedflows has been suggested by
several authors (Pickering and Hiscott 1985,Kneller et al. 1991,
Edwards et al. 1994, and references therein). Althoughthe absence
of any significant grain-size break between the
primary-flowdeposits and the reflected-flow deposits argues against
flow reflection,recent laboratory data show that reflected flows
have a velocity on thesame order as the original flow and thus have
the potential to transportsediment deposited by the forward flow
(Kneller et al. 1997). This abilitywould be enhanced by the
potential bulking through erosion of thereflecting flow as it goes
back down the ramp of the obstacle, a possibilitydemonstrated in
experiments with saline water (Garca and Parker 1993).Moreover,
Kneller et al. (1997, see their Fig. 9) have demonstrated
fromlaboratory data that weaker reflected flows (undular internal
bores ofSimpson 1987; type-A bores sensu Edwards 1993; Edwards et
al. 1994)
display orbital fluid motions with a period in the range of a
few seconds,and that these internal waves may affect the bed in
contrast with thosegenerated during the passage of a normal
turbidity current, which donot. Therefore, near-bottom
combined-flow conditions may exist whenturbidity currents undergo
reflection. Furthermore, the profile of velocityvs. time during the
passage of a train of solitary waves has a sinusoidalform with
stepwise reversals and a mean shear velocity declining withtime
(Kneller et al. 1997, see their Fig. 9). This pattern agrees with
thevertical arrangement of the stacked couplets of Fm-1 plus
mudstone (Fm-4), which fit well in a waning pulsing succession of
sedimentation events.During the passage of a wave, traction
progressively wanes while falloutrapidly increases to become
dominant during the inter-bore quiescentperiods when the thin
mudstone (Fm-4) drapes are most probablyproduced (see below;
Edwards 1993; Edwards et al. 1994; Kneller et al.1997). Features of
the remaining facies of this group suggest that Fm-1records the
passage of the strongest bores recorded.
The Fm-2RFm-4 Sequence.The Fm-2RFm-4 sequence is a continu-ation
of the sequence formed by the Fm-1Fm-4 couplets. The
overallcomposite sequence indicates a waning, pulsing flow that
conforms to thepattern of behavior of a train of bores decaying in
time formed byreflected flows and separated by quiescent periods.
In addition, evidenceof flow reversals given by the shear
directions (Fig. 11C), albeit moresubtle than in Fm-1, are present
in Fm-2 to Fm-4. We speculate thatduring quiescent periods the rear
parts of the residual primary flow suffera severe loss of energy,
carried away by the passing bores (Edwards et al.1994; Haughton
1994) and then collapse rather abruptly from a
relativelyhigh-density, nonturbulent flow (most Fm-4 divisions; see
above).
Sedimentary Record of Bores and Synsedimentary
Deformation.Summarizing, the sequence formed of the facies of Group
2 is interpretedas recording deposition from a waning, pulsing flow
that may haveresulted from the passage of decaying undular bores
(moving hydraulicjumps; see Simpson 1997; type A bores, Edwards
1993, Edwards et al.1994; see also Kneller et al. 1997) generated
by flow reflection from thebasin margins. Complex traction plus
fallout, under combined-flowconditions (Fm-1), was succeeded by
fallout alone (Fm-2 to Fm-4) assuccessive internal waves of
progressively lower power swept across thebasin. During the
quiescent periods between waves, the residual forwardflow, with its
energy severely damped, collapsed to form Fm-4 divisions.
The abundant deformation that affects the intervals formed of
Fm-1 toFm-4 facies is interpreted to be syndepositional to
immediately post-depositional, because of its occurrence as
discrete horizons withinturbidite beds, with a downward decrease of
deformation (see Fig. 11C),the plastic rheology displayed, and the
close relationships between thedeformed and the parental materials.
The flat depositional topography,deduced from the correlation (Fig.
4; see also below), precludes an originby slumping or creeping and
points to shearing by an overriding flow.This would account for the
deformed cross-laminae in the Fm-1 divisions(cf. Rust 1968) and the
folds and thrusts observed in some Fm-3 divisions(Fig. 11C). Shear
features principally underlie Fm-1 facies, suggestingthat shear on
the bottom was predominantly produced by the strongestbores. Also,
it is interpreted that the combination of liquification andplastic
deformation accounts for the overall soft-sediment
deformationprocesses that led to the progressive obliteration of
the original sequencesof facies Fm-2 to Fm-4, and finally resulted
in the graded muddysandstonesiltstone and the graded to homogeneous
mudstone capcontaining relics of the constituent facies (Fm-1 and
Fm-2 pseudo-nodules). Because of the complex interaction among
these processes, theirdetailed relationships have not been fully
deciphered and further studiesare necessary. In light of the
preliminary results, two main processesassociated with the
repetitive passage of waves are envisaged: (1)vibration
liquification (or liquefaction; see Nichols 1995) triggered by
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 811J S
R
-
812 E. REMACHA ET AL. J S R
-
cyclic wave loading and (2) shear liquification. Subordinate
seepageliquification (or fluidization) and plastic behavior
features also wouldaccount for obliteration of the primary
features. This process wasparticularly effective where relatively
thick newly deposited Fm-4divisions were present, suggesting
strongly that inter-bore deposits weremetastable during wave
passage. That points to a rapid deposition ofmost of the Fm-2 to
Fm-4 facies, which displayed a thixotropic behaviorduring
vibration.
CALCILUTITES: HEMIPELAGITES OR HEMITURBIDITES?
Calcilutites (Table 1; hemiturbidite facies in Table 3) are
moderatelybioturbated, medium-gray marlslimestones, weathering to a
distinctivewhitish color (Fig. 11E), which commonly lightens
upward, suggesting anincrease in carbonate content. Their
qualitative mineralogical composi-tion is similar to that of the
sediments of the southern-forelandmargincarbonate ramp (Burgui
Marls Formation of Camara and Klimowitz
1985). The biogenic content is scattered and variable from bed
to bed; itconsists mainly of planktonic foraminifera and
coccoliths, and rarebenthic foraminifera (Mutti et al. 1972).
Calcilutite divisions are , 10 cm thick and most commonly
transi-tionally overlie either Boumas Te (F9) (Fig. 11E) or the
uppermost Fm-4division of a bed. The transition is usually gradual
through a significantportion of the terrigenous mudstone division
(more than half of thethickness in the case of Te), although there
is a steeper gradient of thecarbonate content at the top, marking
the base of the calcilutite. Both inthe basin plain and laterally
toward the southern foreland margin,calcilutite divisions also
directly overlie Boumas Tc (primary ripples),thus forming the
mudstone division, or constitute individual beds. In thelatter case
the coarsest biogenic particles are concentrated near the
base,forming a graded or parallel-laminated lower interval (Td
division). Also,there is a relationship between calcilutites and
the underlying facies withinthe same bed. For the dataset of the
four studied sections, 45% of bedswith secondary-flow facies have a
calcilutite cap (mean thickness 3 cm),
FIG. 12.Detailed stratigraphic cross section from sheet-like
lobes (Jaca section, right) to basin plain (Anso section, left)
showing calcilutite beds as the end membersof downcurrent evolution
of beds (e.g., interval between Jaca section beds # 8 and 9). Note
that lobe beds tend to consist of a sandstone division sharply
overlain bya mud division, whereas in basin-plain beds sandstone
division gradually pass into the mudstone division via a siltstone
interval (compare with Fig. 5). Also note theoverall thickening of
thicker lobe beds and the wedging out of the thinner lobe beds
toward the basin plain (e.g., interval between Jaca section beds #
4 and 6). SeeFigure 4 for stratigraphic location.
r
FIG. 11.A) Thick bed displaying secondary-flow facies
(overturned section, stratigraphic top toward lower left corner).
Arrows mark the base and top of the bed.From base to top, bars lie
at the limits between primary flow facies, an Fm-1Fm-2 interval,
the graded muddy sandstonesiltstone, and the mudstone cap. B)
Typicalfield aspect of a graded muddy sandstonesiltstone interval.
The unit between the lower and middle bars, which overlies the
primary facies interval, is a transitional casebetween the sandier
part of the graded muddy sandstonesiltstone unit and the original,
Fm-2-dominated interval; notice the large pseudonodules (bold
arrows). Abovethat, the interval between the middle and upper bars
is formed of Fm-2 and Fm-3 facies showing well developed load
casts, locally detached and forming pseudonodules(white arrow on
the right). Finally, the interval above the upper bar is a thick
graded muddy sandstonesiltstone unit showing deformed fragments of
the immediatelyunderlying interval that draw tight recumbent folds
facing to the left (white arrows on the left). Notice the presence
of small pseudonodules elsewhere (e.g., black arrowsin the upper
part). Diameter of coin is 2.1 cm. C) Exceptionally well-preserved
example of two elemental units, each composed of an Fm-2Fm-3
couplet. Noticesynsedimentary folding and thrusting of the
uppermost Fm-3 laminae in the upper elemental unit marking a
rightward (southward) shearing from overriding flow. Theoverlying
interval in the same bed, out of the field of the photo, is a
pervasively deformed, graded muddy sandstonesiltstone interval.
Coin is 2.5 cm. See Figure 4 forlocation of this bed. D) Thick
mudstone cap formed by stacked Fm-4 elemental units, which are
barely distinguishable. Notice the syndepositional deformation
revealedby the truncation surface and angular relationships of the
lowermost elemental units (lower arrow) and by the recumbently
folded upper elemental unit (upper arrow)marking a drag towards the
right (northwards). Curvilinear, fan-shaped features at the right
correspond to plumose pattern of a joint. Scale in centimeters and
inches. E)Calcilutite-rich, thin bed interval showing the
gradational relationships between Te mudstone (light gray) and the
calcilutite (whitish) divisions within each bed. Sandstoneand
siltstone divisions of beds are dark to medium gray. Scale in
centimeters and inches. The five photographs are from Anso
section.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 813J S
R
-
while only 15% of beds lacking secondary-flow facies have a
calcilutitecap, which is also thinner (mean thickness 2.5 cm).
Calcilutites appear locally in the lobe and become widespread in
thebasin plain. In correlated beds, they evolve from thin marls in
the lobeand lobebasin-plain transition to thicker marlslimestones
in the basinplain, where they can form the whole bed following the
pinch-out of theunderlying terrigenous divisions (Fig. 12).
Furthermore, the downcurrentthinning of Boumas Te (F9) division is
accompanied by a downcurrentthickening of the calcilutite.
Laterally toward the southern forelandmargin, calcilutite divisions
become amalgamated as the terrigenousportions of beds onlap onto
the topographic highs, which are draped bycondensed calcilutite
intervals.
Since the work of Mutti et al. (1972), calcilutites have been
interpreted,not only in the Hecho basin, as hemipelagites, i.e.,
the non-turbiditicsediments in the basin, and considered as one of
the diagnostic features ofthe basin-plain facies association (e.g.,
Mutti and Ricci Lucchi 1972,1975; Rupke 1976; Mutti 1977, 1979;
Mutti and Johns 1979; Nilsen 1984).Notwithstanding, a turbiditic
origin has also been suggested (e.g., Kuenen1964; Van der Lingen
1969; Hesse 1975; Rupke 1975; Stanley andMaldonado 1981; pertinent
papers in Stow and Piper 1984; Stow andWetzel 1990), giving rise to
the term hemiturbidites (Stow and Wetzel1990). We also point to a
turbiditic origin for the calcilutites of thestudied interval,
although this controversy cannot completely be resolvedon the basis
of field observations.
The mode of occurrence of calcilutites and their internal
structuringwhen forming a whole bed (cf. Fauge`res et al. 1984)
strongly suggesta turbiditic origin. Also, bioturbation is less
intense than in truehemipelagites. The downcurrent and lateral
evolution of calcilutitessuggest that the flows, as they dropped
their terrigenous load, began todeposit carbonate sediments, which
can finally form entire beds inthemselves. This means that
calcilutites may constitute the distal endmembers in both the
primary and the secondary facies tracts (endsubfacies of F9 and of
Fm-4; Fig. 8). We suggest that these hemiturbiditesare the result
of hydraulic sorting of the finer and lighter carbonateparticles
toward the upper part of the flow. The lower settling velocities
ofthe carbonate particles, when compared to the terrigenous muds,
mostlikely result from a decreased ability of the former to
flocculate (see Piper1978; Stow et al. 1984). Thus, the
fine-grained carbonate sediment wouldremain in suspension, settling
slowly until the flow ceased. Hydraulicsorting of carbonate
particles affected not only the mud-size population,as we propose
here, but also the coarser (sand grade) population, asdemonstrated
by Fontana et al. (1989), who invoked both density andshape of the
carbonate particles as the factors controlling their
hydraulicsorting.
The qualitative mineralogical composition suggests that the
sedimentsof the foreland-margin carbonate ramp in the south were
the main sourceof the calcilutite sediment. Carbonate sediment was
most probablyincorporated into the flows by bulking through erosion
in two locations:first from the foreland margin adjacent to the
shelf, the canyon, and theinner depositional elements and, later,
when the turbidity currentsobliquely encountered the foreland
margin ramp and underwent re-flection, flowing up and then down the
ramp (see below). Carbonateenrichment of flows during reflection is
supported by the relativeabundance and thickness of calcilutite
divisions in beds with secondaryfacies vs. beds lacking secondary
facies.
RELATIONSHIPS BETWEEN SHEET-LIKE LOBE AND BASIN-PLAIN
ELEMENTS
The bed-by-bed correlation, simplified in the cross section of
Figure 4,demonstrates that the sheet-like lobe and basin-plain
elements aregenetically related, because all of the beds in the
basin-plain element(Anso section) are found in the upcurrent
sheet-like lobes (Jaca section)(see Remacha and Fernandez 2003).
However, only 50% of the lobe beds,
principally the thickest and coarsest-grained beds, extend
downcurrent tothe basin plain (see also Fig. 7). Statistically,
lobe beds thicker than10 cm, containing a clean sandstone division
thicker than 6 cm, havea 95% probability of reaching the basin
plain. Conversely, most of thethin lobe beds thin downcurrent and
wedge out before reaching the basin-plain element; the rate of loss
is about 2.3 percent of thin beds perkilometer. Also, 50% of the
beds that reach the basin plain thickenthrough the lobe-basin plain
transition, this being accomplished mainlyby means of the
homogenized Fm-2 to Fm-4 intervals (i.e., the
dirtysandstonesiltstone and shale divisions; see Table 1; Figs. 4,
7, 12).
The most striking feature in the cross section in Figure 4 is
the even andparallel correlation pattern that extends across both
elements. Thispattern results from the downcurrent thickening of
the thickest beds,which compensates for the downcurrent thinning to
wedging-out trend ofthe thinnest beds (see also Fig. 12). This
highest-frequency mechanism,related to flow reflection, provides a
possible explanation of thecompensation process.
Lateral Facies Evolution
Primary facies divisions are in accordance with Muttis (1992)
faciestract (Fig. 8A, B). Thick (. 10 cm) beds in lobes are
characterized bya high-concentration flow facies interval (F5 to F8
divisions) overlain bydilute-flow deposits (F9 interval). The
latter typically consists of a Tcdivision sharply overlain by a Te
division (Fig. 13). Toward the basinplain, beds progressively lose
the basal high-concentration flow depositsand consist mainly of F9
beds (Tbe to Tce) capped by a hemiturbiditeinterval, or of
top-missing F9 beds (Tbc to Tcd) capped by a secondaryfacies
interval (Fig. 13). Naming beds after their basal facies, the F5
(i.e.,the beds whose basal part is formed of an F5 division), F7,
and F8 bedstend to maintain or slightly decrease in frequency
toward the basin plain.In contrast, F9 (Tc) beds show a marked
decline in frequency toward thebasin plain, accompanied by a
concomitant increase in F9 (Td) beds(Fig. 14). Hemiturbidite beds
in the basin plain are more frequent in theAragues Section,
possibly as a consequence of the marginal position ofthis section
with respect to the basinward, but more axial, Anso
Section.Bed-by-bed correlation shows that lobe beds with
high-density turbidity-current facies (F5 to F8 beds) are more
prone to reach the basin plain,while F9 lobe beds deposited from
low-density currents tend to wedge outbefore reaching the basin
plain (see also Remacha and Fernandez 2003).
Reflected facies, as they become more common toward the basin
plain,vary in character (see Fig. 13); Fm-2 to Fm-4 deposits
acquire moreimportance with respect to Fm-1 divisions. The latter
also tend topreferentially overlie dense primary-flow facies (F5
through F8), i.e., Fm-1 deposits tend to appear in the thicker
beds. Moreover, bed-by-bedcorrelation shows that the ability of
flows to undergo reflection uponreaching the basin plain may be
linked to flow volume and momentum.The larger flows depositing F5,
F7, most of F8 and the thicker (. 12 cmthick) Tce beds in the lobes
underwent reflection in the basin plain, whilethe smaller and more
dilute flows depositing thinner Tce and Tde lobebeds did not,
provided that they reached the basin plain.
DISCUSSION
Processes in Topographically Driven Modified Flows
Considering the evolution of flows, two groups can be
distinguished:simple waning flows, which evolve downcurrent
strictly following Muttis(1992) facies tract, and composite
(pulsing) waning flows, which do not(Fig. 8A, C). Simple waning
flows are the smaller and more dilute flowsthat have an incomplete
basin coverage and do not display any evidenceof reflection
processes. This type of flow is recorded mainly in sheet-likelobes
essentially by F9 beds, which thin downcurrent and wedge outeither
in the transition to basin plain or within the latter element.
814 E. REMACHA ET AL. J S R
-
Composite waning flows are the large-volume and high-density
flows.They have a high to complete basin coverage and the beds that
theydeposited are the main component (volumetrically) of the
basin-plainelement. In the sheet-like lobes they behaved as simple
waning flows,being recorded by beds that on average are 20 cm thick
and contain36.5% of the dense-flow facies (F5 to F8). Composite
waning flowsrecord a spatial change after encountering a
topographic obstacle. Thisobstacle is represented by the southern
foreland-margin ramp, assuggested by the distribution of Fm-1
paleocurrents, which mainlyspread radially away from it (Fig. 9).
After encountering the southernbasin margin, downdip of the
sheet-like lobes, the flows experienceddecoupling of a lower
higher-density part and an upper lower-density part(see Kneller and
McCaffrey 1999). The lower part was deflected to flowparallel to
the basin margin, as deduced from the sole-mark paleocurrents
parallel to the ramp trend (Figs. 9, 15), and continued evolving
witha simple waning behavior. The upper lower-density part of the
flows wasreflected on the ramp, changed their properties to
composite waning, andgave rise to the modified-flow facies. Mainly
on the basis of publishedexperimental data (see references above),
we can gain insight into theseprocesses.
At the foreland margin, a substantial part of the mud-rich,
low-densityturbulent suspension forming most of the thickness of
the forward flowwas forced to travel obliquely up the ramp and then
to be reflected downthe ramp (Fig. 15). During this process, some
entrainment of fine-grainedsediment from the distal carbonate ramp
(Burgui Marls) could have takenplace. At the ramp toe, this process
led to the formation of a bulge ofdenser fluid that was fed
primarily by the flow coming back from theramp but also by the
residual forward flow approaching the foreland
FIG. 13.Markov chain analysis of verticalfacies transitions for
the four discussedsections. Analysis has been done following
themethod of Powers and Easterling (1982).Numbers denote residuals
[(observedtransitions fitted transitions)/sqr fittedtransitions] of
which the positive ones areshown. x2 testing is statistically
significant atthe 99.9986% level. The preferred transitionsdisplay
the highest residuals and are shownwith bold lines.
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 815J S
R
-
margin (Edwards et al. 1994). After a time, the bulge propagated
awayfrom the ramp as a bore (gravity current, or moving internal
hydraulicjump, undercutting the residual forward flow) advancing on
the newlydeposited sediments, and a new bulge started to be formed
(Edwards et al.1994). With time a series of bores with an overall
dispersive behavior wasproduced (Kneller et al. 1997).
The paleocurrents directed to the south (Fig. 11C; see also
Figs. 4 and9) suggest that the flow reflected from the southern
foreland margin,upon encountering the northern margin (poorly
developed submarineorogenic wedge formed of the LarraBoltana plus
Lakora thrusts), wasalso reflected from there by means of weaker
bores. Successive reflectionsby both margins and also from the
western closure of the basin (structural
rise related at least to the OrozBetelu thrust) led to the
completeblocking of the currents, i.e., to ponding.
The suite of secondary facies allows us to deduce that the
strongestrecorded bores are of undular type (type A bores; see
above). Fm-1records the stronger (leading) recorded bores,
reflected mainly from thesouthern margin. Fm-2Fm-3 would result
from the later intermediatebores, from leading bores associated
with less powerful flows, or fromsecond-generation, north-derived
reflections, as shearing below someFm-2 divisions suggests (Fig.
11C). Finally, Fm-3, when forming the baseof an elemental unit and
the uppermost Fm-4 divisions within a bed(Fig. 11D), would
represent the weakest bores, deriving from anyconfining margin. In
spite of the usual synsedimentary deformation and
FIG. 14.Pie diagrams showing thecomposition of each studied
section in terms ofbeds as named after their basal facies (see
textfor details). Notice the downcurrent evolutionfrom lobe (Jaca
section) to basin plain (Ansosection) elements. F9 beds have been
split intosubcategories according to the basal subfacies(cf. Table
3).
FIG. 15.Paleogeographic sketch map showing the dispersal pattern
of both axial and reflected parts of large-volume flows (compare
with Figs. 1B, C, and 9; see textfor details). Bi 5 Binies thrust,
LkCo 5 LakoraCotiella thrust, LaBo 5 LarraBoltana thrust.
816 E. REMACHA ET AL. J S R
-
the multi-source reflections, the number of Fm-1 and Fm-2
divisionssuggests that few (24) stronger (recorded) undular bores
were generated(cf. Edwards 1993; Kneller et al. 1997). Furthermore,
the increasingthickness of the quiescent-period deposits (Fm-4)
suggests that thestronger waves were progressively more spaced.
This agrees with theexperimental results of Kneller et al.
(1997).
Facies-Tract Variation
Topographically driven changes of flow properties suggest a new
faciestract relative to Muttis (1992) (Fig. 8). This variation
concerns the fine-grained facies group. It depends strictly on (1)
high flow volume and (2)a lower, sand-laden, high-density part,
relatively well segregated froma thicker upper, muddy, low-density
part. This facies tract overliesa sharply truncated standard Muttis
(1992) facies tract up to Bouma Tc(F9) and combines facies from
Fm-1 to Fm-4. It is composed of a pulsingsequence comprising the
stacking of divisions Fm-1, Fm-1Fm-4, Fm-2Fm-3Fm-4, Fm-3Fm-4, and
Fm-4 (Fig. 8C) until the completeexhaustion of the flow by means of
the final sedimentation ofa hemiturbidite division cap. Because of
synsedimentary deformation,the deposit overlying Fm-1Fm-4 becomes
reorganized as a muddysandstonesiltstone grading into a mudstone
cap, in which the muddysandstonesiltstone distinctively contains
small soft-sediment deforma-tion structures as floating intraclasts
(Figs. 8D, 11A, B).
Analogs to these facies and to some of the associated
deformeddeposits have been described in other basins from the
Paleozoic to theQuaternary and interpreted as the product of
flow-reflection processes(e.g., Ricci Lucchi and Valmori 1980;
Pickering and Hiscott 1985;Pickering et al. 1989; Marjanac 1990;
Porebski et al. 1991; Haughton1994; Edwards et al. 1994; Kneller
and McCaffrey 1999; McCaffrey andKneller 2001; Roveri et al. 2002).
In some of these cases, a closeresemblance to the facies described
here has been pointed out (Roveriet al. 2002). However, it is
important to emphasize that the ideal faciestract as well as
possible local variations would depend not only on flowfeatures but
also on the degree of evolution of the flow at the point whereit
meets a topographic obstacle, among other factors (see Simpson
1987;Edwards 1993). In summary, the study of flow-reflection
processes andthe resulting facies tract must be considered case by
case. In our study,flows met the ramps obliquely after a distance
varying between some 60and 90 km from the transfer zone. The
sediment load calculated for someof the most outstanding flows
extending the entire length of the basin(between Boltana and near
Pamplona) also allows them to be classified asmud-rich flows (about
3540% sand load).
Frequency of Large-Volume Flows and Maintenance of the Flat
Floor
Within the depositional elements of sandy stages, the sheet-like
lobesdisplay the highest preservation potential of beds. Upcurrent,
in thechannel-lobe transition, this preservation potential is lower
because oferosional processes. Downcurrent from sheet-like lobes,
the thin bedsrelated to small-volume flows tend to wedge out (see
above). Asa consequence, the ratio of thick beds to thin beds in
the sheet-like lobesrecords the frequency of large-volume flows in
the sandy stage, i.e.,frequency of flows that may change their
properties and build the basinplain. This ratio is about 1:4,
shifting to 1:2 in the proximal parts of basinplain.
Therefore, given the sheet geometry of the system and the
continuitybetween sheet-like lobes and basin plain, the main effect
of depositionfrom reflected flows on the floor morphology is to
compensate whatevertopographic lows are created by the stacking of
a discrete number ofsuccessive thin beds. Moreover, the complete
blocking of the currents bythe three confining margins (ponding)
led to the regulation of the flat-topped growing pattern, extending
even as far back as the sheet-likelobes. In the latter element, the
upper parts of the muddy divisions (partof the Te and the
hemiturbidite) of the thick beds (. 10 cm) containinghigh-density
facies may therefore be the final result of ponding thatcontributed
substantially to the overall sheet-like character across
thedepositional zone.
CONCLUSIONS
Basin-plain and lobe elements are closely related because all of
thebasin-plain beds are found in sheet-like lobes. At least 50% of
the flowsbuilding the sheet-like lobes reach the basin-plain
element. These are thelarger turbidity currents, which have a lower
high-density part. Incontrast, the smaller, low-concentration
currents, after having themaximum preservation potential in the
sheet-like lobes, wedge outdowncurrent within the transition to
basin plain or within the latterelement, having a loss ratio of
about 2.3 percent of thin beds perkilometer.
Beds in sheet-like lobes follow Muttis facies tract, i.e.,
correspond tosimple waning flows, whereas 36% of the beds in the
basin plain do not.These basin-plain beds account for 78% of the
volume of the element,forming the bulk of the distal basin
environment, and correspond tocomposite waning flows. This flow
behavior is due to the topographicallydriven changes of
primary-flow properties and provides the diagnosticfeatures of the
basin plain.
TABLE 4.Diagnostic criteria of lobe vs basin-plain beds.
Lobe Basin Plain
Facies in thick beds F5-F7-F8-F9 (Tb-c/e or Te). Reflected flow
(Fm)facies may be present atop exceptionally thick beds.
Rare F5 and F7. F8-F9 and F9 (Tb-e). F9 restricted toTb-c.
Reflected flow (Fm) facies overlie F5 to Tc divisions.
Facies in thin beds F9 (from Tb-e to Te) F9 (from Tb-e to Te)
Some beds display Fm facies atopTc divisions.
Top of sandstone divisions Sharp. Commonly flat, or with thin,
poorlydeveloped ripples. Small (wavelength, 10 cm)2D to 3D
(linguoid) current ripples
Gradational. Always rippled. Large (wavelength , 2 m),sinuous to
3D (hummocky-like), slightly asymmetricalto symmetrical
combined-flow ripples
Siltstone divisions Absent. May be present in ripple troughs
Always presentShale divisions Usually thin (high net-to-gross
ratio). Thick for
exceptionally thick bedsUsually thick (low net-to-gross
ratio)
Calcilutite divisions Rare to absent CommonMiscellaneous Locally
frequent amalgamation of sandstone
divisions by scouring (tabular erosional featuresand local
impact features; Mutti and Normark1987). Scour marks.
Rare amalgamation of sandstone divisions. Tool marks
Criteria in italics are from the present authors: cf. Table 1
and Fig. 8 (reprinted from Remacha and Fernandez 2003; with
permission from Elsevier).
THE TRANSITION BETWEEN SHEET-LIKE LOBES AND BASIN PLAIN 817J S
R
-
At the southern foreland margin, the large, axially evolving
flows wereforced to change their properties as follows: a lower
sand-laden high-density part underwent deflection, flowing
downcurrent parallel to theforeland margin. This part deposited
high-density facies following Muttisfacies tract. The upper part,
more dilute and thicker, was reflected fromthe foreland-margin
ramp, generating a train of undular bores that laterunderwent
multiple flow reflections by the flanking margins, i.e.,
thesouthern foreland margin, the northern margin in the poorly
developedsubmarine orogen and the western closure of the basin. As
a result, thecomplete blocking of the currents (ponding) prevented
flow-out fartherthan the western boundary of the south-central
Pyrenees turbidite basin,indicating that no connection between the
Hecho and the Bay of Biscaydeep-water turbidite basins existed.
The passage of bores is recorded by means of four main facies
typesthat form a pulsing facies sequence combining different
facies: Fm-1,Fm-1Fm-4, Fm-2Fm-3Fm-4, Fm-3Fm-4, and Fm-4, each of
whichmay be repeated. The passage of bores may have produced
vibrationliquification by cyclic wave loading. As a consequence,
syndepositional toimmediately postdepositional soft-sediment
deformation destroyed theoriginal appearance of the deposit
overlying the first relatively thick(greater than a few
centimeters) Fm-4 division. As a result, an overallgraded muddy
sandstone containing distinctive small intraclasts of Fm-2and Fm-3
divisions (pseudonodules and fragments of very thin laminae)merges
into a thick mudstone cap. In the muddy sandstone, the
floatingintraclasts become smaller upward until disappearing.
Calcilutite divisions reported in previous literature as true
hemipela-gites are here envisaged as the product of hydraulic
sorting of carbonateparticles forming the residual sediment load,
which settles at the end ofthe event. Therefore, calcilutites have
a turbidite origin (hemiturbidites).
The bed-by-bed correlation has shown an overall sheet-like
stackingpattern extending between sheet-like lobe and basin-plain
elements.Reflection processes have been envisaged as the factor
responsible forbalancing any topographic low produced by
small-volume flows. Asa result, reflection processes related to the
relatively frequent large flowscontrol the sheet-like aggradational
pattern in both elements studied. Thefrequency is 1 thick bed to 4
thin beds in the sheet-like lobes and 1:2 in thebasin plain.
The relationships between sheet-like lobes and basin plain
presentedhere and the new set of facies described permit an update
of the classicaldiagnostic criteria for recognition of lobe vs.
basin plain (Table 4; see alsoRemacha and Fernandez 2003; cf. Mutti
and Ricci Lucchi 1972, 1975;Mutti 1977; Mutti and Johns 1979).
ACKNOWLEDGMENTS
This research was supported by the Direccion General de
InvestigacionCientfica y Tecnica of the Spanish government (Project
PB94-1312-C02). Wethank A. Gardiner, T.A Hickson, B. Kneller, and
D. Mohrig for theirthorough review of this manuscript, which led to
much better organizationand exposition of data and interpretations.
Editorial work by J.B. Southard isalso highly appreciated.
REFERENCES
ARNOTT, R.W., AND SOUTHARD, J.B., 1990, Exploratory flow-duct
experiments oncombined-flow bed configurations, and some
implications for interpreting storm-event stratifications: Journal
of Sedimentary Petrology, v. 60, p. 211219.
BAAS, J., 1994, A flume study on the development and equilibrium
morphology ofcurrent ripples in very fine sand: Sedimentology, v.
41, p. 185209.
BANERJEE, I., 1996, Populations, trends, and cycles in
combined-flow bedforms: Journalof Sedimentary Petrology, v. 66, p.
868874.
BOUMA, A.H., 1962, Sedimentology of some Flysch deposits; a
graphic approach tofacies interpretation: Amsterdam, Elsevier, 168
p.
BRADLEY, J.B., 1986, Hydraulics and bed material transport at
high fine suspendedsediment concentrations [Ph.D. Thesis]: Colorado
State University, Fort Collins,Colorado, U.S.A., 140 p.
CAMARA, P., AND KLIMOWITZ, J., 1985, Interpretacion geodinamica
de la vertientecentro-occidental surpirenaica (cuencas de
Jaca-Tremp). Estudios Geologicos, v. 41,p. 391404.
CAMPBELL, C.V., 1967, Lamina, laminaset, bed and bedset:
Sedimentology, v. 8, p. 726.EDWARDS, D.A., 1993, Turbidity
Currents: Dynamics, Deposits and Reversals: Berlin,
Springer-Verlag, Lecture Notes in Earth Sciences, v. 44, 175
p.EDWARDS, D.A., LEEDER, M.R., BEST, J.L., AND PANTIN, H.M., 1994,
On experimental
reflected density currents and the interpretation of certain
turbidites: Sedimentology,v. 41, p. 437461.
FACI PARICIO, E., CASTIELLA MURUZABAL, J., DEL VALLE DE
LERSUNDI, J., GARCIA DEDOMINGO, A., DIAZ DE NEIRA, A., SALVANI
DURAN, J.M., CABRA GIL, P., RAMREZ DELPOZO, J., AND MELENDEZ HEVIA,
A., 1997, Mapa Geologico de Navarra. Memoria:Gobierno de Navarra,
Departamento de Obras Publicas, Transportes y Comuni-caciones,
Servicio de Obras Publicas, 142 p.
FAUGE`RES, J.-C., CREMER, M., GONTHIER, E., NOEL, M., AND
POUTIERS, J., 1984, LateQuaternary calcareous clayeysilty muds in
the Obock Trough (Gulf of Aden):hemipelagites or fine-grained
turbidites?, in Stow, D.A.V., and Piper, D.J.W., eds.,Fine-Grained
Sediments: Deep-Water Processes and Facies: Geological Society
ofLondon, Special Publication 15, p. 209222.
FISHER, R.V., 1983, Flow transformations in sediment gravity
flows: Geology, v. 11, p.273274.
FONTANA, D., ZUFFA, G.G., AND GARZANTI, E., 1989, The
interaction of eustacy andtectonism from provenance studies of the
Eocene Hecho Group Turbidite Complex(South-Central Pyrenees,
Spain). Basin Research, v. 2, p. 223237.
GARCA, M.H., AND PARKER, G., 1993, Experiments on the
entrainment of sediment intosuspension by a dense bottom current:
Journal of Geophysical Research, v. 98, p.47934807.
HAUGHTON, P.D.W., 1994, Deposits of deflected and ponded
turbidity currents, SorbasBasin, southeast Spain: Journal of
Sedimentary Research, v. A64, p. 233246.
HESSE, R., 1975, Turbiditic and non-turbiditic mudstone of
Cretaceous flysch sections ofthe East Alps and other basins:
Sedimentology, v. 22, p. 387416.
HESSE, R., AND CHOUGH, S.K., 1980, The Northwest Atlantic
Mid-Ocean Channel of theLabrador Sea: II. Deposition of parallel
laminated levee-muds from the viscoussublayer of low density
turbidity currents: Sedimentology, v. 27, p. 697711.
HISCOTT, R.N., 1994, Traction-carpet stratification in
turbidites-fact or fiction?: Journalof Sedimentary Research, v.
A64, p. 204208.
HISCOTT, R.N., PICKERING, K.T., AND BEEDON, D.R., 1986,
Progressive filling ofa confined Middle Ordovician foreland basin
associated with the Taconic Orogeny,Quebec, Canada, in Allen, P.A.,
and Homewood, P., eds., Foreland Basins:International Association
of Sedimentologists, Special Publication 8, p. 309325.
KNELLER, B.C., 1995, Beyond the turbidite paradigm: physical
models for deposition ofturbidites and their implications for
reservoir prediction, in Hartley, A.J., and Prosser,D.J., eds.,
Characterisation of Deep Marine Clastic Systems: Geological Society
ofLondon, Special Publication 94, p. 3149.
KNELLER, B.C., AND BRANNEY, M.J., 1995, Sustained high-density
turbidity currents andthe deposition of thick massive sands:
Sedimentology, v. 42, p. 607616.
KNELLER, B., AND BUCKEE, C., 2000, The structure and fluid
mechanics of turbiditecurrents: a review of some recent studies and
their geological implications:Sedimentology, v. 47(Supplement 1),
p. 6294.
KNELLER, B., AND MCCAFFREY, W.D., 1999, Depositional effects of
flow nonuniformityand stratification within turbidity currents
approaching a bounding slope: deflection,reflection, and facies
variation: Journal of Sedimentary Research, v. 69, p. 980991.
KNELLER, B., AND MCCAFFREY, W.D., 2003, The interpretation of
vertical sequences inturbidite beds: the influence of longitudinal
flow structure: Journal of SedimentaryResearch, v. 73, p.
706713.
KNELLER, B.C., BENNETT, S.J., AND MCCAFFREY, W.D., 1997,
Velocity and turbulencestructure of density currents and internal
solitary waves: potential sediment transportand the formation of
wave ripples in deep water: Sedimentary Geology, v. 112,
p.235250.
KNELLER, B.C., EDWARDS, D.A., MCCAFFREY, W.D., AND MOORE, R.,
1991, Obliquereflection of turbidity currents: Geology, v. 19, p.
250252.
KOMAR, P.D., 1985, The hydraulic interpretation of turbidites
from their grain sizes andsedimentary structures: Sedimentology, v.
32, p. 395407.
KUENEN, P.H., 1964, Deep-sea sands and ancient turbidites, in
Bouma, A.H., andBrouwer, A., eds., Turbidites: Amsterdam, Elsevier,
p. 333.
KUENEN, P.H., 1967, Experimental transportation and deposition
in circular flume:Report to the 7th International Sedimentological
Congress, 4 p.
LABAUME, P., 1983, Evolution tectono-sedimentaire et
megaturbidites du Bassinturbiditique Eocene sud-Pyreneen
[unpublished Ph.D. Thesis]: Universite des Scienceset Techniques du
Languedoc, Montpellier, France, 195 p.
LABAUME, P., MUTTI, E., AND SEGURET, M., 1987, Megaturbidites: a
depositional modelfrom the Eocene of the SW-Pyrenean Foreland
Basin, Spain: Geo-Marine Letters,v. 7, p. 91101.
LEEDER, M., 1999, Sedimentology and Sedimentary Basins; From
Turbulence toTectonics: Oxford, U.K., Blackwell, 592 p.
LOWE, D.R., 1982, Sediment gravity flows: II. Depositional
models with specialreference to the deposits of high-density
turbidity currents: Journal of SedimentaryPetrology, v. 52, p.
279297.
MARJANAC, T., 1985, Composition and origin of the megabed
containing huge clasts,flysch formation, middle Dalmatia,
Yugoslavia (abstract): International Associationof
Sedimentologists, 6th European Regional Meeting, Lleida, Spain,
Abstracts andposter abstracts, p. 2703.
818 E. REMACHA ET AL. J S R
-
MARJANAC, T., 1990, Reflected sedimentary gravity flows and
their deposits in the flyschof Middle Dalmatia: Sedimentology, v.
37, p. 921930.
MCCAFFREY, W.D., AND KNELLER, B.C., 2001, Process controls on
the development ofstratigraphic trap potential on the margins of
confined turbidite systems, and aids toreservoir evaluation:
American Association of Petroleum Geologists, Bulletin, v. 85,p.
971988.
MCCAVE, I.N., AND JONES, K.P.N., 1988, Deposition of ungraded
muds from high-density non-turbulent turbidity currents: Nature, v.
333, p. 250252.
MIDDLETON, G.V., 1967, Experiments on density and turbidity
currents: III. Depositionof sediments: Canadian Journal of Earth
Sciences, v. 4, p. 475505.
MUTTI, E., 1977, Distinctive thin-bedded turbidite facies and
related depositionalenvironments in the Eocene Hecho Group
(South-Central Pyrenees, Spain):Sedimentology, v. 24, p.
107131.
MUTTI, E., 1979, Turbidites et cones sous-marins profonds, in
Homewood, P., ed.,Sedimentation Detritique (Fluviatile, Littorale
et Marine): Universite de Fribourg,Institut de Geologie, p.
353419.
MUTTI, E., 1985, Turbidite systems and their relations to
depositional sequences, inZuffa, G.G., ed., Provenance of Arenites:
Dordrecht, The Netherlands, Reidel,NATO Advanced Study Institutes
Series, Series C, Mathematical and PhysicalSciences, v. 148, p.
6593.
MUTTI, E., 1992, Turbidite sandstones: AGIP-Istituto di Geologia
Universita` di