Sequence stratigraphic interpretations in the southern Dead Sea basin, Israel
Istvan Csato* Christopher G. St. C. Kendall† Alan E. M. Nairn
University of South Carolina, Department of Geological Sciences, Columbia, South Carolina 29208
Gerald R. Baum§
Rice University, Department of Geology and Geophysics, Houston, Texas, 77005-1892
}
ABSTRACT A chronostratigraphic framework has been developed to date the late Tertiary and Pleistocene fill of the southern Dead Sea basin. This framework is based on a sequence stratigraphic interpretation of seismic data tied to eustatic sea-level curves. The limited biostratigraphic data available for the Dead Sea basin stratigraphy make this chronostratigraphy tentative, but nine third-order sequence boundaries have been identified and related to climatically driven lake-level falls that appear to correlate with eustatic events. In contrast, higher-order sequence boundaries were interpreted to reflect changes in both climate and local tectonics and the consequent sediment supply. The seismic stratigraphic interpretation indicates that fan-delta sediments accumulated at the southern and northern ends of the basin, while lacustrine clastic sediments filled the rest. Interfingering of southern and northern source sediments has been recognized in the northern part of the basin. The changing position of interfingering was interpreted as a response to lake-level changes, the lake-level falls inducing increased sediment flux from the northern source, and lake-level highstands favoring the influx and progradation of sediments from the south. INTRODUCTION The purpose of the study reported in this paper was to interpret the Pleistocene sequence *Present address: MOL Hungarian Oil & Gas Company, Ltd., International Exploration and Production Division, Oktober 23 u. 18, Budapest 1117, Hungary. †Corresponding author; e-mail:
[email protected] §Present address: Texaco, 4800 Fournace Place, Bellaire, Texas 77401.
stratigraphy of the southern Dead Sea basin and to determine (1) if the sedimentary section could be dated by relating the major unconformities seen on seismic profiles and in outcrop to thirdorder eustatic changes, and (2) the character of the sedimentary fill in the subsurface portion of the basin.
migrated seismic lines and palynologic age data from four wells (Fig. 2). These latter ages were defined by Horowitz (1987). Seismic velocity data used for seismic traveltimedepth conversions are those of ten Brink and Ben-Avraham (1989) and Yuval Ben-Gai (1996, personal commun.).
Significance of This Study
GEOLOGIC SETTING
This study provides new insight into the use of sequence stratigraphy to interpret transitional basins. It describes how the sediments that fill these basins can be subdivided into sequences by unconformities and how, through a mix of paleontological and radiometric dates and published sea-level charts, the stratigraphic ages of the sequences can be inferred. It also shows how the geometric relationships of the seismic reflectors can be used to determine how these basins may have been filled as they cycled through changes in base level (lake surfaces) and sedimentation.
Structure of the Dead Sea–Jordan Rift System
Study Area and Data The southern Dead Sea basin is located in Israel (Fig. 1). The southern boundary of the basin and the study area is the steep Iddan fault (ten Brink and Ben-Avraham, 1989). The western boundary of the basin and the study area is the western border fault, which is expressed topographically as a surface escarpment. The eastern edge of the study area is the political boundary between Israel and Jordan, coinciding with the part of the deepest basin. To the north, the southern shore of the Lisan peninsula constitutes the northern edge of the study area. The study area is divided into two distinct portions, southern and northern, by the Amaziahu fault, and stratigraphic correlations were determined separately for these two areas. The data set used in the study consisted of 10
GSA Bulletin; November 1997; v. 108; no. 11; p. 1485–1501; 20 figures.
1485
The Dead Sea–Jordan rift is the northern branch of the extensive African-Levant rift system. The rift links the spreading Red Sea with the collision zone of the Taurus. The rift zone can be subdivided into rift segments by north-south– oriented, en echelon, strike-slip faults. The eastern master faults die out northward, and the western master faults continue their northward trend (Kashai and Croker, 1987). The Dead Sea, Lake Kinneret, and Hula basins are confined between the parallel strike-slip faults (Fig. 1). Three tectonic models have been proposed to explain the origin and structural evolution of the Dead Sea rift zone. The first is the graben model, which considers the Dead Sea rift zone as a full graben bounded by normal faults having vertical displacements (Picard, 1967, 1970). The second assumes that the rift originated from the leftlateral movement of the Arabian plate relative to Africa and Sinai, and that the Dead Sea formed as a strike-slip basin (Quennell, 1958). This model has become the most widely accepted (Freund et al., 1970; Zak and Freund, 1981). A third model combines the earlier two models, suggesting that northwest-southeast tension opened the basin diagonally along the strike-slip faults (Vroman, 1973; Horowitz, 1979; Bahat and Rabinovitch, 1983). The seismic and well data now available provide new insights into the structure and stratigraphy of the Dead Sea rift. A considerable thick-
CSATO ET AL.
Stratigraphy of the Dead Sea Area 35
40 re Sutu gros a Z us Taur
Thrust Fold axes 0
200 km
Cyprus
35
Hula
ne Zo ra y lm Pa
Levant Basin
Lake Kinneret
3 5o 3 0 ISRAELI -JORDANIAN BORDER
Cypr ian A rc
Jerico
Jerusalem
Dead Sea
Dead Sea
Ar
no
3 1o 3 0
3 1o 3 0
nS
ink
Study area
30
Sinai
qeq
ult
3 1o 0 0
Sca
rp
a
Fa
Rift
Am
azi
ISRAEL -JORDAN I IA BORDER N
Stud
Red Sea
y Are
3 1o 0 0
Bo
Mt. Sedom
Gulf
of Ela
t
Lisan Peninsula
Idd
an
ahu
Fa
ult
Fa
ult
0
10km
3 5o 3 0
Figure 1. Generalized tectonic map of the Jordan–Dead Sea rift zone (after Kashai and Croker, 1987; Ben-Gai and Ben-Avraham, 1995). Inset is a structural map of the Dead Sea (after ten Brink and Ben-Avraham, 1989).
ness of the Neogene strata has been documented (5–10 km), and many of the structural elements in the basin can now be explained (Neev and Hall, 1979; Ginzburg and Kashai, 1981; Croker 1983; Marcus and Slager, 1985; Kashai and Croker, 1988; Kashai, 1988). Three transverse
1486
faults in the southern Dead Sea (Fig. 2) were identified by ten Brink and Ben-Avraham (1989). These are the Iddan fault, which is the southern border fault of the basin; the Amaziahu fault, a listric fault recognized earlier; and a shear zone (Boqeq fault) south of the Lisan peninsula.
Two major groups of strata have been recognized in the Dead Sea, prerift (pregraben) and synrift (postgraben [Neev and Emery, 1967] or intrarift [Steinitz and Bartov, 1992]). The pregraben sediments are those that were deposited before the onset of major extension, and the postgraben strata constitute the basin-filling Dead Sea Group (Horowitz, 1979). The sediments that form the postgraben fill are largely terrestrial. The lack of diagnostic fauna within these continental sediments of the Dead Sea basin has caused ambiguities in the correlations between the sedimentary sequences and the seismic data of the study area. Figure 3 summarizes the major stratigraphic classifications. Most authors agree that the prerift stage of basin evolution ended after deposition of the Hazeva Formation, which represents an initial red-bed sequence. The Hazeva Formation is interpreted as prerift because it extends beyond the Dead Sea, and it is considered to be late Miocene in age (Garfunkel and Horwitz, 1966; Neev and Emery, 1967; Agnon, 1983; Manspeizer, 1985; Steinitz and Bartov, 1992). The second major evolutionary stage of the Dead Sea basin was the Pliocene–Pleistocene synrift phase. No general consensus has been reached with respect to the correlation and age dating of these Pliocene–Quaternary formations. Some authors regard the Pliocene as a single sedimentary cycle (Neev and Emery, 1967), whereas Horowitz (1979) divided the Pliocene into two parts (Tabianian and Piacenzian), and Marcus and Slager (1985) proposed a three-part division. The rock salt located in the Dead Sea basin has been interpreted to have been deposited during Pliocene time (Neev and Emery, 1967). Horowitz (1979) established the most detailed stratigraphic subdivision for the Pleistocene. He proposed a preglacial Pleistocene interval and a glacial Pleistocene interval that contains four members, matching the European glaciation events of Günz, Mindel Riss, and Würm. The last Würm stage, when the Lisan Formation was deposited, has been studied the most. This formation was deposited in a large basin that was desiccated some 18 ka (Begin et al., 1974). The last period of the Dead Sea evolution, which followed the shrinkage of Lake Lisan, is designated as the post-Lisan stage, and coincides with the formation of the present Dead Sea. The PleistocenePliocene boundary in the classification of Horowitz (1979) has an age of about 2.8 Ma (base of preglacial Pleistocene time). This age for this boundary is used in this paper. The Sedom Formation was defined first by Zak (1967) and Zak et al. (1968); this unit was described initially in the salt diapir of Mount
Geological Society of America Bulletin, November 1997
SEQUENCE STRATIGRAPHY IN THE DEAD SEA BASIN, ISRAEL
Lis
Study area
o 32
Boqe
36o
su
la
q Fa ult
Well
A
Fault & Scarp
729
M
Lake
2
Rift
Sca
rp
63
NOR THER N AR EA
14 Am
SOU THE RN 40
19
Aa
azi
ahu
22
13
Esc
arp
40
70
59
Au
70
6
40
23
7001
ARE A
0
me
nt 10 km
Iddan fault
Figure 2. Location map. The study area extends south of the Lisan Peninsula toward the Arava area. Mount Sedom is an outcrop of salt in the southwest corner of the lake. The seismic lines used in this study are indicated by their number. The four wells used for age data are A—Amiaz-1, M—Melekh Sedom-1, Au—Amaziahu-1, Aa—Arava-1. The Amaziahu fault divides the region into a southern and a northern area.
Sedom. This formation consists mainly of rock salt, sandstone, shale, dolomite, and anhydrite, which are interpreted to have been deposited in a narrow embayment of the Mediterranean Sea that extended into the Jordan–Dead Sea rift zone. Agnon (1981) found Ammonia sp., citing it as evidence for a late Neogene marine brackish-water setting in the southern Dead Sea depression. The age of the Sedom Formation was dated as late Pliocene–Pleistocene by Zak (1968). Many authors consider the Sedom Formation to represent the entire Pliocene Epoch (Neev and Emery, 1967; Manspeizer, 1985; ten Brink and Ben Avraham, 1989), and others have placed the formation in the lower Pliocene (Horowitz, 1979; Steinitz and Bartov, 1992). Agnon (1983) suggested that deposition of the Sedom salt started in Miocene time, coeval with the Hazeva Formation. Stein et al. (1994) conducted a new analysis
Garfunkel and Horowitz (1966), Horowitz (1979), Manspeizer (1985), and Marcos and Slager (1985) as the most probable cause for the unconformities, which reflect prominent changes in the style of sedimentation. As outlined in the following section, we favor the interpretation that these unconformities were produced by changes in lake level. METHODS OF STUDY
Seismic Line Mt. Sedom
Au
nin
DEAD SEA
KEY 729
Pe
3
o 28
Sinai Plate
an
63
o 30
Arabian Plate
32o
Dead Sea
Mediterranean Sea
of the strontium isotopic composition to determine the age of the salt, which was set at 22 ± 4 Ma. This date, however, may be suspect because diagenesis may have altered the primary isotopic composition of the salt. In our study, several unconformities have been defined and are mainly interpreted to match those seen in the outcrops described in previous stratigraphic studies (Fig. 3). These unconformities are (1) the top of the Hazeva Formation; (2) one or two unconformities within the Pliocene; (3) the top of the Pliocene section; (4) four or possibly five unconformities within the Pleistocene; and (5) the top of the Pleistocene section. These unconformities probably represent the most significant chronostratigraphic events that occurred during the filling of the Dead Sea; however, their nature and origin remain controversial. Tectonic and climatic changes were ascribed by
In frontier exploration areas where there is poor biostratigraphic control, sequence stratigraphy is used to constrain the relative ages of the sedimentary fill seen in seismic sections (Brown et al., 1995). In our study of the southern Dead Sea basin, second-order sequence boundaries and (where possible) maximum flooding surfaces were identified and separated from third-order unconformities within seismic sections, well sections, and outcrops. The second-order unconformities were used to bracket the ages of thirdorder unconformities. The intervening third-order unconformities were then matched in magnitude with, and assigned to, the timing of the intervening third-order sea-level positions of Haq et al. (1987). Where possible, the dating of these features was determined from biostratigraphy, but in the cases where biostratigraphic control was lacking, we used the timing of the inflection points on sea-level charts (Haq et al., 1987; TGS Offshore Geophysical Company, 1989), bracketed by biostratigraphic dates. The internal geometries of sequences were then described in terms of their progradational, aggradational, and/or retrogradational character. The basement beneath the synrift strata and some unconformities was correlated on seismic sections and mapped across the study area. This mapping generated insight into the geometrical relationships of the stratigraphy. Qualitative interpretation of seismic reflection characteristics (wavelength, frequency, continuity, reflection geometry) allowed seismic facies analysis. The depths of age boundaries determined by palynologic analysis (Horowitz, 1987) were converted to traveltime, tied to seismic sections, and correlated along the seismic sections. In this way, a tentative determination of the Pliocene-Quaternary boundary in the basin was possible. Seismic onlap charts were constructed from the observed unconformities and correlated to eustatic curves to extend the chronostratigraphy. RESULTS Correlation of the Basement The basement of the basin fill was identified and correlated as a seismic surface that separates stratified reflectors from chaotic and less-
Geological Society of America Bulletin, November 1997
1487
CSATO ET AL.
acter, as is seen on line 4019. Basinward from the margins, the basement deepens and becomes flatter northeast of line 7014. This general trend is disturbed by basement faults between line 4019 and 4022. The Amaziahu fault is on this northeastward deepening basement slope. Few data are available north of line 7001. The deepest part of the basin is between line 729 and the northern shear zone: from here, the basement rises into the Boqeq fault zone. A similar decrease in the depth of basement is predicted beneath Mount Sedom toward the western border zone. Correlation of Sequence Boundaries
Figure 3. Summary chart and tentative correlation of formations in the Dead Sea area. G&H—Garfunkel and Horowitz (1966); N&E—Neev and Emery (1967); M—Manspeizer (1985); A—Agnon (1983); S&B—Steinitz and Bartov (1992); M&S—Marcus and Slager (1985).
continuous ones. As can be seen on line 7013 (Fig. 4, A and B), the most distinct reflector was defined as the basement. The strong reflectors extend from about 2.1 to 2.8 s in a southeastward direction. To the northwest along the western border fault, the basement rises very steeply. The northwest-southeast–oriented seismic lines show that most of the basin-filling sediments onlap this border fault, suggesting that the basin fill was synrift strata. This is clearer between the A4 and A5 surfaces, when subsidence rates appear to have been greatest. Lines 4019 (Fig. 5, A and B) and 4023 (Fig. 6, A and B) join and are oriented longitudinally. The basement on these has a steeply inclined surface; its deepest part is over the southern border fault, where it is marked by short wavelength, and there are slightly scattered reflectors on the basinward side. The morphology of the basement here, particularly on line 4019, suggests that it is composed of blocks separated by normal faults. Adjacent to the southern border fault, a small half-graben structure can be observed in the basement (Fig. 5, A and B), and there is evidence of inversion between surfaces A4 to A1. Northeast of this half
1488
graben, a full graben can be recognized, and from here the basement gradually deepens to the northeast. The faults indicated in Figure 5B probably continue beneath the A4 surface. On line 596, the basement surface is easily identifiable below the hanging wall, where strong continuous reflectors mark the basement (Fig. 7, A and B). The correlation suggests that the basement dips northeast, matching the observations of lines 4019–4023. On line 633 (Fig. 8, A and B), the basement first slightly deepens from the southwest, and then is more elevated toward the northeast. The overlying reflectors onlap the basement surface. The onlapping indicates that the tilting or erosion of the basement was prior to the deposition of the overlying layers. The reflector pattern below the basement surface is well stratified in the southwestern part of the section, but it becomes increasingly chaotic to the northeast. The chaotic pattern is attributed to the large zone of tectonically disturbed material that occupies the northeastern half of the section (Boqeq fault zone). The basement map (Fig. 9) shows that the southern border of the basin has a stepped char-
South of the Amaziahu fault, six unconformities were identified within the sedimentary fill of the basin. These surfaces, from the youngest to the oldest, are designated #A1 to #A6, respectively. The key section for their identification was line 4019 (Fig. 5, A–C). The youngest sequence boundary (#A1) is a distinct surface that separates continuous, strong reflectors from the underlying weak reflector zone. Age data from the Amaziahu-1 well (Fig. 6C) show that this unconformity corresponds with the base of the Lisan unit. The Lisan Formation and the underlying Hamarmar or Samra Formations are referred to in this paper as the Lisan unit (Fig. 3). This unit is equivalent to a suite of slightly northward dipping reflectors. A marked onlap pattern defines the sequence boundary on top of #A1. Within sequence #A1, six additional higherorder sequences having five sequence boundaries can be identified (Fig. 5, A and B). One sequence boundary was identified within sequence #A2, #A3, and #A4; in #A5 two and in #A6 three more high-order sequences were interpreted. The identification and correlation of these high-order sequences was limited where the resolution of the seismic data was poor. The most accurate correlations were made between sequences #A1 and #A2, where the reflections are quite distinct. The high-order sequences could not be correlated far across the area because the quality of the seismic data deteriorated and the many local geologic effects disrupted their continuity. Correlation across the Amaziahu fault is equivocal, so that correlations for the northern area were independent of the southern area. North of the Amaziahu fault, nine unconformities, #B1 to #B9, were distinguished. The key lines used in their correlation were 596 (Fig. 7, A and B) and 633 (Fig. 8, A and B). Seismic maps of the unconformity surfaces provided insight into the architectural geometry of the sedimentary fill. Maps for surfaces #A4 and #B2 are presented in Figures 10 and 11. The border faults to the west and southwest, the interior
Geological Society of America Bulletin, November 1997
Line 7013
SE
Line 4019
Line 4022
1.0
1.0
2.0
2.0
3.0
3.0
4.0
4.0
TWT (sec)
NW
A NW
Line 7013
SE
Line 4019
Line 4022
#A2 #A2
#A3
#A3
#A4
1.0
#A4
1.0
#A6
2.0
2.0
3.0
3.0
4.0
4.0
Fan deposits
B Well Aa
#A3
#A2 #A4
#A3 #A4 #A5
M.M
1.0
Quaternary
#A2
(sec)
SE
2.0
#A6
Miocene-Pliocene
Line 7013
U. Miocene
NW
TWT (sec)
#A5
Figure 4. (A) Seismic line 7013. (B) Interpretation of seismic line 7013. A very strong reflector contrast indicates the basement in the middle of the section. The basement elevates along a steep border fault to the northwest. The packages of reflectors bounded by dashed lines are interpreted as fan deltas. A vertical fault cuts the younger sequences. (C) Facies interpretation of line 7013. The ages indicated in the Arava-1 well were deter-mined from palynologic analyses of Horowitz (1987). The Quaternary-Pliocene boundary is assumed in this study to match unconformity #A4.
3.0 Border Fault Fans 4.0
C
Quaternary-Pliocene boundary assumed in this study
Axial Fans
0
1km
Geological Society of America Bulletin, November 1997
1489
Line 4019
Line 7014
NE
1
1
2
2
3
3
0
TWT (sec)
Line 7013
SW
5km
A Line 7013
SW
Line 4019
Line 7014
Quaternary
A0/1 A1/1
NE
#A1 A1/2 A1/3 A1/4 A1/5
#A2 Quaternary
A2/1
#A3
1
A3/1
TWT (sec)
#A4
1
#A5 A5/1 A5/2
2 #A6
A6/1 A6/2 A6/3
3
Top of Transgressive ST Top of Lowstand ST High-order unconformities 3rd order unconformities Fan deposits Basement
Miocene-Pliocene
A4/1
2
3
Figure 5. (A) Seismic line 4019. (B) Interpretation of seismic line 4019, which crosses the southern edge of the basin. The packages of reflectors stacked along the steep border fault are interpreted as fan deltas. The fanshaped packages above the basement are interpreted as basement-floor fans that interfinger with finer grained or salty lacustrine sediments basinward. A smaller half graben and full graben can be seen in the basement. (C) Facies interpretation of seismic line 4019. ST is systems tract.
Basement Trough ? 0
5km
B SW
Line 4019
NE A0/1
#A1
A1/1 A1/2
A1/3 A1/4 A1/5
#A2
A2/1
#A3
1
1
A3/1
TWT (sec)
#A4 A4/1
#A5 A5/1 A5/2
2
2 A6/1
#A6
A6/2 A6/3
3
C 1490
Lowstand ST
Fan Delta
Transgressive ST Highstand ST
Basin Floor Fan Lacustrine clastics and salt
3
High-order unconformities 3rd order unconformities
Geological Society of America Bulletin, November 1997
SEQUENCE STRATIGRAPHY IN THE DEAD SEA BASIN, ISRAEL
Salt Deposits and Formation of the Amaziahu Fault
1.0-
-2.0
2.0-
-3.0
3.0-
0
TWT (sec)
NE
-1.0
1km
A
Line 4023
NE H0/1
A0/1
#A1
#B1
A1/1 A1/2 H1/1
A1/4
#B2
A1/5
#A2
H2/1
A2/1
-1.0
H2/2
#B3
1.0H3/1
#A3
H3/2 H3/3
A3/1 F3/1
H3/4
Pliocene
#A4 A4/1 F4/1
-2.0
A5/1
H4/2
#B5
#A5
2.0A5/2
A6/1
#B4
H4/1
F5/1
#A6 #B6
A6/2
TWT (sec)
A1/3
Quaternary
Amaziahu Fault
Quaternary
SW
Miocene ?
Significant salt structures are present in the southern Dead Sea basin. The largest salt diapir is Mount Sedom. The exact extension of the salt body cannot be determined from the seismic coverage, but it seems that the thickness of the salt gradually decreases southward away from Mount Sedom. It probably diminishes between lines 7013 and 7014. The salt is present at the Amaziahu fault, but there is no direct seismic evidence that it extends to the southern margin of the basin. The northward and eastward distributions of the salt also remain unclear. Examples of this problem can be seen on lines 596 (Fig. 7, A and B) and 633 (Fig. 8, A and B). In the interpretation presented here, the #B7 surface covers the salt formation. In order to construct an accurate map of the salt, more seismic lines with well control are needed. The Amaziahu fault is thought to have resulted from a combination of salt and extensional movements. This interpretation is in accordance with the work of Neev and Emery (1967) and ten Brink and Ben-Avraham (1989). In the study reported in this paper, an initial detachment growth fault is interpreted, which started to develop above the northeastward-sloping basement (Fig. 12). The fault was initiated by the rapid extension and subsidence of the basin and the gravitational instability on the basement slope. The onset of the faulting can be traced in the stratigraphic record. Sequences above the #B7 surface within the hanging wall have an increasing thickness toward the fault (see line 596 in Fig. 7, A and B). The fault movement mobilized the salt that had been deposited early in the basin’s history. This salt started to migrate upward, primarily along the fault plane. The growth fault carried the salt up to the sediment-water interface, where it dissolved. Part of the salt was trapped in the fault
Line 4023
SW
Salt ?
?
A6/3
-3.0
F4/1
Top of Transgressive ST Top of Lowstand ST High-order unconformities 3rd-order unconformities
H1/1
Local unconformity on the footwall Local unconformity on the hanging wall
3.0-
Mio-Pliocene
faults, the Amaziahu fault, and the salt structures have had a great influence on the morphology of sequences. The surfaces form a concave, “U” form geometry that deepens toward the Amaziahu fault (see maps in Figs. 10 and 11); however, these maps are equivocal because there are few data. A thick sedimentary package pinches out against the diapir of Mount Sedom. The Amaziahu fault and its antithetic fault at the northern end of line 596 break the basin fill into two major domains. The sequence boundaries are elevated from the Amaziahu fault to the antithetic fault. Northeast of the antithetic fault (hinge zone), the surfaces tend to deepen. Both the footwall and hanging wall of the Amaziahu fault are demarked by the intersection of concave sequence boundary surfaces that are elevated away from the fault.
Basement 0
1km
B Figure 6. (A) Seismic line 4023. (B) Interpretation of seismic line 4023. This line crosses the Amaziahu fault. An exact correlation between unconformities south (#A) and north of the fault (#B) cannot be made. Local unconformities that are not correlatable across the basin are indicated as F on the footwall and H on the hanging wall. A salt diapir is probably close to the base of the fault. ST is systems tract.
Geological Society of America Bulletin, November 1997
1491
CSATO ET AL.
Line 4023
SW
Well Au
NE Lisan Hamarmar/Samra
A0/1
#A1
H0/1
#B1
A1/1 A1/2 H1/1
A1/4
#B2
Quaternary
A1/5
#A2 A2/1
-1.0 #A3
H2/1 H2/2
1.0-
#B3
H3/1
H3/2 H3/3
A3/1 F3/1
#B4 H4/1
Pliocene
H4/2
#A5
#B5
A5/1
2.0A5/2
F5/1
?
A6/1
#A6 A6/2
A6/3
Quaternary
-2.0
Miocene ?
Pliocene
A4/1 F4/1
Salt
?
#B6
Salt ? ?
? F4/1 H1/1
Local unconformity on the footwall Local unconformity on the hanging wall
High-order unconformities 3rd order unconformities
C
?
?
?
?
-3.0
3.0-
Basement 1km
0
TWT (sec)
H3/4
#A4
Mio-Pliocene
A1/3
Quaternary
Amaziahu Fault
zone as a salt diapir. This salt-withdrawal mechanism contributed greatly to the creation of accommodation space for sedimentation around the Amaziahu fault. The increased space formed by the withdrawal of material from below caused a collapse of the overlying strata. The detachment surface of the Amaziahu listric fault marks the base of, or lies within, the salt formation. This collapse feature is seen on line 596 (Fig. 7, A and B). According to the interpretation presented here, the bulk of the salt has remained close to its original position at the northern end of line 596 below surface #B6. The collapse occurred between the two ends of this section. The fault zone at the northern end of the section was a mechanical response of the collapse. The sedimentary mass was forced to fracture at the location where the salt remained at its original thickness. Seismic Facies
Figure 6. (C) Line-drawing interpretation of seismic line 4023. The ages in the Amaziahu-1 well were determined from the palynologic analyses of Horowitz (1987). The base of the Hamarmar Formation correlates well with the #A1 and #B1 unconformities. The Quaternary-Pliocene boundary cannot be correlated because it has been tectonically disturbed. Line 596
NE
Line 7001
Line 632
1
1
2
2
3
3
4
4
TWT (sec)
SW
A Line 596
SW
NE
Line 7001
Amaziahu Fault
Line 632
#B1 #B2 #B3 #B4
1 #B5
#B6
2
2 #B7
Salt ? 3
Basement
#B8
TWT (sec)
1
3
Salt ?
4
4
B Figure 7. (A) Seismic line 596. (B) Interpretation of seismic line 596. A salt diapir is interpreted within the zone of the Amaziahu fault, and a salt-clastic formation is proposed below the unconformity #B8. Series of high-angle normal faults step down northeastward in the basement. The arrows indicate the onlap terminations along unconformities.
1492
Facies in the Southern Area. Seismic lines 7013 (Fig. 4, A and B) and 4019 (Fig. 5, A, B, and C) were important to the determination of the facies assemblages in the southern part of the basin. On line 4019 (Fig. 5C), close to the southern border fault, the reflection pattern is slightly chaotic from the basement up to the #A5 surface. The boundary between this package of reflections and the elevated basement is the southern border fault. Basinward, the reflectors become continuous and easy to correlate. The facies adjacent to the border fault are interpreted as fan deltas, and the continuous reflectors are interpreted as lacustrine sand-shale facies. This interpretation is supported by the wedge-like, basinward-thinning seismic geometries adjacent to the border fault and the fact that fan deltas are common facies associations in other synrift settings (Nemec and Steel, 1988). The reflections show interfingering between the lacustrine strata and the more chaotic fan deltas; that is, six deltaic tongues interfinger with lacustrine basinal sediments. The horizontal lacustrine reflectors onlap the fan-delta tongues. The fan-delta tongues or lobes represent episodic sedimentation, which may have been a response to fluctuating sediment supply or changing base level. The correlation of sequence boundaries #A5 and #A6 was adjusted to the succession of the fan-delta tongues. Sequence boundary #A5 marks the top of the fan-delta complex. The reflections onlap and then overlap the basement to the south. This stratigraphy exhibits an overall transgressive event. Line 7013 (Fig. 4, A–C) is perpendicular to line 4019 (Fig. 5, A–C) at the position of the fan deltas. Several fan-shaped packages can be distinguished on this profile. Two fans are on the western border fault and represent proximal fans derived from the west. The remainder of the sed-
Geological Society of America Bulletin, November 1997
SEQUENCE STRATIGRAPHY IN THE DEAD SEA BASIN, ISRAEL
Part of Line 633 NE
Line 729
1
1
2
2
3
3
4
4
TWT(sec)
SW
A Part of Line 633 SW
NE
Line 729
#B1 #B2 1
1
#B3 #B4 #B5
2
2
#B7
B6/1
B7/2
B7/1
#B8 3
B8/1
3
TWT(sec)
B5/1
#B6
B8/2
#B9 B9/1
B9/2
4
4 Top of Transgressive ST Top of Lowstand ST 3rd order sequence boundary Higher order sequence boundary
Detail in Figure 16a,b
0
5km
B Figure 8. (A) Part of seismic line 633. (B) Interpretation for part of seismic line 633. The fault zone at the northeastern end of the section is the Boqeq shear zone. The strata comprise a series of progradational lacustrine sequences. ST is systems tract.
iments below #A4 fill the inner portion of the graben and may have been transported from the south, parallel to line 4019. Figure 4C shows a possible interpretation of these stacked fan-delta bodies, which are partly on the western border fault and partly fill the graben space. Interfingering of Facies in the Northern Area. Seismic section 633 provides key information on the stratigraphy of the northern area.
Figure 8B shows the sequence stratigraphic interpretation of the section. The basin-fill stratigraphy begins with the northward-prograding sedimentary unit below sequence boundary #B9. This sequence can be further divided by two higher-order packages. This unit may be partly coeval with the Hazeva Formation. Lowstands of lake level produced sediments that onlapped the clinoforms formed by the prograding delta. A
transgression covered the delta, creating a condensed section. A subsequent highstand is represented by the deposition of additional progradational sediments on top of this condensed section. The shape of the progradational body became more elongate and less sigmoidal toward the younger sequences. The younger reflectors may represent deeper water conditions in the form of lacustrine, open-water shales. Figure 13 shows the interpreted systems tracts on line 633. Truncations of time lines are observed in two places, and these probably were a response to sublacustrine erosion. Section 633 (Figs. 8B and 13) shows an interesting interfingering stratigraphy. Reflectors on the northeastern side of the section dip toward the southwest, indicating that these sediments were derived from the northeast. Figure 14 shows the interpreted interfingering zone between the two sedimentary systems together with the unconformities. The interfingering front moves back and forth through time. Two major advances and three retreats occurred before the development of the #B6 unconformity. After the formation of sequence #B6, the northeastern system gradually extended farther southward in this section. Sediment transport occurred from the southwest even after #B2 time, as indicated by the downlapping reflections above #B2 on line 596 (Fig. 7, A and B); however, in the area of line 633, the northeastern system was more dominant. The major southwestward advances of sediments took place following formation of unconformities (i.e., following lake-level falls). The first major advance occurred above unconformity #B9. The second advance followed the formation of unconformity #B8. It appears that the relative dominance of sediment supply from the two directions was linked to lake-level fluctuations. This observation can be explained by the location of this area. The northeastern edge of the basin is more proximal than the southern. Consequently, a lake-level fall initiated a greater sediment influx from the closer northern margin (Lisan peninsula) than from the southern shores. Figure 14 also shows that the relative proportions of the two sedimentary filling systems changed as a function of time. The change in the dominance of sediment supply direction is reflected in the thicknesses of the sequences. Sequences #B9 to #B6 are thicker toward the southwestern end of the section. Conversely, sequence #B4 has a uniform thickness at both ends of the section. Above #B4, the sequences are thicker toward the northeastern end than at the southwestern end of the profile. The most obvious difference in thickness is displayed in the uppermost sequence. On the basis of these variations in thickness, three major sedimentary units
Geological Society of America Bulletin, November 1997
1493
CSATO ET AL.
Seismic map of Basement 63
Normal Fault
Map: Unconformity#A4
Lis an Pe nin su la
Salt
3
Shear zone 0.8
Lis an Pe nin su la
Boqeq Fault
Contour line in seconds
3.6
Fault zone Proposed Fault Zone
3.7 3.8
Dead Sea Normal Fault
A
633
DEAD SEA
Well 0.8
Contour line in seconds
Rift sca rp
2
1
2.0
0 1.
23 40
az
iah
22
22
1.6 1.5
uF
au
lt
1.4
1.
2.5
13
Am
1.7
1
70
3.0
19 40
0.5
40
13
1.8
40
40
0.4
3.1
Au
19
14
0.3
40
23
2.9 2.8
0 0 0.9 0.8 0.7 .6 .5
70
3.0
1.5
14
2.7
70
M
2
596
596
3.5
700
63
Border Fault
3.7
1.0
70
Mt. Sedom
3.7 3.6
Mt. Sedom
63
3.8
1
729
A
3.5
2.5
700
3.0
Bord er Fa ult
Rift Sc arp
729
1.3
3.3
1.2
10 km
0
1.3
Aa
1.1 1.1
1.0
0
10 km
0.
0.8
9
0. 7
Figure 9. Seismic reflection map of the basement of the graben fill. Iso–two-way traveltime lines are in seconds.
can be defined that capture the effects of the relative dominance of the two sedimentary filling systems. Unit I is nearly identical to sequence #B9, the bulk of the sediments having been derived from the southwest. Unit II is composed of sequences #B8 to #B6 and reflects the time when the southwestern system was still dominant, and the northeastern was also important. Unit III encompasses the sequences above the #B6 unconformity, and marks the third stage of basin fill history, when the dominant sediment transport was from the northeast. Seismic Facies in the Northern Area. The quality of section 633 allows seismic facies analysis along this profile. A summary of this facies classification is shown in Figure 15. The basis of the classification combines the sediment transport direction, the systems tract position within a sequence, and the seismic reflection character. The two sedimentary filling systems (transported from
1494
Figure 10. Seismic map of unconformity #A4. Note that the iso–reflection time lines form a concave surface that deepens toward the Amaziahu fault on its footwall side.
the south and north) are clearly distinguished, and they interfinger in the syncline between them. In the southern system, two types of lowstand facies are recognized. Ls1 (lowstand, southern origin) consists of mounded, short-wavelength reflections, which may represent basal fans. Facies Ls2 is not mounded, but is composed of continuous reflections and is situated in the syncline. This facies is present only in the younger sequences. In the northern system, only one lowstand facies, Ln1 (Lowstand, northern origin), was recognized by chaotic, short-wavelength reflections. Transgressive system tracts in the southern system always consist of a condensed section (Ts1). In the northern system, transgressive system tracts can be either a thick backstepping package (Tn1) or a condensed section (Tn2). Two main end members of facies are present in the highstand systems tract of the southern sys-
tem. One has a mounded shape and progradational reflection patterns (Hs1). The length of the reflections is always greater than that of facies Ls1. The other highstand facies is the long, continuous, flat sigmoidal progradational reflection assemblage (Hs2). This facies is present in the upper sequences. On the opposite side, the northern system contains three highstand facies: a condensed section (Hn1), a short wavelength progradational facies (Hn2), and a long wavelength progradational facies (Hn3). Facies Hn3 and Hs2 are identical. Figure 16 (A and B) illustrates facies distribution in a part of section 633. At the bottom of the sedimentary fill, the first facies is the short-wavelength fan-shape reflector package defined as the lowstand systems tract (Ls1). The reflections have a progradational character suggesting a prograding fan-complex facies. The highstand systems tract in the lowest sequence is also com-
Geological Society of America Bulletin, November 1997
SEQUENCE STRATIGRAPHY IN THE DEAD SEA BASIN, ISRAEL
Map: Unconformity#B2
0.4 0.2
Lis an Pe nin su la
0.3 0.5
Salt
0.6
SALT
Fault zone
A
Proposed Fault Zone
0.7
633
Shear zone
A
Dead Sea
0.8
Normal Fault
Well 0.8
Contour line in seconds
Rift sca rp
63
M
B
2
0.6 0.8
Border Fault
1
596
700
Mt. Sedom
729 A
Au 0.7
23
14
1.0 0.8 0.9
40
70
Am
az
22
Fa
Downlap
ul
t
10 km
0
40
19
Aa
hu
C
Onlap
ia
13
40
70
Figure 11. Seismic map of unconformity #B2. The extension of the salt is not known. The iso–reflection time lines form a concave surface that deepens toward the Amaziahu fault, similar to the footwall configuration.
posed of mounded reflections, but the wavelength here is greater (Hs1). The transgressive systems tract (Ts1) between these facies is a condensed section. This succession of systems tracts can be traced upward. The highstand systems tracts in the upper two sequences have different seismic patterns (Hs2). The reflections are long and continuous, but not mounded. The overall shape of the sigmoid progradation is flatter than that of the reflectors below. Sedimentary Model Figures 17 and 18 are longitudinal cross sections across the basin that summarize the character of the sedimentary fill. Figure 17 was constructed from seismic sections 4022, 596, and 633. The sizes of the gaps between lines are to
Figure 12. A hypothetical explanation for the mechanism of the Amaziahu fault. Stage A: The initial conditions before faulting. Stage B: After deposition of additional sediments, the section fractured and became a growth fault. Stage C: Some of the salt became mobilized and started to migrate upward in the fault zone. The withdrawal of salt created space, and the deposited strata collapsed. ST is systems tract.
scale. The tectonic elements in the basement and in the sedimentary fill, together with the observed unconformities, are indicated on the section. The deepest part of the basin is on line 633, south of the Boqeq fault zone. The basement gradually deepens from both sides of the basin to this part of the section. The interpreted tectonic elements in the basement show that the basement extension was not homogeneous along the section; some grabens and half grabens formed near the southern end of the basin. The faults in the southern segment of the basin (area of 4022) are related to the extension of the basement. The Amaziahu fault, however, developed above the northeastsloping basement. The sequence boundaries dip regionally from south to north, indicating an overall south-north sediment transport into the basin. The Amaziahu fault interrupts this trend, creating
considerable local accommodation space by the collapse of the hanging-wall mass. Figure 18 shows the sedimentary model for the basin through sections 4019, 4023, 596, and 633. Fan deltas were deposited at the two edges of the basin. The bottom of the basin was the site of sedimentation of lacustrine basin-floor fans and sediments rich in salt basinward. Later lacustrine shale sedimentation prevailed in the basin. Near the southern basin margin, onlapping slope fans accumulated in response to lakelevel fluctuations. On the footwall side of the Amaziahu fault, onlapping sediments fill tectonic synclines and the wedge-shaped space between tectonic anticlines and the fault plane. The interfingering of the southern and the northern filling systems can be seen at the northeastern end of the basin.
Geological Society of America Bulletin, November 1997
1495
CSATO ET AL.
Part of Line 633 SW
NE
#B1 #B2
1
#B3 #B4
2 #B7 #B8
TWT(sec)
#B5 #B6
3 #B9
4
Top of Transgressive ST
Sediments transported from SW
Top of Lowstand ST 3rd order sequence boundaries
Sediments transported from NE
Higher order sequence boundaries
Erosional surface 0
5km
Figure 13. Sequence stratigraphic interpretation of seismic line 633. Shading patterns indicate the two major sedimentary systems filling the basin from the southwest and northeast. ST is systems tract.
Significance of Unconformities Sequence stratigraphic analysis of seismic data has enabled the recognition and correlation of key surfaces in the southern Dead Sea basin. However, whether the lake-level fluctuations represented by these surfaces are correlatable to eustatic events remains unknown. Without proper biostratigraphic markers, age dating through the stratigraphy is inaccurate. The age of the basement can be estimated only from outcrop studies. The lowermost unit that accumulated prior to synrift subsidence is the Hazeva Formation. Shahar (1973) attributed an early Miocene age to the lower Hazeva Formation, and a late Miocene to Pliocene age to the upper Hazeva Formation. Garfunkel et al. (1974) described the Raham Conglomerates north of the Gulf of Elat as the initial synrift sediments that are coeval with the Hazeva Formation. A middle Miocene or older age was proposed for this formation. Using the data of Horowitz (1987), the lowest sedimentary sequence of the seismic sections described in this study (below #A6 and #B9) may be of late Miocene age and is partly coeval with the Hazeva Formation. The age of the basement of the basin was estimated to be 6 Ma. Sequence boundaries #A1 and #B1 most likely are correlative and have the bestdefined age representing the basal Lisan unconformity of 70 ka (Begin et al., 1974; Horowitz, 1979). The Pleistocene-Pliocene boundary has not
1496
been identified accurately in the basin strata; ten Brink and Ben-Avraham (1989) estimated its position on the basis of fault geometries. The surface they identified as the Pliocene-Pleistocene boundary is identical to sequence boundaries #B7 and #A2 of this study. Horowitz (1987) used palynologic analyses to determine the PlioceneQuaternary boundary in wells. The palynologic ages do not tie to the same reflections (i.e., time surfaces) across the basin, so further seismic velocity and age data are necessary to determine accurate depths of ages in the basin. In this study, palynologic data from the Amaziahu-1 well were used to place the Pliocene-Pleistocene boundary between the #A3 and #A5 surfaces on the footwall side of the fault. The palynologic age from the Melekh Sedom-1 well (Horowitz, 1987) and the work of ten Brink and Ben-Avraham (1989) were used to approximate the age of the hangingwall side. It was assumed that the PliocenePleistocene boundary is near #B7. Figures 19 and 20 show correlations between the interpreted onlap curves from the Dead Sea basin and published sea-level charts for the Pleistocene and Pliocene. Horowitz (1979) constructed a sea-level chart for the Mediterranean Sea for Quaternary time that was compared to a chart constructed by TGS Offshore Geophysical Company (1989) and to others published by Beard et al. (1982) for the Gulf of Mexico, and to that of Wornardt and Vail (1991). The Mediterranean sea-level curve of Horowitz
(1979) records sea-level falls coeval with those of the Gulf of Mexico. The base and middle of the preglacial Pleistocene, and the Günz, Mindel, and Riss glacials are marked by sea-level falls. The youngest glacial period, the Würm, has two sea-level falls, one at its base and another in its middle. The excellent correspondence between the Mediterranean and Gulf of Mexico charts offers a consistent basis for correlation of the observed-onlap charts. The onlap chart was constructed from section 4019 (southern area) (Fig. 5, A–C), and from section 633 (northern area) (Fig. 8, A and B) and are shown in the middle of Figure 20. The two charts are not equivalent. The southern chart contains more sequence boundaries in the Pleistocene than does the northern chart. This difference reflects the fact that line 4019 is in a marginal portion of the basin, and consequently, there is a better stratigraphic record of the lake-level fluctuations. However, more higher-frequency sequence boundaries were identified in the northern area for the Miocene–Pliocene and Pleistocene, where the quality of line 633 had better resolution than line 4019. The 9 sequence boundaries in the northern area (#B1–#B9) divide the time span into 10 intervals that represent, on average, about 600 k.y. each, whereas the 6 sequence boundaries in the southern area divide the stratigraphy into intervals of 860 k.y. Using the Vail et al. (1991) classification, these sequences are considered
Geological Society of America Bulletin, November 1997
Part of Line 633
Front of interfingering zone
SW
NE
Quaternary
#B1 #B2
III
1
III
#B3 #B4
Pliocene
2 #B7
II #B8
II
TWT(sec)
#B5 #B6
3 #B9 I
I 4 I
3rd order sequence boundaries
II
III
Sedimentary systems
Higher order sequence boundaries
0
5km
Figure 14. Interfingering of the southern and northern lacustrine strata in seismic line 633. The heavy line marks the front of the interfingering zone. Three subsequent systems were distinguished on the basis of the overall geometry of the stacked sequences. System I represents the oldest package formed when the bulk of the sediments came from the south. In system II the amounts of sediments transported from the south and north were balanced. During the deposition of system III, the northern source became more dominant in this area. The position of interfingering shows a strong correlation with lake-level falls, which are represented by unconformities. The evolution through systems I–III shows a trend; that is, increasing dominance of the northern sedimentary source.
Facies Model SW
TST
Hs2
HST
Hn3
TST
LST
HST
Tn2
Ts1
TST
TST
Tn1
Hs2
HST LST HST
NE
Line 633
Ls2
Ls1 Ts1
Ls2
Ls1 Ts1
Ln2
Hn2
Hs1
Hn1 Tn1
Ln1 Ln1
Ls1 Ls1
LST
Top of Transgressive ST Top of Lowstand ST Unconformity Transported from South Ls1
Mounded, short wavelength fans
Ls2
Continuous reflections
Ts1
Condensed section
South
HST
Hs1
Mounded, progradational, long wavelength fans
TST
Hs2
Continuous, progradational
LST
Transported from North
North
Hs2
Hn3
Ln1
Chaotic, short wavelength
Hs1
Hn2
Tn1
Backstepping, short wavelength
Hn1
Tn2
Condensed section
Hn1
Condensed section
Hn2
Progradational, short wavelength
Hn3
Continuous, progradational
Ts1
Tn2
Hn1
Tn2
Ln1
Hs1
Figure 15. Facies model for seismic line 633, northern area. The definition of facies was based on systems tract analysis, the direction from which the sediment was transported, and the characteristic geometry and reflection pattern. LST—lowstand systems tract; TST— transgressive systems tract; HST—highstand systems tract.
Tn1 Ls2 Ls1
Ln1
Geological Society of America Bulletin, November 1997
1497
Detail in Line 633
SW
NE
3
4
4
TWT (sec)
3
0
A SW
Figure 16. (A) Detail of seismic line 633. Location is marked in Figure 9B. (B) Interpretation for detail of seismic line 633. The Ls1 and Hs1 facies show characteristic fan shapes and progradational patterns. The reflectors of Hs2 are continuous and have downlap terminations. See Figure 15 for abbreviations. ST is systems tract.
1km
Detail in Line 633
NE
#B8 B8/1
Ts1
3
3
Ls1
B8/2
Hs2
Ts1
Hs1
Hs1
Ts1
Ls1
TWT (sec)
Hs2
Ls1
#B9
B9/1 B9/2
Ts1
Ts1
Ls1
Hs1
4
Ls1
Top of Transgressive ST
0
4
1km
Top of Lowstand ST Higher order unconformity
B
3rd order unconformity
Amaziahu Fault
SW #A1
1
#B4
2#A5
#A6
NE
#B2 #B3
#A3 1 #A4
TWT(sec)
1.0 2.0 3.0 4.0
Boqeq Fault Zone #B1
#A2
2
?
3
3
?
?
? 4
#B5 #B6 #B7 #B8
? ?
4
Line 4022
#B9
Line 596
Salt
0
Line 633
5km
N Rift S
carp
402
Mt. Sedom
2
596
azi
ahu
Fau l
t
Cross Section
ni ns u
la
Am
633
Li sa n
10km
Pe
Dead Sea 0
Figure 17. Cross section constructed from seismic lines 4022, 596, and 633. The correlated third-order sequence boundaries are indicated (#A1–#A6, #B1–#B9). No accurate correlation can be made across the Amaziahu fault because of the lack of age data. Most of the sediments were transported from the south; the sequences tend to deepen and thicken toward the north. Some of the sediments in the northeastern side were transported from the north.
1498
Geological Society of America Bulletin, November 1997
Sedimentary Model, Southern Dead Sea Basin Amaziahu Fault
SW
Boqeq Fault Zone
NE
#B1 4
#A1 #A4
#A5 1
#B4 2
#B5
3
#B6 #B7
? Salt ?
3.0
2
9
1
7
6 2-
#A6
3
3
?
?
?
8 ?
#B8 4
4.0
Line 4019
4
Line 4023
#B9
Line 596
1
Fan delta
5
Onlapping facies above tectonic anticline, footwall side
8
Interfingering Lowstand Slope Fans, transported from SW and NE
2
Basin Floor Fan
6
Onlapping facies above tectonic syncline, footwall side
9
Progradational Facies from NE
3
Lacustrine clastics and salt
7
Onlapping facies above tectonic anticline, hanging wall side
4
Lowstand Slope Fan
Line 633
0
5km
Salt
Lacustrine clay-sand
N 40
Rift S
19
carp
Mt. Sedom 40
Cross Section Fa
ult
23
azi
ahu
596
ns
ul
a
Am
633
Pe
ni
Dead Sea
sa
n
10km
Li
0
Figure 18. Cross section constructed from seismic lines 4019, 4023, 596, and 633, and facies interpretations.
Eustatic sea-level chart after Haq et al. 1987 Rise
Quaternary
Seismic onlap curves Southern Area Northern Area Fall
0.8 1.3
#A2 1.65
Pliocene
Piacenzian
3.5
#B4
1.6 2.0
A2/1
2.4
#A3
2.7
A3/1 3.0
#A4
3.4
A4/1 A4/2
3.8 4.0 4.2
#A5
#B5 B5/1
#B6 B6/1
#B7 B7/1 B7/2
#B8
Zanclean 5.2
Miocene
TWT(sec)
1
#A3
2.0
#B2 #B3
5
#A2
1.0
Messinian 6.3
B8/1
5.0
A5/1
5.5
A5/2
5.8
?
#A6
B8/2
? #B9
6.3
Tortonian
Figure 19. Tentative correlation of observed unconformities to the eustatic curve of Haq et al. (1987) in Pliocene time.
Geological Society of America Bulletin, November 1997
1499
CSATO ET AL.
Ages (Ma) 0
Mediterranean sea-level changes after Horowitz 1979 elevation in meters Würm
Southern area Northern area -140 -120 #B1 #A1
5 16
Riss-Würm
Onlap chart in the S. Dead Sea basin
Eustatically driven coastal onlap charts for the Gulf of Mexico after TGS
after Beard et al. 1982
A1/1
Riss
Sangamonian
-90 A1/2
Mindel-Riss
after Wornardt & Vail. 1991
Wisconsian
#B2
35 Illinoian
Mindel
-60
A1/4
Günz-Mindel
1
Glacial Pleistocene
A1/3
#B3
Yarmouthian
55
A1/5 Kansan
-30
Günz
#A2
#B4
85 #B5
2
Aftonian
B5/1 -10
#A3
#B6 Nebraskan
Calabrian
Preglacial Pleistocene
A2/1
120
A3/1
#A4
B6/1
#B7
3 Correlation Figure 20. Tentative correlations of observed unconformities to those of the Mediterranean Sea and the Gulf of Mexico in Quaternary time. The unconformities in the southern part of the basin and in the northern part of the basin are shown. The interpreted third-order unconformities appear to correlate with the major eustatic sea-level falls.
third-order cycles. The higher-order sequences are attributed to fourth-order cycles. Only the uppermost base Würm surface can be dated accurately, and the ages of all the other sequence boundaries are tentative. In the northern area, the Günz, Mindel, Riss, and Würm glacials were correlatable as third-order sequences. The preglacial strata are composed of three thirdorder sequences. In the southern area, the #A2 surface was assumed to match the base of the Günz, and the
1500
A1/1–A1/5 higher-order sequence boundaries correspond to individual glacial and interglacial events. Sequence boundaries #A3 and #A4 are within and at the base of preglacial time. The Miocene–Pliocene correlation is shown in Figure 19. In both areas, the time span below the Pliocene-Pleistocene boundary is divided into three units, by sequence boundaries #A5 and #A6 in the southern area, and by #B8 and #B9 in the northern area. In Pliocene–Miocene record, the observed fourth-order surfaces provide more
details than the eustatic chart. Only the eustatic sea-level fall at 4.2 Ma cannot be correlated to any observed third-order sequence boundary. One of the observed fourth-order sequence boundaries may be coeval with this event. In conclusion, the observed third-order sequence boundaries in both Pleistocene and Pliocene–Miocene time correspond fairly well with the eustatic cycles. The correlation between observed-onlap charts and eustatic curves is used in our study as
Geological Society of America Bulletin, November 1997
SEQUENCE STRATIGRAPHY IN THE DEAD SEA BASIN, ISRAEL
a possible solution for age dating the sequence boundaries identified on seismic profiles, and thus dating lake-level changes in the Dead Sea. The lack of accurate age data prevents a better and more certain correlation, but the use of sealevel charts may alleviate this problem. The relatively good correspondence between the observed third-order surfaces and eustatic sea-level curves suggests that the same climatic changes that affected eustasy also affected Neogene lake levels in the Dead Sea. Because the ages of the lake-level falls and rises cannot be determined, it is possible they may not have exactly the same timing as the eustatic changes (Csato, 1995). Although the third-order events of the Dead Sea appear to follow the same thirdorder events of the world’s oceans, the timing of the lake-level cycles could have a delayed offset from the eustatic cycles. This was suggested by a computer model that simulated the fill of the basin. The best fit to the simulated sedimentary fill was obtained when the lake levels were offset from the eustatic cycles. The fourth-order cyclicity in the Dead Sea may be a eustatic signal. In the southern area, the clear correlation between the A1/1–A1/5 fourth-order unconformities with the glacials and interglacials suggests that those fourth-order sequences were produced as a consequence of the global climatic changes. Other high-order cycles may be at least partly caused by tectonic effects associated with the subsidence and structural deformation of the basin. CONCLUSIONS (1) Nine sequence boundaries were identified and correlated within the north and south sections of the southern Dead Sea basin. (2) These sequence boundaries have been related to changes in lake level that are inferred to have been driven by global climatic change. (3) The ages of these unconformities were inferred from well data, outcrop, and an apparent match to eustatic sea-level charts. Improved biostratigraphic markers are needed to confirm these inferences. (4) Seismic geometries within each sequence were related to specific sedimentary depositional systems, including fan deltas at the basin margin and lacustrine and evaporite deposition at the basin center. (5) Sediments from the north and south ends of the basin interfinger, and the position of this interfingering appears to have changed location in response to the lake level. Southward advances were related to lake-level falls and sediments that prograde southward being just above the sequence boundary; northward advances
overlie the southward ones and were interpreted to have occurred at lake-level highs. ACKNOWLEDGMENTS We thank Y. Druckman, the Israeli Petroleum Commissioner, and M. Goldberg, former Petroleum Commissioner at the Israeli Ministry of Energy and Infrastructure, for releasing the data used in the study. We also thank especially Y. Ben-Gai, who organized the data retrieval and assisted with this study. This paper was improved by the reviews of E. L. Kashai, Neil Summer, Joel Watkins, Joe Lambiase, and Steve Darobek. REFERENCES CITED Agnon, A., 1981, Neogene development of the western margin of the southern Dead Sea depression: Geological Survey of Israel Current Research, p. 27–29. Agnon, A., 1983, An attempted revision of the Neogene stratigraphy in the Dead Sea Valley: Israel Geological Society Annual Meeting Abstracts, p. 1. Bahat, D., and Rabinovitch, A., 1983, The initiation of the Dead Sea rift: Journal of Geology, v. 91, p. 317–332. Beard, J. H., Sangree, J. B., and Smith, L. A., 1982, Quaternary chronology, paleoclimate, depositional sequences, and eustatic cycles: American Association of Petroleum Geologists Bulletin, v. 66, p. 158–169. Begin, Z. B., Ehrlich, A., and Nathan,Y., 1974, Lake Lisan, the Pleistocene precursor of the Dead Sea: Geological Survey of Israel Bulletin 63, 30 p. Ben-Gai,Y., and Ben-Avraham, Z., 1995, Tectonic processes in offshore northern Israel and the evolution of the Carmel structure: Marine and Petroleum Geology, v. 12, p. 533–548. Brown, L. F., Jr., Benson, J. M., Brink, G. J., Doherty, S., Jollands, A., Jungslager, E. H. A., Keenan, J. H. G., Muntingh, A., and van Wyk, N. J. S., 1995, Sequence stratigraphy in offshore South African divergent basins: An atlas on exploration for Cretaceous lowstand traps by Soekor (Pty) Ltd., American Association of Petroleum Geologists Studies in Geology 41. Croker, P. F., 1983, Geophysical data from the Dead Sea–Jordan rift valley: Geophysics, v. 49, p. 628. Csato, I., 1995, Sequence stratigraphic interpretations and modeling in lacustrine rift basins: Southern Dead Sea basin, Israel and Pannonian basin, Hungary [Ph.D. dissert.]: Columbia, University of South Carolina, 273 p. Freund, R., Garfunkel, Z., Zak, I., Goldberg, M., Weissbrod, T., and Derin, B., 1970, The shear along the Dead Sea rift: Royal Society of London Philosophical Transactions, ser. A, v. 267, p. 107–130. Garfunkel, Z., and Horowitz, A., 1966, The upper Tertiary and Quaternary morphology of the Negev, Israel: Israel Journal of Earth-Sciences, v. 15, p. 101–117. Garfunkel, Z., Bartov, J., Eyal, Y., and Steinitz, G., 1974, Raham Conglomerate—New evidence for Neogene tectonism in the southern part of the Dead Sea Rift: Geological Magazine, v. 111, p. 55–64. Ginzburg, A., and Kashai, E., 1981, Seismic measurements in the southern Dead Sea: Tectonophysics, v. 80, p. 67–80. Horowitz, A., 1979, The Quaternary of Israel: New York, Academic Press, 394 p. Horowitz, A., 1987, Palynological evidence for the age and rate of sedimentation along the Dead Sea Rift, and structural implications: Tectonophysics, v. 141, p. 107–115. Kashai, E. L., 1988, A review of the relations between the tectonics, sedimentation and petroleum occurrences of the Dead Sea–Jordan rift system, in Manspeizer, W., ed., Triassic-Jurassic rifting, continental breakup and the origin of the Atlantic Ocean and passive margins, Part B: Amsterdam, Netherlands, Elsevier, Developments in Geotectonics, v. 22, p. 883–909.
Kashai, E. L., and Croker, P. F., 1987, Structural geometry and evolution of the Dead Sea–Jordan rift system as deduced from new subsurface data: Tectonophysics, v. 141, p. 33–60. Katzman, R., ten Brink, U. S., and Lin, J., 1995, Three-dimensional modeling of pull-apart basins: Implications for the tectonics of the Dead Sea basin: Journal of Geophysical Research, v. 100, p. 6295–6312. Manspeizer, W., 1985, The Dead Sea rift: Impact of climate and tectonism on Pleistocene and Holocene sedimentation, in Biddle, K. T., and Christie-Blick, N., eds., Strike-slip deformation, basin formation, and sedimentation: Society of Economic Paleontologists and Mineralogists Special Publication 37, p. 143–158. Marcus, E., and Slager, J., 1985, The sedimentary magmatic sequence of the Zemah 1 well (Jordan–Dead Sea rift, Israel) and its emplacement in time and space: Israel Journal of Earth-Sciences, v. 34, p. 1–10. Neev, D., and Emery, K. O., 1967, The Dead Sea: Depositional processes and environments of evaporites: Geological Survey of Israel Bulletin no. 41, 112 p. Neev, D., and Hall, J. K., 1979, Geophysical investigations in the Dead Sea: Sedimentary Geology, v. 23, p. 209–238. Nemec, W., and Steel, R. J., eds., 1988, Fan deltas: Sedimentology and tectonic settings: Glasgow, United Kingdom, Blackie and Son, 444 p. Picard, L., 1967, Thoughts on the graben system in the Levant: Geological Survey of Canada Special Paper 66, p. 22–32. Picard, L., 1970, On Afro-Arabian graben tectonics: Geologische Rundschau, v. 59, p. 337–381. Quennell, A. M., 1958, The structural and geomorphic evolution of the Dead Sea Rift: Geological Society of London Quarterly Journal, v. 114, p. 1–24. Shahar, Y., 1973, The Hazeva Formation in the Oron-Ef-e syncline, Israel: Israel Journal of Earth-Sciences, v. 22, p. 31–49. Stein, M., Agnon, A., Starinsky, A., Raab, M., Katz, A., and Zak, I., 1994, What is the “Age” of the Sedom Formation?: Israel Geological Society Annual Meeting Abstracts, p. 108. Steinitz, G., and Bartov, Y., 1992, The Miocene–Pleistocene history of the Dead Sea segment of the rift in light of K-Ar ages of basalts: Israel Journal of Earth-Sciences, v. 40, p. 199–208. ten Brink, U. S., and Ben-Avraham, Z., 1989, The anatomy of a pull-apart basin: Seismic reflection observations of the Dead Sea basin: Tectonics, v. 8, p. 333–350. ten Brink, U. S., Ben-Avraham, Z., Bell, R. E., Hassouneh, M., Coleman, D. F., Andreasen, G., Tibor, G., and Coakley, B., 1993, Structure of the Dead Sea pull-apart basin from gravity analyses: Journal of Geophysical Research, v. 98, p. 21–894. TGS Offshore Geophysical Company, 1989, Gulf of Mexico chronostratigraphic correlation chart: Houston, Texas, GECO-PRAKLA Exploration Services. Vail, P. R., Audemard, F., Bowman, S. A., Eisner, P. N., and Perez-Cruz, C., 1991, The stratigraphic signatures of tectonics, eustasy and sedimentology—An overview, in Einsele, G., Ricken, W., and Seilacher, A., eds., Cycles and events in stratigraphy: Berlin, Springer-Verlag, p. 617–659. Vroman, A. J., 1973, Is a compromise between the theories of tension and of shear for the origin of the Jordan–Dead Sea trench possible?: Israel Journal of Earth-Sciences, v. 22, p. 491–539. Wornardt, W. W., and Vail, P. R., 1991, Revision of the Pleistocene Cycles and their application to sequence stratigraphy and shelf and slope sediments in the Gulf of Mexico: Gulf Coast Association of Geological Sciences Transactions, v. 41, p. 719–744. Zak, I., 1967, The geology of Mount Sedom [Ph.D. dissert.]: Jerusalem, Israel, Hebrew University, 208 p. Zak, I., Karcz, I., and Key, C. A., 1968, Significance of some sedimentary structure from Mount Sedom: Israel Journal of Earth-Sciences, v. 17, p. 1–8. Zak, I., and Freund, R., 1981, Asymmetry and basin migration in the Dead Sea Rift: Tectonophysics, v. 80, p. 27–38. MANUSCRIPT RECEIVED BY THE SOCIETY JANUARY 22, 1996 REVISED MANUSCRIPT RECEIVED FEBRUARY 17, 1997 MANUSCRIPT ACCEPTED APRIL 26, 1997 Printed in U.S.A.
Geological Society of America Bulletin, November 1997
1501
