Can the Sedimentary Record be Dated from a Sea-Level Chart? Examples from the Aptian of the U.A.E. and Alaska
Published:January 01, 2000
- PDF LinkChapter PDF
Christopher G. St. C. Kendall, Abdulrahman S. Alsharhan, Kurt Johnston, Sean R. Ryan, 2000. "Can the Sedimentary Record be Dated from a Sea-Level Chart? Examples from the Aptian of the U.A.E. and Alaska", Middle East Models of Jurassic/Cretaceous Carbonate Systems, Abdulrahman S. Alsharhan, Robert W. Scott
Download citation file:
Ten depositional shoaling-upward cycles have been identified in the Aptian Shuaiba Formation of the United Arab Emirates (U.A.E.). A similar number of cycles have been recognized in the National Petroleum Reserve of Alaska (NPRA). Similarities of these cycles with the onlapping shelf geometries of the Neogene of the Bahamas suggest that the sequence geometries of Aptian strata of the NPRA and the U.A.E. are a response to high-frequency changes in eustatic sea-level position. Because the Aptian cycles of the NPRA match similarly dated, events in the U.A.E., it is suggested that where biostratigraphic data are poor the sedimentary section can be tentatively dated by relating the geometries of the shelf margin to the character of the coastal onlap curve and its coincident sea-level chart. Thus, a sea-level chart might be used at locations for which biostratigraphic data is sparse to determine and constrain preliminary depositional models for specific time intervals.
With this in mind, two biostratigraphic models for dating the Shuaiba Formation and the Bab Member were tested against the sea-level curve of Haq et al. (1987) using a sedimentary simulation. The results were ambiguous because both biostratigraphic models could not be matched. Also in both cases, the simulation suggested that just prior to the deposition of the Bab Member the basin margin was uplifted and then subsided, causing a local relative sea-level fall followed by a rise, an event not found on the sea-level chart of Haq et al. (1987). Additionally, the sedimentary simulation supports the position that the Aptian in the U.A.E. is bounded by erosional unconformity surfaces and contains higher-frequency cycles.
When making sequence stratigraphic interpretations, sequences are identified on the basis of bounding unconformities and the onlapping and offlapping relationship of sedimentary geometries (Vail et al. 1977; Vail and Todd, 1981; Van Wagoner et al., 1987; Vail, 1988; Posamentier and Vail, 1988; Posamentier et al., 1988). For stratigraphic sections in which fossil evidence is sparse, the hypothesis is proposed that the ages of these sequences can be inferred by matching the size, shape, and position of the onlapping sequence geometries to the relative magnitudes and geometries of the sea-level events on an eustatic chart (Kendall et al., 1997a) (Fig. 1).
In other words, the events on the sea-level chart reported by Haq et al. (1987) have produced unique responses in the sedimentary section through time. Each sea-level cycle transgressed the basin margins by different amounts. This produced onlapping sequences of sedimentary fill that should have a unique onlap position on the shelf margin that, when compared to the position of the adjacent underlying and overlying sequences, can be directly related to the shape of the sea-level chart. We tested this working hypothesis using the Haq et al. (1987) sea-level chart with a simulation, and we have used the results of this work to date the geologic section of the Neogene of the Bahamas. We then confirmed these dates with biostratigraphic data (Lidz and McNeil, 1998; Kendall and Sen, 1998).
A major problem with this test is that, realistically, eustatic sea-level does not alone control the position of the sea surface with respect to the adjacent sediment surface. Both sediment supply and accumulation rates as well as tectonics affect this accommodation too! So, contrary to the findings described above, if a series of chronostratigraphic charts is created from seismic profiles at different positions across a shelf margin, the position and onlapping character of sequences with similar ages can be expected to vary from one geographic locale to another, reflecting the different tectonic and sedimentary histories at each location. Recognizing that there is some variability in the onlapping geometries produced by variations in local tectonics and sedimentation, we selected the profiles for this study because they had simple tectonic and sedimentary histories and because their sequence geometries appeared to be largely a response to the major sea-level events. Simulations developed to recreate these sections used the sea-level values of Haq et al. (1987) as a known variable, and rates of tectonic movement were uniform for a simulation run. The sedimentation accumulation rates were changed only as a last resort to achieve a better geometric match to the section. The resulting onlapping sequence geometries of the simulation reflected the effects of the eustatic events used in the simulation and, we believe, provided realistic solutions.
To set the stage for application of the hypothesis to the Lower Cretaceous of the National Petroleum Reserve of Alaska (NPRA) and the United Arab Emirates (U.A.E.), this paper begins by describing the test of the hypothesis made by Kendall and Sen (1998). For this purpose, they dated the Neogene section of the Bahamas by matching the shapes of the onlapping sequence geometries to the sea-level chart of Haqet al. (1987). These dates were confirmed with biostratigraphic data. We then report on our studies of the Aptian-Albian sections of the NPRA of Alaska and of the U.A.E.
The rationale for picking these latter areas and profiles was that though they represent a less well defined part of the geological column than the Neogene, the strata are of similar age and are associated with sufficient data to enable further examination of our hypothesis. Both data sets have seismic profiles that contain evidence of onlapping and prograding basin-margin sequences.
The NPRA seismic data set shows a clearer set of high-frequency sequences but has imprecise biostratigraphic control, while the U.A.E. seismic section shows poorer resolution of the high-frequency sequences but is associated with more detailed biostratigraphy. This paper focuses more on the U.A.E. because this was the area for which we had the most data.
For both areas we were able to demonstrate, through the medium of a sedimentary simulation, that eustatic events produced distinct sedimentary responses. The products of these eustatic events were used to provide preliminary dates for parts of the sections, suggesting that in a region where biostratigraphic data are sparse, a sea-level chart coupled with sedimentary simulations could be used as a powerful tool for interpreting the sedimentary section. The major assumption to the hypothesis is that subsidence and sedimentation are uniform so that for the duration of the simulation the eustatic sea-level cycles produce distinct and identifiable sequence geometries.
Initial Test Of The Hypothesis
The above hypothesis was tested with a study of the Neogene section of the Bahamian Banks (Eberli et al., 1994; Kendall and Sen, 1998). To achieve this, a seismic line from the Bahamas (Fig. 2) was interpreted by first identifying the second-order type 1 unconformities (Fig. 3). These second-order unconformities were then related to the second-order events on the sea-level chart of Haq et al. (1987). Then, counting the third-order unconformities between the second-order ones, the shapes of these third-order cycles were matched to events on the sea-level chart. The ages of the sea-level events were then given to the third order sequence boundaries that had been interpreted on the seismic profile.
A sedimentary simulation, Sedpak (Eberli et al., 1994), was then used to test the match between the ages of the cycles and the interpretation of the sequences on the seismic (Kendall and Sen, 1998). Kendall and Sen (1998) iteratively changed the inputs to the simulation, including trying a variety of different rates of tectonic subsidence and sedimentation, and sea-level models. The sedimentary simulation that provided the best match to the seismic interpretation eventually used as inputs constant rates of tectonic subsidence and carbonate accumulation, taking the timing and magnitude of the sea-level history from the Haq et al. (1987) chart.
The resulting position of the onlapping geometries of the simulation (Fig. 4) and those of the interpreted seismic data (Fig. 3) are similar. The sequence boundaries were tied to biostratigraphic data (Lidz and McNeil, 1998; Kendall and Sen, 1998) which suggests that the Haq et al. (1987), sea-level chart for the Neogene was accurate. The authors were able to identify individual eustatic sea-level events, determine their magnitude, and establish the timing of sequences identified on the seismic profile. In fact they argued that, if the rates of subsidence are low in an area and the rates of sedimentation were high, the stratigraphic signa I produced by sea-level change should be clear. This model assumed that the rate of carbonate sedimentation was such that shoreward accommodation filled to sea level and that the sediment surface on the shelf could be taken as a proxy for sea level. Not only that, but if the rates of subsidence and carbonate accumulation were constant for several cycles of eustatic sea level, then the frequency and amplitude of the geometries of the onlapping carbonate bodies of the shelf were the product of the frequency and size of the cycles of eustatic sea level. So, if a simulation with a low rate of constant subsidence and a high rate of constant carbonate accumulation is used with the Haq et al. (1987) curve and produces cross sections that match that of the interpreted seismic, then the height of the eustatic events could be correct and their ages might be read from the eustatic chart. Neverthe-less, this conclusion is dependent on the models proposed for the seismic and sedimentary geometries. Essentially, use of the sea-level chart cannot be divorced from confirmatory data from biostratigraphy, and should be considered only as a first pass until better biostratigraphic data become available.
Dating The Aptian Of Alaska By Sea-Level Chart
Using the hypothesis described above we examined and modeled two Aptian to Albian geological sections. We first examined the Lower Cretaceous Torok and Nannashuk Formations in the NPRA area north of the Brooks Mountain Range (Kendall et al., 1997b; Ryan, 1997). Here a large wedge of deltaic sediments has prograded northwestward across the northern era tonic edge of Alaska (Bird and Molenaar, 1992). TheNanushuk Formation consists of the proximal sandier facies of the Torok Formation prograding complex. The depositional sequence framework for the Torok and Nanushuk Formations was established using regional seismic (Fig. 5) and well data, and reveals an eastward-prograding shelf edge that contains a series of type 1 unconformities (Fig. 6). It is our interpretation that high-amplitude third-order eustatic events were superimposed on the longer-term second-order sea-level events. Collectively these eustatic events created the characteristic geometric patterns of the depositional margin, including packages of onlapping highstand sig-moidal reflectors and downlapping lowstand systems tracts. These depositional geometries were simulated in order to determine the relationship between the sequences and the variables affecting relative sea level (sea-level position, clastic supply, and subsidence). Though the sections thin onto the cratonic margin to the north, the general lack of faulting and the uniformity of the onlapping sediment thicknesses suggest that changes in base level were driven by eustasy.
Ryan (1997) identified the second-order, type 1 unconformities on seismic profiles within the Aptian–Albian part of the section and correlated these to the second-order, type 1 sea-level falls on the sea-level curve of Haq et al. (1987). He proposed that the unconformities were of the same ages as the sea-level falls. He then identified unconformities in the third-order sequences between these second-order, type 1 events and matched these to the third-order sea-level events on the Haq et al. (1987) chart, and assumed that the ages of these intermediate cycles, or sequences, were those seen on the chart. After using a variety of inputs to the simulation, he iteratively converged on using sea-level variations from the Haq et al. (1987) chart, eventually assuming a simple model of constant sedimentation and constant subsidence. This latter model of tectonic behavior was in part based on the uniformity of the thickness of the onlapping sequence geometries. The simulation output was matched to the seismic by iteratively tweaking rates of constant sedimentation and of constant subsidence until the simulation output matched the seismic (Fig. 6). The biostratigraphic ages for the National Petroleum Reserve of Alaska were used to bracket portions of the Aptian-Albian. The modeling of the intermediate section confirmed the probable existence of major second-order and third-order eustatic sea-level variations in the Aptian-Albian. It demonstrated that these major eustatic changes were probably the primary control for the development of the stratal geometries in the Torok and Nanushuk Formations.
Use of the Sea-Level Chart for Dating the Aptian of the U.A.E.
We next tested the hypothesis in the U. A.E., where the Aptian-Albian sedimentary section has a more detailed biostratigraphy than in the NPRA. In the U.A.E. the Lower Cretaceous is made up of major cycles of carbonate deposition (Fig. 7). These were terminated with an unconformity in the Turonian that coincided with a major marine regression. The Lower Cretaceous accumulated in an intercra tonic shallow basin (Fig. 8) that was developed on the Arabian Shield back from the southern margin of the Tethys Sea. Its final character was imposed by the onset of the collision of the Asian Plate with the Arabian Shield, which began to crumple in the Turonian (Alsharhan and Nairn, 1988).
Initially a basin with a relatively flat bottom was formed. It was rimmed by shallow-water carbonates that climaxed with rudistid reefs (Fig. 8). These prograded into the basin, producing large variations in the thickness between the sediments of the basin and those of the shelf (Aldabal and Alsharhan, 1989) (Fig. 1). By the end of the deposition of the Aptian Shuaiba Formation, the thickness of the shelf deposits was approximately 400 meters, and at the center of the basin sediment thicknesses were on the order of 100 to 150 meters (Azzam and Taher, 1995). It is proposed that the progradational shelf margin formed as a series of lowstand, transgressive, and highstand carbonate wedges that extended out from the edge of the basin. These eventually filled the margins of this basin with a rudistid facies. The depositional model for the Shuaiba Formation suggests that shortly after the deposition of the Kharaib Formation a transgression occurred, the water deepened a little, and a shallow-water shelfal limestone with Lithocodium sp. accumulated (Alsharhan, 1985). Progressively through the Aptian these slightly deeper shelf sediments shoaled upward and eventually acquired a cap of rudistid reefs (Calavan et al., 1992). Finally the buildups were exposed and then onlapped by the Albian Nahr Umr Formation (Alsharhan, 1991). On the basis of lithology, diagenesis, and fauna, we interpret these lower Shuaiba carbonate bodies as not having reached sea level. We propose that they were prevented from reaching the air/water interface by the onset of the anoxic conditions produced when the basin became isolated. Later, by the end of Shuaiba Formation deposition, the carbonate platform reached sea level. The platform margin is represented by an enormous buildup created by the accumulation of rudistids and their debris. The seismic line (Fig. 9) shows the progradational character of this margin.
The characterization of the biostratigraphy by Simmons (1994), Witt and Gökdag (1994), Calavan et al. (1992), Fischer et al. (1994), and Kendall et al. (1997b), suggests that at least two chronostratigraphic models constrain the interpretation of the timing of the Shuaiba carbonate wedges. The first model was that of Alsharhan and Kendall (1991), Simmons (1994), and Witt and Gökdag (1994), in which the base of the Shuaiba Formation was correlated with the lower part of the Aptian and matched with the 112 Ma transgression of the Haq et al. (1987). The rest of the Shuaiba Formation was modeled as a series of progradational units that terminated at 108 Ma just before the major sea-level drop at 107.5 Ma. This drop was followed by the deposition of the onlapping Nahr Umr Formation (Figs. 2, 10).
The second group of chronostratigraphic interpretations that we modeled were those of Calavan et al. (1992) and Fischer et al. (1994). They interpreted that deposition of the Shuaiba Formation spanned from approximately 113.5 Ma to 108 Ma (Fig. 1). Thus, in contrast to Alsharhan and Kendall (1991), the base of the Shuaiba was placed at 113.5 Ma, and the lower Shuaiba was interpreted to represent deposition during a slightly higher sea level, followed by a drop in sea level at 112 Ma. During deposition of the rest of the Shuaiba Formation two further progradational units were identified and dated from 112 Ma to 109.5 Ma and from 109.5 Ma, to approximately 108 Ma, after which a major sea-level drop occurred. This drop was followed by the deposition of the onlapping Nahr Umr Formation (Figs. 1, 9).
We tested both biostratigraphic models against the Haq et al. (1987) curve. In both group one and group two models (112 Ma to 108 Ma, and 113.5 Ma to 108Ma), to simulate the geometric shapes of the carbonate bodies responding to sea level, we counted the number of well-log lithologic cycles that we saw in the Shuaiba Formation. We then assumed that each carbonate cycle accumulated over a period of around 40,000 to 100,000 years and that the coincident sea-level variations had a sinusoidal character. For the time intervals involved in both models, we superimposed these fourth-order sinusoidal high-frequency events onto the chart of Haq et al. (1987), producing what we considered to be the sea-level curve responsible for the various geometries that can be seen in the Shuaiba Formation. We then ran the Sedpak simulation using a basin with a flat bottom. This shape was assumed on the basis of the changes in thickness in the shelf margin and basin, and on a water depth of no more than 50 meters. Carbonate sedimentation and subsidence were varied but kept as constant as possible, so the character of the geometries responded most to the variations in sea level. The results were a match to the geometries that can be seen on both the seismic and stratigraphic cross sections that have been developed for the Shuaiba Formation (Figs. 10, 11).
We assumed that eustasy, carbonate production rates, and tectonics were the major variables controlling sequence geometry in the Shuaiba Formation. The sea-level curve used in the simulation was modified from Haq et al. (1987). We added a fourth-order, 100,000 year cycle to it. The timing of the resulting sequences is assumed to be equal to that of this sea-level event. We then assumed that the sea-level curve that we developed for this study was reasonable and needed no further adjustments. Carbonate production and tectonics in the simulation were iteratively varied through time to acquire the best match between the interpretation and the simulation. When we kept these parameters constant, the resulting simulations did not convincingly match the seismic interpretation.
The most accurate simulation of the Shuaiba Formation sequence stratigraphy during the interval from 112 Ma to 108 Ma were achieved with the following input variables. From 112 to 110 Ma, the carbonate production rate was set at 35 cm/1000 years close the air/sea surface. This rate declined quickly with depth. During the same interval, the subsidence rates was at 20 cm/1000 years. The results of the simulation (Fig. 10) show a fairly rapid change from a ramp margin to a well defined shelf edge by the end of Shuaiba Formation deposition. The depositional sequences of the Shuaiba Formation aggrade with minor progradation at the shelf edge.
At 110 Ma the shelf was uplifted at a rate of 5 cm/1000 years to maintain the position of sea level below the shelf edge throughout the deposition of the Bab Member. Because of the lowered sea level, Bab carbonate production was reduced to 3cm/1000 years at the surface and declined slowly to near zero at 100 meters depth. This simulation resulted two progradational sequences. The first progradational sequence ranges in age from 112 to 110 Ma and onlaps the Shuaiba slope. Following the tectonic uplift and a resulting relative sea-level fall at 110 Ma, the Bab Member onlaps onto the lower Shuaiba Formation margin and progrades out into the basin.
The change from subsidence to uplift occurred fairly rapidly. A coincident uplift was required to maintain the geometries of the Bab sequences, which were restricted to deposition below the Shuaiba shelf edge. Without uplift, Bab deposition would have onlapped onto the shelf, and the stratigraphic evidence shows no evidence of this.
In contrast, for the 113.5 Ma to 108 Ma Shuaiba Formation model, the best simulation inputs and results were as follows (Fig. 11). From 113.5 to 112 Ma a carbonate production rate was set at 90 cm/1000 years at the sea surface. This rate declined sharply with water depth. During the same interval, the subsidence rates began at 40 cm/1000 years, and near the end of this interval it declined abruptly to 1 cm/1000 years. The results of the simulation (Fig. 11) show a fairly rapid change from a ramp margin to a well defined shelf edge by the end of Shuaiba deposition. The Shuaiba depositional sequences aggrade with minor progradation of the shelf edge.
During the second simulation run following a major sea-level drop at 112 Ma, the shelf was uplifted at a rate of 3 cm/1000 years to maintain the position of sea level below the shelf edge throughout the deposition of the Bab Member. Because of the restriction of the sea surface during Bab deposition, carbonate production was dramatically less than it was during Shuaiba deposition. Carbonate production was 6 cm/1000 years at the sea surface, decreasing slowly to near zero at 100 meters water depth. The simulation produced two progradational sequences. The first ranged in age from 112 to 109.3 Ma and onlapped the Shuaiba slope. Following a smaller sea-level fall at 109.3 Ma, a second sequence, the upper part of the Bab Member, onlapped onto the lower sequence before prograding farther out into the basin.
The change from subsidence to uplift occurred fairly rapidly, and coincided with the 112 Ma sea-level fall. A coincident uplift was required to maintain the geometries of the Bab sequences, which were restricted to deposi tionbelow the Shuaiba shelf edge. Without uplift, Bab deposition would have onlapped onto the shelf.
From the above discussion it can seen that simulation of the section using two chronostratigraphic models of the Aptian section in the U.A.E. does not immediately resolve which model is more reasonable. In both simulations, however, the Shuaiba Formation was bounded by an erosional unconformity and contained higher-frequency cycles that were the function of the time steps used. The shape and stacking patterns of these higher-frequency cycles appears to be influenced by their position within the associated third-order cycles. Thus at the base of the first third-order cycle, the higher-frequency cycles were more like those in a lowstand sequence. Upward in the third-order sequence these higher-frequency cycles became more transgressive in character. In the upper part of the third-order cycles the higher-frequency cycles had the character of a highstand systems tract. In the adjacent intershelf basin, both cycles contain lowstand and highstand systems tracts that accumulated in a deeper and more open marine setting dominated by deposition of argillaceous limestone.
From the seismic data, we know that the Bab Member is restricted to the basin. Two explanations for the restricted nature of the Bab sediments are proposed:
The tectonic uplift was required to maintain depositional integrity. The cause of the uplift is not known but may be related to the collision of the Arabian plate with the Asian continent.
The amount of coastal onlap after 112 Ma, and hence, the scale of the transgression, used in both the simulations was not correct. If this latter is true, then a reduction in the magnitude of post–112 Ma coastal onlap should produce the same geometries with a subsidence rate that changed very little.
Both explanations are equally valid from a standpoint of the geologic evidence, and neither is affected by the two different biostratigraphic models. If anything, the dating of the Shuaiba Formation and Bab Member as between 112 Ma and 108 Ma, or between 113.5 Ma and 108 Ma, supports the proposed tectonic uplift to explain the character of the downdip but onlapping Bab geometries.
It was demonstrated that the hypothesis that events on the sea-level chart of Haq et al. (1987) produced unique responses in the sedimentary section which could potentially be used to date that section. To this end a sea-level chart was used, in conjunction with a simulation, to date the geologic section of the Neogene of the Bahamas. These dates were confirmed with biostratigraphic data (Kendall and Sen 1998).
Unconformities bounding the third-order sequences of the seismic sections of the Aptian-Albian of the National Petroleum Reserve of Alaska were matched to the third-order sea-level events on the Haq et al. (1987) chart. A sedimentary simulation using sea-level variations from this chart, and a simple model of constant sedimentation and constant subsidence, was matched to this seismic section. The modeling of the section confirmed the existence of major second-order and third-order eustatic sea-level variations in the Aptian-Albian and demonstrated that these major eustatic changes might be the primary control for the development of the stratal geometries in the Torok and Nanushuk Formations.
The Aptian of the U.A.E. has been dated by two biostratigraphic schemes. The use of the Haq et al. (1987) sea-level chart to determine which of these two models was more likely was unsuccessful, but the simulation used in the test suggests that the basin margin was uplifted just before the deposition of the Bab Member.
It has been argued in this paper that by combining strati-graphic interpretations with a sedimentary computer simulation, the ages of eustatic sea-level events, and their sizes, can be tentatively coupled to confirm the ages of the sequence boundaries in interpreted sections, but the results are ambiguous. Not only that, but further tests should be made on the effect of eustasy, tectonics, and sedimentation on the distribution of sequence geometries. Further it is also possible to hypothesize on the initial basin shape, and its accommodation history (the latter by combining tectonic and eustatic history), and the sediment accumulation history of the basin. Finally it is also possible to identify and predict geometries and lithofacies variations within the sequences and to test current and future sequence stratigraphic interpretations and models built from well cross sections and seismic. The problem is that any model that simulates the sedimentary geometry from the geologic record is unquestionably non-unique, and dependent upon data resolution to establish the best interpretation for the geology. So for the techniques outlined in the paper to have merit there is a need for further study if the hypothesis is to be more widely accepted.
We would like to express our appreciation for the help and constructive criticism of Colin Plank, Animikh Sen, James Kxiapp, Michael Simmons, Robert Scott, and Dennis Kerr, who have all read and, in many cases, offered extensive suggestions for the improvement of the original manuscript of this paper. We take responsibility for the contents of the current paper, and none of these people can be blamed for the mista kes we may have made.
Figures & Tables
Middle East Models of Jurassic/Cretaceous Carbonate Systems
This volume will interest tectonic modelers, stratigraphers, sedimentologists, and explorationists. It is the product of the international conference of “Jurassic/Cretaceous Carbonate Platform-Basin Systems, Middle East Models” that was convened in December 1997 jointly by SEPM (Society for Sedimentary Geology) and the United Arab Emirates University in Al Ain, United Arab Emirates. The twenty-three papers present new data and interpretations arranged in three sections: 1) sequence stratigraphy, cyclostratigraphy, chronostratigraphy, and tectonic influences, 2) depositional and diagenetic models of carbonate platforms, and 3) hydrocarbon habitat and exploration/development case studies. New tectonic models of the Arabian Basin, new stratigraphic and sequence stratigraphic reference sections, new geochemical and source rock data, and new reservoir data are presented. New geologic models make this set of papers relevant to geoscientists working outside of Arabia also.