One of the main assumptions of seismic sequence stratigraphy is that seismic reflections follow depositional surfaces. As such they are essentially time lines, and have chronostratigraphic significance (Vail et al., 1977). High-resolution seismic data from the Equatorial Pacific were some of the first data sets in the public domain to corroborate this assumption (Mayer, 1979a, 1979b; Mayer et al., 1986). However, several studies of seismic modeling show that under certain circumstances these preconditions might sometimes be violated (Rudolph et al., 1989; Schlager and Stafleu, 1993; Stafleu et al., 1994; Anselmetti et al., 1997). A good seismic data set and good core coverage along seismic lines are needed to evaluate the time consistency of seismic reflections. Very few data sets fulfill these criteria. The high-resolution seismic Line 106 along the Bahamas Transect and the five continuously cored drill holes of Leg 166 in combination with the precise time-to-depth conversion from the check-shot survey provide such a data set. Dating of the seismic reflections allows for a test of chronostratigraphic significance of the seismic reflections and, thus, the seismic sequence stratigraphic concept for the investigated carbonate platform margin setting.
The ages of the drilled cores of Leg 166 were determined by the shipboard party (Eberli, Swart, Malone, et al., 1997) using calcareous nannofossils and planktonic foraminifers. Postcruise work refined these ages. For every site across the entire transect, the average maximum age offset of all seismic sequence boundaries amounts to 0.32 Ma (Table 1; Anselmetti et al., in press) and thus lies within the combined range of seismic and biostratigraphic resolution. It is remarkable that all Neogene seismic horizons can be traced from the middle of the Straits of Florida up to the Great Bahama Bank with high chronostratigraphic accuracy, thus confirming the concept that seismic reflections represent time lines. Consequently, even the drift deposits maintain their biostratigraphic integrity although ocean currents transported the sediment to the drift. The seismic reflections thus correlate across the depositional facies transitions and boundaries without changing their chronostratigraphic positions.
The well-defined ages of the sequence boundaries along the Bahamas Transect (Table 1) can be compared to the eustatic sea-level curve of Haq et al. (1987) (Fig. 2). The eustatic curve has been adjusted to the Berggren et al. (1995) time scale to allow a comparison to the ages of the sequence boundaries along the Bahamas Transect that were also dated with the Berggren et al. (1995) time scale. Approximately two-thirds of the ages of sequence boundaries coincide well with proposed sea-level falls on the eustatic sea-level curve. Because local tectonics and sediment supply influence the position and resolution of sea-level falls, this correlation is remarkably high. Most of the known major sea-level falls in the Neogene are recorded in the strata along the Bahamas Transect. Examples include the sea-level fall in the earliest Miocene, the first two falls in the middle Miocene, the fall at the Miocene/Pliocene boundary, and most of the Pliocene-Pleistocene major sea-level falls (Fig. 2). There are also, however, some notable differences. Sequence Boundary O is a well-expressed boundary positioned on an overall sea-level rise on the Haq et al. (1987) curve (Fig. 2). This sea-level rise, 18 m.y. ago, has been recognized in the South Atlantic (Abreu and Anderson, 1998) and along the New Jersey margin (Miller et al., 1998). Other sequence boundary ages differ only slightly from the proposed eustatic lows. Most of these shifts are toward younger ages and fall within early rises of the eustatic curve. At this point, it is not possible to evaluate whether the ages on the eustatic curve need to be adjusted or if the ages along the Bahamas Transect are shifted because of local causes. At any rate, the precise dating along the Bahamas Transect has produced an important data set to assess global synchrony of third-order sea-level changes. Other excellent data sets such as the New Jersey margin transect are needed to complete this task.