We focus mainly on the geothermal data obtained at five drill sites (Sites 1003 through 1007) located along an east-west transect of the western margin of the Great Bahama Bank (Fig. 1). Figure 2 shows plots of temperature vs. depth and thermal conductivity vs. depth for these five sites. It also shows plots of the temperature against the cumulative thermal resistance (Bullard plots after Bullard, 1939) for the same sites. The thermal resistance of a depth interval is defined by the thickness of the interval divided by the thermal conductivity.
The geothermal profiles at these five drill sites show significant curvature in the upper 50-100 mbsf (Fig. 2). At Site 1003, the profile is slightly concave-upward in the upper 35-70 mbsf and is concave-downward in the depths below. At Site 1004, the uppermost two temperature points deviate positively from the linear trend of the lower points and the profile is concave upward. At Site 1005, the uppermost two points deviate negatively and the geothermal profile in the upper 60 mbsf is concave upward. Site 1006 exhibits a concave-upward profile most clearly from 30-100 mbsf. Closely spaced temperature measurements are available from this site from the Adara/APC deployments. Some temperature determinations at shallow depths have relatively large uncertainties (~1°C) because the corer sometimes moved during measurement in sediments that were relatively soft. At Site 1007, the uppermost two points deviate negatively from the linear trend. All the sites show some degree of concave-upward curvature except for Site 1007.
The curvature is also seen in the Bullard plots (Fig. 2), which should make a straight line for a steady state thermal regime. For all the sites, it is not possible to draw a straight line within the uncertainties of the individual temperature measurements. Site 1006, which yields the most detailed profile of the bottom-hole temperature, demonstrated the curvature most clearly. The curvature of the Bullard plots suggests that the perturbation of the temperature profile cannot be accounted for by the downhole variation in thermal conductivity. Either the thermal regime is not in steady state or there is a significant, convective component in the total heat budget at these relatively shallow depth ranges. More specifically, three mechanisms can be responsible for a concave-upward geothermal profile. The first is a recent increase in sedimentation rate. The second is recent increase in bottom water temperature (BWT). The third is influx of seawater into the sediments. We discuss these possibilities in the next section.
At all sites except Site 1003, the points below 70-100 mbsf in the Bullard plots show a linear trend clearly. The conductive heat flow can be determined from these deep portions of the measurements. We obtain the heat flow as the slope of the Bullard plot (Fig. 2). We calculate the standard error for each heat-flow determination using Student t-distribution with 90% confidence level. In Figure 2, the data points used in the heat-flow determination are shown as closed circles and those omitted are shown as plus signs. The heat-flow values at these sites show a very systematic variation along the transect (Table 2). From the west to east, heat flow decreases upslope. The lateral variation in heat flow is discussed in the next section.