Site 1203, at 50°57.00'N, 167°44.40'E, is located toward the central region of the summit area of Detroit Seamount (Fig. F7). This position was initially selected along an available, high-exaggeration (12–15:1) single-channel (analog) seismic reflection profile collected in 1988. An underway geophysical survey was conducted to gather digital seismic reflection data to more adequately characterize the stratigraphic and structural setting of the site; high-resolution profiles were collected using a single 80-in³ water gun firing at a rate of 6 s. Basement was clearly imaged, and Site 1203 was positioned along the crossing points of three survey lines.

Hole 1203A was spudded at a water depth of 2593 m. In the vicinity of the selected site, an acoustically prominent basement reflection is overlain by a 400- to 500-m-thick carapace of sedimentary deposits. The greater part of this sequence consists of late Oligocene and younger beds of the Meiji drift (Rea et al., 1995). Coring began at a depth of 300 meters below seafloor (mbsf) in diatom and nannofossil ooze beds of late, middle, and early Miocene age. At ˜390 mbsf, diatomaceous material gave way to chalk with abundant but poorly preserved nannofossils of late Oligocene age. Lower Eocene (upper part of Zone NP12; ˜51 Ma) (Berggren et al., 1995) chalky and sandy-silty sediment immediately overlies basaltic lava flows of the basement rock sequence. Beds of volcaniclastic sediment and chalk within the upper part of the cored basement complex contain Campanian nannofossils assigned to Zones CC22–CC23, the estimated age of which is 71.3–76.0 Ma. Toward the base of the basement section (˜400 m into basement or ˜865 mbsf), nannofossils characteristic of Zone CC22 were identified, indicating an age of 75–76 Ma (Berggren et al., 1995).

We reached basement at 462 mbsf. The underlying ˜453 m of cored basement consists of 18 lava units and 14 volcaniclastic interbeds (Fig. F8). The average recovery in basement was 56.5%. The upper part of the basement sequence defines deposition or emplacement in a distal environment relative to eruptive centers and at relatively shallow water depths. It is characterized by nonvesicular pillow lavas and thick, sparsely vesicular pahoehoe lava flow units interbedded with volcaniclastic sedimentary sequences of primary and resedimented basaltic tuff (ash fall deposits) and vitric siltstone and sandstone.

The lower part of the basement succession is dominated by highly vesicular compound pahoehoe lavas (up to 65 m thick) and includes subordinate lapillistone (i.e., scoria fall deposits), pillow lava, hyaloclastite tuff, and breccia, along with thin vitric siltstone to sandstone sequences. The highly vesicular pahoehoe flows and the lapilli scoria deposits are characteristic of eruption and emplacement in a subaerial setting close to a source vent. The presence of pillow lavas, hyaloclastite tuff, and marine vitric sandstone and siltstone within the sequence indicates emplacement (or deposition) in water (Fig. F9). These contrasting indicators imply that the depositional environment extended from shallow marine to land, an interpretation consistent with an early subaqueous emplacement and then subsequent emergence of the lava section as it thickened.

Similarities are noted between the Site 1203 basalt units and those recovered during Leg 145 at Sites 883 and 884, positioned, respectively, on the summit and at the base of the eastern flank of Detroit Seamount (Rea et al., 1995). For example, a plagioclase-phyric basalt containing centimeter-size glomerocrystic plagioclase phenocrysts (Fig. F10) is similar to seven of the ten igneous units described from Site 884. Differences include the presence of olivine-rich zones (that also contain Cr spinel) in Hole 1203A (Fig. F11).

Geochemically, tholeiitic to transitional basalt is present at the top of the sequence and alkali basalt occurs intercalated lower down (Fig. F12). Some alkali basalt lavas are geochemically distinct in that they have a substantially lower Ti/Zr ratio (Fig. F13). Whereas Na and K are subject to mobility during alteration, at Site 1203 the designation of basalt units as alkalic is supported by elevated abundances of Ti, Zr, and Y relative to the tholeiitic basalt. The Site 1203 alkalic basalt, which erupted as subaerial pahoehoe lavas, may be analogous to the dominantly alkalic postshield-stage lavas that erupt as Hawaiian volcanoes migrate away from the hotspot. However, the shift upsection to tholeiitic pillow lavas is not easily understood with reference to the evolution of single Hawaiian volcanoes. This sequence could have developed from interfingering of lava flows from two distinct volcanic centers that were in different stages of growth. This possibility is supported by the observation that at least one flow unit within the lower alkalic section is compositionally related to lavas in the upper tholeiitic section.

The igneous rocks recovered at Site 1203 have undergone secondary alteration and weathering. Evidence for pervasive low-temperature alteration is exhibited by vesicle- and vein-filling secondary minerals. The degree of alteration increases downsection (Fig. F14). Mineral paragenesis in the upper part of the hole is dominated by calcite, Fe oxyhydroxide, and brown and green (saponite) clay. Associated secondary minerals are sulfide (pyrite), blue-green clay (celadonite), and zeolite. Most vesicles are filled with calcite or saponite. Near the bottom of the hole (˜415 m into basement) vesicles are mostly filled with zeolite and Fe oxyhydroxide.

Bulk density measurements for that part of the sedimentary section cored (300–462 mbsf) show an increase with depth from ˜1.5 to 2.3 g/cm³. This gradient correlates with an overall decrease in porosity downhole, from ˜65% to ˜40%, which probably largely reflects increasing compaction, although the proportion of calcareous material with respect to diatomaceous debris also increases below ˜390 mbsf.

In the basement section, variations in index properties, gamma ray counts, and thermal conductivity correspond to the alternation of volcaniclastic sediment and basaltic lava flows. In the volcaniclastic sediment, bulk density and thermal conductivity are generally low, whereas porosity is high (>40%). Conversely, in the basalt units, bulk density and thermal conductivity are high and porosity correspondingly low (<20%). For the basalt, natural gamma ray measurements uncorrected for background radiation generally range between ˜15 and 30 counts per second. To orient core segments to their in situ position, digital photos of 200 m of whole-round cores were taken to compare with Formation MicroScanner (FMS) logging images.

Logging operations in Hole 1203A were extensive, including the collection of downhole natural gamma ray, density, porosity, electrical resistivity, and temperature data with the triple combination (triple combo) tool and FMS and velocity measurements in a second tool string. Downhole magnetometer data were also collected with the Goettingen borehole magnetometer (GBM) in a third run (Figs. F15, F16). Excellent data quality and repeatability were observed along the entire section during the three runs. Basaltic sections are characterized by high electrical resistivity (up to 10 m), low porosity (<0.5%), high density (up to 2.5 g/cm³), and low natural gamma ray (<20 gAPI). In contrast, sediment and volcaniclastic units exhibit low resistivity, high porosity, and high natural gamma ray counts. FMS electrical images are of high quality and can be used to distinguish pillow basalt and more massive units (Fig. F17). The borehole magnetometer, which employed three fluxgate sensors and an innovative fiber-optic sensor to record tool rotation, yielded data that can be used to identify sequences of basalt and volcaniclastic sediment (Fig. F18). The strength of the recorded vertical to the horizontal component of the anomalous field suggests that within the basalt section the combined remanent and induced magnetic field has an inclination >45°.

A total of 258 discrete samples were taken from cores of volcanic basement rock. These samples were measured for natural remanent magnetization and then were alternating-field demagnetized to 80 mT in 5- and 10-mT steps. Inclination, declination, and intensity were measured and orthogonal vector plots employed to determine the stability of remanence and the number of magnetic components present (Figs. F19, F20). Principal component analysis was used to determine the characteristic remanent magnetization direction. All samples exhibit normal polarity.

The average inclination of volcaniclastic units is 54.7° (+3.1°/–6.4°; 95% confidence level), which is a minimum value because compaction processes in sediment can rotate the remanent magnetic vector toward a reading shallower than originally set (Fig. F21). The minimum paleolatitude is thus 35.2° (+3.2°/–5.9°). Paleomagnetic inclination data for the lava flows were grouped according to flow unit and averaged. The means of the individual flow units were averaged to determine an overall mean inclination for the thickness of the basement section penetrated. Based on 16 units, the average inclination is 48.0° (+6.8°/–10.1°), a value that corresponds to a paleolatitude of 29.0° (+6.3°/–7.7°) (Fig. F22). This reading will change with improved age control (several flows may together represent a short period of time) and shore-based thermal demagnetization studies to address evidence that a high-coercivity component of magnetization present in some units was not adequately demagnetized with the alternating-field treatments applied.

The significant result of shipboard measurement is that the range of preliminary mean paleolatitude determinations extracted from volcaniclastic sediment and lava flow units (29°–35°) and the distributions of the data are distinct from the value predicted by the fixed hotspot model (i.e., ˜19°N).