Previous   |    Next

doi:10.2204/iodp.proc.344.207.2017

Results and discussion

The data are listed in Table T1. Figures F2, F3, F4, F5, and F6 show the K, Rb, and Cs concentration-depth profiles and their cross-element mixing plots.

Overriding plate sites

Site U1380

Site U1380 was drilled to investigate the deeper portions of the upper slope sequence and the underlying wedge sediments. The sedimentary succession recovered from the framework wedge consists of alternating terrestrial upper slope turbidites and shelf sequence, with some input from deltaic sediments. As mentioned above, this site is complementary to the adjacent Site U1378 drilled during expedition 334 to ~480 mbsf; therefore, Site U1380 was washed to ~450 mbsf and drilled to ~800 mbsf. Because of cementation at >600 mbsf, pore fluids were only recovered between ~450 and ~600 mbsf.

Chloride concentrations of the pore fluids are fresher than modern seawater by 30%–36%, suggesting communication with a fluid that migrated from a deeper source where the temperature is >60°C, high enough to support clay dehydration reactions (e.g., Perry and Hower, 1970; Bekins et al., 1994, and references therein). The in situ temperature at the depth range of the Site U1380 pore fluid samples is just 25°–32°C, lower than the required temperature for the smectite to illite (S/I) transformation reaction. K concentrations as well decrease with depth but to a greater extent than the observed Cl dilution (Table T1), indicating that, in addition to dilution, K is also involved in fluid-mineral reactions, most likely in the S/I transformation reaction. The concentration-depth profile of Rb, normalized to Cl, mimics that of K concentration (Fig. F2). The Cs concentration profile (also normalized to Cl) is also similar to the K and Rb profiles to ~540 mbsf; at >540 mbsf, however, Cs concentration abruptly increases. Based on just one datum point it is impossible to conclude if this increase is a sample handling problem or a real datum point that suggests an in situ diagenetic reaction that releases Cs.

Site U1413

Site U1413 was also drilled in the middle slope region along a 3-D seismic line (Bangs et al., 2013). The objectives were to determine the nature, composition, and physical properties of the slope sediment and to constrain the fluid regime of the slope sediments. The upper ~150 mbsf is a slump. Three lithostratigraphic units were distinguished in the sediment section. At this site the Cl concentrations somewhat decrease with depth below the slump, suggesting minor lateral flow of fresher fluid in the more sandy horizons at ~500 mbsf. Several tephra layers were recovered at the top of the section and between 135 and 180 mbsf.

Similar to Cl concentrations K concentrations also gradually decrease with depth and could be related to the decrease in the Cl concentration; some of the decrease in K concentration could as well be caused by tephra alteration to illite/smectite or to zeolites.

The Rb depth-concentration profile (normalized to Cl) (Fig. F3) is very similar to that of K; Rb and K may be involved in the same diagenetic reactions. Cesium concentrations, however, remain constant with depth except for an increase at ~500 mbsf, which is very similar to the increase in Cs observed at Site U1380 at a similar depth, where some sand layers are present. The cause of this increase is as yet unclear, and could be caused by some lateral flow of a fluid enriched in Cs.

Site U1412

At Site U1412, drilled on the prism toe, borehole instability precluded sampling the décollement and recovering the underthrust sediment and igneous basement. Sample recovery was very poor especially in the deeper parts of the section below the bottom simulating reflection (BSR) at ~200 mbsf. Neither Cl nor K concentrations vary significantly with depth except for the artifact caused by localized gas hydrate dissociation in some upper section horizons. The Rb concentration-depth profile and the K/Rb mixing plot (Fig. F4), although similar above the BSR, indicate some increase in Rb but not in K concentration at depth; carbonate recrystallization would exclude Rb but not K. The Cs profile deviates even more from the K concentration-depth profile—Cs concentrations gradually increase with depth from ~100 mbsf, suggesting diffusional communication with a deeper Cs-enriched fluid.

All the above overriding plate sites show communication with a deeper Cs-enriched fluid, suggesting that the most incompatible element—Cs—is not involved in any of the most important overriding plate sediment diagenetic reactions; hence, Cs is being expelled into the fluid phase that is circulating at greater depths. At Site U1412, the deepest pore fluid samples analyzed closest to the décollement, are also enriched in Rb, possibly suggesting communication with the underthrust section, which may be enriched in both Rb and Cs.

Incoming plate sites

The incoming plate Sites U1381 and U1414 were cored to characterize the nature of the sediments and oceanic crust entering the seismogenic zone.

Site U1381

Although cored during Expedition 334, Site U1381 was cored again during Expedition 344 in order to acquire higher-resolution data on the incoming plate. No significant variations in the Cl concentrations with depth are observed. The K concentrations, however, decrease with depth, most likely caused by ash alteration to clay minerals and possibly to zeolites. Both Rb and Cs concentrations decrease in the uppermost part of the section (Fig. F5); the ash alteration reactions to clay minerals and particularly possibly to zeolites could be responsible for the observed uptake of Rb. The cause of the decrease in Cs concentration is as yet unclear but could be caused by lateral fluid flow.

Several of the geochemical profiles at this site, especially of Sr and sulfate concentrations acquired shipboard, suggest that the pore fluids in the deeper sediment section at this site are influenced by diffusional communication with a fluid in the igneous basement (see the Input Site U1381 chapter [Harris et al., 2013c]), altered seawater. This was also observed, for example, at Sites U1039 and U1253 drilled offshore Nicoya Peninsula during ODP Legs 170 and 205, respectively (Silver et al., 2000; Morris, Villinger, Klaus, et al., 2003). Both Rb and Cs concentrations increase at greater depths at Site U1381. This suggests that the basement fluid is enriched in both Rb and Cs.

Site U1414

Site U1414 Cl concentrations do not vary with depth. Similar to Site U1381, K concentrations decrease with depth; some of the tephra alteration to clay minerals (and possibly zeolites) must be responsible for this observation. The Rb and Cs depth-concentration profiles (Fig. F6) suggest some uptake in the uppermost ~100 m; below, however, the concentrations increase with depth and show a sharp increase at ~300 mbsf, at the boundary between two lithologic units (Unit II/III boundary; Harris et al., 2013). A seismic reflector is observed at this depth, where carbonate cementation increases abruptly. No pore fluids could be recovered from the lowermost ~65 m of the sediment section, which is characterized by a sequence of lithified, calcareous, and siliceous cemented silt- and sandstone. Large increases in the concentrations of both Rb and Cs in the deepest pore fluids recovered, above the recrystallized section, indicates that during intense recrystallization and cementation of calcareous and siliceous oozes these two incompatible elements are being expelled into the fluid phase; hence, they could be used as a proxy for fluid flow at greater depths in forearcs.

The sediment cemented basal section prevents communication with the basement fluid observed at Site U1381.

Top of page   |    Previous   |    Next