IODP Proceedings    Volume contents     Search


Site summaries

Hole 395A

Investigating coupled geochemical and microbial processes in active aquifers within the upper oceanic crust is the main science goal of Expedition 336, and the primary objective is initiating multilevel CORKs. We planned to install one observatory in the 664 m deep Hole 395A in the southeastern part of North Pond. Hole 395A was drilled during Leg 45 in 1975–1976, was logged repeatedly, and was then equipped with a first-generation CORK in 1997 during Leg 174B (Becker, Malone, et al., 1998). Hole 395A is located in an area of exceptionally low conductive heat flow (Langseth et al., 1992) due to cooling of the uppermost basement by cold seawater, which recharges basement and is inferred to flow underneath the sediment cover in a northerly direction.

At the beginning of our operations at Site 395, the old Leg 174B CORK, including the entire 603 m internal string with thermistors, a data logger, and pressure sensors, was successfully pulled out of Hole 395A and secured on board the R/V JOIDES Resolution. The pressure and temperature data were downloaded, the thermistors were cut out of the string, and sections of the string were sampled for microbiological analyses. Further microbiological samples were obtained from the CORK’s ROV platform and wellhead elements. The hole was then logged with a new in situ deep UV (<250 nm) fluorescence tool for detecting microbial life in ocean floor boreholes—the DEBI-t. Other logging data obtained included spectral gamma ray and temperature. A rock ledge in the borehole at ~180 mbsf had to be bridged by lowering the logging bit to ~198 mbsf, but then an open-hole section of 405.7 m was logged (total depth reached was 603.5 m). The lowermost ~50 m of the hole was not logged because it was found to be filled with rubble during Leg 174B. The logging results are consistent with the data obtained by Bartetzko et al. (2001) and allow the distribution of massive basalt, pillow basalts, altered lava flows, and rubble zones (sedimentary breccia and hyaloclastite) to be distinguished.

A 530 m long, multilevel CORK observatory was assembled to perform long-term coupled microbiological, biogeochemical, and hydrological experiments. Assembling the observatory entailed preparing osmotically driven fluid samplers, microbial incubation experiments, seven temperature sensors, and two oxygen sensors. Packers at midpoint depths of 111, 149, and 463 mbsf were installed to isolate the borehole into three intervals characterized by different thermal and fluid flow regimes. Umbilicals containing fluid sampling lines attached to the outside of the CORK casing were designed to reach depths of 120, 430, and 506 mbsf. OsmoSamplers for fluid geochemistry and microbiology were lowered on Spectra rope inside the slotted or perforated CORK casing to sample four intervals: 112–140, 233–261, 409–438, and 491–527 mbsf. The CORK wellhead was instrumented with sensors for monitoring pressures in the four zones isolated by packers and with OsmoSamplers for retrieving fluid samples from the lowermost zone.

The assembly and installation proceeded well until the CORK head broke off during the final step of releasing the CORK running tool. The CORK head experienced forces that bent the wellhead and severed its 5 inch pipe ~4 m below the top of the reentry cone, parting the Spectra rope and the umbilicals and leaving the downhole tool string in place. On the basis of the portion of the CORK wellhead recovered, the upper end of the remaining 5 inch diameter cup packer subassembly near the seafloor (5 inch pipe mandrel) is not completely rounded; however, it may be open enough to allow recovery of the internal downhole samplers, sensors, and experiments in the future. Several stainless steel tubes likely extend above the cup packers and the top of the 5 inch casing. Damage to the stabilizing fins above the cup packers suggests that they may have been too large in diameter to enter the throat of the reentry cone (DSDP documentation indicated a 24 inch diameter, now thought to be less), which may have been the root cause of the installation failure. Indeed, similar damage was observed on the Leg 174B CORK that was recovered (it, too, did not fully land). The CORK pressure logging system was recovered along with the broken-off wellhead. The recorded data do not definitively resolve whether the downhole CORK packers actually inflated; however, no data are otherwise available to suggest that the packers did not inflate as intended. A plan is being formulated to recover the downhole instrument string in 4 y with an ROV.

Site U1382

Hole U1382A was drilled 50 m west of Hole 395A at 22°45.353′N, 46°04.891′W, in 4483 m water depth. The primary objective for Hole U1382A was to install a CORK to perform long-term coupled microbiological, biogeochemical, and hydrological experiments in uppermost basaltic crust in this area of very low conductive heat flow. Coring and downhole logging of basement were also conducted.

After the reentry cone with 53 m of 16 inch casing was jetted in, the hole was deepened by drilling with a 14¾ inch tricone bit to 110 mbsf without coring. Basement was encountered at 90 mbsf, and 3 m of basement was penetrated in 30 min. The interval from 93 to 99 mbsf was drilled very quickly and is inferred to be sediment; however, the underlying formation to 110 mbsf was drilled slowly (2–3 m/h) without significant torque. Casing (10¾ inch) was installed and successfully cemented to 102 mbsf. RCB coring recovered basement from 110 to 210 mbsf (Cores 336-U1382A-2R through 12R). In total, 32 m of core was retrieved, with recovery rates ranging from 15% to 63%. This succession resembles the lithostratigraphy encountered in Holes 395 and 395A and provided excellent sampling material for various microbiological and petrologic studies.

The shipboard petrologists divided the core into eight lithologic units, comprising numerous subunits. Major unit boundaries are defined by contacts between massive and pillowed flows and interlayered sedimentary units. Each major lava flow unit consists of several cooling units, which are recognized by glassy or variolitic margins or marked changes in grain size. Results from thin section studies reveal a large range of grain sizes (glassy to medium grained) and diverse textures (aphanitic to subophitic or intersertal). Basalts are either aphyric or plagioclase-olivine-phyric and have <3% vesicles. Phenocryst contents range up to 25%, with plagioclase being more abundant than olivine. All of the volcanic rocks recovered from Hole U1382A are affected only by low-temperature alteration by seawater, manifesting as replacement of groundmass and phenocrysts, vesicle filling, glassy margin replacement, and vein formation with adjacent brown alteration halos. Chilled margins often show advanced palagonitization, which develops as blotchy alteration texture following the primary variolitic texture of the mesostasis. The extent of alteration ranges up to 20%, with clay (smectite and celadonite) being the most abundant secondary phase, followed by Fe oxyhydroxides and minor zeolites and carbonates. The recovered section has between 13 and 20 veins/m, with vein thickness being usually <0.2 mm. A sedimentary unit in Cores 336-U1382A-8R and 9R features a variety of clasts, including plutonic and mantle rocks. The peridotites are weakly serpentinized harzburgites and lherzolites with a protogranular texture. The intensity of deformation of the gabbroic lithologies ranges from undeformed to mylonitic. Minor cataclastic deformation of the peridotites has led to the development of carbonate-filled vein networks, along which the rocks have been subjected to oxidative alteration, resulting in the breakdown of olivine to clay, oxide, and carbonate.

Physical properties measurements reveal typical P-wave velocities for these lithologies and a correlation between sonic velocity and porosity of the basalt. Elevated potassium and uranium concentrations in the oxidatively altered part of the core were revealed by natural gamma radiation (NGR) core scanning. Thermal conductivity also reflected the typical values associated with basalt and peridotite and showed small variations with depth.

Whole-rock geochemistry analyses reveal systematic differences in compositions between aphyric and porphyritic basalt, which are due to plagioclase accumulation in the porphyritic basalt. The aphyric basalts show a liquid line of descent that is controlled by the fractional crystallization of olivine. As the extent of alteration increases, loss on ignition values and potassium concentrations also increase. Immobile trace element ratios (Zr/Y and Ti/Zr) indicate that parental magma compositions for the basalts above and below the sedimentary unit are different from each other. Petrographically and geochemically, the basalts correspond to the uppermost lithologic units identified in Hole 395A. Likewise, a sedimentary unit with varied plutonic and mantle rocks was also observed in Hole 395A.

A primary objective of basement coring was to obtain samples for microbiological analysis. We collected 46 hard rock and 2 sediment whole-round samples for these studies (11% of core recovered). Samples were preserved for ship-based (deep UV fluorescence scanning, culturing and enrichment, and fluorescent microsphere analysis) and shore-based (DNA and RNA analysis, fluorescence in situ hybridization (FISH), cell counting, and isotopic analysis) studies. Generally, 1–3 microbiology hard rock samples were collected from every core section. Hard rock samples span a range of lithologic units, alteration states, and presence of chilled margins, and some contain veins or fractures. Additionally, a few recovered plastic bags that held the fluorescent microsphere solutions in the core catcher were collected as a contamination check via DNA analysis.

An open-hole section of 105.61 m was logged over a period of ~19.5 h with two tool strings (the adapted microbiology combination I [AMC I] and the FMS-HNGS). Downhole log measurements include natural total and spectral gamma ray, temperature, density, electrical resistivity, electrical images, and deep UV–induced fluorescence (acquired with the new DEBI-t). The borehole remained in good condition throughout logging, and no obvious tight spots were encountered in open hole. Integration of core and log measurements and observations showed excellent correspondence between potassium concentrations provided by shipboard NGR, the spectral gamma ray logging tool, and whole-rock geochemical analyses. FMS data were combined with images of the external surfaces of whole-round cores. Prominent veins with alteration halos in cores from the massive flows were matched with fractures in the FMS images. Also, logging results constrain the depth of the peridotite interval from 165 to 167 mbsf (based on density and low K/U ratios).

Downhole hydrologic (packer) tests failed because ship heave up to 3 m prevented the packer from sealing in the casing for >10 min.

A CORK to monitor and sample a single interval in uppermost basement was successfully installed in Hole U1382A. The 210 m deep hole was sealed with a 189 m long CORK with nine external umbilicals and a retrievable internal instrument string. The umbilicals include one for pressure monitoring, two for microbiological sampling, and six for fluid sampling. The retrievable internal instrument string comprises several osmotic pump-driven samplers for basement fluids and microorganisms as well as enrichment experiments, an oxygen probe, and a thermistor with data recorder. The samplers and probes extend from 146 to 174 mbsf and are kept in position by a 150 lb sinker bar at 177 m. A pressure gauge and a fast-pumping OsmoSampler are situated in the wellhead and monitor/​sample fluids from 162 mbsf.

Site U1383

Site U1383 (prospectus Site NP-2) is located 5.9 km northeast of Site U1382 in an area of elevated conductive heat flow in 4414 m water depth. The primary objective at Site U1383 was to install a multilevel CORK for long-term coupled microbiological, biogeochemical, and hydrological experiments in uppermost basaltic crust. Basement coring, downhole logging, and hydrologic experiments were also planned.

Hole U1383B at 22°48.1328′N, 46°03.1556′W, was prepared for drilling 500 m into basement by installing a reentry cone with 20 inch casing extending to 35 mbsf. We then prepared the hole for 16 inch casing by drilling an 18½ inch hole to 68 mbsf; the sediment/​basement interface is at 53 mbsf. After installing and cementing the 16 inch casing to 54 mbsf, we started to prepare the hole for 10¾ inch casing by drilling a hole into basement with a 14¾ inch tricone bit. We decided to abandon Hole U1383B after this tricone bit failed at 89.8 mbsf, resulting in large parts of the bit being left in the hole. Although we did not choose to deepen this hole, it remains a viable CORK hole because it has a completely functional reentry cone and casing system with ~35 m of accessible basement. A ROV landing platform was therefore installed in the reentry cone to facilitate future ROV operations, which will include installation of an instrumented plug in Hole U1383B.

Hole U1383C (22°48.1241′N, 46°03.1662′W) was started at the site of the original jet-in test, 25 m southwest of Hole U1383B. The primary objective was installing a multilevel CORK in the uppermost ~300 m of basement. The ultimate configuration of the CORK in Hole U1383C was to be determined by the depth of basement penetration and the downhole logging results. A reentry cone with 16 inch casing was installed to 34.8 mbsf, and a 14¾ inch hole was drilled to 69.5 mbsf for the 10¾ inch casing string. The sediment/​basement interface was encountered at 38.3 mbsf. After cementing the 10¾ inch casing at 60.4 mbsf, drilling in Hole U1383C proceeded with a RCB bit from 69.5 to 331.5 mbsf. From this 262 m long interval, 50.3 m of core was recovered (19.2%). Rocks are glassy to fine-grained basalts with variable phenocryst (plagioclase and olivine) contents. Three major lithologic units were distinguished on the basis of primary texture and phenocryst abundance. From 69.5 to 127 mbsf, the core consists of microcrystalline to fine-grained, sparsely plagioclase-phyric basalt with abundant glassy margins and numerous intervals of hard interflow limestone. From 127 to 164 mbsf, massive plagioclase-olivine-phyric basalts occurs, occasionally hosting limestone (with and without basalt clasts) as fracture fill. Below 164 mbsf, glassy to variolitic to cryptocrystalline basalts (most likely pillow flows) predominate, and limestone is largely missing. Each of these three main lithologic units is divided into numerous subunits on the basis of hyaloclastite layers and rare tectonic breccias. The overall abundance of glass is noticeably greater than that in Hole U1382A, and the extent of palagonitization ranges from weak to moderate. Basalts are avesicular to sparsely vesicular and show vesicle fills of clay, zeolite (mainly phillipsite), calcium carbonate, and Fe oxyhydroxide. Brownish alteration halos commonly track veins filled with clay or carbonate and zeolite. Within Unit 3, a gradational change from glassy, to variolitic with abundant hyaloclastite layers, to more massive microcrystalline, to fine-grained basalt with rare glassy margins can be observed.

Although the hyaloclastites are noticeably palagonitized throughout the hole, the extent of background alteration appears to decrease downsection. Vein densities average 33 veins/m and increase somewhat downsection to 50 veins/m. Zeolite veins are abundant in the upper section of the drilled interval, whereas carbonate veins predominate in the lowermost part. Sparse vesicles are filled with zeolite and clay.

Physical properties measurements reveal tight correlations between sonic velocity, density, and porosity of the basalt. These correlations reflect changes in physical properties as a result of low-temperature alteration. In basalts with up to 40% alteration, P-wave velocities and bulk densities as low as 4750 m/s and 2.43 g/cm3 were recorded. These most altered basalts also show exceptionally high porosities (up to 16.6%). Despite the locally high alteration intensity, NGR core scanning revealed fairly low average potassium and uranium concentrations (e.g., 0.19 ± 0.05 wt% K). Fresh basalts show physical properties and K concentrations typical for mid-ocean ridge basalt.

Whole-rock geochemical data show that basalts from Units 1 and 2 systematically differ in Zr/Y and Zr/Ti ratios from basalts from Unit 3. The compositional variability in the different units is primarily due to fractionation of olivine, but some trends (gain of K, loss of Mg) are also related to increased alteration intensity. The porphyritic basalts do not show the distinct plagioclase accumulation signature revealed in similar basalts from Holes U1382A and 395A. Correlations between Site U1383 and Sites U1382 and 395 based on geochemical composition cannot be made for individual flow units, indicating that the sites belong to different volcanic centers that were fed by mantle sources with variable compositions.

Hard rock samples for microbiological analysis were collected from every RCB core from Hole U1383C. Roughly 12% (6.11 m total) of core recovered from Hole U1383C was taken as whole-round samples from the core splitting room and dedicated to microbiological analysis. The 79 hard rock microbiology samples span a range of lithologic units, alteration states, and presence of chilled margins, and some contain at least one vein or fracture. Additionally, a few background contamination samples were collected for shore-based DNA analysis, including recovered plastic bags that held the fluorescent microsphere solutions in the core catcher.

Samples recovered were preserved for shore-based DNA and RNA analysis, shore-based FISH and cell counting, ship-based culturing and enrichments, shore-based isotopic analysis, and ship-based fluorescent microsphere analysis. Microspheres were used during all coring operations to help in evaluating core contamination. The enrichment and cultivation experiments initiated include carbon fixation incubations; carbon and nitrogen cycling experiments; enrichments for methanogens, sulfate reducers, sulfide oxidizers, and nitrate-reducing iron-oxidizing bacteria; and enrichments for heterotrophic metabolisms.

Deep UV (<250 nm) scanning of hard rock materials was used to identify sample regions with concentrations of biomass and organic material.

Wireline logging data include natural total and spectral gamma ray, density, compressional velocity, electrical images, and deep UV–induced fluorescence (acquired with the new DEBI-t) of a 274.5 m section of open hole. Lithologic Unit 1 is characterized by variable caliper, density, and sonic velocity values. Gamma ray intensities are generally low but increase in the bottom part of the unit. Lithologic Unit 2 has a uniform caliper and high densities and apparent sonic velocities and shows high-resistivity massive flows with fractures in the FMS images. The upper section of lithologic Unit 3 (153–166 mbsf) is characterized by decreases in density, apparent resistivity, and velocity and an increase in gamma ray intensity. This interval corresponds to thin flows with interpillow/​flow sediments and tectonic breccia. From 166 mbsf to the bottom of the hole, the logging data reveal fairly uniform values for density, velocity, and apparent resistivity. Areas with peaks in gamma ray intensity correspond to intervals with abundant hyaloclastite in the recovered core (in particular around 175 mbsf and from 220 to 250 mbsf).

Drill string packer experiments were attempted in Hole U1383C to assess the transmissivity and average permeability of open-hole zones bounded by the bottom of the hole at 331.5 mbsf and three different packer inflation seats at 53, 141, and 197 mbsf. The packer experiments were not successful.

In preparation for the CORK observatory, Hole U1383C was cased through the 38.3 m thick sediment section with 10¾ inch casing to 60.4 mbsf and through a 14¾ inch rathole to 69.5 mbsf. After the hole was RCB cored to 331.5 mbsf and cleaned in five wiper trips, there was no noticeable fill, which was verified during logging. The CORK screens and downhole instrument string targeted three zones, selected mainly on the basis of recovered core and the caliper log. An upper zone extends from the combination packer and landing seat at 58.4 mbsf in the casing to the first open-hole packer and landing seat at 145.7 mbsf. Within this section the miniscreens are centered at 100 mbsf, with the slotted portion of the casing extending from 76 to 129 mbsf. The middle zone extends to an open-hole packer and landing seat at 199.9 mbsf. Within this section the screens are centered at 162 mbsf, with the slotted casing extending from 146 to 181 mbsf. The deep zone reaches to the bottom of the hole (331.5 mbsf) with miniscreens centered at 203 mbsf. The miniscreens are connected to the wellhead by five umbilicals with stainless steel or Tefzel internal tubing that are strapped to the outside of the casing. The downhole tool string consists of six different OsmoSampler packages, two dissolved oxygen sensors and recorders (one in the shallow zone and one in the middle zone), two miniature temperature recorders, sinker bars, sealing plugs, and interspersed sections of ⅜ inch (0.95 cm) Spectra line. The wellhead is instrumented with a pressure logger monitoring each of the three horizons and bottom seawater and a fast-flow OsmoSampler with both standard and microbiological sampling packages. The CORK extends to 247.6 mbsf, yet leaves the bottom portion of the hole open for future logging and access (247.6–331.5 mbsf).

Sediment and basement contact coring (Holes U1383D, U1383E, U1382B, and U1384A)

This section summarizes the results of APC sediment coring and XCB coring of the sediment–basement transition at Sites U1383 (prospectus Site NP-2), U1382 (near Hole 395A), and U1384 (prospectus Site NP-1).

In Hole U1383D, 44.3 m of sediment was cored, of which the lowermost 1 m was XCB cored through basalt and limestone-cemented breccia (0.76 m of basement was recovered). At nearby Hole U1383E, 44.2 m of sediment and 1 m of basaltic basement was cored (0.3 m of basement was recovered). The basalts are aphyric and slightly to moderately altered. They are distinct from the uppermost basaltic flow that was RCB cored in Hole U1383C and hence represent a different lithologic unit.

Hole U1382B was drilled midway between Holes 395A and U1382A. A total of 90.0 m of sediment was APC cored, and another 8.8 m was cored with the XCB, recovering a piece of basalt and countless millimeter- to centimeter-sized pebbles of completely altered plutonic and ultramafic rocks at the basement/​sediment interface. These rocks are interpreted to be part of the sedimentary breccia overlying the massive basalt of Unit 1 cored in Hole U1382A.

Coring in Hole U1384A recovered 93.5 m of sediment overlying 0.58 m of basalt and limestone-cemented breccia. The basalts are aphyric and sparsely vesicular with glassy to variolitic to microcrystalline groundmass. They are between 3% and 10% altered and display brown alteration halos along clay veins and fractures.

The sediments at all sites consist of foraminifer nannofossil ooze with layers of foraminiferal sand. The bottom several meters of the sedimentary pile are brown and appear rich in clay. Sediments from Hole U1382B show moderately rounded rock fragments concentrated in layers or dispersed in the ooze. These fragments range from coarse sand to pebble in grain size and consist of serpentinized mantle peridotite, gabbro, troctolite, and basalt. Both XCB cores from Hole U1382B also contain coarse sediment with predominantly serpentinite clasts, including soapstone and talc-tremolite schist. The occurrence of these rock fragments is consistent with the polymict sedimentary breccia recovered during basement drilling in Holes 395A and U1382A. The deformed and metasomatized lithologies encountered in Hole U1382B corroborate the hypothesis that this material was transported to the Site U1382 area in North Pond by mass wasting events and that its source is an oceanic core complex, probably in the southern rift mountains. Layers of foraminiferal sand are abundant in all holes and many show erosional bases and normal-graded bedding, suggesting that they represent deposits of turbidity currents.

Each of the four holes cored was intensively sampled for microbiological and interstitial water analyses. The sampling program was similar for each hole. In total, we collected 167 whole-round samples for interstitial waters and 691 whole-round samples for microbiological analysis. Sampling density was increased in the bottom sections. Pore waters in these basal sediments are dominated by diffusion of components from the basement fluids into the sedimentary pile, and they allow estimation of basement fluid compositions by extrapolation. Whole-round cores were preserved for shore-based molecular analysis to provide a detailed description of the microbial community. Ship-based enrichment cultures were established to enrich for multiple metabolic functional groups. These cultures will be analyzed on shore for both metabolic activity and community composition. Sediment sections remaining after whole-round sampling were analyzed for oxygen concentration using optodes. Hard rock samples were sectioned and allocated following previous strategies established during the hard rock drilling phase of the expedition. Multiple basalt samples were provided for RNA/DNA, geochemistry, and culture analysis.

An extensive high-resolution physical properties data set was obtained for all holes. This data set includes whole-round, split-core, and discrete measurements of magnetic susceptibility, velocity, density, porosity, NGR, and resistivity.

Atmospheric and surface-water microbiology studies

Investigations of atmospheric and surface-water microbiology were conducted by a shipboard atmospheric microbiologist during Expedition 336 for a project titled “Long-range transatlantic transport of microorganisms in clouds of African desert dust: a study of atmospheric microbiology, chemistry, and the influence of desert dust on surface-water microbial ecology.” A full description of this project and the shipboard results are presented in a paper in this volume (Griffin, 2012).

Atmospheric samples were collected in transit and while on-station daily from 19 September to 11 November 2011 (54 sample days). As a result of incubation-period constraints, the resultant report included data collected from 19 September to 1 November (44 sample days). These data indicate that atmospheric particulate matter moving over the Atlantic Ocean from Europe and Africa carries culturable microorganisms and that colony-forming unit concentrations in these air masses correlate with particulate concentrations. Further, these data demonstrate a “fertilization effect,” where surface-water prokaryote concentrations correlate to near-surface atmospheric particulate concentrations. Numerous episodes of African dust presence occurred in the near-surface atmosphere over the period of the study. Many of the high-volume membrane filter particulate/​aerosol loads produced dark and light orange coloration similar to that of desert surface soils of the Sahara and Sahel. These samples will be evaluated postexpedition for chemical and biological composition. Dust deposition was so heavy at times that dust particles pooled into naturally occurring wind traps located on the roof of the R/V JOIDES Resolution, which enabled collection into resealable plastic bags. Particle count data and the strength of correlations with surface-water communities and atmospheric communities will be evaluated again after chemical analyses of the high-volume filters. These analyses will be needed to verify sources and to determine if the particulate loads on the light gray to dark gray filters are an artifact of diesel exhaust or biomass burning in Europe or Africa. These data, along with data such as wind speed, wind direction, and ship orientation, will be valuable in determining which air masses and their sources influenced these communities. Obviously, additional (shore-based) data and more careful interpretation are needed to accurately evaluate the true extent of the observed relationships outlined here and for those yet to be determined. Additional shore-based studies will include the identification of isolates and nonculture-based studies to determine community composition and the reaction of various species of bacteria to surface-water dust deposition.