The volcanic section at Site 1136 consists of three sedimentary lithologies with volcanic components (Units III, IV, and V) (Table T5) overlying three lava flows (Units 1, 2, and 3) (Table T6). Unit III is a very light brown foraminifer-bearing calcareous ooze with <5% disseminated palagonitized basaltic glass shards. Unit IV is a brown zeolitic calcareous volcanic clayey sand, which contains 10%-15% palagonitized basaltic glass. Unit V is a dark greenish gray carbonate-bearing zeolitic silty clay with <5% palagonitized basaltic glass, but with abundant authigenic clinoptilolite and heulandite. The three underlying lavas are of tholeiitic composition, and all are sparsely to moderately phyric basalts containing phenocrysts of plagioclase, clinopyroxene, and olivine. We have identified flow units on the basis of their internal textures, internal vesicle distribution patterns, and contact relationships. Volcaniclastic sediments immediately overlying the lavas were deposited at 105-107 Ma (see "Biostratigraphy").
The lowermost few meters of Unit III (Core 183-1136A-9R) contain <5% disseminated silt-sized palagonitized basaltic glass shards. The abundance of shards decreases upward from the base of the unit. Shards are dominantly oblate to tabular, but blocky and equant shapes are also common. There is little evidence of rounding, and shards have conchoidal and stepped surfaces. No microcrystallites or vesicles were observed; however, shards are small (<0.3 mm). The glass ranges from pale golden brown or dark brown to dark red. Some shards have a green rim, a few micrometers wide, of a more crystalline phase, which we interpret to be an iron-rich smectite clay mineral (nontronite) after palagonite. This observation is supported by XRD analysis of similar glassy material from lower in the stratigraphic section (Core 183-1136A-10R). A few shards show an internal polygonal (jigsaw fit) fracture texture that is consistent with quenching of glass in water.
The brown color of the volcanic clayey sand interval (Core 183-1136A-10R) is attributed to the 10%-15% silt to sand-sized yellow-brown palagonitized glass component. Texturally, the sand is composed of silt-sized aggregates of zeolites, calcareous microfossils, and volcanic shards forming a matrix enclosing sands dominated by palagonitized volcanic glass. Glass shards are angular to subrounded, equant, and blocky with conchoidal and stepped surfaces. We observed microcrystallites or vesicles in these shards. Some shards have green nontronite clay mineral rims.
A reverse graded interval at the top of Section 183-1136A-10R-1, 0-25 cm, shows a transition from medium to fine sand. The lower part of Section 183-1136A-10R-1, 25-87 cm, is fine sand and is internally massive (Fig. F6). Four subrounded centimeter-sized basaltic pebbles are distributed in this interval.
The clast-to-matrix ratio of the clayey-sand is ~70% sand and 30% silt and clay. The sand-sized component is dominated by palagonitized basaltic glass fragments with scattered opaque grains (10%-15%), authigenic glauconite grains, which are commonly casts of foraminifer tests (<5%), and calcareous microfossils (5%-10%). The silt- and clay-sized fraction is dominated by the zeolites clinoptilolite and heulandite (60%-70%), palagonitized basaltic glass shards (5%-10%), and a mixture of opaque grains, glauconite casts, and calcareous microfossils (10%-20%). The rounding of enclosed clasts and mixed nature of the sediment suggests that the clayey sand is epiclastic.
One piece of brown and white silicified mafic volcanic breccia is preserved at the base of Unit IV (Section 183-1136A-10R-1, 87-92 cm) (Fig. F6). Petrographic observations show that clasts in the breccia are internally brecciated. Domains that were once glassy are now replaced by opaline silica and sheaves of chalcedony, which are stained with iron oxide (hematite) and oxyhydroxide (goethite). Late-stage quartz veins further disrupt the silicified material. Silicification of sediments rich in volcanic glass may be caused by hydrothermal alteration after emplacement.
The gray-colored zeolitic silty-clay in Cores 183-1136A-11R through 14R does not contain a high proportion of palagonitized volcanic glass (<5%) and is principally composed of silt-sized aggregates of clay-sized zeolites and biogenic carbonate (fossil fragments) (see "Lithostratigraphy"). In the marine environment, zeolites commonly form in situ in the presence of altered volcanic glass. Fine-grained well-formed crystals of authigenic clinoptilolite and heulandite are extremely abundant in Unit V, suggesting a greater proportion of volcanic glass in these fine-grained sediments when they were originally deposited.
In the lower part of the section, especially Core 183-1136A-14R, bivalve shell material and other fossil detritus are mixed with the volcaniclastic sediments, indicating reworking. An articulated bivalve implies that reworking was relatively gentle and that the sediment mixture has not been transported far. In Core 183-1136A-14R green iron-rich clay (nontronite and saponite) and mica (celadonite) are present in halos around fossil fragments, in permeable silt layers, and along fractures (Fig. F7).
We made seven principal observations about the palagonitized volcanic glass and its distribution in the calcareous ooze in Core 183-1136A-9R: (1) the glass is variably oxidized, (2) glass shards retain angular to subangular shapes, (3) glass shards do not have vesicles or microcrystallites, (4) there is more glassy material closer to the base of the unit, (5) glass shards are almost wholly palagonitized, (6) some shards have clay mineral (nontronite) rims, and (7) some have internal polygonal fractures.
The proposed mechanism for distributing these shards into accumulating calcareous microfossil-rich sediment is that flow resedimentation elutriated fine-grained particles from mafic volcanic sediments forming a cloud of suspended material above the flow. These fine particles may be transported for short distances in the water column and will eventually settle on adjacent sediment. This mechanism is supported by the identification of flow resedimented volcanic clayey sands in Core 183-1136A-10R. Angular to subangular shards imply little abrasion during transport.
Whether the glassy material is originally derived from a subaerial or subaqueous environment is equivocal. The variation in extent of oxidation of the shards suggests subaerial exposure, but the lack of preserved vesicles, equant blocky shard shapes, and internal polygonal fracturing of grains are consistent with fragmentation during subaqueous quenching. The glass may have more than one origin in epiclastic sediments.
Devitrification of volcanic glass commences when shards start to cool, and immersion in seawater enhances the process of palagonitization and ultimately the formation of more crystalline phyllosilicate minerals. This process imparts the yellow-brown color to the preserved glass shards and explains the green nontronite rims on some shards.
The brown zeolitic calcareous volcanic clayey sand has a mixture of grain types, including marine microfossils indicating deposition in a marine environment. Absence of lamination in most of the clayey sand suggests limited traction during deposition, implying formation either by rapid deposition from suspension or deposition from a highly concentrated sediment flow. The preferred mechanism for emplacement is as a subaqueously deposited mass flow. The upper reverse-graded bed may have been formed during waning of flow when there was more grain-grain contact.
The variation in degrees of oxidation of the shards and the brown color of the silicified mafic volcanic breccia suggests subaerial exposure of the glassy material from which the volcanic clayey sand and breccia were derived; however, equant blocky shards in the clayey sand are consistent with interaction with water at the time of formation. A range in clast shape from angular to subrounded implies that some grains have been abraded before or during transport. Incorporation of a few subrounded basaltic pebbles supports this observation.
Euhedral tabular to prismatic heulandite and clinoptilolite are abundant in the clay-sized fraction and appear to have precipitated in situ. If clay-sized zeolites form at the expense of sand-sized basaltic glass shards similar to those preserved in Unit IV, the material initially deposited may have been a fine sand. Iron from altered basaltic volcanic glass and potassium from seawater contribute to the formation of authigenic glauconite.
Fine grained silty clay in Cores 183-1136A-11R through 14R suggests a quiet environment of deposition. Shallow marine fauna in Core 183-1136A-14R, including an articulated mollusk, implies deposition in, or reworking from, a shallow marine environment. Unit V is dominated by clay-sized euhedral calcium and potassium bearing zeolite minerals. Authigenic clinoptilolite and heulandite form in the marine environment when the activity of silica and of the alkali cations are high and pH in the pore fluids is low. Celadonite and iron-bearing smectite clay minerals (nontronite and saponite) in veins, pore spaces, and halos around fossil fragments or burrows suggest postdepositional alteration, possibly at elevated temperatures (Fig. F7) (see "Lithostratigraphy" and "Alteration and Weathering").
Unit III was deposited in a midbathyal to bathyal marine environment. Unit IV was reworked, probably into a deep neritic to midbathyal marine environment. Unit V was deposited in a shallow neritic environment, below wave base (see "Lithostratigraphy," Table T3). It appears that a progressive increase in water depth with time is reflected in the epiclastic succession and overlying calcareous sediments. This corresponds with a decreasing volcanic signature in the sediments with increasing water depth. Basaltic volcanic components in the epiclastic sediments at this site may be genetically associated with the effusive volcanic activity immediately preceding their deposition.
In Cores 183-1136A-15R through 19R, we recovered basalt from three separate lava flows (Fig. F8). In this section we focus on physical structures preserved in the core. Their petrography and geochemistry are described in "Igneous Petrology". Table T6 lists the recorded and estimated original thickness of the flows.
The upper portion of the first flow is highly altered (see "Alteration and Weathering"), and no vesicular flow top material was recovered. However, drillers noted a distinct drop in the drilling rate at 133 mbsf (see "Operations"). Three thin, subplanar, vesicular regions are present in Core 183-1136A-15R-1 at 49, 98, and 128 cm (see Fig. F8 for vertical distribution and Fig. F9 for one example). These regions contain up to 20 vol% elongate, angular vesicles ranging from 0.5 to 4 mm in diameter. The petrographic texture of the groundmass crystals and bulk composition of the lava changes distinctly in these regions (see "Igneous Petrology"). Another discrete vesicular patch, at interval 183-1136A-15R-2, 10-15 cm, is more subvertical. The vesicles inside this patch are similar to those in the subplanar regions, but are slightly more rounded in shape. The remainder of the lava has 0.5%-0.1% spherical vesicles with 1-5 mm diameter.
Vesicularity gradually increases near the base of the flow, changing from 0.1% in interval 183-1136A-16R-2, 0-80 cm, to 10% in interval 183-1136A-16R-2, 135-152 cm. These vesicles are moderately to highly spherical and rounded and are not associated with a groundmass with distinct petrographic texture. However, in interval 183-1136A-16R-2, 135-145 cm, a second population of vesicles appears in the form of one-vesicle-thick wispy bands. These are very small (diameter = <0.3 mm) and angular.
The first recovery of the second lava flow is in Core 183-1136A-16R-2, 0 cm, and the flow extends to almost the base of Core 183-1136A-19R. We used several macroscopic criteria to distinguish Units 1 and 2, and this division has been confirmed by thin-section examination (see "Igneous Petrology"). The macroscopic criteria used in hand sample are discussed in detail in the interpretive part of this section (see "Distinction of Units 1 and 2").
Most of the lava is relatively dense, containing 0.1-0.5 vol% spherical vesicles 1-5 mm in diameter. Subplanar vesicular zones similar to those in Unit 1 are scattered through most of Unit 2 but are most common in the upper parts of Cores 183-1136A-17R, 18R, and 19R. Generally, these vesicular zones become less common and thinner deeper in the flow (Fig. F8). Examination of the individual pieces indicates that the core preferentially split along these vesicular zones. Another important observation is that most of the subplanar vesicular zones and elongate vesicles in Section 183-1136A-16R-2 and Core 183-1136A-17R dip at 40°-60°. A 1.5-cm "megavesicle" is present in Section 183-1136A-17R-1, 70 cm.
In Section 183-1136A-19R-2, vesicularity gradually increases from 1% from 0-16 cm to 10%-15% from 80-93 cm (Fig. F8). The mean vesicle size decreases downward over this same distance, but the number density of vesicles increases strongly (from <0.01/cm2 to 30/cm2), producing a net increase in vesicularity. The lowermost 2 to 4 cm of the above described region also has a very fine grained groundmass. The frequency and degree of fracturing marginally increases in this same region.
The volcanic breccia recovered at the base of Hole 1136A in the interval 183-1136A-19R-2, 93-149 cm, is the third lava unit. Clasts range from 1 cm to larger than the core diameter (7 cm). These clasts have irregular shapes with moderate sphericity and are subangular to rounded with fluidal margins in places (Fig. F10). However, no signs of plastic deformation are observed. Vesiculation is variable, but generally 10-15 vol% with nonspherical, angular vesicles more prevalent in the upper 10 cm. In places, clasts are in close contact and could be agglutinated, but surfaces adjacent to void spaces have irregular apophyses and protrusions. Void spaces are infilled by a dark gray-green clay-bearing material, most likely dominated by Fe-rich smectite clay minerals (saponite and nontronite) formed after volcanic glass. No sedimentary structures are preserved in the infilling material. The pieces of lava are held together by a matrix of dark-colored smectite clays, which are likely to be the product of altering of fine-grained volcanic glass.
In discussing our interpretations of features within the lavas, we proceed from smaller to the larger features. Correspondingly, there are increasing levels of interpretation and decreasing levels of certainty.
Most lava recovered from Hole 1136A consists of dense, almost nonvesicular lava typical of the massive interior of inflated or ponded lavas. The subplanar vesicular zones with associated distinct petrographic groundmass appear to be identical to horizontal vesicle sheets (HVSs) exposed in subaerial basalt flows, but the 5- to 7-cm width of the core is insufficient to define sheets. The subvertical vesicular zone at interval 183-1136A-15R-2, 10-15 cm, is probably a vesicle cylinder. Such features are usually found in the massive interiors of flows, and HVSs, in particular, are typically concentrated near the top of the massive interior. Megavesicles, such as observed at Section 183-1136A-17R-1, 70 cm, also indicate the upper part of the massive interior. From these vesicle features, we conclude that most recovered rock from Units 1 and 2 is from the massive interiors of inflated or ponded flows.
The other key vesicle feature is the increase in vesicularity toward the bases of both Units 1 and 2 (Fig. F11). These profiles are similar to the bases of inflated pahoehoe flows in the Columbia River flood basalt province (L. Keszthelyi and Th. Thordarson, unpubl. data). The thin zone with a high density of small vesicles is the chilled base of Unit 2. We interpret the wispy features in interval 183-1136A-16R-2, 0-80 cm, to be differentiated material (mesostatis) just starting to segregate from the lava when it was frozen in place. The trapping of these wispy segregation veins by the cooling front and the increasing vesicularity strongly suggest that the base of Unit 1 is within a few centimeters of Sample 183-1136A-16R-2, 145 cm. However, the actual chill margin was not recovered.
Vesicularity and vesicle number density correlate closely (Fig. F8). Vesicle size and vesicularity also generally correlate, but some of the most interesting vesicle features do not. In particular, the chilled margin at the base of Unit 2 has high vesicularity but very small vesicles.
We initially separated Units 1 and 2 on the basis of vesicle structures. Although the vesicular base of Unit 1 was a good indicator, it is possible to confuse vesicular bases with horizontal vesicle zones, which can form in the upper vesicular crust of large inflated pahoehoe flows (see "Physical Volcanology" in the "Explanatory Notes" chapter). The most convincing macroscopic evidence that Unit 1 and 2 are separate lava flows comes from the orientation of HVSs and other usually subhorizontal features. Within Unit 1, these features dip about 20°, but in the upper part of Unit 2 they dip 40°-60°. In the lower part of Unit 2, they again dip ~20°. Dip of HVSs should not vary like this deep within a single flow. See "Structural Geology" for a more detailed description of these dipping features and their interpretation. As noted earlier, the igneous petrology confirms that Units 1 and 2 are different lavas (see "Igneous Petrology").
The volcanic breccia below the chilled base of Unit 2 does not fit any textbook examples. The cauliflower-like clasts are not the typical jagged, spinose aa clinker, nor are they like Hawaiian slab pahoehoe. No evidence, such as long stretched vesicles, was observed that would suggest the recovered lava clasts were extensively distorted while in a plastic state. The rounding of the clasts suggests some movement and/or reworking. The individual clasts are not highly fractured as is typical of hyaloclastites or peperites.
We interpret the clay matrix between the clasts as altered granulated or pulverized glassy volcanic silt and sand. Such material is present in quantity in quench fragmented hyaloclastitic rubble and in subaerial aa flow tops.
All that we can confidently conclude is that Unit 3 is a part of a disrupted flow top. Although similar rubbly flow tops are found in the Columbia River Basalts and Iceland (L. Keszthelyi, unpubl. data; Th. Thordarson, unpubl. data), the interpretation of these transitional lava types with no Hawaiian analog is problematic.
Based on our interpretation of the vertical distribution of vesicles and study of inflated flows elsewhere, it is possible to make tentative conclusions about original flow thicknesses (Table T6) and which portions of the core were not recovered. This is shown in diagram form in Figure F8. The general rule used in determining original flow thicknesses is that the massive interior of inflated flows typically makes up 40%-60% of the total flow thickness. The 6.23 m of recovered massive core from Unit 1 and ~13 m from Unit 2 suggests total original flow thicknesses of ~12 and ~25 m, respectively. Correcting for the possible ~20° local dip (Fig. F36; see "Structural Geology") lowers these values to 11 and 23 m, which is well within the errors of this crude estimate. There is insufficient unrecovered core to accommodate the >20 m estimated thickness of Unit 2. The vesicular flow top would be thinner in proportion to the massive interior if the flow was emplaced relatively rapidly for an inflated flow. Alternatively, there might have been sufficient time between the emplacement of Units 1 and 2 for significant erosion. However, the most likely explanation, given the highly tilted HVSs in the top of Unit 2, is that there is significant topography to the top of Unit 2. Topography of 3-5 m is very common on inflated sheet flows (e.g., Self et al., 1997). In contrast, there is sufficient nonrecovery to accommodate the full estimated thickness of Unit 1, if we assume that the contact felt by the drillers at 133 mbsf represented a reduction in alteration and vesicularity, rather than the very top of the flow. All that can be concluded about the thickness of Unit 3 is that it should be >1 m and is probably much greater.
As noted above, the observed internal structures and flow thicknesses are very similar to continental flood basalt lava flows interpreted to have been emplaced as inflated pahoehoe sheet flows (e.g., Self et al., 1997). Flows 10-20 m thick are very difficult to produce on slopes >1°, unless they are confined by topography, so it is likely that these flows were originally emplaced on gentle slopes over a period of several months. The vesicularity and morphology of the recovered lava is consistent with emplacement in a subaerial environment.