Site 1254 is located on the lower trench slope of the Costa Rica forearc; targeted coring was conducted to penetrate through a fault zone within the forearc prism, as well as through the plate boundary décollement and into the uppermost underthrust sediments close to the location of Site 1040. Figure F8 shows that the section recovered at Site 1254 overlaps with that sampled at Site 1040, spanning a sparsely dated sequence that is presumed to be Pliocene-Pleistocene in age (Shipboard Scientific Party, 1997b). The section recovered at Site 1254 largely comprises massive dark greenish gray claystones and silty claystones with minor volumes of interbedded volcanic ashes, fine sands, and silts. Figure F9 shows a typical example of the sediment facies, with a thin diffuse altered volcanic ash layer interbedded within homogeneous claystone. This example is from a relatively coherent piece of core, yet the sediments and sedimentary rocks recovered often show moderate to extreme degrees of drilling disturbance so that it is commonly difficult to tell if the absence of sedimentary structures in the cut core surface reflects a true absence in the original section or whether they were present but have been homogenized by drilling. Nonetheless, coherent fragments of more lithified sedimentary rocks do indicate that much of the section is either massive or slightly mottled, suggestive of moderate bioturbation. There is no evidence for strong bottom-current activity in the sediment depositional process at Site 1254.
Although fine-grained sediment dominates the sequence, this material is typically not well sorted, with an array of clay, silt, and more rarely, fine sand grains visible to microscopic analysis (Fig. F10A). In addition, rare clasts of shallow-water carbonate rock are found as blocks in a muddy matrix. Together these characteristics suggest rapid deposition by way of debris or gravity flows as the dominant mode of sedimentation. There is a general trend downsection to more development of sandy layers. Cores 205-1254A-8R through 11R contain thin silty and sandy layers that are often strongly disrupted by drilling, with a soft sediment deformation character, indicating that the sediment was largely unconsolidated prior to recovery. Major coherent intervals of fine- to medium-grained sand were described in intervals 205-1254A-10R-4, 91-128 cm (315 mbsf); 12R-1, 37-52 cm (329 mbsf); 12R-3, 75-115 cm (333 mbsf); 14R-3, 85-97 cm (352 mbsf); 14R-7, 40-109 cm (356 mbsf); and 16R-4, 31-72 cm (367 mbsf). These sands are mostly massive and although subject to heavy drilling disturbance, especially in Cores 205-1254A-12R and 14R, the sand appears to have been partially lithified prior to recovery. Parallel laminations are noted locally (e.g., interval 205-1254A-12R-3, 74-91 cm) (Fig. F11), suggestive of current activity during deposition of at least that unit.
The variation in lithology downsection can be broadly assessed from data derived from microscopic analysis of smear slides, presented graphically in Figure F12 (see "Site 1254 Smear Slides"). The dominance of clay minerals within the sequence is readily apparent, as is the downcore decrease in volcanic ash. Variability in the sandy grade component is not well tracked by smear slide analysis, which tends to underestimate its abundance. The downhole increase in sand is more accurately shown by core description, shown in Figure F8. All samples showing abundant levels of volcanic glass are present above 250 mbsf, where the shards are usually found as very clear angular grains that have apparently not yet suffered heavy alteration (Fig. F10B). Volcanic material is, however, present at low (<10%) and moderate levels (<30%) in most of the sediment, sometimes still fresh (Fig. F10C) but also often heavily altered to palagonite or even zeolites deeper (>250 mbsf) in the section. In these cases, the grains show rounded geometries and are often cloudy in transmitted light. The continental provenance of the sediments cored in Hole 1254A is clear from the abundance of quartz and feldspar grains and also from the bright brownish red biotite mica flakes that are found at all stratigraphic levels. Biotite composes <5% of the total sediment mass. The quartz and feldspar grains are partially delivered to the forearc by volcanic ash fall, but the subrounded character of many of the grains indicates that these have experienced significant transport and are largely reworked from terrestrial source terrains, which themselves are mostly volcanic.
The biogenic component of the sediment is very low (<5%) and restricted to occasional nannofossils above 200 mbsf and below 360 mbsf. More significantly, below Section 205-1254A-15R-2, 142 cm (360.62 mbsf), the proportion of diatoms increases sharply (>10%) (Fig. F10E). The appearance of diatoms is considered important for understanding the structure of the forearc prism because the uppermost sedimentary unit in the subducting Cocos plate stratigraphy (Subunit U1A) recorded high percentages of diatom abundance (Shipboard Scientific Party, 1997a). It should be noted that although 360.62 mbsf marks the clearest lithologic break between underthrust and forearc wedge sediments, the structural base of the décollement lies at 364.2 mbsf (see "Structural Geology"). Although diatom abundance increases most sharply below 360 mbsf, it should be noted that diatoms forming up to 2%-3% of the total sediment are noted in sediments from Core 205-1254A-14R (Fig. F10D). Although this may indicate a structural break higher in the section, we suggest that it could instead reflect either original sediment variability within the forearc wedge strata or a tectonic interleaving of underthrust sediment above the principal boundary.
The downsection transition from diatom-free to diatom-rich sediment noted in the smear slides lies just above the décollement and represents a passage from forearc wedge sediment into sediment originally deposited on the Cocos plate in the Middle America Trench. In the core, the lithologic break is seen as a slight change of color from a dark-colored more clay-rich sediment above to a lighter more silty claystone below (Fig. F5).
Evidence for significant downslope resedimentation in the sediments above the lithostratigraphic unit boundary at 360 mbsf occurs in the form of occasional blocks of strongly lithified tan-colored carbonate rock (e.g., Fig. F13). The most prominent examples are found in intervals 205-1254A-7R-4, 42-45 cm (208.72 mbsf); 7R-4, 52-55 cm (208.82 mbsf); and 14R-CC, 20-23 cm (356.06 mbsf). The blocks are angular and <3 cm thick. Under microscopic thin section analysis, the lithologies present in the cobble recovered in Sample 205-1254A-14R-CC, 20-23 cm, were revealed to be a peloidal limestone interbedded with a micritic cemented sandstone. Figure F14A shows the peloids with no micritic matrix but instead a sparry calcite cement that shows the fanning pattern characteristic of precipitation with a fully saturated subtidal environment (Fig. F14C) (Flügel, 1982). The lack of dissolution of the peloids and the regular form of the calcite spar laths argues against significant flow of the pore fluid during cementation. The lack of a muddy matrix indicates that sedimentation occurred in a current-swept setting, whereas the lack of compaction of the peloids, typically interpreted as fecal pellets from shrimp or crablike organisms, constrains cementation and lithification to have occurred early after deposition and before burial compaction could occur. Such pelsparite limestone facies are most common in shallow shelf regions, above storm wave base, and below the intertidal zone (i.e., ~10-30 m water depth) (Wilson, 1974).
Further evidence for a shallow-water depositional environment is found in the identification of shallow-water bivalve shell fragments (Fig. F14B) and small gastropods (Fig. F14D), found within a medium-grade sandstone. This sandstone is composed largely of grains of altered volcanic rock with a significant subsidiary component of quartz and is well sorted, consistent with a current-swept shallow marine setting. The lithified carbonate cobbles are interpreted as having been transported as coherent clasts within muddy debris flows into the trench from their origin within the coastal zone. The enclosing muddy matrix sediments must represent debris flow deposits in which these blocks were carried into the lower trench slope area. Thus, the carbonate cobbles confirm the hypothesis that the section is dominated by reworked gravity and debris flow units, rather than being the product of a gentle hemipelagic rain and earlier accretion to the forearc wedge.
Compared to the sequence of well-preserved tephra found at Sites 1039 and 1253 on the subducting Cocos plate, little well-preserved tephra stratigraphy is found at Site 1254. Although occasional thin altered ash layers are recognized, they are rare, typically <2 cm thick and often completely altered to claystone. More coherent volcanic ash layers recognized in the core are listed in Table T2. Tephras are mostly seen in the upper cored intervals (155-215 mbsf) and are either structureless bands or occasionally show normal grading. The original volcanic glass shards are often altered to palagonite. Volumetrically the tephra represent <1% of the total section. However, much larger volumes of volcanic detritus falling in the forearc appear to be largely reworked into the dominant silt-clay sediment. Given the relative lack of evidence for bioturbation, we suggest that much of this reworking occurs during debris flow events that emplace the bulk of the section.
Two larger coherent ash layers were recovered at Site 1254. The thinner unit is identified in interval 205-1254A-5R-8, 14-20 cm (193.49 mbsf), and is marked by being light gray, homogeneous, and quite deformed by drilling. The larger tephra is a reverse-graded layer, recovered in Sample 205-1254A-8R-8, 22-65 cm (222.37-222.80 mbsf) (Fig. F15). This is colored light gray in its basal section and grades into lighter greenish gray coarse sand-grade material toward the top, where it has a sharp contact with the overlying dark greenish gray claystone. Both of the thicker ashes preserve relatively fresh glass shards and are interpreted as the product of primary air fall deposition, followed by settling through the water column.
The base of the tephra recovered in Sample 205-1254A-8R-8, 22-65 cm, was not recovered because this bed was at the base of Core 205-1254A-8R, the last core taken in the upper cored interval at this site. The study of available core results in a minimum thickness estimate of 43 cm. Because Site 1254 is ~150 km from the nearest arc volcano, this thickness at this range indicates that this must have been a very large eruption. For comparison, the 1980 eruption of Mount Saint Helens resulted in a tephra <2 cm thick at a distance of 150 km (Sarna-Wojcicki et al., 1981). The Site 1254 ash exceeds the Pleistocene eruption of Crater Lake that deposited the Mazama ash layer, which is only ~10 cm thick at a range of 150 km (Williams and Goles, 1968). The Minoan ash from Santorini Volcano provides the closest analog in being 25 cm thick at this distance from the source (Watkins et al., 1978).
Given the incomplete recovery of the tephra in Sample 205-1254A-8R-8, 22-65 cm, a more meaningful comparison between tephra can be made using grain size data. Grains within this unit are up to 400 µm in diameter with a median of ~100 µm. Following Ninkovich et al. (1978), the maximum grain size relative to the distance from the eruption location implies an eruption event comparable to or slightly larger than that which emplaced the Minoan tephra from Santorini Volcano (Fig. F16A). A comparison of the median grain size to the distance from the eruptive center shows that the tephra in Sample 205-1254A-8R-8, 22-65 cm, represents a large event, but one that is less than the predicted upper limit for terrestrial volcanism defined by Walker (1971). Consequently, we conclude that the tephra in Sample 205-1254A-8R-8, 22-65 cm, represents deposition from a very large event, yet one that is resolvably less powerful than the greatest known events, such as the Toba Volcano eruption (Sumatra, 75 ka). Because the section at Site 1254 is not dated by biostratigraphy or paleomagnetic work, we cannot unambiguously correlate this layer with any of the tephra found at Sites 1039 and 1253.
Major and trace element analyses of 15 sediment samples and 1 volcanic ash were conducted aboard the ship following the methods described in "Inorganic Geochemistry" in the "Explanatory Notes" chapter. Major, minor, and trace element analyses are presented in Table T3. The tephra was run on the ICP-AES separately from the sediments, and the results are discussed below. Analyses were made on freeze-dried bulk samples that are a mixture of sediment particles and sea salts (see "Inorganic Geochemistry" in the "Explanatory Notes" chapter). The data have not yet been corrected for porosity and pore water contribution to the bulk sediment composition.
The ICP-AES geochemical data at this forearc site can be useful for (1) assessing whether the sediments in the overthrust wedge are similar to those in the subducting section and testing models for sediment accretion; (2) determining the degree of chemical heterogeneity in the wedge; and (3) constraining the provenance of the clastic sediments composing the wedge.
Figure F17 shows variations in TiO2, Al2O3, SiO2, and Ti/Al with depth below seafloor. Volcanic detritus is nearly ubiquitous at low concentrations in much of the sediment recovered, and the tephra analyzed at Site 1254 has much lower aluminum and titanium concentrations than the surrounding wedge sediments. This distinction suggests that reworked tephras may account for only a relatively moderate proportion of the total sediment in the wedge.
Major element compositions are seen to change across the lithostratigraphic boundary from wedge to underthrust sediment. Although the alkali elements show little variability overall (Fig. F18), there is a slight decrease in TiO2, Al2O3, and SiO2 below the lithostratigraphic unit boundary. Within the range of elemental values recorded at Site 1040, these changes are small but likely reflect the higher proportion of biogenic hemipelagic material in the underthrust Subunit U1A compared to the more clastic sedimentary rocks in Subunit P1B (Fig. F8). Higher silica concentrations in the wedge may signify more advanced diagenesis because the formation of clay minerals and zeolites resulting from volcanic glass alteration will tend to concentrate silica. The internal silica source from diatoms may, however, supply the needed silica. The shift to generally higher potassium in the sediments of the wedge may also be consistent with this process (Fig. F18). Within the wedge the major element chemistry is remarkably constant, a characteristic also noted at Site 1040 (Shipboard Scientific Party, 1997b). The lack of variation is interpreted to indicate a relatively uniform sediment type with little interbedded biogenic sediment. The sediment is likely derived from the same source throughout the time of deposition and has experienced a large degree of homogenization during emplacement, as might be expected from the debris flow depositional process favored on the basis of the sedimentary structures and grain size character.
Figure F19 plots CaO, MgO, Sr, and Ba against depth for the sediments and volcanic ash at Site 1254. Levels of CaO and Sr remain constant with depth, suggesting a relatively constant low proportion of carbonate content throughout the wedge and top of the underthrust section. The lithostratigraphic unit boundary at 360 mbsf is best highlighted in terms of the trace element Ba, as also recognized at Site 1040. Ba rises sharply below 360 mbsf. Although higher concentrations of Ba are sometimes attributed to the presence of volcanic materials in the sediment, the evidence from the SiO2 and K2O concentrations instead favors higher volcanic influence in the wedge (Subunit P1B). Smear slide analysis did not identify a large increase in volcanic material below the level of the lithostratigraphic unit boundary. Consequently, we interpret the increased Ba concentration in the sediment below 360 mbsf (Subunit U1A) as a function of higher barite content in the hemipelagic units.
Major and trace element analyses of a single tephra (Sample 205-1254A-8R-8, 22-65 cm) are presented in Table T3. The volcanic glass appeared fresh under microscope inspection, suggesting that the bulk sediment analysis should be close to a melt composition. Although the analytical total only just exceeds 92%, this may not necessarily imply strong alteration because the magmatic water content of high silica arc glass can reach up to ~5% (Burnham and Jahns, 1962), meaning that only ~3% of the missing total may be attributable to postdepositional hydration. In terms of its major elements, the tephra in Sample 205-1254A-8R-8, 22-65 cm, is classified as a dacite using the scheme of Cox et al. (1979) and as a subalkalic magma under the classification of Miyashiro (1978). The geochemical discrimination plots of Middlemost (1975) show that whereas the ash is calc-alkaline, it is sufficiently K2O rich to be classified as medium K in that scheme (Fig. F20). These basic characteristics are consistent with the presumed source of the tephra in the volcanic front of Costa Rica.
Diagenesis appears to have proceeded to only a moderate degree at Site 1254, with the exception of the well-cemented redeposited carbonate blocks, which experienced early diagenesis further upslope. No soupy sediment was recovered, and the sediment was typically lithified, while still being easy to scratch or gouge with metal instruments (i.e., a chalk-type consolidation). Clearly some cementation had occurred, as evidenced by the recorded brittle deformation of the sediment (e.g., the development of polished joint planes, crosscutting original bedding fabric). Such features are detailed in "Structural Geology". Nonetheless, some of the sand units (e.g., intervals 205-1254A-10R-4, 91-128 cm; 315 mbsf) show strong plastic drilling disturbance that is incompatible with a lithified state prior to drilling over certain intervals above the décollement.
Cores 205-1254A-11R through 14R (319-357 mbsf) show development of occasional carbonate concretions that are often focused on burrow structures. These tan-colored bodies are irregular, usually >5 cm in diameter, and show a similar degree of consolidation as the surrounding sediment, unlike the redeposited carbonate blocks (Fig. F21). Rare authigenic pyrite is restricted to the upper cored section (150-185 mbsf) where it is present as tiny (<20-µm diameter) irregular grains dispersed in the claystones.
The combined results of X-ray diffraction (XRD) analyses of randomly oriented bulk sediment powders from Sites 1040 and 1254 are presented. Shipboard Leg 170 XRD data were not evaluated previously, nor were mineral peaks identified. Data from Leg 170 diffractograms were analyzed for the first time during Leg 205 and are interpreted and discussed here. Site 1040 peak intensity and peak area data are listed in Table T4, with peak area ratios listed in Table T5. Site 1254 peak intensity and peak area data are listed in Table T6, and the corresponding peak area ratios are listed in Table T7.
Cristobalite/quartz peak area ratio data have been used to monitor opal transformation during diagenesis (e.g., Mitsui and Taguchi, 1977) because the proportion of cristobalite is known to increase with temperature (e.g., Gambhir et al., 1999). The cool borehole temperatures (<10°C at the base) encountered at Site 1254 are below those required for in situ formation of cristobalite in this location. Volcanic ash diagenesis to cristobalite could be an alternative source. The origin of the cristobalite in Subunit U1A is not resolvable based on the existing analyses aboard the ship. However, the difference in cristobalite/quartz ratios between prism and underthrust units is notable; the average peak area ratio in the prism units is 0.68, and the average peak area ratio in the two uppermost underthrust units is 1.05. This difference suggests little or no accretion of hemipelagic sediments in Subunit U1A into the forearc wedge.
Clay separates were obtained for nineteen sediment samples from Hole 1254A. Oriented aggregate slides were generated as described in "Lithostratigraphy" in the "Explanatory Notes" chapter. Figure F22 presents relative weight percentages for smectite, illite, and kaolinite in the samples; Tables T8 and T9 portray the peak intensity and peak area data and calculated relative weight percentages for the three clay minerals considered.
Relative weight percentages of smectite, illite, and kaolinite/chlorite in the clay splits vary downhole. Smectite ranges from 77 to 93 wt%, illite ranges from 0 to 3.67 wt%, and kaolinite/chlorite ranges from 5 to 20 wt%. The maximum proportion of smectite (93 wt%) occurs at 350.81 mbsf in Sample 205-1254A-14R-3, 0-40 cm, taken from interstitial water squeeze cake trimmings. The maximum percentage for illite (4 wt%) occurs at 364.5 mbsf in Sample 205-1254A-16R-2, 20-21 cm (noted in "Structural Geology") in part of a brittle shear zone. The maximum percentage for kaolinite/chlorite (20 wt%) occurs at 360.55 mbsf in Sample 205-1254A-15R-3, 0-44 cm, taken from interstitial water squeeze cake trimmings.
The sediments and sedimentary rocks recovered at Site 1254 are dominated by structureless claystones with variable subsidiary quantities of silt and rare interbedded volcanic ashes and sandstones. Below 360 mbsf, these sediments contain abundant diatoms and are assigned to a different lithostratigraphic unit, interpreted to represent Cocos plate hemipelagic sediment underthrust under the forearc wedge. Redeposited blocks of shallow-water limestones, lithified prior to incorporation within mudstones, are consistent with sedimentation by fluidized gravity and debris flows. Sediment provenance indicates erosion of volcanic and quartz-mica-bearing basement rocks onshore, with no evidence for the incorporation of hemipelagic sediments into the forearc wedge.
A single large tephra was identified in Sample 205-1254A-8R-8, 22-65 cm, characterized by chemistry consistent with a source within the Costa Rica arc volcanic front. Grain size and thickness characteristics of the tephra indicate that it was the product of a giant eruption similar in size to that associated with the Minoan tephra of Santorini Volcano. The age of the deposit is unknown.