The Juan de Fuca Ridge (JdFR), which produces ocean crust at a half-rate of 29 mm/yr, is located a few hundred kilometers off the western coast of North America. The topographic relief of the ridge produces a barrier to terrigenous turbidite sediment supplied from Pleistocene glacial sources along the continental margin. This has resulted in the accumulation of an onlapping layer of sediment that buries the eastern flank of the JdFR within a few tens of kilometers of the ridge crest, where the crust at some locations is less than 1 Ma. This area provided several targets for studying ridge-flank hydrothermal processes during Leg 168 (Davis, Fisher, Firth, et al., 1997). Three distinct fluid-flow regimes were evaluated by drilling a transect normal to the JdFR (Fig. 1). These were (1) a transition zone between sediment-free (permitting open hydrothermal circulation) and sediment-covered (hydrologically sealed) igneous crust, named the Hydrothermal Transition (HT) area; (2) an area where rugged basement topography, large lateral variations in sediment thickness, and small, isolated basement outcrops are all inferred to influence the patterns and rates of fluid flow in and out of the basement, named the Rough Basement (RB) area; and (3) an area where a uniform and regionally continuous cover of sediments over less rugged basement topography largely prevents crust-ocean fluid exchange and impedes heat flow, named the Buried Basement (BB) transect. Ten drilling sites were designed to sample sediments, rocks, and fluid across these transects (Fig. 1).
Altered basalts investigated in this study come from the top of the volcanic basement in nine of these boreholes where the present-day temperatures determined during Leg 168 vary from 15° to 64°C (Davis, Fisher, Firth, et al., 1997). Progressive changes in alteration style are related to the increased age, closed-system nature, and alteration temperature of the samples (Davis, Fisher, Firth, et al., 1997; Marescotti et al., Chap. 10, this volume).
All of the igneous rocks recovered during Leg 168 exhibit secondary alteration affects, forming pervasively or as discrete fracture-parallel alteration haloes. Secondary minerals, most commonly clays, iron oxyhydroxides, and carbonates, are found lining or filling vesicles, coating fracture surfaces as veins, and replacing phenocrysts and groundmass. The total amount of alteration varies widely from about 1 to 24 volume percent secondary minerals. In general, the more easterly samples, which are also presently at higher temperatures, exhibit more alteration (see Marescotti et al., Chap. 10, this volume).
Saponite is cryptocrystalline or fibrous. It is typically green to blue in hand specimen and pale brown to yellowish and olive green in thin section (Davis, Fisher, Firth, et al., 1997). Celadonite is the descriptive term used for the cryptocrystalline to fibrous, strongly green colored clay in thin sections. Saponite forms characteristically in the less altered interiors of rock pieces, usually associated only with secondary pyrite. Celadonite is more typically present in the oxidized haloes of rock pieces, associated with Fe(O,OH)x. In the transitional zone between oxidizing and reducing areas, considerable overlap may result in sequential vesicle and vein fills. The observed sequences vary from saponite followed by celadonite to celadonite followed by saponite. Iddingsite (a clay-Fe(O,OH)x mixture) commonly fills vesicles after celadonite. Photomicrographs of a variety of vesicles and veins showing the various filling sequences are shown in Davis, Fisher, Firth, et al. (1997) and discussed more fully in Marescotti et al. (Chap. 10, this volume).
X-ray diffraction results confirm that the darker green clays in veins and vesicles are trioctahedral varieties (d(060) = 1.528-1.540 Å) that expand upon glycolation and collapse on heating (Fig. 2). Electron microprobe analyses (Table 1; Fig. 3) confirm that they are saponite with very low interlayer potassium and variable Fe/(Fe + Mg). The formulas of bright green celadonitic clays (Table 2; Fig. 3) trend toward ideal celadonite, yet all exhibit a deficit of interlayer potassium. The formulas listed in Table 2 contain between 1.2 and 1.6 interlayer cations for eight tetrahedral cations, whereas the ideal celadonite formula requires two interlayer cations. This suggests either that a physical mixture with saponite or iron oxyhydroxide was actually analyzed, or that the material is really a poorly crystallized precursor to ideal celadonite (often referred to as protoceladonite [Adamson and Richards, 1990]). The compositions of saponite and celadonitic clays from Leg 168 are comparable to those of clays from many other altered seafloor basalts (e.g., Adamson and Richards, 1990; Alt, 1993; Desprairies et al., 1989; Gillis et al., 1992; Laverne et al., 1996; Teagle et al., 1996).
Trace elements were determined in a small number of samples that provided enough material for hand picking, digestion, and conventional nebulization in an ICP-MS (Table 3). In one specimen containing a 1-mm-thick saponite vein, a polished slab was prepared and studied by means of laser ablation ICP-MS (Table 4). Trace elements that were determined in saponite by both techniques exhibit good agreement. The strontium concentration, which was determined to be between 93 and 303 ppm with six different ablations, was determined by nebulization ICP-MS to vary from 0.4 to 140 ppm. Lanthanum and U concentrations were below the detection limits of around 16 ppm and 8 ppm by LA-ICP-MS, respectively, and they varied from 0.1 to 5.5 ppm La and 0.1 to 24 ppm U by nebulization ICP-MS. The cerium concentrations were below detection limits (<13 ppm) to 64 ppm by LA-ICP-MS, and 0.1-15 ppm by nebulization ICP-MS. The rubidium concentrations were below detection limits (<2 ppm) to 21 ppm by LA-ICP-MS, and around 9 ppm by nebulization ICP-MS. In all cases, the ranges determined for each element overlap.
Relative to an average MORB, the trace element patterns for the saponite show enrichment in a number of large-ion lithophile and high field-strength elements (Fig. 4A). However, normalized to the trace element concentrations in the host basalt (M. Constantin, pers. comm., 1998), the enrichment is not as strong (Fig. 4B). The saponite, in general, exhibits mild enrichment in Ba, Rb, Th, and U; depletion or no fractionation in Nb, the REE, and Sr; and either enrichment or depletion with respect to Zr and Hf.