Volatile Release to the Environment during Formation of the Kerguelen Plateau

The release of volatiles such as CO₂, S, Cl, and F accompanying eruption of enormous volumes of basaltic magma during formation of the Kerguelen Plateau and Broken Ridge probably had significant environmental consequences. The role of subaerial volcanic eruptions (Barron, Larsen, et al., 1989; Schlich, Wise, et al., 1989; Coffin, 1992; Coffin, Frey, Wallace, et al., 2000) is significant in this regard because basaltic magmas are nearly completely degassed during subaerial eruptions, whereas hydrostatic pressure inhibits vesiculation and degassing of relatively soluble volatile components (H₂O, S, Cl, and F) during deepwater submarine eruptions (Moore, 1970; Moore and Fabbi, 1971; Unni and Schilling, 1978). In contrast to the more soluble magmatic volatiles, the very low solubility of CO₂ causes it to be mostly degassed even at abyssal depths (Gerlach, 1989).

Determining the volatile fluxes from Kerguelen Plateau magmatism to the environment is difficult because of relatively poor knowledge about primary magmatic volatile contents, magma volume fluxes through time, and the proportion of volcanic activity that was subaerial or in a shallow submarine environment. Nevertheless, it is possible using existing data to make order of magnitude estimates for volatile release in order to consider potential environmental effects. The magma output rate of the Kerguelen hotspot between 120 and 95 Ma, when the SKP, CKP, and Broken Ridge were formed, was ~1 km³/yr (Coffin et al., 2002) (Fig. F3). The magma output rate appears to have decreased after that time (95 to ~25 Ma) to values of ~0.1 km³/yr. Based on these output rates, the S contents of enriched Site 1140 glasses, and assuming that 50% of the SKP, CKP, and Broken Ridge formed by subaerial eruptions (Coffin, 1992), the estimated S flux to the atmosphere from 120 to 95 Ma would have been 1.4 x 10¹² g S/yr (Wallace, 2002). This value is ~13% of the current global flux of S to the atmosphere from subaerial volcanic eruptions (Andres and Kasgnoc, 1998).

An important factor that would have increased the environmental consequences of volcanic S release during formation of the Kerguelen Plateau is the high latitude at which the plateau formed (Frey et al., 2000a). In most basaltic eruptions, released volatiles remain in the troposphere, but at high latitudes, the tropopause is relatively low. As a result, large mass flux basaltic fissure eruption plumes have the potential to transport SO₂ and other volatiles into the stratosphere (Stothers et al., 1986; Self et al., 1998). Sulfuric acid aerosol particles that form in the stratosphere after such eruptions have a longer residence time and greater global dispersal than if the SO₂ remains in the troposphere, thereby resulting in greater effects on climate and atmospheric chemistry. Although the value of 1.4 x 10¹² g S/yr estimated above for the period of Kerguelen Plateau formation from 120 to 95 Ma is only ~13% of the current global flux of S from subaerial volcanic eruptions, it is comparable in magnitude to the estimate of ~0.2 x 10¹² to 2 x 10¹² g S/yr injected into the stratosphere by explosive eruptions (Pyle et al., 1996).

The magma output rates of Coffin et al. (2002) can also be used to estimate the magnitude of volcanic CO₂ release during formation of the SKP, CKP, and Broken Ridge. Assuming a range in primary CO₂ contents based on normal MORB, enriched MORB, and ocean-island basalt magmas (~0.2-0.6 wt% CO₂) leads to a range of estimates from 0.5 x 10¹³ to 1.4 x 10¹³ g CO₂/yr released from Kerguelen Plateau volcanism from 120 to 95 Ma. This is ~5% to 10% of the current global annual flux of CO₂ released by mid-ocean-ridge and subduction zone volcanism (Gerlach, 1991; Varekamp et al., 1992). Although the magnitude of CO₂ release from Kerguelen Plateau volcanism appears to be relatively small compared to the total global volcanic flux, a recent reconstruction of atmospheric CO₂ variations during the last 300 m.y. (Retallack, 2001) shows two periods of pronounced CO₂ increase that correspond to the times of formation, respectively, of the SKP and CKP. It should be pointed out, however, that the latter period also partially overlaps with formation of both the Caribbean LIP and the Madagascar flood basalt (Coffin and Eldholm, 1994; Eldholm and Coffin, 2000).

Highly explosive felsic eruptions, such as those that produced the pyroclastic deposits on Elan Bank, Skiff Bank, and the CKP, can also inject both particulate material and volatiles (SO₂ and CO₂) directly into the stratosphere (McCormick et al., 1995). The previously unrecognized significant volume of explosive felsic volcanism that occurred when the Kerguelen Plateau and Broken Ridge were subaerial would have further contributed to the effects of this plume volcanism on global climate and environment (Frey et al., 2000a). The total volume of felsic volcanic rocks is poorly constrained, but Leg 183 drilling results indicate that they account for a significant fraction of the volcanic deposits erupted during the final stages of magmatism at several locations on the Kerguelen Plateau.

Analyses of rhyolitic glass inclusions trapped in quartz and sanidine phenocrysts in the tuff at Site 1137 on Elan Bank indicate dissolved magmatic H₂O concentrations of 2 to 6 wt% (Wallace et al., 2000). Such concentrations are sufficient to sustain a powerful Plinian eruption column that can reach stratospheric altitudes (Wilson et al., 1980). Although dissolved S contents of the inclusions are relatively low (P.J. Wallace, unpubl. data), the presence of significant CO₂ makes it likely that the felsic magmas were vapor saturated before eruption. This is important because the vapor phase may contain significant S (e.g., Scaillet et al., 1998). As a result, eruptions of vapor-saturated silicic magmas typically release large amounts of SO₂ derived from the vapor phase, despite the very low concentrations of dissolved S in such magmas (Wallace, 2001).

In addition to the potential environmental effects of subaerial volcanism, submarine volcanism and hydrothermal activity during formation of the Kerguelen Plateau and Broken Ridge may also have had important effects on ocean chemistry. Duncan (2002) notes that the seawater Sr isotopic evolution curve exhibits a decline from ~122 to 112 Ma and suggests that this results from large contributions of relatively unradiogenic Sr from hydrothermal activity. Formation of much of the submarine Ontong Java Plateau in the western Pacific at ~122 Ma (Tejada et al., 1996) may also have contributed significantly to seawater Sr changes. However, the importance of hydrothermal activity during formation of oceanic plateaus has not been established. For the Kerguelen Plateau and Broken Ridge a potentially major source of Sr is weathering and erosion of subaerial basalt flows and subsequent riverine transport of Sr to the oceans. Duncan (2002) concludes that if the decrease in seawater radiogenic Sr from 122 to 112 Ma is caused by Kerguelen Plateau hydrothermal activity, then significant submarine volcanism must have ceased between ~115 and ~93 Ma. The latter is the next time period when a sharp decrease in the seawater Sr isotopic evolution curve is seen. Based on new radiometric ages, Duncan (2002) also suggests that rapid submarine construction of the SKP by 118-119 Ma may have contributed to a global environmental crisis recorded by widespread marine black shales (global anoxic event OAE1). Formation of much of the submarine Ontong Java Plateau in the western Pacific at ~122 Ma (Tejada et al., 1996) may also have contributed significantly to this anoxic event.

A wide variety of mafic subaerial lava types were recovered during Leg 183. A review of the characteristics of these lava types and a refined method for classifying flows recovered in core based on macroscopic textural features is provided by Keszthelyi (this volume). Excluding flow units that cannot be classified because of poor core recovery or intense alteration, Keszthelyi (this volume) concludes that of 30 flow units from Sites 1136, 1137, 1138, and 1139, 7% are slab pahoehoe, 13% are aa, 27% are pahoehoe, and 53% are rubbly pahoehoe. The last of these—rubbly pahoehoe—refers to a lava type that has a flow top composed of broken pieces of smaller pahoehoe lobes. Keszthelyi considers rubbly pahoehoe to be distinct from slab and other recognized types of pahoehoe and suggests that it is a common flow type in Iceland and the Columbia River Basalt Group.

Kurnosov et al. (this volume) show that the subaerial basalts have been variably altered by low-temperature hydrothermal processes (<120°C), and some have been affected by subaerial weathering. Important secondary minerals associated with the hydrothermal alteration include smectites and chlorite minerals. Alteration of the subaerial basalts appears to be related primarily to horizontal fluid flow within permeable contact zones between lava flows (Kurnosov et al., this volume). No soil horizons between basalt flows were recovered at any of the Leg 183 drill sites (Coffin, Frey, Wallace, et al., 2000), but this could be due to the difficulty of recovering thin layers of soft material between basalt during rotary core barrel coring rather than the real absence of soils.