Salty Matters

Sulphate and oxygenation levels across the Proterozoic

Proterozoic (2.5 – 0.542 Ga) saline sediments encompass significant transitions in evaporite style and chemistry within an evolving atmospheric and oceanic framework. Lithospheric changes tie to a cooling and biologically-evolving earth as earth-scale plate tectonics move to a system set comparable with that operating today. The Proterozoic eon is divided into three eras: the Paleoproterozoic, Mesoproterozoic and Neoproterozoic. Thick sequences of halite are only found as the actual bedded salts in sediments of the Neoproterozoic (and the Phanerozoic), while calcium sulphate residues and beds occur in all three Proterozoic eras, especially in parts of the Paleoproterozoic and the Neoproterozoic.

Paleoproterozoic era sediments (2.5-1.6 Ga) contain isotopic evidence the first significant oxygenation event in the world's atmosphere, largely driven by the increasing dominance of cyanobacterial photosynthesis. Neoproterozoic sediments (1.0 – 0.542Ga) contain evidence of the second oxygenation event, which is associated with the evolution of widespread multicellular life and CaCO3/siliceous carapaces. By the end of the Neoproterozoic, the world oceans had chemistries, temperatures and salinities similar to those of the Phanerozoic (Blamey et al., 2016). The intervening Mesoproterozoic (1.6- 1.0 Ga) retains evaporitic residues with aspects of both the late Archean and the Phanerozoic.

The oxygenation of Earth’s atmosphere-ocean system occurred in two steps: 1) the Paleoproterozoic “Great Oxygenation Event” (GOE ≈ 2.3 Ga), which refers to the transition from a pervasively reducing Earth-surface system to one with an oxygenated atmosphere and oxygenated shallow seas, and 2) the “Neoproterozoic Oxygenation Event” (NOE), when the Earth’s atmosphere and ocean are understood to have become persistently oxygenated down to the deep ocean bottom (Turner and Bekker, 2016; Scott et al., 2014). The GOE is indicated by a Proterozoic carbon isotope anomaly known as the “Lomagundi event,” a positive carbon isotope excursion between ca. 2.22 and 2.06 Ga, interpreted to be the result of high organic carbon burial and attendant accumulation of atmospheric oxygen (Figure 1; Bekker and Holland, 2012)). A long interval spanning the remainder of the Paleoproterozoic and much of the Mesoproterozoic followed the Lomagundi event, typified by lower levels of atmospheric oxygen and little variation in carbon isotope values. This ended in the late Neoproterozoic with dramatic fluctuations, of escalating magnitude, in the biogeochemical carbon cycle and attendant fluctuations within an overall increasing oxygen content (Figure 2). By the end of the Neoproterozoic, not just shallow shelf waters but much of the deep-ocean water column was consistently oxygenated (see article 1 in this series)

Chert evolution

 We have already seen how low levels of oxygen and high levels of CO2 in the Archean favoured the precipitation of nahcolite, and its hydrothermal silica association, atop subsealevel isolated saline sumps in microcontinents and island arcs in a saline waterworld (article, 28 August 2016). The hydrothermally-dominated silica of the silica-rich Archean oceans is reflected in more negative ³⁰Si isotope values in widespread marine cherts of that time, compared with most Proterozoic cherts (Chakrabarti et al., 2012; see also Figure 3b). However, shorter-term fluctuating levels of atmospheric oxygen in the Proterozoic also influenced drop-out salinities for gypsum in Mesoproterozoic marine brines. In brines derived from the modern, well-oxygenated world oceans, as in evaporite successions deposited throughout the Phanerozoic, the sulphate minerals gypsum and anhydrite precipitated from evaporating seawater after aragonite or calcite, but before halite (see 26 August, 2015, blog for more detail). At lower seawater sulphate levels across the Archean and much of the Proterozoic, gypsum and anhydrite precipitated after halite, even at Na and Cl concentrations similar to those of the modern ocean.

This is why some post-Lomagundi, Paleoproterozoic marine evaporite successions show clear evidence of halite precipitation before gypsum or anhydrite or even an absence of gypsum or anhydrite with halite (e.g., ≈ 1.88 Ga Stark Formation; Pope and Grotzinger, 2003). The post-halite precipitation of calcium sulphate is construed as evidence for a limited marine sulphate reservoir and little atmospheric oxygen (Scott et al., 2014). In contrast, Lomagundi- age sedimentary successions contain evidence for sulphate precipitation before halite (Melezhik et al., 2005; Bekker et al., 2006; Schröder et al., 2008). Sulphur isotope values of marine sulphates (in CaSO4, barite) and sulphides in marine pyrite also record expansion and contraction of anoxic oceanic settings. That is, a higher burial rate of pyrite in anoxic settings is indicated by a positive shift in the sulphur isotope values of sulphates, whereas ocean oxygenation creates a negative shift in values (e.g., Claypool et al., 1980; Strauss, 1997). Furthermore, expansion of the area of anoxic oceanic settings decreases the size of the seawater sulphate reservoir, resulting in more variable sulphur isotope values of sulphate evaporites, barites, and other carbonate-associated sulphates (Figure 2; Kah et al., 2004). This applies in particular in the Mesoproterozoic when only tshallow oceanic waters were consistently oxygenated.

When we look at silica mobility and chert styles across a Proterozoic milieu of evolving oxygen and sulphate levels we see some aspects similar to the Phanerozoic and others more akin to the high-silica oceans of the Archean. Maliva et al. (2005) and Perry and 2014 show that the latter part of the Paleoproterozoic era (post-Lomagundi) is marked by the end of widespread primary and early diagenetic silica precipitation in normal marine subtidal environments. However, silica precipitation continued apace in the deeper marine in waters that were still anoxic. The Paleoproterozoic is defined by the “rusting” of the shallower parts (shelves and upper slopes) of the world’s ocea,n as dissolved oxygen levels increased and the accumulation of widespread Banded Iron Formations (BIFs) occurred, including the huge deposits of NW Australia.

So where and when do we see nodular cauliflower chert after sulphate in the Proterozoic?

Some of the oldest silicified nodular sulphates with cauliflower chert textures and actual relict anhydrite are found in the Huronian Gordon Lake Formation (≈2.4 Ga; Chandler, 1988). The nodules are commonest in mud chip breccia at the base of sandstone, siltstone and mudstone upward-fining storm cycles. Anhydrite nodule relicts are composed of a mosaic and meshworks of blocky crystal laths. Earlier laths are preserved in silicified outer rims of many cauliflower chert nodules, with texture alignments similar to that of Recent displacive sabkha anhydrite nodules. Completely silicified nodules are composed of megaquartz, some calcite-cored, or of jasper, with replacement textures identical those documented the Phanerozoic by Milliken (1979). Similarly-textured cauliflower cherts are found in the Mallapunyah Formation (1650 Ma) in the Paleoproertozoic sediments of the McArthur Basin in the Northern Territory of Australia (Warren, 1999). Thin sections through 5-30cm diameter Mallapunyah nodules (in a redbed host) can still retain small relict highly-birefringent laths of anhydrite, but most of the former felted cores to the nodules are now composed of mimetic silica. There are also older sedimentary chert nodules in the McArthur Basin succession located in units below the level of the Mallapunyah, but they are smooth walled not rugose-surfaced features.

Unfortunately, the term cauliflower chert is loosely defined and is used to describe cemented features in Meosproterozoic and earlier sediments and metasediments, which are not true calcium sulphate evaporite replacements.Although termed cauliflower features, they do not have surface textures resembling the florets of a cauliflower (see article 2 in this series - July 31, 2016). For example, aggregates and clusters of growth-aligned barite crystals in the Archean of South Africa are described as cauliflowers when they should be described as bladed, palmate crystal aggregates (Reimers and Heinrich, 1997). Interstingly, Chowns and Elkins (1974) in a study of cauliflower cherts occurrences across the USA list no examples older than Cambrian. Using a tighter definition of cauliflower chert and recognising that this term should not be interchangeable with crocodile-skin chert it seems that Proterozoic occurrences of cauliflower chert nodules largely mirror times when oxygen levels were sufficiently high in the world's ocean to allow sulphate in solution. In the Paleoproterozoic and Mesoproterozoic only the upper parts of the ocean column, including waters covering the world's continental shelves (and derived evaporite basins) were sufficiently oxygenated to allow the formation of cauliflower chert after nodular anhydrite. However in some Neoproterozoic basins, especially if located in sumps in a highly-restricted brine layered seafloor, the levels of anoxia in the ponded bottom brines facilitated the accumulation of laminar microporous chert in association with evaporites or their early replacements

Primary laminated hypersaline silica chert in an evaporite basin at the Precambrian-Cambrian boundary

An organic-rich laminated porous chert known as the Athel or Al Shamou silicilyte consists of up to 90% microcrystalline quartz along with dolomite, magnesite, anhydrite and halite (Rajaibi et al., 2015). It occurs at the Precambrian-Cambrian boundary in the subsurface of the South Oman Salt Basin, Sultanate of Oman, where it acts as a light-oil reservoir  (Ramseyer et al., 2013; Amthor et al., 2005). Fully encased in variably halokinetic salt masses, it was first discovered during the 1990's hydrocarbon exploration activities of Petroleum Development Oman. This laminated microporous and variably fractured chert, has its source of silica and its mode of precipitation tied to an anoxic, sulphur-rich, stagnant and highly saline basin. Its homogeneous silica distribution and high Si isotope values (avg. d³⁰Si = +0.83 ± 0.28), coupled with a low molar Ge/Si ratio (<0.25 x 10^-6) in its microcrystalline quartz matrix imply dissolved silica in concentrated seawater as the Si source, and hydrothermal or biogenic (e.g. sponge-derived) silica are excluded.

Silica precipitation from a seawater-sourced brine was likely the result of a dramatic increase in salinity in response to halokinetic salt dissolution atop and adjacent to the edges of transtensional depressions on a deep basin floor in the South Oman Salt basin, thus markedly reducing the solubility of amorphous silica in these brine-filled seawater depressions. This saturation triggered the formation of silica gel. The gel accumulated at the base of a brine-layer covered basin floor, forming a soft silica-rich layer bound into bacterial mats, giving rise to its fine-scale lamination. The mean number of laminae in this laminated chert is ca. 32 per year suggesting that layering is non-annual and controlled by processes such as fluctuations in nutrient supply, lunar driven re-mixing or diagenetic segregation. The transformation of the silica-gel to microcrystalline quartz occurred below 45°C indicating a less than -4.5‰ d¹⁸O composition of the pore-water during microcrystalline quartz formation. The  microporous hydrocarbon filled nature of this ancient chert and the fact the hydrocarbon-filled micropores are still distinct after more than 500 million years after they filled (Figure 4d, e) is  why when artificially fractured the silicilyte can act as a hydrocarbon reservoir (See Rajaibi et al. 2015 and Warren, 2016; Chapter 10 for a summary of relevant literature).
 

The Caballos Novaculite outcrops in the Marathon Uplift of Texas, while its lithologic and time equivalent, the Arkansas Novaculite, outcrops in the Ouachita Mountains of Arkansas and Oklahoma (Figure 5). Novaculite chert) in outcrop is very resistant to erosion so that layers of novaculite stand out as characteristic ridges and dip slopes in the Ouachita and Marathon mountains (Figure 5). This outcrop forms and its hard abrasive nature gives it the name novaculite, which in its Latin root novacula, means razor-stone. When some novaculite is fractured in the subsurface, there is sufficient connected porosity to form a fractured reservoir play, as in Arkansas and Texas. There, some 30 years ago, oil and gas fields such as Isom Springs in Oklahoma and McKay Creek, Pinion and Thistle fields in West Texas were discovered in the Caballos and Arkansas novaculite-chert. The chert reservoir is most productive when it is highly fractured, occurs within complex thrust faults and has had any enclosed carbonate material leached from its chert matrix, so creating microporosity (Figure 6; Godo et al., 2011).