Banded iron formations (BIFs) have been at the center of many debates in geology, especially regarding the early (i.e., Archean and Paleoproterozoic) Earth and its surface environments. BIFs are chemical sedimentary rocks that have an anomalously high iron content (>15 wt% Fe) and typically contain layers of chert (Klein, 2005). BIFs have potential value as proxies for marine chemistry (e.g., Viehmann et al., 2015) and life (e.g., Konhauser et al., 2002) at the time of their deposition. Detailed reviews of their characteristics and proposed depositional models are readily available (e.g., Klein, 2005; Beukes and Gutzmer, 2008; Bekker et al., 2010). What stands out of these characteristics are their limited occurrence in time (ca. 3.8–1.8 Ga and ca. 750 Ma) and evolving stratigraphic settings through time. The deposition of glacially associated BIFs (ca. 750 Ma) has been satisfactorily explained as being caused by the Snowball Earth event (Klein and Beukes, 1993; Klein 2005). However, the deposition of the older greenstone belt–hosted (ca. 3.8–2.5 Ga) and craton margin (ca. 3.0–1.8 Ga) BIFs, especially before the Great Oxidation Event (GOE; ca. 2.45–2.32 Ga; Bekker et al., 2004; Holland, 2005), is more challenging to explain.

The first challenge in pre-GOE BIF deposition relates to the transport of dissolved ferrous iron (Fe2+) in the ocean. In contrast to today’s highly oxygenated oceans, dissolved ferrous iron transport would have been possible prior to the GOE, when Earth’s surface environments were generally considered to be anoxic (Holland, 2005). In addition, a greater abundance of large hydrothermal plumes added large amounts of ferrous iron to the early oceans (Bekker et al., 2010). The next challenge related to BIFs is explaining the mechanisms that precipitated the iron. Although this could have been a non-redox process in some instances (e.g., Rasmussen et al., 2015), the general consensus is that the iron was originally precipitated as ferric (Fe3+) oxyhydroxides following oxidation of the ferrous iron (e.g., Bekker et al., 2010; Smith et al., 2013). Oxidation mechanisms that have been proposed include: UV photo-oxidation (Cairns-Smith, 1978); free oxygen formed by photosynthetic bacteria (Klein and Beukes, 1993); and iron-oxidizing bacteria (i.e., anoxic photoferrotrophs or micro-oxic chemilithoautotrophs; Konhauser et al., 2002). With the UV photo-oxidation of iron being shown to be ineffective (Konhauser et al., 2007), the importance of what BIFs can tell us about life on early Earth has been coming to the forefront.

The hypothesis of the precipitation of iron in BIFs by iron-oxidizing bacteria has been gaining popularity in the last decade as it can explain the oxidation of iron in completely anoxic environments. This hypothesis is supported by laboratory studies on photoferrotrophs (e.g., Kappler et al., 2005), as well as the fact that the carbonates in BIFs have depleted δ13C values, suggesting the source was organic carbon related to the bacterial life (Beukes and Gutzmer, 2008; Bekker et al., 2010). Photoferrotrophs require sunlight, implying that they lived in the photic zone of early Earth’s oceans. However, the pre-GOE Earth lacked an ozone layer, as indicated by sulfur mass independent fractionation (SMIF; Farquhar et al., 2000; Pavlov and Kasting, 2002). The UV radiation would therefore have had catastrophic effects on any unprotected bacterial cells living in the photic zone. This problem lies at the heart of the paper by Gauger et al. (2015) in this issue of Geology (p. 1067). It has been proposed that planktonic life could have used solutes in the water column as sunscreen against UV radiation (see Gauger et al., 2015, and references therein). Gauger et al. tested this hypothesis on photoferrotrophs to see if precipitating ferric oxyhydroxides can act as sunscreen. By growing bacterial cultures in the presence and absence of UV light and ferrous iron, respectively, they conclusively showed the following: UV radiation negatively affects the growth of bacterial cultures in the absence of ferrous iron, and in the presence of precipitating ferric oxyhydroxide nanoparticles, the UV radiation had little effect on bacterial growth. This positive result provides evidence on how iron-oxidizing bacterial colonies could have survived in the shallow ocean of the early Earth. It should be noted that the authors state that the “cells…are in close proximity to the produced nanoparticular minerals, although the cell surfaces remain mostly free from precipitates.” This would suggest that further research is required to determine the suspension dynamics of the ferric oxyhydroxide nanoparticles and their longer-term efficiency as sunscreen in an open marine environment.

However, one should always proceed with caution when interpreting the precipitation mechanisms in BIFs. For example, some pre-GOE chemical sedimentary sequences show evidence for primary to early diagenetic hematite deposited below wave base, as well as BIF deposition associated with deepwater, stromatolite-free carbonates, suggesting deposition below the photic zone (e.g., Beukes and Gutzmer, 2008; Hoashi et al., 2009; Smith at el., 2013). This implies that the ferrous iron never reached the photic zone, excluding photoferrotrophs. However, increasing evidence for “whiffs of oxygen” in pre-GOE oceans (e.g., Anbar et al., 2007; Crowe et al., 2013; Planavsky et al., 2014; Stüeken et al., 2015) suggests micro-oxic chemolithoautotrophs, that do not require sunlight and therefore did not need protection from UV radiation, could have played a major role in BIF deposition,.

The research done on the role of bacteria in the deposition of BIFs illustrates an important point: one should always exercise great caution when studying any BIF occurrence, as their characteristics and depositional settings vary. This makes a “one-size-fits-all” depositional model unlikely. However, Gauger et al. should be commended on proving a hypothesis on what was likely a significant BIF depositional mechanism.

Besides their importance with regard to life on early Earth, BIFs are also of great economic importance, as the majority of super-large iron ore deposits formed from precursor BIFs. These high-grade (>60 wt% Fe) ores consist mostly of hematite with significant goethite. Recent advances have been made in dating the hematite using (U-Th)/21Ne, (U-Th)/He, and 4He/3He radiometric techniques. Being able to time ore-forming events is of great value, as it can then be correlated to regional geological events, leading to the identification of larger-scale ore-forming events. In this issue of Geology, Farley and McKeon (2015, p. 1083) used (U-Th)/21Ne hematite dating to delineate two ore-forming events at ca. 772 and 453 Ma, respectively, in the Gogebic iron range, Michigan, USA. What was interesting to note, was that these ages occur at times of tectonic quiescence in the region. Furthermore, the authors used 4He/3He spectra inverse modeling to determine the cooling history of the ores, which was then used to interpret depth of ore formation, erosional rates, and the unroofing associated with Pleistocene glaciation. From these conclusions, one can see the following contributions of this study: the need to identify fluid-mobilizing events in the region to account for ore formation between ca. 800 and 400 Ma, and a better understanding of the landscape evolution of the region during the Phanerozoic.

The methodology, results, and conclusions by Farley and McKeon open many avenues for investigation into BIF-hosted iron ore regions. For example, the supergene-enriched iron ores from the Griqualand West region in South Africa are associated with a regional erosional unconformity that transects numerous units within the Transvaal Supergroup (Van Schalkwyk and Beukes, 1986). The Transvaal Supergroup is of great geological significance, as it marks, among other paleoenvironmental events, the GOE (e.g., Bekker et al., 2004). Determining the depositional age of the Transvaal Supergroup has been problematic due to contradicting ages determined for volcanic units, dikes, and carbonates (e.g., Cornell et al., 1996; Bau et al., 1999; Wiggering and Beukes, 1990; Kampmann et al., 2015). Determining the age of the iron ores, and therefore also the age of the regional unconformity, would provide further geochronological constraints. Furthermore, the 4He/3He spectra inverse modeling could be applied in regions where iron ore formation on craton margins was initiated by collisional tectonics, with the cooling history of the ore providing the post-collisional thermal evolution of the region.

The dating of hematite could also add great value to the study of pristine BIFs. The occurrence of what is interpreted to be primary to early diagenetic dusty hematite in pre-GOE chemical sedimentary rocks is a matter of debate, especially with regard to redox processes and free molecular oxygen at those times (Hoashi et al., 2009; Smith et al., 2013; Rasmussen et al., 2015). Being able to get accurate ages for such dusty hematite could determine whether its crystallization falls within known depositional age ranges, or significantly post-dates deposition, thereby resolving a key issue related to BIFs. In addition, BIFs are used as stratigraphic marker beds in many sedimentary sequences, and diagenetic hematite ages can be used to add further depositional age constraints. However, further refinement will be required to minimize sample dilution and contamination due to lower hematite abundances in BIFs when compared to ore.

From the discussion here, it becomes clear that the work by Gauger et al. and Farley and McKeon has contributed not only to our current understanding of the life and times of BIFs, respectively, but has also opened numerous avenues for further investigation into this unique lithology.

I thank Nic Beukes, Bruce Cairncross, and Clarisa Vorster for their valuable comments and input. DST-NRF CIMERA is thanked for providing partial financial support.