Microborings reveal alternating agitation, resting and sleeping stages of modern marine ooids

Ooids are abundant carbonate grains throughout much of Earth's history, but their formation is not well understood. Here, an in‐depth study of microbial bioerosion features of Holocene ooids from the Schooner Cays ooid shoals (Great Bahama Bank, Eleuthera, Bahamas) and the Shalil al Ud ooid shoals in the Arabian/Persian Gulf (Abu Dhabi, United Arab Emirates) is presented. No obvious differences were found in ooid size distribution, cortex layer thickness, the composition of nuclei or euendolithic community when comparing ooids from both locations. Microendolithic borings are present in most studied ooid surfaces, but the intensity of (micro‐)bioerosion varies significantly. Applying an epoxy vacuum cast‐embedding technique allowed the identification of ichnotaxa and their inferred producers (various genera of diatoms, cyanobacteria, coccolithophores and unspecified bacteria). Euendolithic taxa have specific low‐light tolerances and light optima. This implies that information about the relative bathymetry (seafloor versus burial within an ooid shoal) and ecology for ooid cortex formation can be obtained via the presence or absence of their respective ichnotaxa. The history of a statistically significant number of ooid cortices can be translated into dune dynamics and the temporal variations thereof by allocating the inferred index producer to a defined burial or light penetration zone. In this context, ooid formation can be divided into four stages: (i) an agitation stage in the water column, characterized by the colonization of grains by photoautotrophs; (ii) a resting stage, characterized by temporary burial of the ooid, leading to immobilization and a shift towards heterotrophs; (iii) a sleeping stage, characterized by prolonged burial and colonization by organotrophs; and (iv) a reactivation stage, characterized by a resurfacing of the ooid and a subsequent shift towards photoautotrophs. The sleeping stage is presumably a stage of ooid degradation where bioerosion, mainly by heterotrophic fungi and bacteria is particularly active.

Ooids are abundant carbonate grains throughout much of Earth's history, but their formation is not well understood.Here, an in-depth study of microbial bioerosion features of Holocene ooids from the Schooner Cays ooid shoals (Great Bahama Bank, Eleuthera, Bahamas) and the Shalil al Ud ooid shoals in the Arabian/Persian Gulf (Abu Dhabi, United Arab Emirates) is presented.No obvious differences were found in ooid size distribution, cortex layer thickness, the composition of nuclei or euendolithic community when comparing ooids from both locations.Microendolithic borings are present in most studied ooid surfaces, but the intensity of (micro-)bioerosion varies significantly.Applying an epoxy vacuum cast-embedding technique allowed the identification of ichnotaxa and their inferred producers (various genera of diatoms, cyanobacteria, coccolithophores and unspecified bacteria).Euendolithic taxa have specific low-light tolerances and light optima.This implies that information about the relative bathymetry (seafloor versus burial within an ooid shoal) and ecology for ooid cortex formation can be obtained via the presence or absence of their respective ichnotaxa.The history of a statistically significant number of ooid cortices can be translated into dune dynamics and the temporal variations thereof by allocating the inferred index producer to a defined burial or light penetration zone.In this context, ooid formation can be divided into four stages: (i) an agitation stage in the water column, characterized by the colonization of grains by photoautotrophs; (ii) a resting stage, characterized by temporary burial of the ooid, leading to immobilization and a shift towards heterotrophs; (iii) a sleeping stage, characterized by prolonged burial and colonization by organotrophs;
Despite more than a century of research, ooid formation and alteration in marine settings remains insufficiently understood.The influence of abiotic factors, including hydrodynamic level, pH, temperature and salinity, and of biotic factors, including geomicrobiological/organomineral processes, is complex (see reviews in Simone, 1980;Diaz & Eberli, 2019;Harris et al., 2019).Suarez-Gonzalez & Reitner (2021) even observed the static formation of aragonitic ooids in microbial mats and attributed their formation to both biological [extracellular polymeric substance (EPS) degradation] and environmental factors (variations in supersaturation in the biomat).In addition, Diaz et al. (2014) presented compelling arguments for biologically influenced CaCO 3 precipitation in modern marine ooids, fostered by the synergistic effect of microbes with the potential to dissolve and/or induce micritization on carbonate substrates.
This paper investigates microendolithic borings on the surface and within the cortex of modern marine ooids from the Schooner Cays shoals (Great Bahama Bank) and the outer lagoon (Shalil al Ud) north-east of Abu-Dhabi (the Gulf).The aims of this paper are: (i) to provide detailed documentation of microendolithic borings in ooid cortices; (ii) to propose an ichnotaxonomic identification of the microendolithic traces and their inferred producers; (iii) to assess the burial depth distribution and cross-cutting relationship of microendoliths within the cortex; and (iv) to reconstruct the chronology of ooid cortex layer formation and bioerosion events in the context of geomicrobial activity and previous ooid formation hypotheses.This work has wide implications, because elucidating the stratigraphy of microendolithic boring intervals within ooid cortices offers the potential to more quantitatively assess ooid growth, resting and sleeping stages in modern carbonate environments.Moreover, data shown here also have a bearing on understanding ooid shoal dynamics, including the short-term migration of sand waves at the top of the shoal and the resulting long-term effects.

Schooner Cays ooid shoals
The Schooner Cays tidal-bar belt (Fig. 1A and C) is located north of Exuma Sound on the Great Bahama Bank, east of the Eleuthera Islands.It extends over 20 × 56 km and is formed by Holocene ooid sands.Various authors (e.g.Ball, 1967;Dravis, 1979;Harris et al., 2011Harris et al., , 2014;;Purkis & Harris, 2017;Rush & Rankey, 2017;Purkis et al., 2019) provided detailed descriptions of the physiography, sedimentology, size and evolution of the tidal bar belt, with implications for physical and hydrodynamic controls on deposition as well as stratigraphic and morphometric correlations to other sand bodies.Individual tidal bars consist of up to 1 km wide shoals of ooid sands.The crests of the shoals are partially exposed during low tide for several hours and submerged during high tide.The average water depth atop the shoals during mid-tidal levels is less than 1 m.Salinity at the Great Bahama Bank is 38 to 42‰ and therefore slightly elevated compared to open marine settings (Whitaker & Smart, 1990).The Schooner Cays proper represent stabilized carbonate islands atop several ooid-sand shoals.The Schooner Cays ooid sands (Fig. 2A to E) consist of whitish ooid grains with average diameters between 0.18 mm and 1.0 mm (Fig. 2A).Radiocarbon dating suggests that the ooid formation started approximately 100 to 2800 years BP (Duguid et al., 2010).These large deviations can be explained by the fact that tidal currents still actively formed and transported these ooids.

The Gulf (Shalil al Ud ooid shoals)
The shallow (average depth = 36 m) southwestern Gulf (Fig. 1B and D) overlies a lowgradient continental basement and represents a modern subtropical epeiric sea with active carbonate sedimentation (Edinger et al., 2002;Lokier & Fiorini, 2016;Pederson et al., 2021).As part of the south-western Gulf, the Abu Dhabi coastal area comprises a mosaic of shallow lagoons (0 to 10 m deep) and small barrier islands.The complex geography is inherited from eustatic fluctuations during Quaternary glaciation events (Evans et al., 1969;Stevens et al., 2014).The climate in Abu Dhabi is hot and arid, with summer temperatures reaching over 50°C and littoral humidity reaching 100% (Kinsman, 1964;Lokier & Fiorini, 2016).The average rainfall is 72 mm/year, but the average evaporation is 2.75 m/year (Lokier & Fiorini, 2016).The coastal tidal range is small (1 to 2 m), but storm surges often occur due to strong seasonal north-north-westerly winds (Shamals).Seawater salinity in the Gulf is about 40 to 45‰ (Ge et al., 2020a).There has been no riverine influx along the Southern coast of the Gulf since the Early Pleistocene (Arboit et al., 2022).
Ooidal shoals and ooid deltas in the outer lagoon (Shalil al Ud) extend to an area of more than 83 km 2 , forming in a high-energy setting influenced by strong tidal activity.At low tide, the crests of shoals are subaerially exposed for several hours, while at high tide, the shoals are submerged and influenced by tidal currents and wind-generated waves.At these times, ooids are actively formed and transported.The intershoal seafloor is covered by seagrass, which provides a stabilizing effect.The seafloor sediment consists of ooids (60 to 90 vol.%), gastropods, calcareous algae, echinoderms and foraminifera (Ge et al., 2020a(Ge et al., , 2020b)).The Shalil al Ud ooid sands (Fig. 2F to J) consist of white-tan carbonate grains with diameters between 0.18 mm and 2.0 mm (Fig. 2F).

METHODOLOGY Fieldwork
Samples were collected during low tide in March 2016 (Shalil al Ud lagoon, Abu Dhabi coastline) and May 2019 (Schooner Cays, Eleuthera Island, Great Bahama Bank; Fig. 1).In both cases, samples were taken from the crests of active ooid shoals.
Raman spectroscopy was used in further preparation to determine the mineralogy of ooid cortices and nuclei on a test basis.For a more comprehensive identification of mineralogy, cathodoluminescence microscopy (CL) was used to obtain information on mineralogy throughout the thin section based on characteristic luminescence colours (see Hoffmann et al., 2016;Goldstein et al., 2017).Both methods confirmed that undisturbed ooid cortex layers are usually aragonitic, while cement filling empty boreholes is calcitic.The mineralogy of the nuclei varies according to their composition, but calcitic nuclei predominate (for details, including instrumentation, see Appendix S1-S3).
About 50% of the ooids with a diameter of 0.25 to 0.50 mm were fractured using a mortar.Subsamples (fragments) of ooids from both locations were chemically etched to enhance the ooid microfabrics.The ooids were treated separately with Di-Na-EDTA (0.27 M), HCl (0.1 M) and acetic acid (5 M), with which the ooid sand was soaked for 10 min, 30 min and 60 min; 2 h, 4 h, 6 h and 24 h; and seven days, respectively.After treatment, each ooid subsample was rinsed with distilled H 2 O, dried, and mounted onto a sample holder.Overall, 24 ooid samples containing fragmented and etched ooids and 10 samples of untreated ooid sand were investigated.The number of ooid subsamples on a sample holder varied between five to 20 grains/ fragments.
The remaining ooids of the size fraction 0.25 to 0.50 mm and those with a size of 0.50 to 0.63 mm were used for the vacuum castembedding technique, whereby four casts were made.This technique was applied to visualize the three-dimensional morphology of microborings and assess the penetration depth of the tunnel systems within the ooid cortex.A low-viscosity epoxy resin was used to infiltrate bioerosion features under vacuum conditions.After curing the epoxy resin, the samples were cut to obtain cross-sections of the ooid structure.Subsequently, ooid cross-sections were treated with 5% HCl for 20 s, rinsed with DI water and dried.In an alternative approach, ooids were completely dissolved using 5% HCl, exhibiting the epoxy-filled microborings' positive relief (see Wisshak, 2012, for detailed methodology).
An earlier attempt to identify ichnospecies on etched thin sections did not yield satisfactory results.However, when used correctly, thin section etching can be a comparatively inexpensive alternative to vacuum casts or for preliminary studies (see Salamon et al., 2019).
Different euendolithic producers create distinct tunnel shapes and sizes, ranging from an assemblage of thin galleries to large solitary tunnels.Specific ichnotaxa were identified by: (i) the appearance of nodular cavities or branching; (ii) the formation of clusters, separate chambers or swellings; as well as (iii) differences in the morphology of tunnel cross-sections or the tunnel terminations and the validity of ichnotaxa was checked back with the annotated list of bioerosion ichnotaxa compiled by Wisshak et al. (2019).

Optical techniques
Petrographic thin sections from the different grain size classes of both sampling sites were investigated using transmitted light, crossed polarized light, UV, green and blue epifluorescence.Thin section analysis allowed for the differentiation of grain types and internal structures.Properties such as ooid size, shape and boundaries provide information about the mineralogical composition of a sample and its mode of formation.Structure and composition were described under transmitted light [Olympus BX 51 (Olympus Corporation, Tokyp, Japan); Leica DM4500P (Leica, Wetzlar, Germany)].To detect fluorescence patterns in the ooid cortices and nuclei, a Leica DM4500P microscope, equipped with filter cube I3 (excitation filter BP 450-490, dichromatic mirror: 510 nm, suppression filter LP 515) and cube A (excitation filter BP 340-380, dichromatic mirror: 400 nm, suppression filter LP 425) was used.Imaging was performed using a Canon EOS 60D (Canon, Inc., Tokyo, Japan).
The etched ooids were studied under a scanning electron microscope (SEM) Gemini 2 Merlin HR-FESEM (Carl Zeiss AG, Jena, Germany), equipped with a Schottky field emitter serving as an electron source, depicting compositional and topographical properties of the ooids at the nanometre scale.High-resolution images were collected with the E-T and SE in-lens detector under ultrahigh vacuum conditions with an acceleration voltage of 1 to 20 kV and a probe current of 25 to 40 pA.Loose grains were sputter-coated with gold using a Cressington 208HR (Cressington Scientific Instruments, Watford, UK) and thin sections were coated with Ó 2023 The Authors.Sedimentology published by John Wiley & Sons Ltd on behalf of International Association of Sedimentologists., Sedimentology, 71, 359-382 carbon using a Quorum Q150T (Quorum Technologies Inc., Puslinch, ON, Canada) to avoid charging effects.

Ooids and ooid fabrics
No systematic differences in shape or size were found when comparing sand grains from the two sampling sites.Ooids display a variety of morphologies ranging from nearly spherical to elongated-ovoid (Fig. 2).Granulometry reveals that the majority of the ooids (>50%) from both locations were 0.25 to 0.5 mm in size.Under the binocular microscope, the ooids appear whitish, light brown or greyish, displaying a dull polished surface.Under transmitted light, ooid cortices are beige to grey with discernible laminae (between three and 45), while ooid nuclei (usually foraminifera, mollusc fragments, unspecified bioclasts or non-carbonate grains) are generally dark-brown (Fig. 2B and G).All ooid cortices show unevenly distributed dark brown or translucent, round-totubular patches of varying size and abundance (see Fig. 2B, orange arrows).
Under cross-polarized light, intact ooid cortices display bright pearlescent colours (Fig. 2C  and H).Thicker cortices further portray pseudouniaxial extinction crosses, with darker bands occurring at the 0°, 90°, 180°and 270°positions, representing the predominantly tangential (c-axis) orientation of the rods, typical for modern marine ooids (Milliman et al., 1974).Epifluorescence reveals successions of darker and lighter shades of green (or blue) for individual cortex layers or a sequence of layers, while micritic areas and peloid nuclei are nonfluorescent or weakly fluorescent (Fig. 2D, E and I to J).
Under the SEM, the surface of ooids comprises batten-like needles, rods and nanograins, with dimensions ranging from 0.1 to 1.0 μm (Figs 3A, 3B, 4A and 4B).Rods and needles are aligned tangentially to the laminae but are oriented randomly within a lamina (Fig. 3B).Besides rods, the ooid cortex also consists of larger, irregular crystallites with a length of 2 to 10 μm (Fig. 3C and D).

Microendolithic borings
Microendolithic borings are present in most of the studied ooid surfaces, but the intensity of (micro-)bioerosion varies significantly between ooids, independent of the study site.While some grains display only localized features of microendolithic activity, the structural integrity of the surface layer and internal laminae of others is largely destroyed, and the cortical layers are obliterated (Fig. 3E and F).Carbonate cements on bioerosion cavity walls (Fig. 3G) exhibit a broad spectrum of morphologies.The most common are fibrous, acicular, or needle-shaped cement (Fig. 3H).Less commonly, columnar morphologies are present.Extracellular polymeric substances (EPS) of microbial origins are observed in the vicinity of bioerosion features at the ooid surface and the cortex layers (Fig. 4).The EPS-biofilm appears as smooth layers covering substantial portions of the ooid surface and occurs as long strands or torn filaments stretching between rods or crystals (Fig. 4A to F). Desiccated remnants of microbial EPS commonly extend between cavity walls.Rods or nanograins are prevalent within or on the EPS surface or occur in clusters at the margins of biofilm layers (Fig. 4A and D).Rods appear to preferentially form in net-like structures that resemble honeycomb or alveolar sacs.Various microbial organisms and collapsed microbial sheaths occur in bioerosion cavities, commonly in direct proximity to EPS.These include (in descending order of abundance) various genera of diatoms, cyanobacteria, coccolithophores and unidentified bacilli bacteria (Fig. 4C and F to H).
Observations of the vacuum casts reveal a range of microbioerosion traces (referred to as ichnotaxa) in cortical layers, characterized by distinctive sizes and shapes (Fig. 5).Microborings are most accessible in the outer cortex, particularly due to numerous open borings (Fig. 5A to C).Most microborings have a tubular to vermiform shape with distinct constrictions (Fig. 5D to F).For practical purposes, borings can be roughly separated into three categories based on diameter sizes: (i) small (ca <5 μm); (ii) intermediate (ca 5 to 15 μm); and (iii) large (>15 μm).Borings appear as single horizontal or vertical strings in larger clusters or wide galleries.Some ooids display super-surficial but extensive fan-shaped bioerosional features (cf.Fig. 5B).
Single microbioerosion traces extend into the ooid cortex and nucleus as much as 50 μm.Ooid nuclei show distinct patterns of microborings.Multiple ichnotaxa are generally present in a single ooid, resulting in a rather irregular Microborings and modern marine ooids 365 distribution of cortical microborings (Fig. 5B).Irregular clusters of small (ca 1 μm in diameter) round to oval tunnels and galleries are found close to tubular borings.Thin strings, lacking constrictions (<1 μm in diameter), can be observed crossing or overlaying larger (>1 μm) ichnotaxa (Fig. 5G and H).
The SEM analysis on the vacuum casting samples reveals that ooids from both localities harbour a diverse assemblage of ichnotaxa, mainly represented by a total of 12 ichnogenera (Table 1; Fig. 6).Table 1 lists concise information about ichnotaxa inferences, characteristic features and environmental implications.

Ooid microfabrics and microborer activity
Mechanical abrasion in agitated depositional settings dictates the orientation of rods of the outermost ooid cortex layers, and leads to polishing and compaction of the ooid surface.The ooid surface is prone to bioerosion caused by microbial organisms such as cyanobacteria, algae or fungi, for example, microendoliths, that chemically dissolve the outmost cortex layer.
Microborers leave shallow, horizontal or vertical tunnels in the surface layers.The latter suggests reworking of the underlying fabric with extensive tubular boring networks and developing widespread subsurface voids that likely compromise the ooid's structural integrity and resistance to physical abrasion.This could result in cavity collapse due to ooid transport and collision with other ooids.The collapse of these cavities results in the formation of irregular pits at the ooid surface.The partial loss of the hard ooid surface leads to increased microbial colonization of internal cortical layers, because the microendolithic communities can penetrate deeper into the ooid and become protected from both mechanical abrasion and predators.
Sequences of cortical layers with varying degrees of microborings and micritization, i.e. intact versus bored and micritized layers, point to the alternation of periods with undisturbed (or less disturbed) ooid cortex growth ('colonization window') followed by periods with a microbial reworking of the outer layers.This results in competition between eroders and constructors at the ooid surface, with the development of irregular cement crystals precipitating in open borings and cavities.The secondary precipitates develop after the rods have formed but in their immediate vicinity.Still, it remains uncertain whether the same microorganisms responsible Furthermore, it cannot be completely ruled out that biofilms excreted by different producers would not also lead to the formation of variable secondary precipitates.Some hypotheses abound seeking to explain this geochemical paradox, whereas concurrent physiological activities (i.e.photosynthesis and respiration) of endolithic cyanobacteria are an antagonistic effect, thereby promoting both calcification and bioerosion of the substrate (see McCutcheon et al., 2016).While some attribute this dichotomy to a temporal and/ or spatial separation of photosynthesis and boring activitiestriggering the production of alkali and acid metabolites (Macintyre et al., 2000;Reid & Macintyre, 2000) others propose a mechanism based on the active transport of Ca 2+ ions via calcium pumps (Garcia-Pichel et al., 2010).

Ichnotaxa and their corresponding light dependence in shallow waters and the uppermost sediment column
The observed microendolithic traces can be attributed to cyanobacteria (photoautotrophic organisms), with Fascichnus frutex, F. dactylus, F. rogus, Eurygonum nodosum and Scolecia Orange: tubular strings without constrictions; purple: vertical borings; red: filled bore traces in the ooid cortex; dark green: cluster of tubular tunnels, light green: string tubular tunnels; light blue: roundish features with a smooth surface; dark blue: nodular roundish features; yellow: threads attached to larger features.The black polygon indicates the surface of the ooid, while the white polygon marks the ooid cross-section.(G) and (H) Narrow threads (yellow arrows) overlying other (larger) ichnotaxa, indicating episodic occurrence of different producers.Microborings and modern marine ooids 369 filosa being the most abundant ichnotaxa in the ooids from both locations studied (see Table 1; Fig. 6).The study presented here confirms previous work (e.g.Harris et al., 1979;Gektidis, 1997;Gektidis et al., 2007;Radtke & Golubic, 2011).Microendoliths are attributed to the benthic community and are most active at the sedimentwater interface and inside the shallow sediment column (see May & Perkins, 1979;Golubic et al., 1984).Since distinct euendolithic taxa are known to have specific low-light tolerances and light optima, information about the relative bathymetry and ecology for ooid cortex formation can be obtained by noting the presence or absence of their respective index ichnotaxa by allocating the inferred index producer to the euphotic (>1% surface illumination), dysphotic (0.01 to 1% surface illumination; respiration rates exceed photosynthesis rates), or aphotic (<0.01%surface illumination; inhibited photosynthesis) zone (see del Giorgio & Peters, 1994;Glaub, 1994;Glaub et al., 2001Glaub et al., , 2007;;Wisshak, 2012).
In the supratidal and intertidal zones, cyanobacteria generally dominate.In the shallow subtidal zone, these organisms are joined by septate chlorophytes (Glaub et al., 2007).As an example, the ichnotaxon Ichnoreticulina elegans, produced by the siphonal green alga Ostreobium quekettii and other species in that genus, and Scolecia filosa produced by the cyanobacterium Plectonema terebrans, are attributed to the dysphotic zone, while heterotrophic organisms (for example, fungi such as Dodgella priscus), which are independent of light exposure, generally occur in the aphotic zone and are prominently represented by Saccomorpha clava, a key ichnotaxon of the aphotic index microboring ichnocoenosis (Glaub, 1994;Försterra et al., 2005;Glaub et al., 2007, see Table 1).
Particle size and composition determine light penetration in the uppermost portion of the sediment column (Colijn & de Jonge, 1984;Sullivan & Moncreiff, 1988;Pinckney & Zingmark, 1993).The abundance and vertical distribution of benthic biomass, for example, microalgae in sediments, depends on factors such as temperature, resuspension, nutrient and light availability, salinity, pH, desiccation, grazing and bioturbation (e.g.Cartaxana et al., 2006;Ichimi et al., 2008;Wisshak, 2012).Fenchel & Straarup (1971) noted a rapid decrease in irradiance to 50% at a sediment depth of 0.65 mm for particle sizes of up to 1 mm (see Ichimi et al., 2008).Generally, the larger the sediment particles, the deeper the light penetrates the uppermost sediment column  (Gomiou, 1967;K ühl et al., 1994;Yallop et al., 1994).Given the ooid size of ≤2 mm at the formation sites, an abundance of larger particles (for example, shell fragments), and reflective or even partly translucent grain surfaces, it is inferred that the depth at which light intensity is reduced by 50%, is deeper than 0.65 mm and may reach down to several centimetres (see Gomiou, 1967).At the studied sites, turbulent currents result in ooids experiencing periods of burial to various depths, either within ooid shoals or intershoal domains, punctuated by episodes of resuspension and transportation at the sediment surface over a time of minutes to hours.This is relevant because it implies that while the ichnocoenosis of a given ooid in the water column or at the sediment surface may correspond to the shallow euphotic zone, the ichnocoenosis, when buried deep enough, may shift towards one that is typical of the dysphotic to aphotic zone, with only the upper limit of the aphotic zone defined.For clarification, a given ooid buried at depths of approximately >30 cm to >1 m below the ooid shoal seafloor may be infested by microorganisms typical for the aphotic zone.This implies that microendolithic ichnotaxa can be applied as a proxy for ooid burial depth, even though both the Schooner Cays and the Shalil al Ud lagoon are characterized by an intertidal to shallow subtidal setting and clear water conditions.The studied ooid shoals are exposed during low tide and flooded during high tide.Intershoal domains are, depending on the location considered, in the shallow subtidal range, but intershoal channels of some metres of water depth also occur.
Consequently, the terms euphotic, dysphotic and aphotic, as used in the classical subdivision based on the light penetration depth of the water column (see Glaub et al., 2001Glaub et al., , 2007;;Wisshak et al., 2011), are not fully applicable in this context and have instead been substituted by light penetration levels (Levels 1 to 3, see Fig. 7).
Level 1 corresponds to the area between the sea surface and the immediate vicinity of the seafloor at a depth with approximately 1% of surface light (Glaub et al., 2001;see Wisshak, 2012).It refers to a region where the ooids are suspended in the water column and are exposed to high levels of illumination, and is characterized by euendolithic organisms usually attributed to the shallow euphotic realm (shallow I-II).Level 2 refers to an area where the ooid periodically resides within the uppermost region of the sediment column but not for an extended period, as the ooid may be exhumed due to wave and tidal activity.It is characterized by euendoliths in the euphotic to deep euphotic zone.Level 3 refers to a region where the ooid is buried within the sediment pile at, relatively speaking, greater depth.There it can no longer be resurfaced by the diurnal tide and average wave action and therefore remains idle for a prolonged period.Level 3 is characterized by euendoliths that prevail in the dysphotic to aphotic zone (see Fig. 7).
Depending on how long the ooid remains in one of these levels, the number of (cyanobacterial) borings in the ooid cortex and their resultant orientation will vary.Given that many euendolithic cyanobacteria are excellent palaeodepth indicators as they are stenobathic, showing a limited vertical distribution restricted to a narrow depth range (Glaub et al., 2001), it is possible to connect the orientation of borings to the proposed light penetration zones.While vertical borings seem the most common in Level 1 (corresponding to the intertidal zone), horizontal borings are more typical in Level 2 (corresponding to the subtidal zone, see review in Glaub et al., 2001;Wisshak, 2012).
Based on these findings, the migration of ooids between different light penetration zones in the shallow marine realm derived from the form and orientation of cyanobacterial borings can be linked to different stages in ooid cortex growth, including an agitation stage, a resting stage, a sleeping stage and a reactivation stage (cf.Davies et al., 1978;Duguid et al., 2010;Anderson et al., 2020).

Relation of oxic-anoxic zonation of ooid cortices with bioeroder distribution
Ooids can be considered as mini bioreactor sedimentary systems in which the outside is akin to the sediment surface where diagenetic processes are dominated by aerobic processes, and layers further into the centre of the ooid represent areas further removed and experiencing more anoxic regimes dominated by bacterial nitrate and sulphate reduction (Fig. 8).Evidence of these geochemical processes was provided not only by genetic analyses (Diaz et al., 2013a(Diaz et al., , 2013b) ) but also by the isolation of remnants of these geochemical reactions, such as elevated nitrate concentrations and δ 15 N values and lower SO 2À 4 /Cl À ratios (Diaz et al., 2015).As these processes proceed, diffusion gradients become established, with ions diffusing into and out of the ooid, supplying and removing reactants.Such zones may not always be consistent in their location because the ooids may be buried in a zone where the external environment might favour the development of more anoxic microenvironments within the ooid cortex, and gradients may be reversed.Such diagenetic processes then may be mirrored by the infestation by the endolithic organisms, with photosynthetic biota occupying the zones characterized by aerobic processes, while fungi and anaerobic biota, including anoxygenic phototrophs (for example, green sulphur bacteria, purple bacteria  marks the area between the sea surface and the very top of the sand dunes.At Level 1, traces of photosynthesizers prevail, so that a parallel can be drawn to the range of the shallow euphotic realm (shallow I + II) of the standard oceanographic zones.Level 2 (twilight zone) marks an area ranging from between the sand dunes to shallow burial.The illumination at Level 2 is highly fluctuating so that this level was equated to the shallow III to deep euphotic zone.Level 3 (no light zone) marks an episode of deep burial with no illumination.Based on the ichnotaxa found (suboxic chemotrophs and fungi), Level 3 was set as a synonym to the dysphotic to aphotic realm.The degree of light transmission in the sediment pile depends on the grain size (K ühl et al., 1994;Yallop et al., 1994).The residence time in the different zones depends mainly on wind and wave action, and tidal currents.

Ooid formation model based on agitation, resting and sleeping stages
The following relationship between ooid formation stages and microbial interaction at Level 1 (high light), Level 2 (twilight) and Level 3 (no light) is proposed (Fig. 9).
Agitation/suspension stage -Level 1, corresponding to shallow euphotic zone I-II This stage is situated at the interface between the water column and the seafloor, i.e. at the surface of the ooid shoal.Cyanobacteria and other microbes colonize the surface of carbonate (i.e.skeletal fragments) and non-carbonate particles (i.e.faecal pellets and quartz) that later become the ooid nuclei.As colonization takes place, microbial exudates surround the nuclei, forming an EPS-biofilm thatupon degradationacts as locations for the initial steps of crystal nucleation through the amorphous calcium carbonate (ACC) phase (Duguid et al., 2010;Diaz et al., 2017).As aggregates of ACC form, secondary nucleation promotes the formation of the first cortical layer comprising aragonite needles.During this stage, ooids are constantly agitated due to wave and current activity, moving freely in the water column and at the shoal surface.Rolling and saltation at the sediment surface and collision with other grains result in the flattening (tamping down) of newly formed and randomly orientated aragonite Fig. 8.The sketch of an ooid illustrates the relationship between oxic (white) and anoxic (purple) zones and their associated ichnotaxa, which is indicated by a colour code.Note that not all discovered borehole features could be assigned to a particular ichnotaxon, and some of the ichnotaxa were likely not discovered.The colour code for the different ichnotaxa is shown in the upper right and was also used for the ichnotaxa distribution in Fig. 7. Most traces in the oxic zones are attributed to cyanobacteria (various Fascichnus ichnospecies, Eurygonum nodosum and Scolecia filosa) followed by chlorophytes (Ichnoreticulina elegans, Cavernula pediculata) and rare fungi (Saccomorpha sphaerula).The suboxic regime is characterized by chemotrophs (for example, Orthogonum lineare) and fungi (for example, Saccomorpha clava).The producer is unknown in some cases (for example, Planobola macrogota).
needles.This results in tangential crystal orientation of the needles and polishing and compaction of the cortex layer that forms the ooid surface.The presence of cyanobacteria produces mainly vertical tunnels.At this stage, microendolithic communities are influenced by both photic and hydrodynamic factors (Glaub et al., 2001) and chlorophyte colonization of the ooid surfaces is strongly limited due to constant movement and mechanical erosion of the ooids (Gektidis, 1997).
Resting stage -Level 2, corresponding to shallow III to deep euphotic Temporary burial of the ooid in the uppermost shoal or intershoal setting impedes ooid growth (Gektidis, 1997;Diaz et al., 2017;Anderson et al., 2020).The immobilization of ooids leads to colonizing cortices by microorganisms preferring stable environmental conditions.Euendolithic assemblages change their composition and boring behaviour (Glaub et al., 2001).In contrast to the shallow I and II zones, here, horizontal boring patterns dominate as a result of decreased illumination, possibly indicating the initiation of a shift from phototrophic to heterotrophic organisms.Based on the ichnotaxa analysis of the casts, the main microbial colonizers in the ooid cortices continue to be cyanobacteria, accompanied by increasing numbers of chlorophytes, rhodophytes and fungi.

Sleeping stage -Level 3, corresponding to dysphotic to aphotic
The ooid is buried in the upper sediment column (centimetres to a few metres) of the ooid shoal for an extended period, and no sunlight penetrates to this depth.Nevertheless, this stage is characterized by increased microbial bioerosion.Numerous but shallow borings with a horizontal orientation occur.The crystallization of ACC to aragonite needles continues.Moreover, direct carbonate precipitation from porewater, in combination with microbial interactions, is likely.Bore traces are partially occluded with carbonate cement crystals.Microbial fingerprints on the casts identify heterotrophs (i.e.fungi) dominating the microborer spectrum.The abundance of organotrophs is expected to rise with burial depth.Photosynthesizing organisms are absent due to the prolonged residence of the ooid within the sediment pile, as light access is limited or unavailable.If ooids are transported to the interdune zones stabilized by seagrass, the sleeping stage may last for an extensive period of time.It seems likely that, in this case, only dune migration may lead to a reactivation of ooid development.
Reactivation stage -Levels 1-2, corresponding to the euphotic zone Hydrodynamically-induced dune migration or stir-up by storms results in the (re-)exposure of the ooids to the seafloor and, potentially, into the water column.Due to renewed light availability, microbial communities shift from mainly fungi and organotrophs to cyanobacteria and photosynthetic protists.This stage is characterized by the transition from a horizontal to a vertical orientation of borings, clearly representing the shift in the microendolithic community.The dominant microborers are cyanobacteria, with rare chlorophytes.
Based on the increasing amount of euendolithic assemblages associated with increased bioerosion with depth (cf.Glaub et al., 2001Glaub et al., , 2007)), it is hypothesized that the production of cortical layers by ACC nucleation outweighs bioerosion only during the suspension/agitation phase in the shallow I-II zones.These two stages are therefore referred to as active ooid growth phases.When the ooids enter the sleeping stage, cortical layer destruction by bioerosion predominates over layer formation.Due to (shallow) burial and the resulting lower light availability, it is suggested that fungi dominate during the sleeping stage, while microalgae dominate during the agitation, resting and reactivation stages.
On that note, there are fungal metabolisms that can induce increases in alkalinity (for example, physicochemical degassing of respired CO 2 ; Bindschedler et al., 2016), organic acid oxidation (Guggiari et al., 2011), urea mineralization (Burbank et al., 2011) and nitrate assimilation (Hou et al., 2011), and therefore can potentially contribute to carbonate precipitation.However, according to other studies (e.g.Priess et al., 2000), fungi are also aggressive decomposers of carbonate substrates, to the point where fungi even parasitise algae (Schneider, 1976;Le Campion-Alsumard et al., 1995), a process that is likely to occur during the resting and, especially, during the sleeping stage, as documented by characteristic bioerosion traces.Still, bioerosion cannot be attributed to fungal communities alone, because endolithic cyanobacteria and diatoms actively destroy the cortex by boring into the substrate, and organisms such as aerobic heterotrophic bacteria, aerobic sulphide oxidizers and nitrifying bacteria participate in the erosion of the Microborings and modern marine ooids 375 grain through the production of acids as byproducts of their metabolisms (Diaz et al., 2013a(Diaz et al., , 2013b(Diaz et al., , 2014(Diaz et al., , 2015(Diaz et al., , 2017)).
Another aspect of ooid cortex growth is the occlusion of bore tunnels at a cortex depth of approximately 100 μm, which passively contributes to the ooid cortex.This process, however, does not lead to an increase in cortical mass, as it merely fills the previously drilled holes and, therefore, cannot be described as cortex growth.
Calcified remnants of meniscus-like structures were observed, which consisted entirely of rods (see Fig. 3E and F).These meniscus structures may indicate a former capillary water bridge between two layers of the ooid cortex, suggesting air in a cavity (Fl ügel, 1979;Folk & Lynch, 2001).However, other studies suggest that organic biofilms in a completely phreatic (saturated) environment can also form a meniscus-like structure, so this observation should not be used as the sole indicator for vadose environments or subaerial exposure (Webb et al., 1999;Hillgärtner et al., 2001).Because of this, menisci in ooid cortices may suggest either inorganic mineralization by precipitation from a porewater in a supratidal environment or mineralization from biofilm residues serving as nucleation templates, the latter possibility being the more likely (cf.Fig. 4A to B and D to E).

Implications for ooid formation and evolution
The extent to which microorganisms with different light preferences contribute to cortex formation or destruction is still controversial.Previous models (e.g.Davies et al., 1978;Duguid et al., 2010;Diaz et al., 2014Diaz et al., , 2017) ) largely focused on geochemical, geomicrobiological or organomineralogical processes involved in cortex formation without characterizing microendolithic communities on ooid surfaces in terms of their preferred habitats (but see Vogel et al., 2000;Gektidis et al., 2007;Glaub et al., 2007).On that note, few, if any, contributions to the broad field of ooid research address the subject of light penetration into the sediment column and how light availability affects the composition of ooid microbial communities.It is hypothesized that there is permanent competition between contemporary microendolithic borers, EPS biofilm production (for example, by biofilm bacteria, diatoms or cyanobacteria), and the precipitation of ACC (inducing the precipitation of aragonite) from biofilms (Diaz et al., 2017).Microbes (diatoms, cyanobacteria, biofilm bacteria, fungi) produce EPS, which then serve as a nucleation template for rods.This formation mechanism is referred to as biologically influenced organomineralization (cf.Dupraz et al., 2009;Diaz et al., 2017).Besides the EPS-mediated mineralization, microbes could also actively contribute to cortex accretion via metabolic pathways (i.e.photosynthesis, denitrification, sulphate reduction, ammonification and anaerobic sulphate oxidation) that generate alkalinity conditions favourable to carbonate precipitation (Dupraz et al., 2009;Diaz et al., 2014Diaz et al., , 2015)).Accordingly, ooid cortex formation is facilitated by a mixture of biologically influenced and biologically induced organomineralization processes.
In contrast to this constructive function of microbes during ooid cortex formation, others (e.g.Sumner & Grotzinger, 1993;Duguid et al., 2010) acknowledge the involvement of microbes in ooid evolution but argue for an exclusively destructive role.It follows that these authors distinguish between a phase of cortex formation (abiogenic) and a phase of cortex destruction by microendolithic borers.According to Duguid et al. (2010), abiogenic ACC precipitation occurs during a 'stationary phase' (i.e. the stage of cortex formation), whereas microbes such as Solentia or Hyella contribute solely to cortex layer Fig. 9. Modern ooid growth and cortex infestation model (after Davies et al., 1978, andAnderson et al., 2020, with respect to the results of Duguid et al., 2010, andDiaz et al., 2017).The four different stages of ooid cortex growth include an agitation stage at Level 1, characterized by very little microboring activity and the formation of amorphous calcium carbonate (ACC) nucleating from extrapolymeric substance (EPS) biofilms on the grain surface; a resting stage at Level 2, characterized by an increased physical abrasion and microborers; a sleeping stage at Level 3, characterized by burial in the sediment with not light and strong bioperforation of the cortical layers; and a reactivation stage where the ooids are resurfaced to Levels 1 and 2, and subsequently characterized by a shift in the microbial community from heterotrophs (mainly fungi) to photosynthesizers.While aerobic processes prevail in the outmost cortex layers, suboxic conditions can develop in microenvironments in deeper cortex layers due to bacterial nitrate and sulphate reduction.Anaerobic processes take place predominantly during the sleeping stage, favoured by the absence of light and the generally lower oxygen supply.Microborings and modern marine ooids 377 destruction.Based on their observations, microbes do not play a primary role in ooid genesis.This does not align with the findings presented here, which suggest that microbes play a role in ooid cortex formation through microbial metabolic pathways (i.e.photosynthesis) and EPS-mediated mechanismthe latter providing nucleation sites for mineral precipitation.
In addition, this study suggests that bioerosion outweighs precipitation only during the sleeping stage ('stationary stage' of Duguid et al., 2010).In this context, it is argued that the term 'sleeping stage' implies a stasis in ooid development.
In light of the data presented here, this stage is characterized by active cortex destruction.This raises the question of whether the term sleeping stage is appropriate or should be replaced by the more-appropriate term 'destruction stage'.
On a different note, this study could also hold implications for forming fossil ooids or ooids in environments other than the shallow marine realm (for example, lacustrine and deep marine).If sufficiently preserved, a detailed examination of the microbial communities in the cortex could help to reconstruct the palaeoenvironmental conditions under which fossil ooids formed.

CONCLUSIONS
Features of microbial bioerosion in Holocene ooids from two sampling sitesthe Schooner Cays ooid shoals (Great Bahama Bank, Eleuthera, Bahamas) and the Shalil al Ud ooid shoals (Abu Dhabi, United Arab Emirates)were investigated, and vacuum casting was successfully applied, allowing the precise identification of ichnotaxa and their inferred producer as well as their assignment to different ooid formation stages.
The abundance of microendolithic traces varies greatly between grains, suggesting a fluctuating residence time of the ooids at different surficial or shallow burial stages at both sampling sites.Using ichnotaxa as proxies for residence time on the dune surface or within the sediment column depends on the dominance of bioerosion traces efficiently assigned to heterotrophic fungi or photoautotrophic algae.Work shown here differentiates between a high light zone (Level 1) in the water column and above the sediment surface, a twilight zone (Level 2) at the sediment-water interface, constantly affected by shallow burial and resuspension, and a no light zone (Level 3) inside the dune sediment column, characterized by a prolonged burial with an unspecified lower depth limit that might extend to several metres.This approach provides insight into the distribution and abundance of euendolithic activity on ooids during the established growth stages.This paper links: (i) the stages of ooid formation; (ii) light availability at the seafloor and in the uppermost sediment column during each stage; and (iii) the resulting composition and distribution of the succeeding ichnotaxa reflecting different microbial communities on and in the ooids.This allows reconstructing the various formation and destruction cycles of a given ooid.Data shown here differentiate an agitation/suspension stage, a resting stage, a sleeping stage (or 'destruction' stage, as proposed here) and a reactivation stage.When applied systematically and for a statistically significant number of ooids, this approach has significant potential to better understand ooid dune dynamics and temporal changes thereof.

Fig. 1 .
Fig. 1.Location of sampling sites indicated by red stars.(A) Map of the Great Bahama Bank with sampling sites on Schooner Cays ooid shoals, located north of Exuma Sound and east of the Eleuthera Islands.(B) Eastern Gulf, with indication of outer lagoon of Abu-Dhabi (Shalil al Ud, ca 20 km north-east of Abu Dhabi city).Small inset on top locates the sampling sites on the world map.(C) Drone images of the Schooner Cays ooid shoals, Bahamas (image courtesy of Drew Dalton).(D) Field image of the Abu-Dhabi outer lagoon shoals (Shalil al Ud), Eastern Gulf.

Fig. 2 .
Fig. 2. Reflected and transmitted light microscopic images of Schooner Cays (A) to (E) and Shalil al Ud (F) to (J) ooids, hereafter referred to as GBB (Great Bahama Bank, Schooner Cays) and AD (Abu Dhabi, Shalil al Ud).(A) Binocular microscope image of subsample of ooidal sands from GBB. (B) to (E) Thin section photomicrograph images of an ooid from the GBB sampling site.(B) GBB ooid under transmitted light.Note the prominent vertical microendolithic borings in the cortex (orange arrows) and the <20 cortical laminae.(C) Same as (B), but under polarized light, with non-micritized parts of the cortex showing bright pearlescent colours (black arrow).(D) and (E) Same as (A) and (B) but under fluorescence light.Note the bright fluorescence colours of the unspecified nucleus (mollusc fragment?), and the clearly visible rim cement (yellow arrows).(F) Binocular microscope image of subsample of ooidal sands taken from AD. (G) to (J) Thin section photomicrograph images of various ooids from the AD sampling site.Note nuclei and 4 to 15+, in part micritized, laminae.(G) Transmitted light image showing nuclei (foraminifera, blue arrow, chambers filled with blocky spar, pink arrow) and beige cortex layers.(H) Same as (B) but under crossed polarizers.Note the bright pearlescent colours of non-micritized laminae (black arrow).(I) and (J) Same as (G) and (H) but under fluorescence light.Note the bright fluorescence colours of the biogenic materials (for example, foraminifera) non-micritized layers, and rim cements (yellow arrows).

Fig. 3 .
Fig. 3. Scanning electron microscopy (SEM) images of distinct features and microbioerosion traces in marine ooids, accentuated by etching agents -Great Bahama Bank (GBB): A, E, F, G, H; Abu Dhabi (AD): B, C, D. (A) Magnification of an ooid surface.Note the combination of rods, nanograins and cement forming the solid surface.(B) Cortical layers (etched with 0.1 M HCl, 6 h).Note the random orientation of rods between horizontally arranged planes with a much denser packing of rods.(C) and (D) Vertical bridges of randomly orientated rods (white arrows) connecting cortical layers (5 M acetic acid, 30 min).Note the larger crystallites, attributed to microendolithic bore traces, which accumulate on top of or in between finer rods (green arrows).(E) Ramified network of microendolithic borings disrupting an ooid surface.(F) Sequential changes of nearly undisturbed cortical layers and intervals of heavy microendolithic reworking (etched with 0.27 M EDTA, 1 h).(G) Cavities beneath the outer surface.Older borings closer to the nucleus are filled with carbonate.(H) Aragonite needles developing in surficial cavities.

Fig. 4 .
Fig. 4. Scanning electron microscopy (SEM) images of biofilms and microbes on marine ooids -Great Bahama Bank (GBB); A-D, G-H; Abu Dhabi (AD): E, F. (A) Extracellular polymeric substance (EPS) biofilm blanket with amorphous calcium carbonate (ACC) and strands extending over and between ooid surface irregularities (cement, boreholes).(B) Thin reticulate EPS strands stretching between single rods.(C) Collapsed microbial filament protruding from a vertical borehole.(D) Biofilm blanket covering a cavity.The marginal thickening (yellow arrow) and the graininess (blue arrow) in the centre of the blanket indicate the nucleation of ACC.(E) EPS, rods and larger acicular crystals forming honeycomb-like structures.(F) Variety of microbial life (for example, pennate diatom, green arrow; coccolith, blue arrow) in direct proximity to EPS (yellow arrows) inside surface cavities.(G) Microbial filament surrounded by desiccated EPS.(H) Coccobacillus bacteria, characterized by short rods and oval shape morphology.

Fig. 5 .
Fig. 5. Scanning electron microscopy (SEM) images of endocasts of different ichnotaxa in the ooid cortex -Great Bahama Bank (GBB): A, C, E, G; Abu Dhabi (AD): B, D, F. (A) Average ooid showing the common extent of microendolithic damage to the cortex.(B) Inhomogeneous distribution of the microboring pattern.Note the differentiation between horizontal tubes and cauliflower-like domes.Note also the superficial fan-shaped bioerosion traces (blue rectangle).(C) Dimensional difference between ichnotaxa (left ooid = medium; right = large).(D) Filled vertical microborings (white arrow) and preparation residues (black arrow) in an ooid cortex.Note the dimensional difference between the layer forming rods and the needle to plate cement filling the boreholes.(E) and (F) Various ichnotaxa marked by arrows.Orange: tubular strings without constrictions; purple: vertical borings; red: filled bore traces in the ooid cortex; dark green: cluster of tubular tunnels, light green: string tubular tunnels; light blue: roundish features with a smooth surface; dark blue: nodular roundish features; yellow: threads attached to larger features.The black polygon indicates the surface of the ooid, while the white polygon marks the ooid cross-section.(G) and (H) Narrow threads (yellow arrows) overlying other (larger) ichnotaxa, indicating episodic occurrence of different producers.

Fig. 7 .
Fig.7.Division of light penetration zones (Levels) based on euendolithic occurrence.Level 1 (high light zone) marks the area between the sea surface and the very top of the sand dunes.At Level 1, traces of photosynthesizers prevail, so that a parallel can be drawn to the range of the shallow euphotic realm (shallow I + II) of the standard oceanographic zones.Level 2 (twilight zone) marks an area ranging from between the sand dunes to shallow burial.The illumination at Level 2 is highly fluctuating so that this level was equated to the shallow III to deep euphotic zone.Level 3 (no light zone) marks an episode of deep burial with no illumination.Based on the ichnotaxa found (suboxic chemotrophs and fungi), Level 3 was set as a synonym to the dysphotic to aphotic realm.The degree of light transmission in the sediment pile depends on the grain size(K ühl et al., 1994;Yallop et al., 1994).The residence time in the different zones depends mainly on wind and wave action, and tidal currents.

Ó
2023 The Authors.Sedimentology published by John Wiley & Sons Ltd on behalf of International Association of Sedimentologists., Sedimentology, 71, 359-382 Microborings and modern marine ooids 373 and Chloroflexi) primarily colonize anoxic or microaerophilic areas on the grain.