Search

Lower Triassic limestones contain giant ooids (>2 mm) along with other precipitated carbonate textures more typical of Precambrian strata. These features appear to have resulted from changes in seawater chemistry associated with the end-Permian mass extinction, but quantifying the carbonate chemistry of Early Triassic seawater has remained challenging. To constrain seawater carbonate saturation state, dissolved inorganic carbon, alkalinity, and pH, we applied a physicochemical model of ooid formation constrained by new size data on Lower Triassic ooids from south China, finding that the Triassic giant ooids require a higher carbonate saturation state than typifies modern sites of ooid formation. Model calculations indicate that Early Triassic oceans were at least seven times supersaturated with respect to aragonite and calcite. When combined with independent constraints on atmospheric pCO2 and oceanic [Ca2+], these findings require that Early Triassic oceans had more than twice the modern levels of dissolved inorganic carbon and alkalinity and a pH near 7.6. Such conditions may have played a role in inhibiting the recovery of skeletal animals and algae during Early Triassic time.

The upper Cambrian Wilberns Formation in central Texas records deposition on a low‐gradient shelf within a mixed carbonate–siliciclastic tidal‐flat system that changes offshore to subtidal shelf and open‐marine oolitic skeletal shoals with large microbial mounds. Siliciclastic sediment is interpreted to have been delivered to the tidal flat by aeolian processes because of the narrow range in grain size and paucity of clay. Tidal influence is dominant as evidenced by reversing currents and desiccation on the tidal flat, and megaripples with reversing current indicators in offshore shoals. Intraclastic conglomerates were deposited in broad channels on the tidal flats during storm surges. Microbialite deposition is interpreted to be controlled by accommodation favouring amalgamated thin biostromes developed in the tidal flat vs. larger mounds with greater synoptic relief in the offshore, and current energy resulting in preferential elongation of offshore mounds in a NE–SW orientation. Intertidal mounds and biostromes grew in the presence of significant siliciclastic flux and trapped it within their structure, whereas offshore large buildups incorporated little siliciclastic component. Oolite and skeletal grainstone formed in tide agitated shoals associated with large subtidal microbial mounds. Storms extensively recycled and redistributed skeletal and oolitic sands from the offshore shoals across the shelf as thin sand sheets. Spatial mixing of siliciclastic and carbonate sediment occurred across the tidal flat and shelf. Low‐frequency and intermediate‐frequency stratigraphic cycles were driven by shifts in the shoreline and changes in rate of siliciclastic flux in response to relative sea‐level fluctuation. Random facies stacking and the lack of metre‐scale cyclicity are interpreted to reflect stratigraphic incompleteness and an episodic signal introduced by storms.

Faults are commonly the largest fractures in a fracture network, forming the backbone of the permeability system in low-permeability rocks. Faults in outcrops of the Austin Chalk — an important oil and gas reservoir in south Texas, Louisiana, and westernmost Mississippi — exemplify deformation and fluid histories of subseismic-scale faults in mechanically layered chalk-dominated strata. Exposures of near-horizontal Austin Chalk in southwest Texas contain northwest- and southeast-dipping normal faults with exposed heights of meters to tens of meters and displacements of centimeters to decimeters. Local fault dips are steep to vertical through chalk and limestone beds and moderate through marl, mud rock, and clay-rich ash beds, producing refracted fault profiles. Failure surface characteristics indicate various failure modes, including tensile, hybrid, shear, and (locally) compactive shear behavior depending on the host lithology. Dilational zones associated with steep fault segments host calcite veins. Crack–seal textures in fault zone veins record repeated reactivation of refracted fault profiles. Fluid inclusion and stable isotope geochemistry analyses of fault veins indicate formation at depths of 1.4 to 2.9 km (4593 to 9514 ft) and possibly of 4.2 km (13,780 ft). Fluids include locally sourced saline waters and externally sourced waters and oil in rocks that did not reach oil maturation conditions. Inclusions of 30° API to greater than 50° API oil are found in veins 30 km (19 mi) updip from the Eagle Ford Formation oil zone. These results illustrate the migration of oil and other fluids within subseismic-scale Austin Chalk faults and fundamental control of lithology on fault zone architecture.

Foraminifera are important components of tropical marine benthic ecosystems and their recovery pattern from the end-Permian mass extinction can yield insights into the Mesozoic history of this group. Here we report the calcareous and agglutinated foraminifera recovered from five measured stratigraphic sections on the Great Bank of Guizhou, an uppermost Permian to Upper Triassic isolated carbonate platform in the Nanpanjiang Basin, south China. The material contains >100 Triassic species, including three that are newly described (Arenovidalina weiin. sp.,Meandrospira?enosin. sp., andSpinoendotebanella lehrmannin. gen., n. sp.), ranging from Griesbachian (Induan) to Cordevolian (Carnian) age. The species belong to the classes Miliolata, Textulariata, Fusulinata, Nodosariata, and to an unknown class housing all aragonitic forms of the orders Involutinida and Robertinida. Based on previously established conodont zones and carbon isotope chemostratigraphy, the Griesbachian (early Induan) through Illyrian (late Anisian) interval has been subdivided into 12 foraminiferal zones and two unnamed intervals devoid of foraminifera. Following the extinction at the Permian-Triassic boundary, habitable ecological niches of Griesbachian age were invaded by disaster taxa that subsequently became extinct during the Dienerian (late Induan) and left no younger descendants. The disaster taxa were replaced by Lazarus taxa with Permian origins, which were then decimated by the Smithian-Spathian (mid-Olenekian) boundary crisis. The tempo of recovery appears to have been modulated by environmental changes during the Griesbachian through Smithian that involved both climate change and expansion of anoxic ocean bottom waters. Uninterrupted and lasting recovery of benthic foraminifera did not begin until the Spathian.UUID:http://zoobank.org/2a6e9061-b163-402a-9098-8765a80576b3

Facies of the Boquillas and equivalent Eagle Ford Formation can be assigned to two associations: a hydrodynamically recycled association and a pelagic association. The hydrodynamically recycled association is composed of lenticular to continuous cross-laminated lime packstone–grainstone, heterolithic facies, and intraclastic grainstone conglomerate with intercalated calcareous mudrock. Sharp bedding contacts, abundant scours, ripple cross-lamination, oscillatory-current indicators, hummocky cross-bedding, and rounded clasts indicate recycling of the bottom sediments by storms and bottom currents above storm wave base (SWB). The pelagic association is composed of continuous lime packstone to grainstone and chalk beds with intercalated calcareous mudrock. Gradational contacts and planar lamination indicate dominant deposition by pelagic rain, but scour surfaces, coarse-fine alternations, and ripple cross-laminae indicate the influence of bottom currents below SWB.Vertical shifts between the hydrodynamically recycled association, interpreted to represent storm agitation of the seafloor during falls in sea level, alternate with the pelagic association and define four sequence boundaries and three depositional sequences with a lower-order periodicity (third-order periodicity, ca. average 2.4 My). Sequence boundaries correspond approximately to member boundaries recognized by gamma-ray profiles. Spectral analysis reveals higher-frequency mudrock–carbonate cycles prominent in the pelagic association, likely resulting from Milankovitch climate forcing of planktonic carbonate productivity versus clay flux. Comparison of gamma-corrected spectra with absolute age constraints reveals that the Ernst Tinaja section (in Big Bend National Park) is less than 28% complete. This is interpreted to result from abundant depositional hiatuses and erosional gaps that occur at the bed scale and at member boundaries within and bounding the Eagle Ford Formation.We demonstrate that deep-marine pelagic deposits in epicratonic settings may be subject to a complex interaction of depositional processes including pelagic rain and recycling by bottom currents and wave-generated currents modulated by low-frequency sea-level fluctuation and high-frequency fluctuations in sediment supply.

We have developed and validated a new approach to upscale lithology and porosity-type fractions from thin sections to cores using dual energy and multiscale computed tomography (CT). A new rock-typing approach (genetic rock typing [GRT]) is proposed to upscale ⇋diagenetic mineral and diagenetic pore-type fractions, from thin sections to the core domain, eventually to create a diagenesis and porosity types logs. An extensive set of short cores from Mason County (Texas) provides a representative sample set of Late Cambrian microbial buildups and their interbuildup sediments to test the GRT approach. GRTs were defined by using a dolomite log as a proxy for diagenesis and the average percentage of dolomite from each observed depositional facies (buildup interior, buildup rind, and interbuildup sediment) as a cutoff. Dolomite, diagenetic calcite, and diagenetic porosity fractions are summed to form a diagenesis log, which captures depositional facies and the diagenetic overprint at a 0.5 mm resolution. The diagenesis log was subdivided based on the number of pore-throat size classes within each GRT and provided a framework to distribute porosity-type fractions from thin sections to log form. A high correlation coefficient is observed when the predicted extent of diagenetic alteration from the log is compared with that quantified for each thin section using image processing ([Formula: see text]). Multiscale CT imaging and dual-energy-derived logs could be directly linked to well-log photoelectric factor and bulk-density logs. This approach thus has the ability to span six orders of magnitude in resolution (500–0.0005 mm). The diagenesis log can be used to extrapolate porosity-type fractions from thin sections to logs, from which qualitative geologic interpretations can be generally translated into quantitative values.

The growing digitization of fossil images has vastly improved and broadened the potential application of big data and machine learning, particularly computer vision, in paleontology. Recent studies show that machine learning is capable of approaching human abilities of classifying images, and with the increase in computational power and visual data, it stands to reason that it can match human ability but at much greater efficiency in the near future. Here we demonstrate this potential of using deep learning to identify skeletal grains at different levels of the Linnaean taxonomic hierarchy. Our approach was two-pronged. First, we built a database of skeletal grain images spanning a wide range of animal phyla and classes and used this database to train the model. We used a Python-based method to automate image recognition and extraction from published sources. Second, we developed a deep learning algorithm that can attach multiple labels to a single image. Conventionally, deep learning is used to predict a single class from an image; here, we adopted a Branch Convolutional Neural Network (B-CNN) technique to classify multiple taxonomic levels for a single skeletal grain image. Using this method, we achieved over 90% accuracy for both the coarse, phylum-level recognition and the fine, class-level recognition across diverse skeletal grains (6 phyla and 15 classes). Furthermore, we found that image augmentation improves the overall accuracy. This tool has potential applications in geology ranging from biostratigraphy to paleo-bathymetry, paleoecology, and microfacies analysis. Further improvement of the algorithm and expansion of the training dataset will continue to narrow the efficiency gap between human expertise and machine learning.

A negative shift in the calcium isotopic composition of marine carbonate rocks spanning the end-Permian extinction horizon in South China has been used to argue for an ocean acidification event coincident with mass extinction. This interpretation has proven controversial, both because the excursion has not been demonstrated across multiple, widely separated localities, and because modeling results of coupled carbon and calcium isotope records illustrate that calcium cycle imbalances alone cannot account for the full magnitude of the isotope excursion. Here, we further test potential controls on the Permian-Triassic calcium isotope record by measuring calcium isotope ratios from shallow-marine carbonate successions spanning the Permian-Triassic boundary in Turkey, Italy, and Oman. All measured sections display negative shifts in δ44/40Ca of up to 0.6‰. Consistency in the direction, magnitude, and timing of the calcium isotope excursion across these widely separated localities implies a primary and global δ44/40Ca signature. Based on the results of a coupled box model of the geological carbon and calcium cycles, we interpret the excursion to reflect a series of consequences arising from volcanic CO2 release, including a temporary decrease in seawater δ44/40Ca due to short-lived ocean acidification and a more protracted increase in calcium isotope fractionation associated with a shift toward more primary aragonite in the sediment and, potentially, subsequently elevated carbonate saturation states caused by the persistence of elevated CO2 delivery from volcanism. Locally, changing balances between aragonite and calcite production are sufficient to account for the calcium isotope excursions, but this effect alone does not explain the globally observed negative excursion in the δ13C values of carbonate sediments and organic matter as well. Only a carbon release event and related geochemical consequences are consistent both with calcium and carbon isotope data. The carbon release scenario can also account for oxygen isotope evidence for dramatic and protracted global warming as well as paleontological evidence for the preferential extinction of marine animals most susceptible to acidification, warming, and anoxia.

The geological calcium cycle is linked to the geological carbon cycle through the weathering and burial of carbonate rocks. As a result, calcium (Ca) isotope ratios (44Ca/40Ca, expressed as δ44/40Ca) can help to constrain ancient carbon cycle dynamics if Ca cycle behavior can be reconstructed. However, the δ44/40Ca of carbonate rocks is influenced not only by the δ44/40Ca of seawater but also by diagenetic processes and fractionation associated with carbonate precipitation. In this study, we investigate the dominant controls on carbonate δ44/40Ca in Upper Permian to Middle Triassic limestones (ca. 253 to 244 Ma) from south China and Turkey. This time interval is ideal for assessing controls on Ca isotope ratios in carbonate rocks because fluctuations in seawater δ44/40Ca may be expected based on several large carbon isotope (δ13C) excursions ranging from − 2 to + 8‰. Parallel negative δ13C and δ44/40Ca excursions were previously identified across the end-Permian extinction horizon. Here, we find a second negative excursion in δ44/40Ca of ~ 0.2‰ within Lower Triassic strata in both south China and Turkey; however, this excursion is not synchronous between regions and thus cannot be interpreted to reflect secular change in the δ44/40Ca of global seawater. Additionally, δ44/40Ca values from Turkey are consistently 0.3‰ lower than contemporaneous samples from south China, providing further support for local or regional influences. By measuring δ44/40Ca and Sr concentrations ([Sr]) in two stratigraphic sections located at opposite margins of the Paleo-Tethys Ocean, we can determine whether the data represent global conditions (e.g., secular variations in the δ44/40Ca of seawater) versus local controls (e.g., original mineralogy or diagenetic alteration). The [Sr] and δ44/40Ca data from this study are best described statistically by a log-linear correlation that also exists in many previously published datasets of various geological ages. Using a model of early marine diagenetic water-rock interaction, we illustrate that this general correlation can be explained by the chemical evolution of bulk carbonate sediment samples with different initial mineralogical compositions that subsequently underwent recrystallization. Although early diagenetic resetting and carbonate mineralogy strongly influence the carbonate δ44/40Ca values, the relationship between [Sr] and δ44/40Ca holds potential for reconstructing first-order secular changes in seawater δ44/40Ca composition.

The Late Permian to Late Triassic Great Bank of Guizhou (GBG) in southwest China is one of the few isolated carbonate platforms in the world that exposes an essentially complete record of initiation, development, and drowning across multiple platform-to-basin transects. The platform is exceptionally exposed in cross section at the surface by a faulted syncline that rotated strata to a dip angle of approximately 65°. Platform development spanned the end-Permian extinction and Triassic recovery that marks the transition from Paleozoic to Mesozoic styles of carbonate sediment production, providing a rare opportunity to assess the impact of global changes in carbonate factory types at a single locality. In addition, regional basin controls such as differential siliciclastic sediment input and varied antecedent topography provided mechanisms for lateral variability in platform morphology that can be investigated along exposures in several geographic sectors. Consequently, the GBG preserves a record of temporal and spatial variability in platform architecture that offers an unparalleled opportunity to investigate the controls on isolated carbonate platform morphology. A better understanding of these mechanisms is critical for improving predictive geologic models in exploration and field-development settings. The GBG also serves as a key outcrop analog for Early Triassic oolite reservoirs in the Middle East and China, the steep microbial-boundstone slopes of Carboniferous platforms in Kazakhstan, and the Permian platforms of Texas and New Mexico.

A 1.5 km long natural cliff outcrop of nearly horizontal Eagle Ford Formation in south Texas exposes northwest and southeast dipping normal faults with displacements of 0.01–7 m cutting mudrock, chalk, limestone, and volcanic ash. These faults provide analogs for both natural and hydraulically-induced deformation in the productive Eagle Ford Formation – a major unconventional oil and gas reservoir in south Texas, U.S.A. – and other mechanically layered hydrocarbon reservoirs. Fault dips are steep to vertical through chalk and limestone beds, and moderate through mudrock and clay-rich ash, resulting in refracted fault profiles. Steeply dipping fault segments contain rhombohedral calcite veins that cross the fault zone obliquely, parallel to shear segments in mudrock. The vertical dimensions of the calcite veins correspond to the thickness of offset competent beds with which they are contiguous, and the slip parallel dimension is proportional to fault displacement. Failure surface characteristics, including mixed tensile and shear segments, indicate hybrid failure in chalk and limestone, whereas shear failure predominates in mudrock and ash beds – these changes in failure mode contribute to variation in fault dip. Slip on the shear segments caused dilation of the steeper hybrid segments. Tabular sheets of calcite grew by repeated fault slip, dilation, and cementation. Fluid inclusion and stable isotope geochemistry analyses of fault zone cements indicate episodic reactivation at 1.4–4.2 km depths. The results of these analyses document a dramatic bed-scale lithologic control on fault zone architecture that is directly relevant to the development of porosity and permeability anisotropy along faults.

Fine-grained low permeability sedimentary rocks, such as shale and mudrock, have drawn attention as unconventional hydrocarbon reservoirs. Fracturing – both natural and induced – is extremely important for increasing permeability in otherwise low-permeability rock. We analyze natural extension fracture networks within a complete measured outcrop section of the Ernst Member of the Boquillas Formation in Big Bend National Park, west Texas. Results of bed-center, dip-parallel scanline surveys demonstrate nearly identical fracture strikes and slight variation in dip between mudrock, chalk, and limestone beds. Fracture spacing tends to increase proportional to bed thickness in limestone and chalk beds; however, dramatic differences in fracture spacing are observed in mudrock. A direct relationship is observed between fracture spacing/thickness ratio and rock competence. Vertical fracture penetrations measured from the middle of chalk and limestone beds generally extend to and often beyond bed boundaries into the vertically adjacent mudrock beds. In contrast, fractures in the mudrock beds rarely penetrate beyond the bed boundaries into the adjacent carbonate beds. Consequently, natural bed-perpendicular fracture connectivity through the mechanically layered sequence generally is poor. Fracture connectivity strongly influences permeability architecture, and fracture prediction should consider thin bed-scale control on fracture heights and the strong lithologic control on fracture spacing.

During the earliest Triassic microbial mats flourished in the photic zones of marginal seas, generating widespread microbialites. It has been suggested that anoxic conditions in shallow marine environments, linked to the end‐Permian mass extinction, limited mat‐inhibiting metazoans allowing for this microbialite expansion. The presence of a diverse suite of proxies indicating oxygenated shallow sea‐water conditions (metazoan fossils, biomarkers and redox proxies) from microbialite successions have, however, challenged the inference of anoxic conditions. Here, the distribution and faunal composition of Griesbachian microbialites from China, Iran, Turkey, Armenia, Slovenia and Hungary are investigated to determine the factors that allowed microbialite‐forming microbial mats to flourish following the end‐Permian crisis. The results presented here show that Neotethyan microbial buildups record a unique faunal association due to the presence of keratose sponges, while the Palaeotethyan buildups have a higher proportion of molluscs and the foraminifera Earlandia. The distribution of the faunal components within the microbial fabrics suggests that, except for the keratose sponges and some microconchids, most of the metazoans were transported into the microbial framework via wave currents. The presence of both microbialites and metazoan associations were limited to oxygenated settings, suggesting that a factor other than anoxia resulted in a relaxation of ecological constraints following the mass extinction event. It is inferred that the end‐Permian mass extinction event decreased the diversity and abundance of metazoans to the point of significantly reducing competition, allowing photosynthesis‐based microbial mats to flourish in shallow water settings and resulting in the formation of widespread microbialites., After the end‐Permian mass extinction, microbialites filled the ecological niche previously occupied by metazoan reefs. The factors that allowed microbialite‐forming microbial mats to flourish are, however, hotly debated. By investigating the faunal composition and depositional setting of Permian/Triassic boundary microbialites, we propose that the impact of the extinction event on the abundance of metazoans suppressed the biological controls that were previously excluding microbialite development from subtidal environments.

In this paper, we analyze small scale examples of thrust faults and related folding in outcrops of the Cretaceous Boquillas Formation within Big Bend National Park in west Texas to develop detailed understanding of the fault nucleation and propagation that may aid in the interpretation of larger thrust system structure. Thrust faults in the outcrop have maximum displacements ranging from 0.5 cm to 9 cm within competent limestone beds, and these displacements diminish both upward into anticlines and downward into synclines within the interbedded and weaker mudrock layers. We interpret the faults as having nucleated within the competent units and partially propagated into the less competent units without developing floor or roof thrusts. Faults that continued to propagate resulted in hanging wall anticlines above upwardly propagating fault tips, and footwall synclines beneath downwardly propagating fault tips. The observed structural style may provide insights in the nucleation of faults at the formation scale and the structural development at the mountain-range scale. Decollement or detachment layers may be a consequence rather than cause of thrust ramps through competent units and could be over interpreted from seismic data.

We appreciate Kershaw et al.'s comment and continued interest in our Permian–Triassic research in the Nanpanjiang Basin. Kershaw et al. comment on aspects of our recent study that contradict results from the study of Collin et al. (2009). Specifically their comments center around two questions: (1) whether the truncation surface at the base of the Permian–Triassic microbialite shows evidence for subaerial exposure, and (2) whether data support the model of genesis of the microbialites by upwelling of anoxic, alkaline waters with elevated carbonate saturation from bacterial sulfate reduction (Kershaw et al. 1999, 2007, 2012). We note that our recent study is much broader in scope than the issues addressed in this comment and reply. Our study presented extensive stratigraphic, geochemical, biostratigraphic and petrographic analyses from multiple sections, introduced new sections that have not been presented in the literature, and evaluated multiple models for the genesis of the microbialite and underlying truncation surface. We welcome this opportunity to respond to Kershaw et al.'s comment and to further illuminate studies of the Permian–Triassic boundary in the Nanpanjiang Basin. ### Two Generations of Fibrous Cement Fringe? Maybe, But There is Still no Evidence for Pendant Geometry Kershaw et al. comment that there are two generations of bladed fringing cement adhered to grains and that only the second generation has a pendant geometry. Collin et al. (2009) made the same interpretation. From the photographs in Collin et al. (2009, fig. 5Aa–5Ab) and from our thin section observations, we interpreted this to be a single generation of fibrous cement fringe adhered to grains, with a faint zonation of clear-cloudy-clear alternation within the fringe, in which the cloudy zone contained a greater concentration of fluid inclusions (fig. 1A, 1B). Note that the photos presented in the comment by Kershaw et al. (figs. 3, 4) are also consistent with this interpretation. The entire bladed cement fringe was used …

Delayed Earth system recovery following the end-Permian mass extinction is often attributed to severe ocean anoxia. However, the extent and duration of Early Triassic anoxia remains poorly constrained. Here we use paired records of uranium concentrations ([U]) and (238)U/(235)U isotopic compositions (δ(238)U) of Upper Permian-Upper Triassic marine limestones from China and Turkey to quantify variations in global seafloor redox conditions. We observe abrupt decreases in [U] and δ(238)U across the end-Permian extinction horizon, from ∼3 ppm and -0.15‰ to ∼0.3 ppm and -0.77‰, followed by a gradual return to preextinction values over the subsequent 5 million years. These trends imply a factor of 100 increase in the extent of seafloor anoxia and suggest the presence of a shallow oxygen minimum zone (OMZ) that inhibited the recovery of benthic animal diversity and marine ecosystem function. We hypothesize that in the Early Triassic oceans-characterized by prolonged shallow anoxia that may have impinged onto continental shelves-global biogeochemical cycles and marine ecosystem structure became more sensitive to variation in the position of the OMZ. Under this hypothesis, the Middle Triassic decline in bottom water anoxia, stabilization of biogeochemical cycles, and diversification of marine animals together reflect the development of a deeper and less extensive OMZ, which regulated Earth system recovery following the end-Permian catastrophe.

Carbonate rocks are important archives of past ocean conditions as well as hosts of economic resources such as hydrocarbons, water, and minerals. Geologists typically perform compositional analysis of grain, matrix, cement and pore types in order to interpret depositional environments, diagenetic modification, and reservoir quality of carbonate strata. Such information can be obtained primarily from petrographic analysis, a task that is costly, labor-intensive, and requires in-depth knowledge of carbonate petrology and micropaleontology. Recent studies have leveraged machine learning-based image analysis, including Deep Convolutional Neural Networks (DCNN), to automate description, classification and interpretation of thin sections, subsurface core images and seismic facies, which would accelerate data acquisition and reproducibility for these tasks. In carbonate rocks, this approach has been applied primarily to recognize carbonate lithofacies, and no attempt has been made to individually identify and quantify various types of carbonate grains, matrix, and cement. In this study, the applicability and performance of DCNN-based object detection and image classification approaches are assessed with respect to carbonate compositional analysis. The training data comprised of more than 13,000 individually labelled objects from nearly 4000 carbonate petrographic images. The dataset is grouped into six and nine different classes for the image classification and object detection tasks, respectively. Even with a small and relatively imbalanced training set, the DCNN was able to achieve an F1 score of 92% for image classification and 84% mean precision for object detection by combining one-cycle policy, class weight, and label mixup-smoothing methods. This study highlights the inefficiency of image classification as an approach to replicating human description and classification of carbonate petrography. By contrast, DCNN-based object detection appears capable of approaching human speed and accuracy in the area of carbonate petrography because it is able to individually locate and identify different carbonate components with greater cost-efficiency, speed, and reproducibility than conventional (human) petrographic analysis.

The morphological architecture and distribution of modern and ancient carbonate systems has been shown to follow spatial-self-organization, however, limited studies describe the morphometrics of microbial carbonates. Upper Cambrian microbial-build-ups outcropping in Central Texas, are exposed laterally (plan view), enabling a study of their morphological architecture and spatial distribution. Drone imagery was acquired to capture the outcrop features and develop a digital terrain model (cm scale resolution) for a bedding plane outcrop (600 × 200 m in size). Four scales of microbial growth (S1- few dm, S2- few m, S3- few tens of m, and S4- few hundreds of m) were identified and mapped. A series of morphometric analysis including Ripley's k, univariate, multivariate, and grouping were conducted and results demonstrate that, the scales S1, S2, and S3 display clustering and the spatial organization of microbial-buildups is naturally organized and not random. Further, as the size of the build-ups increases (from S1–S4), the anisotropy (length/width) increases, their shape becomes oblong, and they become aligned (S2–S4) with the inferred regional winds and tide-associated currents (NE-SW according to the present geography). The S1 scale does not align itself with the regional currents; instead, the build-ups behaved as a baffle during growth, and modified the currents locally, leading to preferential alignment at the edges within S2. As the scales increases in sizes (S2, S3, S4), there is competition for space, and due to regional currents, the larger scales preferentially align parallel to high-energy currents. The trends and spatial relationships identified in this study are particularly relevant and provide a scenario for sub-seismic scale heterogeneities for subsurface microbial hydrocarbon reservoirs.

Carbonate platforms grow through the precipitation, transport, and final deposition of carbonate sediment out of seawater. Quantifying the relative contributions of initial production versus subsequent transport in determining the growth rates and geometries of platforms remains a significant challenge. In this study, stratigraphic forward modeling is used to quantify the roles of sediment production, transport, and deposition during each growth stage of a Permian-Triassic carbonate platform with a complex growth history. Parameter optimization and sensitivity analysis show that, within the range of reasonable values, the morphology of the platform is most sensitive to sediment transport, moderately sensitive to maximum carbonate production rate, and least sensitive to the productivity-depth curve. The ramp-to-high relief, steep-sloped platform transition during Early Triassic time can be explained by any factor that limits sediment transport from shallow water areas of high production to the slope and basin. Reefs may play a role in limiting sediment transport on many platforms but other processes, such as early marine cementation or carbonate production along the slope, may be equally capable of yielding this shift in platform geometry. In this particular case, early lithification of ooid and skeletal shoals on the platform margin, perhaps facilitated by unusually high carbonate saturation state of seawater, may have inhibited sediment transport into the basin prior to the development of a reef on the platform margin. Later, Anisian progradation of the platform margin can be explained by the development of a slope factory rather than requiring increased sediment transport from the platform top. The development of an escarpment margin in the Ladinian is mainly influenced by accommodation in the slope profile created by antecedent Anisian topography. A general implication from the model results is that the growth of steep-sloped carbonate platforms lacking slope factories may often be limited by transport of sediment from the platform top to accommodation on the slope rather than by the intrinsic production capacity of the platform top factory.

Thrombolite and stromatolite habitats are becoming increasingly recognized as important refuges for invertebrates during Phanerozoic Oceanic Anoxic Events (OAEs); it is posited that oxygenic photosynthesis by cyanobacteria in these microbialites provided a refuge from anoxic conditions (i.e., the "microbialite refuge" hypothesis). Here, we test this hypothesis by investigating the distribution of ~34, 500 benthic invertebrate fossils found in ~100 samples from a microbialite succession that developed following the latest Permian mass extinction event on the Great Bank of Guizhou (South China), representing microbial (stromatolites and thrombolites) and non-microbial facies. The stromatolites were the least taxonomically diverse facies, and the thrombolites also recorded significantly lower diversities when compared to the non-microbial facies. Based on the distribution and ornamentation of the bioclasts within the thrombolites and stromatolites, the bioclasts are inferred to have been transported and concentrated in the non-microbial fabrics, that is, cavities around the microbial framework. Therefore, many of the identified metazoans from the post-extinction microbialites are not observed to have been living within a microbial mat. Furthermore, the lifestyle of many of the taxa identified from the microbialites was not suited for, or even amenable to, life within a benthic microbial mat. The high diversity of oxygen-dependent metazoans in the non-microbial facies on the Great Bank of Guizhou, and inferences from geochemical records, suggests that the microbialites and benthic communities developed in oxygenated environments, which disproves that the microbes were the source of the oxygenation. Instead, we posit that microbialite successions represent a taphonomic window for exceptional preservation of the biota, similar to a Konzentrat-Lagerstatte, which has allowed for diverse fossil assemblages to be preserved during intervals of poor preservation.