Deep Time Research Articles

Memories of the deep past: The importance of the soil memory concept for paleopedology studies

Memories of the deep past: The importance of the soil memory concept for paleopedology studies

Unravelling the Effects of Ecology and Evolutionary History in the Phenotypic Convergence of Fishes.

Understanding the ecological drivers and limitations of adaptive convergence is a fundamental challenge. Here, we explore how adaptive convergence of planktivorous fishes has been influenced by multiple ecological factors, evolutionary history, and chance. Using ecomorphological data for over 1600 marine species, we integrate pattern-based metrics of convergence with evolutionary model fitting to test whether phenotypic similarities among specialist planktivores exceed expectations under null models and whether ecology, evolutionary history, or their combined effects best explain trait evolution. We find that planktivores are significantly more similar in phenotype than expected. Traits with functional relevance for prey detection and capture, such as eye diameter and lower jaw length, are strongly convergent, while general body size and shape are constrained by deep divisions between clades where the effects of evolutionary history are most pronounced. Since not all traits undergo strong selection toward a convergent ecomorph, their evolutionary trajectories have not entirely overcome ancestral differences in the multivariate trait space, resulting in a specific form of convergence termed conservatism. We show how adaptive responses to feeding ecology intertwine with other ecological pressures (i.e., light environment) and historical contingency to shape fish phenotype evolution over deep time, offering key insights into the generality of phenotypic evolution.

Recursive Worlds

In Recursive Worlds, Sara Walker redefines life as a physical process that structures matter through recursive information across deep time. Challenging conventional physics, she introduces assembly theory to describe how complex objects—including molecules, technologies, and living systems—emerge only through informational lineages. Rather than viewing life as spontaneous or reducible to chemistry, Walker argues that living systems operate beyond a measurable threshold of complexity, constructing new possibility spaces in a combinatorially vast universe. Earth, in this framework, is the largest known object in time, hosting a biosphere and technosphere that are recursively stacked and increasingly capable of perceiving their own origins. As the technosphere grows, it becomes not merely a byproduct of biology but a continuation of planetary evolution, possibly leading to biospheric reproduction. The talk invites a shift in physics toward a model where information, history, and recursive causation are central to understanding life, intelligence, and the future of planetary systems.

Planetary Sapience & Planetary Stupidity: a Bildungsroman for the Baby Noösphere

What might “planetary intelligence” entail? One aspect of intelligence, at least of our sapient sort, is the capacity to be aware of—and thus to be able to author—our own biographies. We are biographical beings. We learn from our pasts, in order to plan for our futures. In some ways, we can view our species as bequeathing such an aptitude to our planet itself, as it ploughs through deep time and twirls through deep space. Through us, a planet becomes autobiographical: capable of gauging its deepest past and tracking the unfolding complexities of the present, so as to select between possibilities and prospects for tomorrow. Through our globe-girdling array of sensors and simulations, our planet might just be becoming responsible for itself. That is, capable of authoring its own future and avoiding catastrophe. But, sapience always presupposes stupidity: the ability to learn from one’s errors entails having made them. No one is born wise, after all. It is a status earned through demonstrating willingness to correct grave—sometimes catastrophic—mistakes. Accordingly, assuming that we might be living during the prehistory of a planetary sapience, this talk explores the necessary episodes of planetary stupidity requisite for this. It presents a history of grand, often shocking, follies: of proposed geoengineering projects and other ambitious schemes, all of which would have been catastrophic had they been fully implemented. But, ultimately, the message is hopeful: stupidity and intelligence are sides of a coin, and inspecting the errors of the past throws into relief just how far we’ve already come in our quest to coalesce forms of sapience at scales extending far beyond our skulls.

Chronoseed: A Generative AI Time Capsule

This paper assembles a design space and defines the parameters for a generative time capsule—one that preserves not only an archive of human knowledge but also the cognitive function of human thought within a neural network. Such a time capsule must persist across vast timescales, including scenarios of civilizational collapse, while providing mechanisms for decoding its contents and executing its program, even in radically altered technological and epistemic contexts. We introduce Chronoseed as a speculative thought experiment, exploring the possibility of an AI time capsule that preserves both stored knowledge and the ability to process it. To propose its form and evaluate its feasibility, we construct a three-dimensional design space—durability, accessibility, and completeness—as a framework for systematically examining how synthetic intelligence might reshape cultural preservation. By engaging with historical time capsules, long-term nuclear waste warning systems, and speculative contributions from science fiction writers, the paper considers the broader challenge of transmitting knowledge and logics of thinking across deep time. It positions Chronoseed’s generative function as an alternative to static repositories, proposing a model in which knowledge remains interactive, adaptive, and capable of being reconstructed by future intelligences. Using the assembled design space and its parameters, the article explores possible physical forms of Chronoseed, balancing preservation with the ability of future discoverers to activate its generative potential. It ultimately identifies DNA embedding as the most viable option for long-term durability and recoverability.

Antikythera

Computation was discovered as much as invented and the implications for how we understand artificial computation are profound. Antikythera is the name of a new philosophy of computation that links an evolutionary perspective on technology, and the emergence of planetary computation in particular, with a generative program for how to orient it as it parents us. The core ideas connect synthetic intelligence—the ongoing artificialization of intelligence from the first symbolic notation to frontier models; recursive simulations—the active dynamic between functional models and the real world; hemispherical stacks—the unchartered geopolitics of computational infrastructures; and planetary sapience—the slow, technologically-mediated process by which planetary systems, from deep time to deep contingency, are made comprehensible. The research seeks to establish a school of thought that links humanities, sciences, and engineering in new catalytic relationships.

Recurrence rates of explosive volcanism in paleo-equatorial Pangaea, and implications for climate near the peak late Paleozoic ice age

The continental Brousse-Broquiès basin (Massif Central, France) preserves a well-exposed section of upper Carboniferous pyroclastic-volcaniclastic lacustrine strata. We present high-precision geochronology and detailed sedimentology that together define a Kasimovian (305.77–306.10 Ma) volcanic recurrence interval of < 10 ky. This is the first quantification of volcanic recurrence from deep (pre-Mesozoic) time, and importantly, represents an unique opportunity to gain insight into the nature of frequent and highly explosive silicic volcanism occurring across eastern equatorial Pangaea during the peak of the late Paleozoic ice age (LPIA). These data emanate from a single, unusually well-exposed and well-constrained site. However, this site is representative of many documented but poorly preserved volcanic centers of coeval or similar ages across west-central Europe. The concentration in both time and space of such highly explosive volcanism in paleoequatorial Pangaea has large implications for the climate system during Earth’s penultimate icehouse. The recurrence rates established here support ongoing climate modeling efforts to test the hypothesis that this volcanism imposed a significant cooling effect near and during peak LPIA conditions prior to Permian icehouse collapse. In the wake of enhanced precision in geochronological tools, the integrated approach we apply for quantifying volcanic recurrence rates has broad utility in Earth’s deep-time record.

Application of data twinning based on deep time series model in smart city traffic flow prediction

This paper introduces an intelligent traffic flow prediction system that combines data twinning and deep learning, aiming to improve the prediction accuracy and model adaptability by integrating grey prediction model (GM(1,1)), long-short-term memory network (LSTM) and particle swarm optimization (PSO). The system construction starts from physical layer data acquisition, deals with missing data through smoothing and interpolation, and applies the GM(1,1) model to construct a digital twin layer for preliminary prediction. The information fusion mechanism combines real-time data and forecast data to optimize the model inputs, and the PSO algorithm is used to optimize the model parameters. The LSTM model is used as a deep time-series model, and combined with the output of the data twin prediction to make a comprehensive prediction, and at the same time, feature engineering is utilized to enhance the model performance. The system implements dynamic feedback adjustment to optimize the model in real time according to the prediction error. Empirical analysis shows that the LSTM model with integrated data twin (LSTM + DT) outperforms the traditional model in prediction accuracy, stability, and generalization ability, especially in peak hours, holidays, and emergency response, and provides a reliable solution for intelligent traffic management despite the long training time.

Technological integration in predicting hypoxemia risk for improved surgical outcomes in Type A aortic dissection.

BackgroundPostoperative hypoxemia is a severe complication in patients undergoing surgery for acute Type A aortic dissection (AAD), with significant impacts on recovery and clinical outcomes. Technological advancements in risk assessment models offer opportunities for early intervention and optimized care.ObjectiveTo develop and validate a technology-driven predictive model for hypoxemia based on clinical and intraoperative risk factors, enhancing postoperative management strategies.MethodsA retrospective cohort of 242 patients was analyzed, including 77 with hypoxemia (PaO2/FiO2 ≤ 200 mmHg) and 165 without. Key clinical variables, intraoperative factors, and postoperative outcomes were examined. Spearman correlation analysis and receiver operating characteristic (ROC) curve analysis were conducted to identify and validate predictive markers.ResultsProlonged time from symptom onset to surgery (>48 h), aortic cross-clamp time, and deep hypothermic circulatory arrest time (DHCA) emerged as the most significant predictors (all p < 0.001). DHCA time demonstrated the highest sensitivity (0.961) and area under the curve (AUC = 0.891). Additional significant predictors included intraoperative blood product use and prolonged mechanical ventilation, with cumulative predictive value for hypoxemia risk.ConclusionThe integration of clinical variables into a technology-enhanced prediction model provides robust early warnings of postoperative hypoxemia risk. Implementing timely surgical interventions and refined intraoperative management can minimize adverse respiratory outcomes, improving recovery in AAD patients.

Shifting, expanding, or contracting? Range movement consequences for biodiversity.

Shifting, expanding, or contracting? Range movement consequences for biodiversity.

Machine Learning and Big Data Mining Reveal Earth's Deep Time Crustal Thickness and Tectonic Evolution: A New Chemical Mohometry Approach

AbstractQuantitative analysis of crustal thickness evolution across deep time poses critical insights into the planet's geological history. It may help uncover new areas with potential critical mineral deposits and reveal the impacts of crustal thickness and elevation changes on the development of the atmosphere, hydrosphere, and biosphere. However, most existing estimation methods are restricted to arc‐related magmas, limiting their broader application. By mining extensive geochemical data from present‐day subduction zones, collision orogenic belts, and non‐subduction‐related intraplate igneous rock samples worldwide, along with their corresponding Moho depths during magmatism, we have developed a machine learning‐based mohometry linking geochemical data to Moho depth, which is universally applicable in reconstructing ancient orogenic systems' paleo‐crustal evolution and tracking complex tectonic histories in both spatial and temporal dimensions. Our novel mohometry model demonstrates robust performance, achieving an R2 of 0.937 and an Root Mean Squared Error of 4.3 km. Feature importance filtering highlights key geochemical proxies, allowing for accurate paleo‐crustal thickness estimation even when many elements are missing. Model validation in southern Tibet and the South China Block, regions characterized by well‐constrained crustal histories and complex tectonic processes, demonstrates its broad applicability. Reconstructed paleo‐crustal thickness records reveal a strong correlation between crustal thickening events and the formation of porphyry ore deposits, offering new insights for mineral exploration in ancient orogens subjected to significant surface erosion. By enabling the reconstruction of crustal thickness across geological timescales, this model enhances our understanding of Earth's internal dynamics and their interactions with surface processes, thereby advancing our comprehension of Earth's geological history.

A hell ant from the Lower Cretaceous of Brazil.

A hell ant from the Lower Cretaceous of Brazil.

Growth disturbances in bivalve shell: Implications for past and future intra-annual scale climate change.

Growth disturbances in bivalve shell: Implications for past and future intra-annual scale climate change.

“A Tough Bitch”: Lynn Margulis and the Gaian Sublime

This essay challenges Bruno Latour’s elegiac pronouncement of the sublime’s death in the Anthropocene, proposing instead a “Gaian sublime” emerging from Lynn Margulis’s radical reconceptualization of planetary life. Analysis of Margulis’s scientific nonfiction reveals how her work on microbial agency and symbiosis disrupts traditional sublime theory’s emphasis on human transcendence and geological spectacle. The essay traces how sublime aesthetics, from Longinus through Burke and Kant to contemporary environmental thought, has historically reinforced racial hierarchies, gender binaries, and human exceptionalism. Margulis’s perspective offers a crucial corrective by revealing Earth’s smallest inhabitants as its most profound transformers, generating sublime experience not through nature’s brute force but through recognition of life’s collaborative creativity across scales and through deep time. This reframing moves beyond both conventional sublime theory and contemporary Anthropocene discourse, demonstrating how scientific understanding might enhance rather than diminish the capacity for awe and wonder. The Gaian sublime thus emerges as both aesthetic category and mode of attention, potentially enabling more ethically attuned relationships with our living planet.

고속도로 교통량 결정요인 탐색 및 딥러닝 모형을 통한 예측

This study aims to explore the main determinants affecting traffic volume on Korean expressways and implement a traffic forecasting model based on them. Through Exploratory Data Analysis of traffic volume data provided by Korea Expressway Corporation, we identified the correlation of major variables and analysed environmental and social factors affecting traffic volume. Based on this, we built deep learning time series prediction models that have advantages in predicting dependent variables with nonlinearity, and compared the performance of each model. The results of the study showed that factors such as precipitation and snowfall in a specific region, month, day of the week, national holidays, and public holidays have a significant impact on highway traffic volume, and Block-rNN, N-HiTS of the deep learning-based models were found to be superior in terms of prediction performance. This study is expected to contribute to the development of efficient policies for highway traffic management and traffic information prediction models.

Sleep architecture and quality and pain experience in individuals with persistent low back pain and asymptomatic controls.

Both pain and sleep have a broad impact on health and well-being. There is a bi-directional association between pain and sleep, but how sleep and pain are associated chronically remains unclear. This study examined the associations of sleep architecture and quality in relation to pain experience and its impact in individuals with persistent low back pain (pLBP) and asymptomatic controls. Participants included individuals in a current episode of low back pain with symptoms impacting function and persisting greater than three months (pLBP group, n = 20) and asymptomatic controls (control group, n = 19). A home sleep test device (Zmachine® Insight + , General Sleep Corporation) was used for three nights to ecologically assess sleep architecture and quality. Pain, psychosocial factors, and lumbar movement control were evaluated using standard testing. The participants were 25.0 ± 4.8 yrs (mean±SD) with similar height, weight, and waist and hip circumferences in the pLBP and control groups. Deep sleep time was shorter (p = 0.034) for the pLBP (1.4 ± 0.4 hr) compared to the control group (1.7 ± 0.3 hr). For the pLBP group, deep sleep time and some sleep quality measures were associated with several pain-related anxiety, daily function, and impact measures independent of total sleep time. These results indicate associations of sleep architecture and quality with pain experience in individuals with pLBP. Of all the sleep stages, the deep sleep stage may be more impacted by chronic pain than other stages.

Phylogenomic Inference Suggests Differential Deep Time Phylogenetic Signals from Nuclear and Organellar Genomes in Gymnosperms.

The living gymnosperms include about 1200 species in five major groups: cycads, ginkgo, gnetophytes, Pinaceae (conifers I), and cupressophytes (conifers II). Molecular phylogenetic studies have yet to reach a unanimously agreed-upon relationship among them. Moreover, cytonuclear phylogenetic incongruence has been repeatedly observed in gymnosperms. We collated a comprehensive dataset from available genomes of 17 gymnosperms across the five major groups and added our own high-quality assembly of a species from Podocarpaceae (the second largest conifer family) to increase sampling width. We used these data to infer reconciled nuclear species phylogenies using two separate methods to ensure the robustness of our conclusions. We also reconstructed organelle phylogenomic trees from 42 mitochondrial and 82 plastid genes from 38 and 289 gymnosperm species across the five major groups, respectively. Our nuclear phylogeny consistently recovers the Ginkgo-cycads clade as the first lineage split from other gymnosperm clades and the Pinaceae as sister to gnetophytes (the Gnepines hypothesis). In contrast, the mitochondrial tree places cycads as the earliest lineage in gymnosperms and gnetophytes as sister to cupressophytes (the Gnecup hypothesis) while the plastomic tree supports the Ginkgo-cycads clade and gnetophytes as the sister to cupressophytes. We also examined the effect of mitochondrial RNA editing sites on the gymnosperm phylogeny by manipulating the nucleotide and amino acid sequences at these sites. Only complete removal of editing sites has an effect on phylogenetic inference, leading to a closer congruence between mitogenomic and nuclear phylogenies. This suggests that RNA editing sites carry a phylogenetic signal with distinct evolutionary traits.

Deep-time gene expression shift reveals an ancient change in avian muscle phenotypes.

Gene duplication is an important process of molecular evolutionary change, though identifying these events and their functional implications remains challenging. Studies on gene duplication more often focus on the presence of paralogous genes within the genomes and less frequently explore shifts in expression. We investigated the evolutionary history of calsequestrin (CASQ), a crucial calcium-binding protein in the junctional sarcoplasmic reticulum of muscle tissues. CASQ exists in jawed vertebrates as subfunctionalized paralogs CASQ1 and CASQ2 expressed primarily in skeletal and cardiac muscles, respectively. We used an enhanced sequence dataset to support initial duplication of CASQl in a jawed fish ancestor prior to the divergence of cartilaginous fishes. Surprisingly, we find CASQ2 is the predominant skeletal muscle paralog in birds, while CASQ1 is either absent or effectively nonfunctional. Changes in the amino acid composition and electronegativity of avian CASQ2 suggest enhancement to calcium-binding properties that preceded the loss of CASQ1. We identify this phenomenon as CASQ2 "synfunctionalization," where one paralog functionally replaces another. While additional studies are needed to fully understand the dynamics of CASQ1 and CASQ2 in bird muscles, the long and consistent history of CASQ subfunctions outside of birds indicate a substantial evolutionary pressure on calcium-cycling processes in muscle tissues, likely connected to increased avian cardiovascular and metabolic demands. Our study provides an important insight into the molecular evolution of birds and shows how gene expression patterns can be comparatively studied across phylum-scale deep time to reveal key evolutionary events.

A geological timescale for bacterial evolution and oxygen adaptation.

Microbial life has dominated Earth's history but left a sparse fossil record, greatly hindering our understanding of evolution in deep time. However, bacterial metabolism has left signatures in the geochemical record, most conspicuously the Great Oxidation Event (GOE). We combine machine learning and phylogenetic reconciliation to infer ancestral bacterial transitions to aerobic lifestyles, linking them to the GOE to calibrate the bacterial time tree. Extant bacterial phyla trace their diversity to the Archaean and Proterozoic, and bacterial families prior to the Phanerozoic. We infer that most bacterial phyla were ancestrally anaerobic and adopted aerobic lifestyles after the GOE. However, in the cyanobacterial ancestor, aerobic metabolism likely predated the GOE, which may have facilitated the evolution of oxygenic photosynthesis.

Reconstructing the phylogeny and evolutionary history of freshwater fishes (Nemacheilidae) across Eurasia since early Eocene

Eurasia has undergone substantial tectonic, geological, and climatic changes throughout the Cenozoic, primarily associated with tectonic plate collisions and a global cooling trend. The evolution of present-day biodiversity unfolded in this dynamic environment, characterised by intricate interactions of abiotic factors. However, comprehensive, large-scale reconstructions illustrating the extent of these influences are lacking. We reconstructed the evolutionary history of the freshwater fish family Nemacheilidae across Eurasia and spanning most of the Cenozoic on the base of 471 specimens representing 279 species and 37 genera plus outgroup samples. Molecular phylogeny using six genes uncovered six major clades within the family, along with numerous unresolved taxonomic issues. Dating of cladogenetic events and ancestral range estimation traced the origin of Nemacheilidae to Indochina around 48 mya. Subsequently, one branch of Nemacheilidae colonised eastern, central, and northern Asia, as well as Europe, while another branch expanded into the Burmese region, the Indian subcontinent, the Near East, and northeast Africa. These expansions were facilitated by tectonic connections, favourable climatic conditions, and orogenic processes. Conversely, aridification emerged as the primary cause of extinction events. Our study marks the first comprehensive reconstruction of the evolution of Eurasian freshwater biodiversity on a continental scale and across deep geological time.