Time-lapse Data Research Articles

An uncertainty-aware Digital Shadow for underground multimodal CO2 storage monitoring

Summary Geological Carbon Storage (GCS) is one of the most viable climate-change mitigating net-negative CO2-emission technologies for large-scale CO2 sequestration. However, subsurface complexities and reservoir heterogeneity demand a systematic approach to uncertainty quantification to ensure both containment and conformance, as well as to optimize operations. As a step toward a Digital Twin for monitoring and control of underground storage, we introduce a new machine-learning-based data-assimilation framework validated on realistic numerical simulations. The proposed Digital Shadow combines Simulation-Based Inference (SBI) with a novel neural adaptation of a recently developed nonlinear ensemble filtering technique. To characterize the posterior distribution of CO2 plume states (saturation and pressure) conditioned on multimodal time-lapse data, consisting of imaged surface seismic and well-log data, a generic recursive scheme is employed, where neural networks are trained on simulated ensembles for the time-advanced state and observations. Once trained, the Digital Shadow infers the state as time-lapse field data become available. Unlike ensemble Kalman filtering, corrections to predicted states are computed via a learned nonlinear prior-to-posterior mapping that supports non-Gaussian statistics and nonlinear models for the dynamics and observations. Training and inference are facilitated by the combined use of conditional invertible neural networks and bespoke physics-based summary statistics. Starting with a probabilistic permeability model derived from a baseline seismic survey, the Digital Shadow is validated against unseen simulated ground-truth time-lapse data. Results show that injection-site-specific uncertainty in permeability can be incorporated into state uncertainty, and the highest reconstruction quality is achieved when conditioning on both seismic and wellbore data. Despite incomplete permeability knowledge, the Digital Shadow accurately tracks the subsurface state throughout a realistic CO2 injection project. This work establishes the first proof-of-concept for an uncertainty-aware, scalable Digital Shadow, laying the foundation for a Digital Twin to optimize underground storage operations.

Improving efficiency of estimating angler effort using low-frequency time-lapse camera data

ABSTRACT Objective Management agencies rely on estimates of angling effort to make many management decisions. Managers frequently use counts of boats, trailers, or anglers alongside traditional interview-based creel methods to inform effort estimates. Use of remote cameras for capturing effort data is increasing, but despite the advances made related to this technology, processing photos from surveys is still very time consuming and limits the implementation of these methods. Additional guidance is needed as to what level of effort is necessary to produce reliable estimates of angler effort while minimizing processing times. Methods In this study, we used images captured at low frequency with time-lapse remote cameras to quantify angling effort based on instantaneous trailer counts at six reservoirs in Arkansas. We determined “true” values of effort based on images taken hourly every day for 3 months. By simulating across a range of sampling frequencies (number of days per month and number of photos per day), we assessed the percent error from the true values. Results We determined that analyzing photos from 12 weekdays and seven weekend days per month at a frequency of one randomly selected photo per day was adequate to produce estimates of effort with less than 20% error 80% of the time (well performing). Counting photos from at least 18 weekdays and eight weekend days per month at a frequency of 2 and 4 photos/d, respectively, produced estimates of effort with less than 10% error 80% of the time (best performing). Conclusions The estimated amount of time required to analyze a month’s worth of photos for the best-performing scenario ranged from 9 min for Jack Nolan Lake to 100 min for Lake Dardanelle. Our work provides guidance to managers on how to most efficiently conduct camera surveys to estimate angler effort.

An alternative formulation for estimating seismic properties of hydrocarbon fluids with sensitivities to AVO, modeling of sonic log, and timelapse seismic

We present new equations and workflow for computing seismic properties of hydrocarbon fluids. More specifically, we provide a complete formulation to compute acoustic velocity, bulk modulus and density, with separate workflows for oil and gas. The formulation combines equations of state and machine-learning-assisted empirical correlations, which are functions of pressure, temperature, molecular weight for oil, and specific gravity for gas. An extensive set of published experimental data is used to test and validate the proposed formulation. The traditional method from Batzle and Wang (BW) is discussed in detail and used for comparisons through theoretical and experimental perspectives. Generally, while the results suggest that both models provide adequate modeling of seismic properties, several cases exist where our formulation provides a more accurate description of experimental data of acoustic velocity in gases and oils. To investigate the potential impact of our formulation on Geophysical applications, we perform an AVO modeling analysis at a hypothetical siliciclastic reservoir considering alternative oil and gas saturation scenarios. For various saturation scenarios, there are significant differences between AVO responses at higher pressures when hydrocarbon properties are computed using our formulation and BW equations. We further assess the impact of the proposed formulation on the modeling of acoustic velocity using well-log data from the Norne field, showing the impact of the different fluid formulations in routine fluid substitution steps. Finally, we present a seismic modeling exercise inspired by a Brazilian pre-salt carbonate reservoir, revealing the importance of fluid properties in time-lapse data analysis.

Colony context and size-dependent compensation mechanisms give rise to variations in nuclear growth trajectories.

Colony context and size-dependent compensation mechanisms give rise to variations in nuclear growth trajectories.

Analisis Faktor Penentu Tingkat Partisipasi Kerja Perempuan di Provinsi Jawa Timur

Women have a strategic role in economic development, but their participation in the job market is often constrained by social, economic, and cultural factors. This study aims to analyze the factors that affect the Female Labor Force Participation Rate (TPAKP) in East Java Province during the period 2014–2023. The variables studied include the minimum wage, the average length of schooling, and the Gross Regional Domestic Product (GDP). The data used is a combined panel data between time-lapse data and cross-data from 38 districts/cities in East Java. This research method uses panel data regression analysis with the Fixed Effect Model (FEM) model. The results of the study show that the minimum wage and the average length of schooling have a positive and significant influence on TPAKP. Rising wages encourage women to enter the job market, while higher education improves skills and job opportunities. In contrast, GDP has a positive but insignificant influence, as structural barriers in the economic sector limit women's participation. The study suggests increasing the minimum wage, investing in skills education and training, as well as gender-inclusive policies to increase women's participation in the job market. This research is also expected to be a reference for further studies that can expand the understanding of TPAKP determinants in other regions.

Polycystic ovary syndrome and morphokinetic embryonic development: a case-control study evaluating 791 embryos.

Polycystic ovary syndrome and morphokinetic embryonic development: a case-control study evaluating 791 embryos.

Embankment Project Monitoring Using the Time-Lapse Transient Electromagnetic Method: Numerical Simulation and Field Applications

To preserve flood control infrastructure, it is essential to quickly detect and accurately identify concealed leakage hazards within embankment projects. In this paper, we propose a novel embankment monitoring method based on the time-lapse transient electromagnetic method and complemented by a theoretical framework for analyzing time-lapse data through the lens of resistivity change rates. A time-lapse model that scrutinizes dynamic response patterns associated with leakage anomalies is constructed, while the efficacy of this methodology is verified through rigorous field experiments. Our research findings reveal a well-defined negative correlation between the resistivity variation rate and the development stage of anomalies. Our proposed method demonstrates enhanced sensitivity in the detection of dynamic evolutionary patterns in latent seepage defects, particularly in low-resistivity environments. Moreover, it successfully delineates both the spatial expansions and electrical property alterations of anomalies, providing a novel technical approach for latent seepage defect monitoring and risk management in embankments.

Impact of high temperature in 2023 and 2024 on spring leaf flush phenology in Japan derived by GCOM-C satellite

In Japan, the exceptionally high temperatures in 2023 and 2024 may have impacted vegetation phenology. We investigated the timing of the spring leaf flush (SOS) of non-evergreen plants using satellite remote sensing. We utilized data from the SGLI sensor on JAXA’s GCOM-C satellite. Compared to NASA’s MODIS sensors, which are widely used for phenology studies, GCOM-C/SGLI offers two advantages: more stable orbital control in recent years and a higher spatial resolution (250 m) in the green channel which allows to calculate the Chlorophyll Carotenoid Index (CCI), a more robust index than NDVI against snowmelt. We defined the onset of SOS as the point where CCI exceeds zero during spring, and estimated SOS over seven years (2018–2024) in Japan, validated with ground-based time-lapse camera data. On average (2018–2022), SOS tended to occur later in northern or inland mountainous areas. Compared to the average, SOS in 2023 and 2024 occurred earlier in most of northern Japan. Specifically, an earlier SOS by 3 to 7 days occurred in the southern Tohoku, Kanto, and Chubu in 2023, and across the entire Tohoku and Hokuriku in 2024. These responses of SOS to temperature anomalies align with previous studies on future phenology in Japan.

An in vivo microscopy dataset for the characterization of leukocyte death

Recent advancements in intravital microscopy have enabled the study of cell death in vivo under various experimental conditions, such as infection and cancer. However, the limited throughput of this technology, together with a lack of openly accessible datasets, affects the development of algorithms for the automatic detection and characterization of cell death, which in turn require the integration of extensive and curated datasets. To address these needs, we present a curated dataset of microscopy videos depicting the death of neutrophils, eosinophils, and dendritic cells, acquired in the spleen and in the lymph node of mice under inflammatory conditions. The dataset provides time-lapse imaging data, along with coordinates in space and time of cell death events displaying apoptotic–like morphodynamics, and 3D reconstruction of the cell morphology at each time point. Altogether, these data will be pivotal for developing computer vision and bioimage analysis methods to advance cell death research.

Enhanced cell tracking using a GAN-based super-resolution video-to-video time-lapse microscopy generative model.

Enhanced cell tracking using a GAN-based super-resolution video-to-video time-lapse microscopy generative model.

Super-resolution optical microscopy reveals accumulation of photoactivable dengue protein (Dendra2-NS2B) in the endoplasmic reticulum

The non-structural protein NS2B plays a critical role in the maturation of Dengue virus (DENV Type 2). The underlying mechanism and the role of NS2B are largely unknown due to the unavailability of its location of activity in the target organelle at single molecule level. This is largely due to the incapability of existing optical microscopes to resolve features beyond the diffraction limit of light (), which is limited to a few hundred nanometers. Existing microscopy techniques are at best useful for ensemble study and details at the single molecule level remain hidden. To enable single-molecule resolution, we investigated the role of NS2B protein in a cellular system using scanning single molecule localization microscopy (scanSMLM). Accordingly, a photoactivable plasmid (Dendra2-NS2B) containing the gene-of-interest (NS2B) was constructed and the same is used to transfect NIH3T3 cells. Both super-resolution and confocal imaging studies suggest the accumulation of NS2B proteins in endoplasmic reticulum (ER). Subsequently, single-molecule analysis is carried out, where a total of NS2B aggregates (both big and small) are noted with an area-spread of , the molecular density of , and an average of NS2B molecules per cluster. Moreover, the super-resolved volume image revealed NS2B clusters spreading across several planes with a few extending up to 5 planes ( from the coverslip). In addition, the collective dynamics of NS2B proteins leading to the formation of clusters is evident from time-lapse super-resolved data that provides conclusive evidence of NS2B accumulation, 24 hrs post-transfection. The present study revealed the dynamics of NS2B single viral protein molecule, and its accumulation at ER that may ultimately lead to organelle-specific drug targeting and help to reduce the rate of dengue infection by disrupting the NS2B accumulates.

Joint full-waveform inversion for time-lapse 4D seismic monitoring

Full-waveform inversion (FWI) can be applied to time-lapse (4D) seismic data for subsurface reservoir monitoring. Usually, baseline and monitor velocity models are built through a data-fitting process, and the difference between the models is used to estimate the time-lapse changes. However, the large-scale (or low-wavenumber) velocity structures of deep reservoirs are mainly obtained by fitting diving waves, which can be unavailable due to limited source-receiver offset apertures. Alternatively, joint FWI (JFWI) assumes a scale separation in the model space such that low wavenumbers can be recovered from diving and reflected waves, in addition to high-wavenumber impedance perturbations. We extend JFWI to time-lapse data and improve the sensitivity of waveform inversion approaches to low-wavenumber velocity changes in deep targets. A central-difference strategy (CD JFWI) is also implemented, which averages four JFWI velocity models (two baseline and two monitor models) to attenuate false changes in the 4D velocity estimate and increases the stability of the approach in data fitting. The synthetic 4D SEAM case study indicates that CD JFWI followed by 4D FWI leads to a more accurate estimate and better resolution of velocity changes than conventional 4D FWI methods in a reservoir zone that is not penetrated by diving waves.

Predicting IVF live -birth probability using time-lapse data: Implications of including or excluding age in a day 2 embryo transfer model.

The primary objective of this study was to develop predictive models for the likelihood of live births following In Vitro Fertilisation (IVF) treatment, based on a retrospective analysis of time-lapse data from Day 2 embryo transfers at Klinikk Hausken, Norway. This analysis encompassed 1,506 IVF treatment cycles, which included 865 single and 641 double embryo transfer cycles, totalling 2,147 embryos transferred. The model covariates included nucleation error, timing of two-cell stage (t2) and duration between t2 and the three-cell stage (t3). The predictive ability was assessed using Area Under Curve (AUC). Generalised Additive Mixed Models (GAMM) were utilised to address clustering effects from Single Embryo Transfers (SET) and Double Embryo Transfers (DETs), as well as the non-linear effects of female age and t2 timings. A stratification of age and model scores demonstrated the impact of incorporating age into the model. The" Base Model, not incorporating age, achieved an AUC of 0.641, while the "Age Model", using maternal age, significantly enhanced AUC to 0.745, as estimated through bootstrap analysis. However, when the Age Model was subjected to average ages across three respective age intervals, the AUC values were comparable to the Base Model, rather than the original Age Model scores. Adjusting the Intracytoplasmic Sperm Injection (ICSI) timing by ± 2 hours, purely as a theoretical exercise, has minimal impacts on model predictions. This highlights the value of including t2 despite fertilisation timing variations between ICSI and IVF. The Age Model did not show superiority in predicting live birth within single treatment cohorts. However, given its distinct AUC values for broader age ranges, the Age Model can serve as a counselling tool on live-birth probabilities. With further validation, we suggest only using the Age Model for general counselling, while the Base Model is preferable for the embryo selection decision support.

Blender tissue cartography: an intuitive tool for the analysis of dynamic 3D microscopy data.

Tissue cartography extracts and cartographically projects surfaces from volumetric biological image data. This turns 3D-into 2D data which is much easier to visualize, analyze, and computationally process. Tissue cartography has proven particularly useful in developmental biology by taking advantage of the sheet-like organization of many biological tissues. However, existing software tools for tissue cartography are limited in the type of geometries they can handle and difficult for non-experts to use and extend. Here, we describe blender_tissue_cartography ( btc ), a tissue cartography add-on for the popular 3D creation software Blender. btc makes tissue cartography user-friendly via a graphical user interface and harnesses powerful algorithms from the computer graphics community for biological image analysis. The btc GUI enables interactive analysis and visualization without requiring any programming expertise, while an accompanying Python library allows expert users to create custom analysis pipelines. Both the add-on and the Python library are highly modular and fully documented, including interactive Jupyter Notebook tutorials. btc features a general-purpose pipeline for time-lapse data in which the user graphically defines a cartographic projection for a single key frame , which is propagated to all other frames via surface-to-surface alignment algorithms. The btc differential geometry module allows mathematically correcting for cartographic distortion, enabling faithful 3D measurements in 2D cartographic projections, including for vector fields like tissue flow fields. We demonstrate btc on diverse and complex tissue shapes from Drosophila , stem-cell-based organoids, Arabidopsis , and zebrafish.

CausalXtract, a flexible pipeline to extract causal effects from live-cell time-lapse imaging data

Live-cell microscopy routinely provides massive amounts of time-lapse images of complex cellular systems under various physiological or therapeutic conditions. However, this wealth of data remains difficult to interpret in terms of causal effects. Here, we describe CausalXtract, a flexible computational pipeline that discovers causal and possibly time-lagged effects from morphodynamic features and cell–cell interactions in live-cell imaging data. CausalXtract methodology combines network-based and information-based frameworks, which is shown to discover causal effects overlooked by classical Granger and Schreiber causality approaches. We showcase the use of CausalXtract to uncover novel causal effects in a tumor-on-chip cellular ecosystem under therapeutically relevant conditions. In particular, we find that cancer-associated fibroblasts directly inhibit cancer cell apoptosis, independently from anticancer treatment. CausalXtract uncovers also multiple antagonistic effects at different time delays. Hence, CausalXtract provides a unique computational tool to interpret live-cell imaging data for a range of fundamental and translational research applications.

CausalXtract, a flexible pipeline to extract causal effects from live-cell time-lapse imaging data.

Live-cell microscopy routinely provides massive amounts of time-lapse images of complex cellular systems under various physiological or therapeutic conditions. However, this wealth of data remains difficult to interpret in terms of causal effects. Here, we describe CausalXtract, a flexible computational pipeline that discovers causal and possibly time-lagged effects from morphodynamic features and cell-cell interactions in live-cell imaging data. CausalXtract methodology combines network-based and information-based frameworks, which is shown to discover causal effects overlooked by classical Granger and Schreiber causality approaches. We showcase the use of CausalXtract to uncover novel causal effects in a tumor-on-chip cellular ecosystem under therapeutically relevant conditions. In particular, we find that cancer-associated fibroblasts directly inhibit cancer cell apoptosis, independently from anticancer treatment. CausalXtract uncovers also multiple antagonistic effects at different time delays. Hence, CausalXtract provides a unique computational tool to interpret live-cell imaging data for a range of fundamental and translational research applications.

Inferring bird communities on remote freshwater lakes through time-lapse imagery

Assessing bird diversity and associated ecological patterns in remote freshwater lakes presents challenges that require innovative approaches. Here, we evaluated the utility of time-lapse images from camera traps for this purpose using two lakes in Haida Gwaii, British Columbia, Canada. We consider four key factors: (1) manual versus automated image processing, (2) data validation through in-person observations, (3) the ability of time-lapse data to capture known ecological patterns, and (4) variation in sampling effort. We find that (1) MegaDetector, a common AI approach, is not effective at detecting birds from time-lapse images—necessitating manual screening, (2) relative bird abundances were correlated between time-lapse and in-person observer data, (3) time-lapse data capture previously documented ecological variation in space and time, and (4) sampling effort per camera trap can be, under certain scenarios, scaled down, but camera trap position and time-lapse frequency greatly influence bird detectability. Our research builds on the few previous studies that use time-lapse imagery to detect birds, and our work is the first to focus on detecting ecological patterns on freshwater lakes in remote landscapes. Camera trap technologies can shed light on avifauna in remote freshwater lakes, but additional developments are needed to maximize utility of such applications.

Physics-Guided Data-Driven Workflow Identifies Bypassed Oil and Optimizes Infill

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 218429, “Bypassed Oil Identification and Infill Optimization by a Physics-Guided Data-Driven Workflow: A North Sea Case Study,” by Babak Moradi, SPE, and Susan Behjat, Three60 Energy, and Sándor Völgyi, Wintershall Dea, et al. The paper has not been peer reviewed. _ The complete paper focuses on the vital task of identifying bypassed oil and locating the remaining oil in mature fields, emphasizing the significance of these activities in sustaining efficient oilfield exploitation. The paper introduces a novel digital hybrid workflow, guided by physics principles and driven by data, that is further enhanced by the integration of 4D time-lapse data. Subject Field X In brownfields, the presence of a considerable number of wells, multiple surveillance measurements, and extensive historical production data present unique opportunities for the application of data-driven techniques. This study used a hybrid approach, physics-informed and data-driven, to expedite the process of mapping the remaining oil, reducing the timeline from months to weeks. Field X was discovered in 2007 on the Utsira High in the Norwegian sector of the North Sea. First oil was produced from the field in November 2015. The field was deposited in a continental half-graben setting. Reservoir quality varies from moderate to very good in marine and aeolian sandstone, while the quality is poorer in alluvial sandstone and conglomerate. Oil also is proven in the underlying basement. The reservoir is at a depth of 1900 m. The field is produced by pressure support from water injection. At the time, the field consisted of 13 production wells and four injection wells. One of the wells was converted from a producer to an injector in 2018. The field has been divided into three main flow units: Asgard, Mid Sands, and Basement. The sand quality in Asgard is superior to other flow units, although the net pay is low (an average of less than 5 m), leading to a smaller oil-in-place volume. Both the northeast of Mid Sands and Basement exhibit poor sand quality. Because the current recovery factor (RF) for Basement is less than 1%, the remaining oil compliant mapping (ROCM) workflow has not been implemented for this flow unit. Almost half of the field’s production comes from the top three producers. More than 35% of the total field production originates from the six commingled wells. One of the top three producers is the largest commingled producer, perforated in Asgard and Mid Sands flow units. The kh multiplier and advanced allocation (AA) workflows are used for back-allocating historical production to the well and layer level. This advanced allocation workflow facilitates the generation of multiple realizations, incorporates surveillance data, and accommodates the variation in water-breakthrough times across different wells and layers. The AA workflow conducted a multiphase allocation by seamlessly integrating a pseudosteady-state rate formulation with fractional flow modeling. This process incorporated permeability profiles at the well level and production logs into the allocation. An attempt was made to determine the water-cut (WCT) -evolution trend for each well-flow unit using a stochastic search engine, aiming to match total phase productions at the well level. This method was applied simultaneously across all wells and flow units, yielding individual production profiles for each well.

The geometric basis of epithelial convergent extension

Shape changes of epithelia during animal development, such as convergent extension, are achieved through the concerted mechanical activity of individual cells. While much is known about the corresponding large-scale tissue flow and its genetic drivers, fundamental questions regarding local control of contractile activity on the cellular scale and its embryo-scale coordination remain open. To address these questions, we develop a quantitative, model-based analysis framework to relate cell geometry to local tension in recently obtained time-lapse imaging data of gastrulating Drosophila embryos. This analysis systematically decomposes cell shape changes and T1 rearrangements into internally driven, active, and externally driven, passive, contributions. Our analysis provides evidence that germ band extension is driven by active T1 processes that self-organize through positive feedback acting on tensions. More generally, our findings suggest that epithelial convergent extension results from the controlled transformation of internal force balance geometry which combines the effects of bottom-up local self-organization with the top-down, embryo-scale regulation by gene expression.

The geometric basis of epithelial convergent extension.

Shape changes of epithelia during animal development, such as convergent extension, are achieved through the concerted mechanical activity of individual cells. While much is known about the corresponding large-scale tissue flow and its genetic drivers, fundamental questions regarding local control of contractile activity on the cellular scale and its embryo-scale coordination remain open. To address these questions, we develop a quantitative, model-based analysis framework to relate cell geometry to local tension in recently obtained time-lapse imaging data of gastrulating Drosophila embryos. This analysis systematically decomposes cell shape changes and T1 rearrangements into internally driven, active, and externally driven, passive, contributions. Our analysis provides evidence that germ band extension is driven by active T1 processes that self-organize through positive feedback acting on tensions. More generally, our findings suggest that epithelial convergent extension results from the controlled transformation of internal force balance geometry which combines the effects of bottom-up local self-organization with the top-down, embryo-scale regulation by gene expression.