Without rigorous attention to the completeness of earthquake catalogs, claims of new discoveries or forecasting skills cannot be deemed credible. Therefore, estimating the completeness magnitude (Mc) is a critical step. Among various approaches, catalog-based methods are the simplest, most straightforward, and most commonly used. However, current evaluation frameworks for these methods lack a unified simulation strategy for generating catalogs that are independent of specific Mc estimation methods. An effective strategy should also be capable of simulating datasets with non-uniform Mc distributions across both spatial and temporal dimensions. In this study, we assess nine catalog-based methods under a robust evaluation framework specifically tailored for this purpose. These methods are tested on datasets with homogeneous and heterogeneous Mc distributions, as well as on real-world earthquake catalogs. The method of b-value stability by Woessner and Wiemer (2005), referred to as MBS-WW in this study, demonstrates the best overall performance. The prior model generated by MBS-WW is used as the foundation for generating an updated Mc map for China with the Bayesian Magnitude of Completeness (BMC) method. We also introduce, BSReLU, an augmented Gutenberg-Richter model with a novel probabilistic framework for modeling. The BSReLU model replaces deterministic estimates of Mc with a probabilistic framework that models the smooth transition in detection likelihood from zero to one as earthquake magnitudes increase. By evaluating the limitations of these foundational catalog-based methods, this study seeks to refine our understanding of their appropriate applications, offering a clearer, unbiased perspective on seismicity through improved observational data quality.

The transport of sea ice over the polar oceans plays an important role in climate. This transport is driven predominantly by turbulent winds, leading to stochastic motion of ice floes. Observed diffusivities and velocity distributions of sea ice deviate by orders of magnitude from Brownian models, making it challenging to predict ice transport. We fully resolve these gaps through stochastic granular simulations that account for interactions between ice floes as they respond to a noisy wind. Using only directly measured quantities as model inputs, we reproduce the dispersion, diffusivity, velocity distribution, and power spectra of sea ice observed in the Fram Strait with remarkable quantitative accuracy. We understand these features as direct consequences of collisions between floes, which rapidly dissipate the energy injected by wind. A kinetic theory provides insights into these dynamics in terms of environmental properties, and makes predictions in close agreement with observations. The ideas and tools developed here pave the way for a new predictive understanding of global sea ice transport in terms of local floe-scale processes.

This review explores the integration of deep learning (DL) with full-waveform inversion (FWI) for enhanced seismic imaging and subsurface characterization. It covers FWI and DL fundamentals, geophysical applications (velocity estimation, deconvolution, tomography), and challenges (model complexity, data quality). The review also outlines future research directions, including hybrid, generative, and physics-informed models for improved accuracy, efficiency, and reliability in subsurface property estimation. The synergy between DL and FWI has the potential to transform geophysics, providing new insights into Earth's subsurface.

The Atlantic Meridional Overturning Circulation (AMOC) and polar ice sheets are coupled tipping elements, allowing for potential cascading tipping events in which tipping is facilitated by their mutual interactions. However, while an AMOC destabilization driven by Greenland Ice Sheet (GIS) meltwater release is well documented, the consequences of a West Antarctic Ice Sheet (WAIS) tipping on the AMOC remain unclear. In the Earth System Model of Intermediate Complexity CLIMBER-X, we perform experiments where meltwater fluxes representing plausible tipping trajectories of the GIS and WAIS are applied. We find that WAIS meltwater input can increase the AMOC resilience to GIS meltwater. In particular, we show that this stabilizing effect can cause the AMOC recovery and, for the first time in a comprehensive model, totally prevent an AMOC collapse. Moreover, we find this stabilzation to occur for ice sheet tipping trajectories that are relevant under high future greenhouse gas emission scenarios.

Context. As the Solar System orbits the Milky Way, it encounters various Galactic environments, including dense regions of the interstellar medium (ISM). These encounters can compress the heliosphere, exposing parts of the Solar System to the ISM, while also increasing the influx of interstellar dust into the Solar System and Earth's atmosphere. The discovery of new Galactic structures, such as the Radcliffe wave, raises the question of whether the Sun has encountered any of them. Aims. The present study investigates the potential passage of the Solar System through the Radcliffe wave gas structure over the past 30 million years (Myr). Methods. We used a sample of 56 high-quality, young ($\leq$ 30 Myr) open clusters associated with a region of interest of the Radcliffe wave to trace its motion back and investigate a potential crossing with the Solar System's past orbit. Results. We find that the Solar System's trajectory intersected the Radcliffe wave in the Orion region. We have constrained the timing of this event to between 18.2 and 11.5 Myr ago, with the closest approach occurring between 14.8 and 12.4 Myr ago. Notably, this period coincides with the Middle Miocene climate transition on Earth, providing an interdisciplinary link with paleoclimatology. The potential impact of the crossing of the Radcliffe wave on the climate on Earth is estimated. This crossing could also lead to anomalies in radionuclide abundances, which is an important research topic in the field of geology and nuclear astrophysics.

Weed management remains a critical challenge in agriculture, where weeds compete with crops for essential resources, leading to significant yield losses. Accurate detection of weeds at various growth stages is crucial for effective management yet challenging for farmers, as it requires identifying different species at multiple growth phases. This research addresses these challenges by utilizing advanced object detection models, specifically, the Detection Transformer (DETR) with a ResNet50 backbone and RetinaNet with a ResNeXt101 backbone, to identify and classify 16 weed species of economic concern across 174 classes, spanning their 11 weeks growth stages from seedling to maturity. A robust dataset comprising 203,567 images was developed, meticulously labeled by species and growth stage. The models were rigorously trained and evaluated, with RetinaNet demonstrating superior performance, achieving a mean Average Precision (mAP) of 0.907 on the training set and 0.904 on the test set, compared to DETR's mAP of 0.854 and 0.840, respectively. RetinaNet also outperformed DETR in recall and inference speed of 7.28 FPS, making it more suitable for real time applications. Both models showed improved accuracy as plants matured. This research provides crucial insights for developing precise, sustainable, and automated weed management strategies, paving the way for real time species specific detection systems and advancing AI-assisted agriculture through continued innovation in model development and early detection accuracy.

Two bolides (2 June 2016 and 4 April 2019) were detected at multiple regional infrasound stations with many of the locations receiving multiple detections. Analysis of the received signals was used to estimate the yield, location and trajectory, and the type of shock that produced the received signal. The results from the infrasound analysis were compared with ground truth information that was collected through other sensing modalities. This multi-modal framework offers an expanded perspective on the processes governing bolide shock generation and propagation. The majority of signal features showed reasonable agreement between the infrasound-based interpretation and the other observational modalities, though the yield estimate from the 2019 bolide was significantly lower using the infrasound detections. There was also evidence suggesting that one of the detections was from a cylindrical shock that was initially propagating upward, which is unusual though not impossible.

Weeds are one of the major reasons for crop yield loss but current weeding practices fail to manage weeds in an efficient and targeted manner. Effective weed management is especially important for crops with high worldwide production such as maize, to maximize crop yield for meeting increasing global demands. Advances in near-sensing and computer vision enable the development of new tools for weed management. Specifically, state-of-the-art segmentation models, coupled with novel sensing technologies, can facilitate timely and accurate weeding and monitoring systems. However, learning-based approaches require annotated data and show a lack of generalization to aerial imaging for different crops. We present a novel dataset for semantic and instance segmentation of crops and weeds in agricultural maize fields. The multispectral UAV-based dataset contains images with RGB, red-edge, and near-infrared bands, a large number of plant instances, dense annotations for maize and four weed classes, and is multitemporal. We provide extensive baseline results for both tasks, including probabilistic methods to quantify prediction uncertainty, improve model calibration, and demonstrate the approach's applicability to out-of-distribution data. The results show the effectiveness of the two additional bands compared to RGB only, and better performance in our target domain than models trained on existing datasets. We hope our dataset advances research on methods and operational systems for fine-grained weed identification, enhancing the robustness and applicability of UAV-based weed management. The dataset and code are available at https://github.com/GFZ/weedsgalore

Meta-instability is an irreversible precursor of earthquakes. To identify the meta-instability precursor of the Yangbi $M_S$ 6.4 earthquake ($99.87^{\circ}\mathrm{E}$, $25.67^{\circ}\mathrm{N}$)that occurred on May 21, 2021, we selected seismic data from the pre-earthquake period between 1 and 21 May. We then calculated the apparent wave velocity ratio and the apparent Poisson\text{'}s ratio within the region of $98.5^{\circ} \mathrm{E}-101^{\circ}\mathrm{E}$, $24.6^{\circ}\mathrm{N}-27.1^{\circ}\mathrm{N}$ and interpolated these values. Our findings revealed that the trends of the fitted straight lines at the maximum and minimum points of the gradient divergences of the apparent wave velocity ratio and apparent Poisson\text{'}s ratio fields are consistent with the source mechanism solution for Sections 1 and 2, respectively. Similarly, the trend of the fitted straight lines at the minimum and maximum points of their values is also consistent with the source mechanism solution for Sections 1 and 2. Positive gradient divergence values indicate energy released, whereas negative values suggest energy absorption. The observed stress state matches the experimentally demonstrated meta-instable state. We propose that this method can be a reference for identifying the meta-instability of strike-slip strong earthquakes with a significant number of foreshocks. For seismically active regions, increasing the number of stations with rich data acquisition will facilitate more convenient stress field analysis.

Large language models (LLMs) have shown significant promise in question-answering (QA) tasks, particularly in retrieval-augmented generation (RAG) scenarios and long-context applications. However, their performance is hindered by noisy reference documents, which often distract from essential information. Despite fine-tuning efforts, Transformer-based architectures struggle to prioritize relevant content. This is evidenced by their tendency to allocate disproportionate attention to irrelevant or later-positioned documents. Recent work proposes the differential attention mechanism to address this issue, but this mechanism is limited by an unsuitable common-mode rejection ratio (CMRR) and high computational costs. Inspired by the operational amplifier (OpAmp), we propose the OpAmp adaptation to address these challenges, which is implemented with adapters efficiently. By integrating the adapter into pre-trained Transformer blocks, our approach enhances focus on the golden context without costly training from scratch. Empirical evaluations on noisy-context benchmarks reveal that our Qwen2.5-OpAmp-72B model, trained with our OpAmp adaptation, surpasses the performance of state-of-the-art LLMs, including DeepSeek-V3 and GPT-4o.

We develop a theory of various kinds of large-scale clustering of inertial particles in a rotating density stratified or inhomogeneous turbulent fluid flows. The large-scale particle clustering occurs in scales which are much larger than the integral scale of turbulence, and it is described in terms of the effective pumping velocity in a turbulent flux of particles. We show that for a fast rotating strongly anisotropic turbulence, the large-scale clustering occurs in the plane perpendicular to rotation axis in the direction of the fluid density stratification. We apply the theory of the large-scale particle clustering for explanation of the formation of planetesimals (progenitors of planets) in accretion protoplanetary discs. We determine the radial profiles of the radial and azimuthal components of the effective pumping velocity of particles which have two maxima corresponding to different regimes of the particle--fluid interactions: at the small radius it is the Stokes regime, while at the larger radius it is the Epstein regime. With the decrease the particle radius, the distance between the maxima increases. This implies that smaller-size particles are concentrated nearby the central body of the accretion disk, while larger-size particles are accumulated far from the central body. The dynamic time of the particle clustering is about $\tau_{\rm dyn} \sim 10^5$--$10^6$ years, while the turbulent diffusion time is about $10^7$ years, that is much larger than the characteristic formation time of large-scale particle clusters ($\sim \tau_{\rm dyn}$).

Full-waveform inversion (FWI) is a method that utilizes seismic data to invert the physical parameters of subsurface media by minimizing the difference between simulated and observed waveforms. Due to its ill-posed nature, FWI is susceptible to getting trapped in local minima. Consequently, various research efforts have attempted to combine neural networks with FWI to stabilize the inversion process. This study presents a simple yet effective training framework that is independent of dataset reliance and requires only moderate pre-training on a simple initial model to stabilize network outputs. During the transfer learning phase, the conventional FWI gradients will simultaneously update both the neural network and the proposed adaptive residual learning module, which learns the residual mapping of large-scale distribution features in the network's output, rather than directly fitting the target mapping. Through this synergistic training paradigm, the proposed algorithm effectively infers the physically-informed prior knowledge into a global representation of stratigraphic distribution, as well as capturing subtle variations in inter-layer velocities within local details, thereby escaping local optima. Evaluating the method on two benchmark models under various conditions, including absent low-frequency data, noise interference, and differing initial models, along with corresponding ablation experiments, consistently demonstrates the superiority of the proposed approach.

The San Andreas Fault system, known for its frequent seismic activity, provides an extensive dataset for earthquake studies. The region's well-instrumented seismic networks have been crucial in advancing research on earthquake statistics, physics, and subsurface Earth structures. In recent years, earthquake data from California has become increasingly valuable for deep learning applications, such as Generalized Phase Detection (GPD) for phase detection and polarity determination, and PhaseNet for phase arrival-time picking. The continuous accumulation of data, particularly those manually labeled by human analysts, serves as an essential resource for advancing both regional and global deep learning models. To support the continued development of machine learning and data mining studies, we have compiled a unified California Earthquake Event Dataset (CEED) that integrates seismic records from the Northern California Earthquake Data Center (NCEDC) and the Southern California Earthquake Data Center (SCEDC). The dataset includes both automatically and manually determined parameters such as earthquake origin time, source location, P/S phase arrivals, first-motion polarities, and ground motion intensity measurements. The dataset is organized in an event-based format organized by year spanning from 2000 to 2024, facilitating cross-referencing with event catalogs and enabling continuous updates in future years. This comprehensive open-access dataset is designed to support diverse applications including developing deep learning models, creating enhanced catalog products, and research into earthquake processes, fault zone structures, and seismic risks.

The nonlinear mechanical responses of rocks and soils to seismic waves play an important role in earthquake physics, influencing ground motion from source to site. Continuous geophysical monitoring, such as ambient noise interferometry, has revealed co-seismic wave speed reductions extending tens of kilometers from earthquake sources. However, the mechanisms governing these changes remain challenging to model, especially at regional scales. Using a nonlinear damage model constrained by laboratory experiments, we develop and apply an open-source 3D discontinuous Galerkin method to simulate regional co-seismic wave speed changes during the 2015 Mw7.8 Gorkha earthquake. We find pronounced spatial variations of co-seismic wave speed reduction, ranging from <0.01% to >50%, particularly close to the source and within the Kathmandu Basin. The most significant reduction occurs within the sedimentary basin and varies with basin depths, while wave speed reductions correlate with the fault slip distribution near the source. By comparing ground motions from simulations with elastic, viscoelastic, elastoplastic, and nonlinear damage rheologies, we demonstrate that the nonlinear damage model effectively captures low-frequency ground motion amplification due to strain-dependent wave speed reductions in soft sediments. We verify the accuracy of our approach through comparisons with analytical solutions and assess its scalability on high-performance computing systems. The model shows near-linear strong and weak scaling up to 2048 nodes, enabling efficient large-scale simulations. Our findings provide a physics-based framework to quantify nonlinear earthquake effects and emphasize the importance of damage-induced wave speed variations for seismic hazard assessment and ground motion predictions.

Patterns of the magnetic signature of ionospheric currents, generated from an empirical model based on satellite observations, are used to build a statistical correlation based model for ionospheric fields. In order to stabilize the dynamics and to take into account the dominant role of the sun, the fields are represented in solar magnetic coordinates. The covariance structure is analyzed and a second order process that approximates the full dynamics is generated. We show that for synthetic data observations located at the Earth observatories, the full ionospheric field pattern as observed on the earth surface can be reconstructed with good precision. As a proof of principle we provide a first application to the inversion based on real observatory data.

Self-supervised learning (SSL) allows training data representations without a supervised signal and has become an important paradigm in machine learning. Most SSL methods employ the cosine similarity between embedding vectors and hence effectively embed data on a hypersphere. While this seemingly implies that embedding norms cannot play any role in SSL, a few recent works have suggested that embedding norms have properties related to network convergence and confidence. In this paper, we resolve this apparent contradiction and systematically establish the embedding norm's role in SSL training. Using theoretical analysis, simulations, and experiments, we show that embedding norms (i) govern SSL convergence rates and (ii) encode network confidence, with smaller norms corresponding to unexpected samples. Additionally, we show that manipulating embedding norms can have large effects on convergence speed. Our findings demonstrate that SSL embedding norms are integral to understanding and optimizing network behavior.

Anomaly detection in time series is essential for industrial monitoring and environmental sensing, yet distinguishing anomalies from complex patterns remains challenging. Existing methods like the Anomaly Transformer and DCdetector have progressed, but they face limitations such as sensitivity to short-term contexts and inefficiency in noisy, non-stationary environments. To overcome these issues, we introduce MAAT, an improved architecture that enhances association discrepancy modeling and reconstruction quality. MAAT features Sparse Attention, efficiently capturing long-range dependencies by focusing on relevant time steps, thereby reducing computational redundancy. Additionally, a Mamba-Selective State Space Model is incorporated into the reconstruction module, utilizing a skip connection and Gated Attention to improve anomaly localization and detection performance. Extensive experiments show that MAAT significantly outperforms previous methods, achieving better anomaly distinguishability and generalization across various time series applications, setting a new standard for unsupervised time series anomaly detection in real-world scenarios.

Pathology plays a critical role in diagnosing a wide range of diseases, yet existing approaches often rely heavily on task-specific models trained on extensive, well-labeled datasets. These methods face sustainability challenges due to the diversity of pathologies and the labor-intensive nature of data collection. To address these limitations, we highlight the need for universal multimodal embeddings that can support multiple downstream tasks. Previous approaches involve fine-tuning CLIP-based models, which handle images and texts separately, limiting their ability to capture complex multimodal relationships. Additionally, these models are evaluated across diverse datasets without a unified benchmark. In this paper, we explore the possibility of applying Multimodal Large Language Models (MLLMs) to generate pathology universal embeddings to address these challenges. Our contributions can be summarized in the following aspects: 1) We propose MLLM4PUE, a novel framework that leverages MLLMs to generate embeddings for various pathology downstream tasks. 2) We further introduce the Pathology Multimodal Embedding Benchmark (PMEB), a comprehensive benchmark designed to assess the quality of pathology multimodal embeddings, which comprises 16 original tasks drawn from 15 datasets. 3) Extensive experimental results demonstrate the superiority of MLLM4PUE, illustrating MLLM-based models can effectively support a wide range of downstream tasks and unify the research direction for foundation models in pathology.

Image Quality Assessment (IQA) measures and predicts perceived image quality by human observers. Although recent studies have highlighted the critical influence that variations in the scale of an image have on its perceived quality, this relationship has not been systematically quantified. To bridge this gap, we introduce the Image Intrinsic Scale (IIS), defined as the largest scale where an image exhibits its highest perceived quality. We also present the Image Intrinsic Scale Assessment (IISA) task, which involves subjectively measuring and predicting the IIS based on human judgments. We develop a subjective annotation methodology and create the IISA-DB dataset, comprising 785 image-IIS pairs annotated by experts in a rigorously controlled crowdsourcing study. Furthermore, we propose WIISA (Weak-labeling for Image Intrinsic Scale Assessment), a strategy that leverages how the IIS of an image varies with downscaling to generate weak labels. Experiments show that applying WIISA during the training of several IQA methods adapted for IISA consistently improves the performance compared to using only ground-truth labels. We will release the code, dataset, and pre-trained models upon acceptance.

In this study, we performed a preliminary mapping of Total Electron Content (TEC) over Ecuador using Global Positioning System (GPS) data. This process entails collecting and analyzing pseudorange observations from multiple GPS receivers nationwide. These receivers record signals from GPS satellites, and by comparing the arrival times of these signals, the number of electrons in the ionosphere can be inferred along the lines of sight between the satellites and the receivers. To perform this process, signal processing algorithms are utilized to calculate TEC values, which are subsequently used to generate two-dimensional color maps that illustrate the spatial distribution of TEC in Ecuador. These maps, created using data from 13 GPS receivers distributed throughout the country, offer a valuable visualization of TEC variability regarding geographic location and time. Focusing on specific days in January 2022, this study aims to analyze patterns and trends in ionospheric electron content across the region. The results revealed an oscillatory pattern in TEC evolution, with intensity peaks sometimes reaching or exceeding 80 TEC units (TECU), while local minima never reach zero values. This preliminary TEC mapping approach over Ecuador using GPS data is crucial for understanding ionospheric dynamics in the region. It may have various applications, including improving the accuracy of GPS navigation, monitoring solar activity, and forecasting ionospheric phenomena that can impact communications and satellite navigation.

The classification of seismic events has been crucial for monitoring underground nuclear explosions and unnatural seismic events as well as natural earthquakes. This research is an attempt to apply different machine learning (ML) algorithms to classify various types of seismic events into chemical explosions, collapses, nuclear explosions, damaging earthquakes, felt earthquakes, generic earthquakes and generic explosions for a dataset obtained from IRIS-DMC. One major objective of this research has been to identify some of the best ML algorithms for such seismic events classification. The ML algorithms we are implementing in this study include logistic regression, support vector machine (SVM), Na\"ive Bayes, random forest, K-nearest neighbors (KNN), decision trees, and linear discriminant analysis. Our implementation of the above ML classifier algorithms required to prepare and preprocess the dataset we obtained so that it will be fit for the ML training and testing applications we sought. After the implementation of the ML algorithms, we were able to classify the seismic event types into seven classes in the dataset, and a comparison of each classifier is made to identify the best algorithm for the seismic data classification. Finally, we made predictions of the different event types using the different classifier algorithms, and evaluated each of the various classifier algorithms for seismic prediction using different evaluation metrics. These evaluation metrics helped us to measure the performance of each algorithm. After implementing the seven ML algorithms and a comparison among those various ML algorithms, it has been demonstrated that the best accuracy among these classifiers happened for the Random Forest (RF) algorithm, with an accuracy of 93.5%.

This study investigates the impact of the G5 geomagnetic storm on Total Electron Content (TEC) derived from the Global Positioning System (GPS) in Gal\'apagos, Ecuador (geographic latitude 0.1807{\deg} S, longitude 78.4678{\deg} W) during May 10-13, 2024. Using vertical TEC (VTEC) data from a single pseudorandom noise (PRN) code, along with the average VTEC from the same PRN collected over the ten days before the storm, referred to as background TEC, to analyze the variations in TEC. Our findings indicate that during the main phase of the storm on May 10-11, 2024, TEC experienced a notable decrease, which contrasts with the typical responses observed in previous storms. This decrease can be attributed to rapid recombination processes and potential plasma instabilities triggered by the storm. In the recovery phase following the main storm, a gradual increase in TEC was observed, illustrating the complex dynamics of the ionosphere in response to geomagnetic disturbances. This study underscores the variability in TEC responses during geomagnetic storms. It highlights the importance of real-time monitoring to improve our understanding of the implications for satellite communication and navigation systems.

Semantic segmentation is one of the core tasks in the field of computer vision, and its goal is to accurately classify each pixel in an image. The traditional Unet model achieves efficient feature extraction and fusion through an encoder-decoder structure, but it still has certain limitations when dealing with complex backgrounds, long-distance dependencies, and multi-scale targets. To this end, this paper proposes an improved Unet model combined with an attention mechanism, introduces channel attention and spatial attention modules, enhances the model's ability to focus on important features, and optimizes skip connections through a multi-scale feature fusion strategy, thereby improving the combination of global semantic information and fine-grained features. The experiment is based on the Cityscapes dataset and compared with classic models such as FCN, SegNet, DeepLabv3+, and PSPNet. The improved model performs well in terms of mIoU and pixel accuracy (PA), reaching 76.5% and 95.3% respectively. The experimental results verify the superiority of this method in dealing with complex scenes and blurred target boundaries. In addition, this paper discusses the potential of the improved model in practical applications and future expansion directions, indicating that it has broad application value in fields such as autonomous driving, remote sensing image analysis, and medical image processing.

Identifying the extent of brain tumors is a significant challenge in brain cancer treatment. The main difficulty is in the approximate detection of tumor size. Magnetic resonance imaging (MRI) has become a critical diagnostic tool. However, manually detecting the boundaries of brain tumors from MRI scans is a labor-intensive task that requires extensive expertise. Deep learning and computer-aided detection techniques have led to notable advances in machine learning for this purpose. In this paper, we propose a modified You Only Look Once (YOLOv8) model to accurately detect the tumors within the MRI images. The proposed model replaced the Non-Maximum Suppression (NMS) algorithm with a Real-Time Detection Transformer (RT- DETR) in the detection head. NMS filters out redundant or overlapping bounding boxes in the detected tumors, but they are hand-designed and pre-set. RT-DETR removes hand-designed components. The second improvement was made by replacing the normal convolution block with ghost convolution. Ghost Convolution reduces computational and memory costs while maintaining high accuracy and enabling faster inference, making it ideal for resource-constrained environments and real-time applications. The third improvement was made by introducing a vision transformer block in the backbone of YOLOv8 to extract context-aware features. We used a publicly available dataset of brain tumors in the proposed model. The proposed model performed better than the original YOLOv8 model and also performed better than other object detectors (Faster R- CNN, Mask R-CNN, YOLO, YOLOv3, YOLOv4, YOLOv5, SSD, RetinaNet, EfficientDet, and DETR). The proposed model achieved 0.91 mAP (mean Average Precision)@0.5.

Accurate forecasting of epidemic infection trajectories is crucial for safeguarding public health. However, limited data availability during emerging outbreaks and the complex interaction between environmental factors and disease dynamics present significant challenges for effective forecasting. In response, we introduce CAPE, a novel epidemic pre-training framework designed to harness extensive disease datasets from diverse regions and integrate environmental factors directly into the modeling process for more informed decision-making on downstream diseases. Based on a covariate adjustment framework, CAPE utilizes pre-training combined with hierarchical environment contrasting to identify universal patterns across diseases while estimating latent environmental influences. We have compiled a diverse collection of epidemic time series datasets and validated the effectiveness of CAPE under various evaluation scenarios, including full-shot, few-shot, zero-shot, cross-location, and cross-disease settings, where it outperforms the leading baseline by an average of 9.9% in full-shot and 14.3% in zero-shot settings. The code will be released upon acceptance.

This study explores the challenges of integrating human visual cue-based dehazing into object detection, given the selective nature of human perception. While human vision adapts dynamically to environmental conditions, computational dehazing does not always enhance detection uniformly. We propose a multi-stage framework where a lightweight detector identifies regions of interest (RoIs), which are then improved via spatial attention-based dehazing before final detection by a heavier model. Though effective in foggy conditions, this approach unexpectedly degrades the performance on clear images. We analyze this phenomenon, investigate possible causes, and offer insights for designing hybrid pipelines that balance enhancement and detection. Our findings highlight the need for selective preprocessing and challenge assumptions about universal benefits from cascading transformations.

Forests function as crucial carbon reservoirs on land, and their carbon sinks can efficiently reduce atmospheric CO2 concentrations and mitigate climate change. Currently, the overall trend for monitoring and assessing forest carbon stocks is to integrate ground monitoring sample data with satellite remote sensing imagery. This style of analysis facilitates large-scale observation. However, these techniques require improvement in accuracy. We used GF-1 WFV and Landsat TM images to analyze Huize County, Qujing City, Yunnan Province in China. Using the style transfer method, we introduced Swin Transformer to extract global features through attention mechanisms, converting the carbon stock estimation into an image translation.

Foundation models refer to deep learning models pretrained on large unlabeled datasets through self-supervised algorithms. In the Earth science and remote sensing communities, there is growing interest in transforming the use of Earth observation data, including satellite and aerial imagery, through foundation models. Various foundation models have been developed for remote sensing, such as those for multispectral, high-resolution, and hyperspectral images, and have demonstrated superior performance on various downstream tasks compared to traditional supervised models. These models are evolving rapidly, with capabilities to handle multispectral, multitemporal, and multisensor data. Most studies use masked autoencoders in combination with Vision Transformers (ViTs) as the backbone for pretraining. While the models showed promising performance, ViTs face challenges, such as quadratic computational scaling with input length, which may limit performance on multiband and multitemporal data with long sequences. This research aims to address these challenges by proposing SatMamba, a new pretraining framework that combines masked autoencoders with State Space Model, offering linear computational scaling. Experiments on high-resolution imagery across various downstream tasks show promising results, paving the way for more efficient foundation models and unlocking the full potential of Earth observation data. The source code is available in https://github.com/mdchuc/HRSFM.

This study addresses the need for accurate and efficient object detection in assistive technologies for visually impaired individuals. We evaluate four real-time object detection algorithms YOLO, SSD, Faster R-CNN, and Mask R-CNN within the context of indoor navigation assistance. Using the Indoor Objects Detection dataset, we analyze detection accuracy, processing speed, and adaptability to indoor environments. Our findings highlight the trade-offs between precision and efficiency, offering insights into selecting optimal algorithms for realtime assistive navigation. This research advances adaptive machine learning applications, enhancing indoor navigation solutions for the visually impaired and promoting accessibility.

Idiom detection using Natural Language Processing (NLP) is the computerized process of recognizing figurative expressions within a text that convey meanings beyond the literal interpretation of the words. While idiom detection has seen significant progress across various languages, the Kurdish language faces a considerable research gap in this area despite the importance of idioms in tasks like machine translation and sentiment analysis. This study addresses idiom detection in Sorani Kurdish by approaching it as a text classification task using deep learning techniques. To tackle this, we developed a dataset containing 10,580 sentences embedding 101 Sorani Kurdish idioms across diverse contexts. Using this dataset, we developed and evaluated three deep learning models: KuBERT-based transformer sequence classification, a Recurrent Convolutional Neural Network (RCNN), and a BiLSTM model with an attention mechanism. The evaluations revealed that the transformer model, the fine-tuned BERT, consistently outperformed the others, achieving nearly 99% accuracy while the RCNN achieved 96.5% and the BiLSTM 80%. These results highlight the effectiveness of Transformer-based architectures in low-resource languages like Kurdish. This research provides a dataset, three optimized models, and insights into idiom detection, laying a foundation for advancing Kurdish NLP.

Self-supervised learning (SSL) aims to find meaningful representations from unlabeled data by encoding semantic similarities through data augmentations. Despite its current popularity, theoretical insights about SSL are still scarce. For example, it is not yet known whether commonly used SSL loss functions can be related to a statistical model, much in the same as OLS, generalized linear models or PCA naturally emerge as maximum likelihood estimates of an underlying generative process. In this short paper, we consider a latent variable statistical model for SSL that exhibits an interesting property: Depending on the informativeness of the data augmentations, the MLE of the model either reduces to PCA, or approaches a simple non-contrastive loss. We analyze the model and also empirically illustrate our findings.

Season and their transitions play a critical role in sharpening ecosystems and human activities, yet traditional classifications, meteorological and astronomical, fail to capture the complexities of biosphere-atmosphere interactions. Conventional definitions often overlook the interplay between climate variables, biosphere processes, and seasonal anticipation, particularly as global climate change disrupts traditional patterns. This study addresses the limitations of current seasonal classification by proposing a framework based on phenological markers such as NDVI, EVI, LAI, fPAR, and the Bowen ratio, using plants as a nature-based sensor of seasonal transitions. Indicators derived from satellite data and ground observations provide robust foundations for defining seasonal boundaries. The normalized daily temperature range (DTRT), validated in crop and orchard regions, is hypothesized as a reliable seasonality index to capture transitions. We demonstrated the alignment of this index with phenological markers across boreal, temperate, and deciduous forests. Analyzing trends, extreme values and inflection points in the seasonality index time series, we established a methodology to identify seasonal onset, duration, and transitions. This universal, scalable classification aligns with current knowledge and perception of seasonal shifts and captures site-specific timing. Findings reveal shifts in the Euro-Mediterranean region, with winters shortening, summers extending, and transitions becoming more pronounced. Effects include the Gulf Stream s influence on milder transitions, urban heat islands accelerating seasonal shifts, and large inland lakes moderating durations. This underscores the importance of understanding seasonal transitions to enable climate change adaptive strategies in agriculture, forestry, urban planning, medicine, trade, marketing, and tourism.

Change detection in remote sensing imagery is a critical technique for Earth observation, primarily focusing on pixel-level segmentation of change regions between bi-temporal images. The essence of pixel-level change detection lies in determining whether corresponding pixels in bi-temporal images have changed. In deep learning, the spatial and channel dimensions of feature maps represent different information from the original images. In this study, we found that in change detection tasks, difference information can be computed not only from the spatial dimension of bi-temporal features but also from the channel dimension. Therefore, we designed the Channel-Spatial Difference Weighting (CSDW) module as an aggregation-distribution mechanism for bi-temporal features in change detection. This module enhances the sensitivity of the change detection model to difference features. Additionally, bi-temporal images share the same geographic location and exhibit strong inter-image correlations. To construct the correlation between bi-temporal images, we designed a decoding structure based on the Layer-Exchange (LE) method to enhance the interaction of bi-temporal features. Comprehensive experiments on the CLCD, PX-CLCD, LEVIR-CD, and S2Looking datasets demonstrate that the proposed LENet model significantly improves change detection performance. The code and pre-trained models will be available at: https://github.com/dyzy41/lenet.

Understanding groundwater dynamics is critical for sustainable water management, particularly as climate extremes intensify. However, the resolutions of existing subsurface observational tools are still inadequate for detailed aquifer monitoring and imaging. We introduce an innovative technique for groundwater monitoring using time-lapse full-waveform inversion, leveraging fiber-optic cables as seismic sensors and vehicular traffic as repetitive seismic sources. Over a two-year period along Sandhill Road, California, this approach captures detailed spatiotemporal S-wave velocity variations, revealing a 2.9% reduction corresponding to a 9.0-meter groundwater table rise after atmospheric-river storms in Water Year 2023. Notably, this approach enables the high-resolution daily analysis of rapid aquifer responses. We observe spatially inhomogeneous velocity changes, with less reduction beneath impervious paved zones than under grassy areas, underscoring the impact of urbanization on the natural recharge of aquifers. Our findings highlight the potential of Vehicle-DAS FWI for high-resolution daily monitoring and quantitative spatiotemporal characterizations of groundwater systems.

Polarimetric GNSS-R systems, equipped with an additional polarization channel, offer enhanced capabilities for separating vegetation and surface scattering effects, thereby improving GNSS-R land remote sensing applications such as soil moisture retrieval in vegetated and forested areas and biomass estimation. However, the effectiveness of these applications relies on accurate calibration of the polarimetric GNSS-R sensor. In the Rongowai mission, a newly developed Next Generation GNSS-R Receiver (NGRx) is installed on a domestic Air New Zealand airplane to collect data during its commercial flights. The NGRx processes multi-GNSS satellite signals simultaneously and utilizes a dual-channel (LHCP and RHCP) antenna, thereby improving spatial coverage and retrieval accuracy. The dual-polarized antenna also provides the possibility to examine the polarimetric GNSS-R system. In this article, a new methodology is developed to calibrate the Level-1 power measurement and the on-board antenna cross-pol gain by comparing measurements from inland lakes and ocean with modeled results. The calibration results in a 34% decrease in the uncertainty in co-pol reflectivity retrieval. The retrieved cross-pol and co-pol reflectivity after calibration are examined by their statistical distribution and spatial mapping with 1.5 km resolution, with multi-land surface types and incidence angles. These results validate the effectiveness of the calibration method and pave the way for future terrestrial science applications.

Dynamic predictive modelling using electronic health record (EHR) data has gained significant attention in recent years. The reliability and trustworthiness of such models depend heavily on the quality of the underlying data, which is, in part, determined by the stages preceding the model development: data extraction from EHR systems and data preparation. In this article, we identified over forty challenges encountered during these stages and provide actionable recommendations for addressing them. These challenges are organized into four categories: cohort definition, outcome definition, feature engineering, and data cleaning. This comprehensive list serves as a practical guide for data extraction engineers and researchers, promoting best practices and improving the quality and real-world applicability of dynamic prediction models in clinical settings.

Ecological communities intrigue researchers seeking to explain the emergence of biodiversity observed in nature. This raises a fundamental question. What sustains the stability and coexistence of species within these ecosystems? Traditional ecological models have largely been based on the assumption that species primarily engage in pairwise interactions. However, interactions in ecological systems may involve groups of three or more individuals, i.e. higher-order interactions. As a result, the question of how the combined effects of pairwise and higher-order interactions shape the stability of large ecological communities remains unresolved. This work addresses this gap by analyzing a model of competitive communities that incorporates both pairwise and higher-order interactions. Using analytical techniques and numerical simulations, we find that higher-order interactions alone are not always sufficient to foster and maintain coexistence. When species are identical (i.e., have the same physiological rates), even a small proportion of higher-order interactions can stabilize their dynamics. However, when more realistic factors, such as varied birth and death rates or complex interaction structures are introduced, a finite fraction of higher-order interactions may not be sufficient to achieve stable coexistence. Our findings challenge the role of higher-order interactions as a universal stabilizing mechanism in ecological communities and open new avenues for research into the interplay of different factors that underpin biodiversity and ecosystem stability.

The number of people living in this agricultural nation of ours, which is surrounded by lush greenery, is growing on a daily basis. As a result of this, the level of arable land is decreasing, as well as residential houses and industrial factories. The food crisis is becoming the main threat for us in the upcoming days. Because on the one hand, the population is increasing, and on the other hand, the amount of food crop production is decreasing due to the attack of diseases. Rice is one of the most significant cultivated crops since it provides food for more than half of the world's population. Bangladesh is dependent on rice (Oryza sativa) as a vital crop for its agriculture, but it faces a significant problem as a result of the ongoing decline in rice yield brought on by common diseases. Early disease detection is the main difficulty in rice crop cultivation. In this paper, we proposed our own dataset, which was collected from the Bangladesh field, and also applied deep learning and transfer learning models for the evaluation of the datasets. We elaborately explain our dataset and also give direction for further research work to serve society using this dataset. We applied a light CNN model and pre-trained InceptionNet-V2, EfficientNet-V2, and MobileNet-V2 models, which achieved 91.5% performance for the EfficientNet-V2 model of this work. The results obtained assaulted other models and even exceeded approaches that are considered to be part of the state of the art. It has been demonstrated by this study that it is possible to precisely and effectively identify diseases that affect rice leaves using this unbiased datasets. After analysis of the performance of different models, the proposed datasets are significant for the society for research work to provide solutions for decreasing rice leaf disease.

This study presents the first observation of a mixed mode of charge transfer during an upward positive flash, which was initiated from the S\"antis Tower in Switzerland. High-speed camera footage, along with current and electric field measurements, revealed a downward-propagating recoil leader connecting to the grounded current-carrying plasma channel at a junction height of < 1 km above the tip of the tower. This event triggered the ``return stroke''-like main pulse associated with Type 1 upward positive flashes, leading us to propose a mixed mode of charge transfer (normally observed in upward negative flashes) as the physical mechanism at play. Furthermore, the observed `Main pulse' shared characteristics with both mixed-mode and M-component-type initial continuous current (ICC) pulses, challenging existing classification criteria, and supporting the notion of a unique mode of charge transfer with a range of junction length-dependent pulse characteristics, as opposed to two distinct modes. The recoil leader itself was accompanied by a sequence of fast electric field pulses indicative of step-like propagation, also an observational first. These findings contribute to improving our understanding of the mechanisms of charge transfer in upward lightning flashes.

We perform large-scale numerical simulations of convection in 3D porous media at Rayleigh-Darcy numbers up to $Ra=8\times10^4$. To investigate the convective mixing of carbon dioxide (CO$_2$) in geological formations, we consider a semi-infinite domain, where the CO$_2$ concentration is constant at the top and no flux is prescribed at bottom. Convection begins with a diffusion-dominated phase, transitions to convection-driven solute finger growth, and ends with a shutdown stage as fingers reach the bottom boundary and the concentration in the system increases. For $Ra \ge 5\times10^3$, we observe a constant-flux regime with dissolution flux stabilizing at 0.019, approximately 13\% higher than in 2D estimates. Finally, we provide a simple and yet accurate physical model describing the mass of solute entering the system throughout the whole mixing process. These findings extend solutal convection insights to 3D and high-$Ra$, improving the reliability of tools predicting the long-term CO$_2$ dynamics in the subsurface.

Ever since the last occurrence of a significant earthquake in the Mentawai megathrust zone in 2000, no significant earthquake events have been recorded, which, according to the earthquake repetition cycle, suggests that the zone is a potential epicenter of future earthquakes. The southern and northern parts of the zone have been struck by a significant earthquake with magnitude M > 8.0; however, in the potential location of the Mentawai Islands, earthquake energy has not been released. This research shows the tectonic activity, velocity, and shift that occurred owing to the thrust of the plate. The information is a vital reference for estimating the epicenter of the earthquake whose energy has not yet been released. We analyzed the tectonic characteristics according to the synthetic aperture radar data and geodetic global positioning system observations. The results show that the Pagai Islands are experiencing consistent tectonic deformations. The northern region of North Pagai and the Northern region of South Pagai are experiencing significant subsidence, while the southwest (SW) region of North Pagai and the south segment of South Pagai are experiencing significant uplift. The government and local authorities can use this information as a guide for developing strategies for disaster preparation.

Dunes are bedforms found on sandy terrains shaped by fluid flow on Earth, Mars, and other celestial bodies. Despite their prevalence, understanding dune dynamics at the grain scale is challenging due to the vast number of grains involved. In this study, we demonstrate a novel approach to estimate the forces acting on individual dune grains using images. By combining subaqueous experiments, high-speed camera recordings, discrete numerical simulations, and a specially trained convolutional neural network, we can quantify these forces with high accuracy. This method represents a breakthrough in studying granular dynamics, offering a new way to measure forces not only on dune grains but also on smaller objects, such as rocks, boulders, rovers, and man-made structures, observed in satellite images of both Earth and Mars. This technique expands our ability to analyze and understand fluid-grain interactions in diverse environments.

Self-supervised learning (SSL) has significantly advanced image representation learning, yet efficiency challenges persist, particularly with adversarial training. Many SSL methods require extensive epochs to achieve convergence, a demand further amplified in adversarial settings. To address this inefficiency, we revisit the robust EMP-SSL framework, emphasizing the importance of increasing the number of crops per image to accelerate learning. Unlike traditional contrastive learning, robust EMP-SSL leverages multi-crop sampling, integrates an invariance term and regularization, and reduces training epochs, enhancing time efficiency. Evaluated with both standard linear classifiers and multi-patch embedding aggregation, robust EMP-SSL provides new insights into SSL evaluation strategies. Our results show that robust crop-based EMP-SSL not only accelerates convergence but also achieves a superior balance between clean accuracy and adversarial robustness, outperforming multi-crop embedding aggregation. Additionally, we extend this approach with free adversarial training in Multi-Crop SSL, introducing the Cost-Free Adversarial Multi-Crop Self-Supervised Learning (CF-AMC-SSL) method. CF-AMC-SSL demonstrates the effectiveness of free adversarial training in reducing training time while simultaneously improving clean accuracy and adversarial robustness. These findings underscore the potential of CF-AMC-SSL for practical SSL applications. Our code is publicly available at https://github.com/softsys4ai/CF-AMC-SSL.

1. An understanding of how biodiversity confers ecosystem stability is crucial in managing ecosystems under major environmental changes. Multiple biodiversity drivers can stabilize ecosystem functions over time. However, we know little about how local environmental conditions can influence these biodiversity drivers, and consequently how they indirectly shape the ecological stability of ecosystems. 2. We hypothesized that environmental factors can have opposite influences (i.e., not necessarily either positive or negative) on the temporal stability of communities in different environmental ranges depending on the biodiversity drivers involved. We tested this novel hypothesis by using data from a 4-year-long field study of submerged macrophyte across a water depth gradient in 8 heterogeneous bays of Erhai lake (with total sample size of 30,071 quadrats), a large lentic system in China. 3. Results indicate that a unimodal pattern of stability in temporal biomass measurements occurred along the water-depth gradient, and that multiple biodiversity drivers (the asynchrony in species dynamics, and the stability of dominant species) generally increased the temporal stability of aquatic primary producers. However, the effect of water depth either increased or decreased the stability of biomass according to the environmental conditions associated with sites along the water depth gradient. 4. Synthesis. These results reveal the influence of local environmental conditions on the biodiversity drivers of stability may help predict the functional consequences of biodiversity change across different scenarios of environmental change.

Remote sensing object detection is particularly challenging due to the high resolution, multi-scale features, and diverse ground object characteristics inherent in satellite and UAV imagery. These challenges necessitate more advanced approaches for effective object detection in such environments. While deep learning methods have achieved remarkable success in remote sensing object detection, they typically rely on large amounts of labeled data. Acquiring sufficient labeled data, particularly for novel or rare objects, is both challenging and time-consuming in remote sensing scenarios, limiting the generalization capabilities of existing models. To address these challenges, few-shot learning (FSL) has emerged as a promising approach, aiming to enable models to learn new classes from limited labeled examples. Building on this concept, few-shot object detection (FSOD) specifically targets object detection challenges in data-limited conditions. However, the generalization capability of FSOD models, particularly in remote sensing, is often constrained by the complex and diverse characteristics of the objects present in such environments. In this paper, we propose the Generalization-Enhanced Few-Shot Object Detection (GE-FSOD) model to improve the generalization capability in remote sensing FSOD tasks. Our model introduces three key innovations: the Cross-Level Fusion Pyramid Attention Network (CFPAN) for enhanced multi-scale feature representation, the Multi-Stage Refinement Region Proposal Network (MRRPN) for more accurate region proposals, and the Generalized Classification Loss (GCL) for improved classification performance in few-shot scenarios. Extensive experiments on the DIOR and NWPU VHR-10 datasets show that our model achieves state-of-the-art performance for few-shot object detection in remote sensing.

Precision agriculture leverages data and machine learning so that farmers can monitor their crops and target interventions precisely. This enables the precision application of herbicide only to weeds, or the precision application of fertilizer only to undernourished crops, rather than to the entire field. The approach promises to maximize yields while minimizing resource use and harm to the surrounding environment. To this end, we propose a hierarchical panoptic segmentation method that simultaneously determines leaf count (as an identifier of plant growth)and locates weeds within an image. In particular, our approach aims to improve the segmentation of smaller instances like the leaves and weeds by incorporating focal loss and boundary loss. Not only does this result in competitive performance, achieving a PQ+ of 81.89 on the standard training set, but we also demonstrate we can improve leaf-counting accuracy with our method. The code is available at https://github.com/madeleinedarbyshire/HierarchicalMask2Former.

This study introduces a novel approach that integrates the magnetic field data correction from the Tianwen-1 Mars mission with a neural network architecture constrained by physical principles derived from Maxwell's equation equations. By employing a Transformer based model capable of efficiently handling sequential data, the method corrects measurement anomalies caused by satellite dynamics, instrument interference, and environmental noise. As a result, it significantly improves both the accuracy and the physical consistency of the calibrated data. Compared to traditional methods that require long data segments and manual intervention often taking weeks or even months to complete this new approach can finish calibration in just minutes to hours, and predictions are made within seconds. This innovation not only accelerates the process of space weather modeling and planetary magnetospheric studies but also provides a robust framework for future planetary exploration and solar wind interaction research.

The proliferation of wearable sensors and monitoring technologies has created an urgent need for standardized sensor placement protocols. While existing standards like SENIAM address specific applications, no comprehensive framework spans different sensing modalities and applications. We present a unified sensor placement standard that ensures the reproducibility and transferability of human movement and physiological data across various systems and research domains. Our framework provides precise anatomical landmarks, coordinate systems, and placement protocols with defined precision levels, compatible with existing data-sharing standards such as the Brain Imaging Data Structure (BIDS) and Heirechciacal Event Descriptors (HED). This framework aims to enhance data quality, reproducibility, and interoperability in applications ranging from lab-based clinical biomechanics to continuous health monitoring in everyday life.

With the increasing use of plastic, the challenges associated with managing plastic waste have become more challenging, emphasizing the need of effective solutions for classification and recycling. This study explores the potential of deep learning, focusing on convolutional neural networks (CNNs) and object detection models like YOLO (You Only Look Once), to tackle this issue using the WaDaBa dataset. The study shows that YOLO- 11m achieved highest accuracy (98.03%) and mAP50 (0.990), with YOLO-11n performing similarly but highest mAP50(0.992). Lightweight models like YOLO-10n trained faster but with lower accuracy, whereas MobileNet V2 showed impressive performance (97.12% accuracy) but fell short in object detection. Our study highlights the potential of deep learning models in transforming how we classify plastic waste, with YOLO models proving to be the most effective. By balancing accuracy and computational efficiency, these models can help to create scalable, impactful solutions in waste management and recycling.

Previous studies of the response of the Moon to gravitational waves have been carried out using analytical or semi-analytical models assuming ideal lunar structures. Such models are advantageous for their high-speed calculation but fail to account for the extremely heterogeneous subsurface and/or interior structures of the Moon. Numerical calculations are needed, but it is challenging to model the topography and lateral heterogeneity of the Moon. In addition, the computational cost is great especially when performing the GW simulation for a long time. As a first step towards overcoming the above difficulties, we employ a two-dimensional finite element method to numerically simulate the lunar response to gravitational waves. We verify our method by comparing our numerical results with those semi-analytical solutions. Based on such comparison, we also analyze the limitation of the two-dimensional simulation. Our work breaks a new way towards the precise simulation of realistic lunar response to gravitational waves in the future and lays down a solid foundation for three-dimensional numerical simulations.