Search Results (46)

This study examines the impact of physicochemical and geological factors on radon concentrations in groundwater throughout the Mexican Altiplano. Geological diversity, uranium deposits, seismic zones, and geothermal areas with high heat flow are all potential factors contributing to the presence of radon in groundwater. To move beyond local-scale assessments, this research employs spatial prediction methodologies that incorporate geological and geochemical variables recognized for their role in radon transport and geogenic potential. Certain properties of radon enable it to serve as an ideal tracer, viz., short half-life, inertness, and higher incidence in groundwater than surface water. Twenty-five variables were analyzed in samples from 135 water wells. Geostatistical techniques, including inverse distance weighted interpolation and kriging, were used in conjunction with multivariate statistical analyses. Salinity and geothermal heat flow are key indicators for determining groundwater origin, revealing a dynamic interplay between geothermal activity and hydrogeochemical evolution, where high temperatures do not necessarily correlate with increased solute concentrations. The occurrence of toxic trace elements such as Cd, Cr, and Pb is primarily governed by lithogenic sources and proximity to mineralized zones. Radon levels in groundwater are mainly influenced by geological and structural features, notably rhyolitic formations and deep hydrothermal systems. These findings underscore the importance of site-specific groundwater examination, combined with spatiotemporal models, to account for uranium–radium dynamics and flow paths, thereby enhancing radiological risk assessment. Full article

This essay examines mass castration allegations within the North Indian guru movement Dera Sacha Sauda. Drawing on court records, public commentary, and prior fieldwork, it traces how surgical procedures served as a mechanism of enforced proximity and devotional binding. Castration here functions less as renunciation than as anatomical control within a system of engineered devotion that sutures followers into machinic forms of loyalty. The essay situates these acts within a broader politics of sacrificial excess, linking them to hijra initiation, Mughal-coded sovereignty, and strategies of masculine containment. What emerges is a devotional regime of irreversible subtraction and a sovereignty staged through ritual overreach. Full article

With the rapid development of video streaming services, adaptive bitrate (ABR) algorithms have become a core technology for ensuring optimal viewing experiences. Traditional ABR strategies, predominantly rule-based or reinforcement learning-driven, typically employ uniform quality assessment metrics that overlook users’ subjective preference differences regarding factors such as video quality and stalling. To address this limitation, this paper proposes an adaptive video bitrate selection system that integrates preference modeling with reinforcement learning. By incorporating a preference learning module, the system models and scores user viewing trajectories, using these scores to replace conventional rewards and guide the training of the Proximal Policy Optimization (PPO) algorithm, thereby achieving policy optimization that better aligns with users’ perceived experiences. Simulation results on DASH network bandwidth traces demonstrate that the proposed optimization method improves overall Quality of Experience (QoE) by over 9% compared to other mainstream algorithms. Full article

Introduction: Deep reinforcement learning (deep RL) integrates the principles of reinforcement learning with deep neural networks, enabling agents to excel in diverse tasks ranging from playing board games such as Go and Chess to controlling robotic systems and autonomous vehicles. By leveraging foundational concepts of value functions, policy optimization, and temporal difference methods, deep RL has rapidly evolved and found applications in areas such as gaming, robotics, finance, and healthcare. Objective: This paper seeks to provide a comprehensive yet accessible overview of the evolution of deep RL and its leading algorithms. It aims to serve both as an introduction for newcomers to the field and as a practical guide for those seeking to select the most appropriate methods for specific problem domains. Methods: We begin by outlining fundamental reinforcement learning principles, followed by an exploration of early tabular Q-learning methods. We then trace the historical development of deep RL, highlighting key milestones such as the advent of deep Q-networks (DQN). The survey extends to policy gradient methods, actor–critic architectures, and state-of-the-art algorithms such as proximal policy optimization, soft actor–critic, and emerging model-based approaches. Throughout, we discuss the current challenges facing deep RL, including issues of sample efficiency, interpretability, and safety, as well as open research questions involving large-scale training, hierarchical architectures, and multi-task learning. Results: Our analysis demonstrates how critical breakthroughs have driven deep RL into increasingly complex application domains. We highlight existing limitations and ongoing bottlenecks, such as high data requirements and the need for more transparent, ethically aligned systems. Finally, we survey potential future directions, highlighting the importance of reliability and ethical considerations for real-world deployments. Full article

The Late Permian was a critical interval in geological history, during which dramatic changes occurred in the Earth’s surface system, and a set of black rock series rich in organic matter and silicon, the Dalong Formation, was deposited in the northeastern Sichuan Basin. We conducted a detailed sedimentological and petrological investigation integrated with (major and trace) element contents in the deep-water sequence of the Xibeixiang and Jianfeng sections. It demonstrates the source of silicon, tectonic background, and sedimentary environment of the Dalong Formation, and explores the influence of hydrothermal activities on organic matter enrichment. The results show that the upper part of the Dalong Formation contained more radiolarians in the Xibeixiang section compared to the Jianfeng section. Hydrothermal proxies such as Eu/Eu*, Al-Fe-Mn diagram, Al/(Al + Fe + Mn), and Lu_N/La_N suggest a biotic origin for the chert in the Dalong Formation in the Xibeixiang and Jianfeng sections, while the Xibeixiang section was slightly affected by hydrothermal activities. The La-Th-Sc diagram and the La/Sc and Ti/Zr crossplots point to a continental island arc and active continental margin origins for the Xibeixiang and Jianfeng sections. Combined with previous research, the silicon of the Dalong Formation in the northeastern Sichuan Basin is mainly derived from biological sources. The Xibeixiang section was affected by a small amount of hydrothermal fluid due to its proximity to the Paleo-Tethys Ocean and continental island arcs. Furthermore, the enrichment of organic matter was predominantly driven by high productivity, with minimal impact from hydrothermal activities. These insights hold significant research value and practical implications for shale gas exploration in the Sichuan Basin. Full article

The monitoring of heavy metals in aquatic ecosystems is of critical importance due to the toxic effects that these elements can have on wildlife and the potential risks that they pose to human health. Rivers situated in close proximity to agricultural regions are particularly susceptible to contamination from a combination of natural and anthropogenic sources. The study of bioaccumulation is of great importance for the early detection of environmental stressors. The combination of electrochemical techniques, such as square-wave anodic stripping voltammetry (SWASV), with automated flow-batch systems represents an efficient and cost-effective approach for the detection of trace metals in environmental samples. This study examines the bioaccumulation of cadmium and lead in Cyprinus carpio, a bioindicator of contamination in the Colorado River, Argentina. The fish were exposed to sublethal metal concentrations for 24, 48, and 96 h. Metal quantification was conducted using a novel automatic flow-batch system with SWASV and a bismuth film electrode. To the best of our knowledge, this constitutes the first application of this methodology on aquatic bioindicators for the assessment of metal accumulation in a natural environment. The technique demonstrated enhanced sensitivity and selectivity for the detection of trace metals. The bioaccumulation results demonstrated an increase in cadmium and lead concentrations in fish liver tissue after 96 h, reaching 10.5 µg g⁻¹ and 11.9 µg g⁻¹, respectively. Validation with inductively coupled plasma–atomic emission spectrometry (ICP-AES) demonstrated a satisfactory correlation, confirming the reliability of the method. This novel electrochemical approach offers enhanced accuracy and efficiency, making it a promising tool for environmental monitoring. The results indicate that Colorado River water is within safe levels for aquatic life regarding these metals. However, continuous monitoring is recommended to detect changes in contamination levels and protect ecosystem health, especially during water crises and under climate change. Full article

Alerts are an essential tool for the detection of anomalies and ensuring the smooth operation of online service systems by promptly notifying engineers of potential issues. However, the increasing scale and complexity of IT infrastructure often result in “alert storms” during system failures, overwhelming engineers with a deluge of often correlated alerts. Therefore, effective alert aggregation is crucial in isolating root causes and accelerating failure resolution. Existing approaches typically rely on either semantic similarity or statistical methods, both of which have significant limitations, such as ignoring causal relationships or struggling to handle infrequent alerts. To overcome these drawbacks, we propose a novel two-phase alert aggregation approach. We employ temporal–spatial clustering to group alerts based on their temporal proximity and spatial attributes. In the second phase, we utilize large language models to trace the cascading effects of service failures and aggregate alerts that share the same root cause. Experimental evaluations on datasets from real-world cloud platforms demonstrate the effectiveness of our method, achieving superior performance compared to traditional aggregation techniques. Full article

A calc–alkaline dioritic–andesitic–dacitic intrusive–volcanic complex of Early Oligocene (30 Ma) age and its Mesozoic sedimentary basement at Recsk host a well-preserved porphyry–skarn–polymetallic carbonate-replacement–epithermal mineral system. The unique occurrence offers an exceptional possibility to study these related mineralization types at a single locality. This study presents the textural–paragenetic, compositional characteristics, and systematics of sulfide mineral assemblages for the porphyry, skarn, and carbonate-replacement ore types, which are currently situated at a depth of 500–1200 m below the present surface. Detailed petrography combined with EPMA analyses of molybdenite, galena, sphalerite, tetrahedrite-group minerals and Bi-bearing sulfosalts allows for the establishment of characteristic mineral and chemical fingerprints for each mineralization type. Rhenium concentration in molybdenite, occurring as rare disseminations and quartz–carbonate veinlets in altered host rocks in all three mineralization types, shows a decreasing trend towards the more distal mineralization types. High Re contents ($\overline{\mathrm{x}}$ = 1.04 wt.%, max. up to 4.47 wt%) are typical for molybdenite from the porphyry mineralization, but Re is not homogeneously distributed, neither within individual molybdenite crystals nor on a mineralization scale. Copper and Se show opposite behavior in molybdenite, both becoming enriched in the more distal mineralization types. Silver, Bi, and Se concentrations increase in galena and tetrahedrite-group minerals, both towards the country rocks, making them the best candidates for vectoring within the whole hydrothermal system. For tetrahedrite-group minerals, Ag, Bi, Se, together with Sb and Zn, are the suitable elements for fingerprinting; all these are significantly enriched in the distal carbonate-replacement mineralization compared to the other, more proximal ore types. Additionally, further trends can be traced within the composition of sulfosalts. Lead-bearing Bi sulfosalts preferentially occur in the polymetallic carbonate-replacement veins, while being under-represented in the skarn and porphyry mineralization. Porphyry mineralization hosts Cu-bearing Bi sulfosalts dominantly, while skarn is characterized by Bi-dominated sulfosalts. Sphalerite, although present in all mineralization types, cannot be used for fingerprinting, vectoring, or thermobarometry based on EPMA measurements only. Trace element contents of sphalerite are low, often below the detection limits of the analyses. This is further complicated by the intense “chalcopyrite disease” occurring throughout the distal mineralization types. All the above-listed major, minor, and trace element ore mineral characteristics enable the characterization of the Recsk ores by mineral geochemical fingerprints, providing a possible vectoring tool in porphyry Cu–(Mo)–Au-mineralized systems. Full article

Serverless computing is a new cloud computing model suitable for providing services in both large cloud and edge clusters. In edge clusters, the autoscaling functions play a key role on serverless platforms as the dynamic scaling of function instances can lead to reduced latency and efficient resource usage, both typical requirements of edge-hosted services. However, a badly configured scaling function can introduce unexpected latency due to so-called “cold start” events or service request losses. In this work, we focus on the optimization of resource-based autoscaling on OpenFaaS, the most-adopted open-source Kubernetes-based serverless platform, leveraging real-world serverless traffic traces. We resort to the reinforcement learning algorithm named Proximal Policy Optimization to dynamically configure the value of the Kubernetes Horizontal Pod Autoscaler, trained on real traffic. This was accomplished via a state space model able to take into account resource consumption, performance values, and time of day. In addition, the reward function definition promotes Service-Level Agreement (SLA) compliance. We evaluate the proposed agent, comparing its performance in terms of average latency, CPU usage, memory usage, and loss percentage with respect to the baseline system. The experimental results show the benefits provided by the proposed agent, obtaining a service time within the SLA while limiting resource consumption and service loss. Full article

Background: Pandemic management and preparedness are more needed than ever before and there is widespread governmental interest in learning from the COVID-19 pandemic in order to ensure the availability of evidence-based Infection Prevention and Control measures. Contact tracing is integral to Infection Prevention and Control, facilitating breaks in the chain of transmission in a targeted way, identifying individuals who have come into contact with an infected person, and providing them with instruction/advice relating to testing, medical advice and/or self-isolation. Aim: This study aims to improve our understanding of the use of contact tracing technologies in healthcare settings. This research seeks to contribute to the field of Infection Prevention and Control by investigating how these technologies can mitigate the spread of nosocomial infections. Ultimately, this study aims to improve the quality and safety of healthcare delivery. Methods: A systematic literature review was conducted, and journal articles investigating the use of contact tracing technologies in healthcare settings were retrieved from databases held on the OvidSP platform between March and September 2022, with no date for a lower limit. Results: In total, 277 studies were retrieved and screened, and 14 studies were finally included in the systematic literature review. Most studies investigated proximity sensing technologies, reporting promising results. However, studies were limited by small sample sizes and confounding factors, revealing contact tracing technologies remain at a nascent stage. Investment in research and development of new testing technologies is necessary to strengthen national and international contact tracing capabilities. Conclusion: This review aims to contribute to those who intend to create robust surveillance systems and implement infectious disease reporting protocols. Full article

In recent years, biomonitoring has gained more attention, particularly when assessing the environmental health of significant areas, such as those near waste-to-energy facilities. These requirements coincide with the chance to detect environmental pollutants using sensitive organisms. Bees were shown to be quite effective in evaluating the presence of certain compounds by analyzing their associated matrices, such as pollen, honey, or wax. In our study, we employed the honey bee (Apis mellifera) as an indicator to initially monitor the vicinity of the waste-to-energy plant in Acerra, which is situated in the Campania region of Italy. The primary aim was to determine whether the facility was accountable for any environmental releases of dioxins or dioxin-like compounds. Then, we assessed the presence of additional pollutants in the same area, including trace elements, polycyclic aromatic hydrocarbons, and pesticides, released by human activities. To obtain further information about environmental quality, a second biomonitoring station was installed near the Caivano S.T.I.R. (Waste Shredding, Sifting, and Packaging Plant). The results showed the dioxin levels did not exceed predetermined limitations at the Acerra site, thus demonstrating the efficacy of the waste-to-energy facility and the bees’ ability to detect the presence of other pollutants. Additionally, this biomonitoring system exhibited sensitivity to environmental variations, thereby enabling the evaluation of xenobiotic flux between two proximate zones and across temporal scales. This pioneering study suggests the advantages of utilizing bees to detect a wide range of contaminants, thereby providing valuable insights into environmental quality and potential health risks for both ecosystems and human populations. Full article

Mineral assemblages containing Cu-Bi sulfosalts, Bi chalcogenides, and Ag-(Au) tellurides have been identified in the mid-Miocene Zhibula Cu skarn deposit, Gangdese Belt, southern Tibet. Different mineral assemblages from three locations in the deposit, including proximal massive garnet skarn, proximal retrogressed pyroxene-dominant skarn in contact with marble, and distal banded garnet–pyroxene skarn hosted in marble, are studied to constrain the evolution of the mineralization. Hypogene bornite contains elevated Bi (mean 6.73 wt.%) and co-exists in proximal andradite skarn with a second bornite with far lower Bi content, carrollite, Au-Ag tellurides (hessite, petzite), and wittichenite. This assemblage indicates formation at relatively high temperatures (>400 °C) and high f_S2 and f_Te2 during prograde-stage mineralization. Assemblages of Bi sulfosalts (wittichenite, aikinite, kupčíkite, and paděraite) and bismuth chalcogenides (e.g., tetradymite) in proximal pyroxene skarn are also indicative of formation at relatively high temperatures, but at relatively lower f_Te2 and f_S2 conditions. Within the reduced distal skarn (chalcopyrite–pyrrhotite-bearing) in marble, cobalt, and nickel occur as discrete minerals: cobaltite, melonite and cobaltic pentlandite. The trace ore mineral signature of the Zhibula skarn and the distributions of precious and critical trace elements such as Ag, Au, Co, Te, Se, and Bi support an evolving magmatic–hydrothermal system in which different parts of the deposit each define ore formation at distinct local physicochemical conditions. This is the first report of kupčíkite and paděraite from a Chinese location. Their compositions are comparable to other occurrences, but conspicuously, they do not form nanoscale intergrowths with one another. Full article

The Pulai gold deposit is one of the most promising gold prospects in the Central Belt of Peninsular Malaysia. It is found within the Permian-Triassic Gua Musang sequence of metasediments and metavolcanics and in a structurally controlled NE-SW major fault. Various ore minerals, including pyrite, arsenopyrite, chalcopyrite, sphalerite, pyrrhotite, and galena are typically associated with this deposit. Four types of pyrite (Pyrite 1, Pyrite 2, Pyrite 3, and Pyrite 4) and two types of arsenopyrite (Arsenopyrite 1 and Arsenopyrite 2) were characterised based on their morphological and textural differences. In this region, gold occurs as refractory gold in the nanoparticle form and in the state of Au⁺ within the structure of sulphides of variable concentrations. Through the detailed laser ablation inductively coupled plasma mass spectrometry (LA-ICPMS) trace element mapping analysis of pyrite and arsenopyrite, the main Au-bearing sulphides were found within vein-hosted Pyrite 4 and Arsenopyrite 2 during late phase mineralisation, while Pyrite 3 had the lowest Au concentration. Two phases of Au enrichment were recorded in Pyrite 4, mainly in the core (2 to 11.7 ppm; average 1.4 ppm) and margin of the grain (0.3 to 8.8 ppm; average 1.2 ppm), whereas the highest Au content was detected in the core of Arsenopyrite 2 (0.3 to 137.1 ppm; mean 31.9 ppm). The enrichment of Au is associated with As, forming a zoning elemental pattern distribution. Other trace elements, including Co, Ni, Sb, Pb, Bi, Cu, and Zn, show systematic variation in their composition between the various types of pyrite and arsenopyrite. For early-phase sulphides, the Au enrichment localised at the margin of Pyrite 2 and Arsenopyrite 1, together with Co–Ni, Pb–Bi–Sb, and Ag in the same oriented pattern, suggesting the remobilisation and redistribution of Au in sulfides. Meanwhile, the late crystallisation phase of vein-hosted sulphides formed a rich Au–As ore zoning pattern in the core of Arsenopyrite 2 and Pyrite 4. The second phase of Au enrichment continued at the margin of Pyrite 4 through the remobilisation and precipitation of Au together with Ni, Co, Sb, Pb, Bi, Ag, and Cu. Subsequent deformation then reactivated the late fluid system with the enrichment of Sr, Ba, Rb, Ag, and Zn along the fractures and outermost rim of Pyrite 4 and Rb–Sr–Ba–Pb–Bi along the rim of Arsenopyrite 2. The Pulai gold prospect is interpreted as an orogenic-style gold mineralisation where arsenic can be used as an indicator for proximity to ore mineralisation in exploration. Full article

The combination of managed aquifer recharge (MAR) with soil-aquifer treatment (SAT) has clear advantages for the future sustainable quality and quantity management of groundwater, especially when using treated wastewater. We built a Marthe flow and transport model of an MAR–SAT system located in a near-shore sand aquifer, for quantifying the influence of environmental factors (climate, tides, and operational conditions) on the coastal hydrosystem with regard to the fate of trace organic compounds (TrOCs). The simulations show the impact of these factors on flow rates and dilution, and thus, on the potential reactivity of TrOCs. The dilution of secondary treated wastewater (STWW) is variable, depending on the operations (feeding from infiltration ponds) and on shore proximity (dilution by saltwater). We show that, close to the ponds and during infiltration, the attenuation of TrOC concentrations can be explained by reactivity. At the natural outlet of the aquifer, the simulated average residence times ranged from about 70 to 500 days, depending upon seasonal dynamics. It is important to study TrOCs at site scale in order to anticipate the effect of natural variations on the SAT and on the fate of TrOCs. Full article

The COVID-19 pandemic forced governments to implement strategies for contact tracing due to the disease’s ease of spread. The Malaysian government has sought to develop and implement a digital contact-tracking application to make it easier and faster to detect the spread; the system has become an integral part of the exit strategy from mandated lockdowns. These applications keep track of the user’s proximity with others who are in the system to inform them early on if they are at a risk of infection. The effectiveness of these applications depends on the willingness of users to install and allow the application to track their location at all times. Therefore, this research aims to identify the factors that would stimulate or slow down the adoption of contact-tracing apps. Full article