Search

Abstract Background It is presently considered that Corynebacterium especially Corynebacterium kroppenstedtii (CK) infection, is one of the important causes of granulomatous lobular mastitis (GLM). However, the pathogen of mastitis in the past two years has been identified as a newly discovered Corynebacterium. But it is unclear whether the pathogen associated with the occurrence of GLM is also this bacterium. Methods GLM female patients with positive bacterial culture in pus specimens from February 2023 to February 2024 who were identified as CK infection by mass spectrometer were selected as the research objects in this study, and the clinical isolates were identified by 16S rDNA sequencing technology to identify the specific pathogen of GLM-related bacterial infection. Subsequently, the clinical characteristics of the patients were compared with those of GLM patients without bacterial infection during the same period, to explore the effect of this particular type of Corynebacterium infection on disease development in GLM patients. Finally, we tested the minimum inhibitory concentration (MIC) values of antibiotics when inhibiting these separation strains in vitro through the E-Test experiment, to evaluate their medicine sensitivity. Results A total of 31 GLM patients initially diagnosed with Corynebacterium kroppenstedtii (CK) infection via MALDI-TOF MS were enrolled in the study. However, subsequent 16S rDNA sequencing revealed that 28 isolates (90.32%) were actually identified as the newly recognized Corynebacterium parakroppenstedtii (CPK). This discovery challenges the conventional belief that CK is the primary pathogen of GLM, suggesting instead that CPK is the predominant pathogen associated with GLM bacterial infections. Comparative analysis of the clinical characteristics between the two groups revealed a significantly higher recurrence rate among CPK-infected GLM patients compared to those without CPK infection, along with elevated prolactin levels (P

Abstract Statistical analysis was conducted on groundwater table data in the Ji Yuan Basin from 2006 to 2018, revealing a continuous downward trend in the groundwater table with imminent depletion of shallow groundwater resources. To ensure the sustainable development of groundwater resources in the area, a quantitative model of groundwater table was successfully constructed using the principles of the Quantitative Theory Type I. This model included seven benchmark variables: rainfall, evaporation, exploitation, hydraulic conductivity, specific yield, lithology of the vadose zone, and land-use type. These factors are key contributors to the decline in groundwater levels in the study area. The model yielded an average residual value of 0.35 m and an adjusted R2 of 0.789, indicating that it can explain 78.9% of the benchmark variables with acceptable accuracy. Additionally, the model successfully simulated four types of groundwater dynamics, with the best results obtained for the rainfall infiltration-exploitation-evaporation type. This study’s findings suggest that the model can be utilized to predict the dynamic characteristics of the groundwater table and guide groundwater extraction activities.

Abstract To investigate the stability of a super-large section chamber group, an analysis was conducted based on the in-situ geological conditions of a super-large section chamber group utilized as a coal gangue separation system at the Longgu Coal Mine. Field measurement and numerical simulation were employed to analyze the failure area and stress state of the surrounding rock under varying chamber spacings. The results indicate that the extent of the plastic zone significantly expands when the spacing between chambers is less than 2.0 times the chamber width. When the distance between chambers is 1.5 times the chamber width, it results in the rock pillars being entirely within the plastic zone. As the chamber spacing decreases, the tangential stress within the rock pillar range increases. When the chamber spacing is 2.5, 2.0, and 1.5 times the chamber width, the maximum tangential stress is 1.19, 1.46, and 1.18 times that in the case of a single chamber, respectively. Based on the displacement analysis, it was observed that as the distance between the chambers decreases, there is a notable increase in the displacement of the pillar sides and chamber top, indicating a higher risk of collapse. Integrating the plastic area and stress analysis allows for the categorization of the rock pillar area into four sections: the broken area, loose area, stable area, and firm area. Drawing upon the theoretical solution of the plastic zone of a circular chamber and the equivalent radius method, an approximate solution for the plastic zone of a non-circular chamber has been provided. Furthermore, the minimum reasonable spacing between chambers in a super-large section chamber group is provided as a distance criterion for the failure of a double chamber.

The variation of coal seam thickness is one of the important factors affecting the pressure relief of large-diameter drilling hole. We have established an elastic model of pressure relief of large-diameter drilling hole based on the principle of equivalent elastic modulus. Taking two working faces of a mine as engineering background, we have studied key issues of large-diameter drilling hole regarding coal seam thickness, including the stress distribution characteristics and energy release laws, and the pressure relief and energy release mechanism. Several phenomena of engineering significance have been revealed by our research. Firstly, when the parameters of large-diameter drilling hole are the same, the stress of primary rock, tangential stress and plastic zone range of thin coal seam are greater than those of thick coal seam. Secondly, the peak stress around the large-diameter drilling hole decreases nonlinearly with the increase of coal seam thickness. When the coal seam thickness increases from 1 m to 9 m, the peak stress drops by 12.9%, from 23.2 MPa to 20.2 MPa. Likewise, the pressure relief and energy release rate also decreases with the increase of coal seam thickness. When the coal seam thickness increases from 1 m to 9 m, the pressure relief and energy release rate of large-diameter drilling hole decreases from 68.7% to 45.8%, indicating that better pressure relief effect can be achieved in thin coal seam. Moreover, when the parameters of large-diameter drilling hole are the same, the increase of coal seam thickness causes decrease of the tangential stress and thereby results in the decrease of the range of coal fragmentation zone and plastic zone, and the amount of the released elastic energy, which reduces the pressure relief energy release rate. The coal seam thicknesses of working faces 1208 and 1203 of a certain mine are 9.08 m and 4.95 m, respectively. Although the same set of large-diameter drilling hole parameters is adopted for these two working faces, the one with thicker coal seam suffers from more serious surrounding rock deformation and failure. By optimizing the parameters of large-diameter drilling hole based on our research, the average amount of drilling cuttings decreased from 2.48 kg/m to 1.76 kg/m. Therefore, we suggest that as the coal seam thickness increases, measures such as reducing drilling spacing and increasing drilling diameter should be considered in order to guarantee the expected pressure relief effect.

Abstract The failure of coal mine overburden is mainly caused by fractures under bending loads. The energy evolution characteristics of coal and rock fractures are closely related to coal mine disasters such as rock burst. To obtain the characteristics of energy release and accumulation of coal and rock under bending load, three‐point bending tests of coal, mudstone, and sandstone were carried out respectively. The strength characteristics and fracture propagation process of coal and rock under bending load were studied. The strain energy evolution rules of coal and rock were calculated and obtained. The fracture mechanism of coal and rock was discussed by analyzing the critical strain energy release rate. The results show that the fracture complexity of sandstone and mudstone is greater than that of coal. The microstructure and its directivity in coal and rock indirectly affect their fracture characteristics through the elastic modulus characteristics. The distribution of parameters such as peak load of fracture, fracture energy, and crack length of coal and mudstone samples is discrete, while that of sandstone samples is concentrated. The deformation energy density of coal and rock basically shows a linear increase trend at the prepeak stage. The deformation energy density evolution characteristics at the postpeak stage are mainly affected by the load drop. It is important to establish the internal relationship between the meso structural characteristics and macro mechanical properties for solving engineering problems.

Abstract Rough set theory places great importance on approximation accuracy, which is used to gauge how well a rough set model describes a target concept. However, traditional approximation accuracy has limitations since it varies with changes in the target concept and cannot evaluate the overall descriptive ability of a rough set model. To overcome this, two types of average approximation accuracy that objectively assess a rough set model’s ability to approximate all information granules is proposed. The first is the relative average approximation accuracy, which is based on all sets in the universe and has several basic properties. The second is the absolute average approximation accuracy, which is based on undefinable sets and has yielded significant conclusions. We also explore the relationship between these two types of average approximation accuracy. Finally, the average approximation accuracy has practical applications in addressing missing attribute values in incomplete information tables.

Background: Fat distribution plays an important role in impaired glucose tolerance. Android adiposity (ANDROID) and gynoid adiposity (GYNOID) have been proven to be linked with insulin resistance. A higher risk of sarcopenia is associated with type 2 diabetes mellitus (T2DM). In this study, ANDROID, GYNOID, and ANDROID to GYNOID ratios (A/G ratios) were evaluated in T2DM patients to determine if they were associated with sarcopenia. Methods: We recruited 1086 T2DM patients, measured skeletal muscle index (SMI), ANDROID, GYNOID, and collected clinical data. Results: T2DM patients with 119 male subjects had sarcopenia (20.24%), and 72 female subjects had sarcopenia (16.51%). All patients with T2DM who had high ANDROID and A/G ratios were at a reduced risk of sarcopenia. The SMI showed a correlation with ANDROID and A/G ratios among subjects with T2DM. Conclusion: ANDROID and A/G ratios are inversely related to sarcopenia in T2DM patients.

Sea surface latent and sensible heat fluxes are crucial components of the air-sea energy exchanges that influence the upper-ocean heat content and the marine atmospheric boundary layer. Due to the limited availability of in situ observations, assessing their impact on Arctic weather and climate has mainly been done using data assimilation products and numerical model simulations. The accuracy of the surface fluxes in numerical models are, however, largely unvalidated. Recent deployments of saildrones, remotely piloted uncrewed surface vehicles, can help bridge this data gap of in situ observations. This study represents an initial effort to validate sea surface latent and sensible heat fluxes over the Pacific sub-Arctic open ocean from three commonly used global reanalysis products (NASA MERRA2, ECMWF ERA5, NOAA CFSR2) against observations by saildrones. In general, fluxes from these reanalysis products and saildrone observations agree well, except for CFSR2 sensible heat fluxes, which exhibit systematic negative biases. Sporadic, very large (greater than two observed standard deviations) discrepancies between fluxes from the reanalysis products and observations do occur. These substantial discrepancies in the reanalysis products primarily result from errors in temperature for sensible heat fluxes and errors in both humidity and wind speed for latent heat fluxes. The results from this study suggest that the sea surface latent and sensible heat fluxes from MERRA2 and ERA5 are reliable in representing the mean features of air-sea exchanges in the sub-Arctic region. Nonetheless, their reliability is limited when used for studies of high-frequency variability, such as synoptic weather events.

Abstract We propose a physics‐informed convolutional decoder (PICD) framework for inverse modeling of heterogenous groundwater flow. PICD stands out as a direct inversion method, eliminating the need for repeated forward model simulations. The framework combines data‐driven and physics‐driven approaches by integrating monitoring data and domain knowledge into the inversion process. PICD utilizes a convolutional decoder to effectively approximate the spatial distribution of hydraulic heads, while Karhunen–Loève expansion (KLE) is employed to parameterize hydraulic conductivities. During the training process, the stochastic vector in KLE and the parameters of the convolutional decoder are adjusted simultaneously to minimize the data‐mismatch and the physical violation. The final optimized stochastic vectors correspond to the estimation of hydraulic conductivities, and the trained convolutional decoder can predict the evolution and distribution of hydraulic heads. Various scenarios of groundwater flow are examined and results demonstrate the framework's capability to accurately estimate heterogeneous hydraulic conductivities and to deliver satisfactory predictions of hydraulic heads, even with sparse measurements.

Background: Domain adaptation (DA) techniques have emerged as a pivotal strategy in addressing the challenges of cross-subject classification. However, traditional DA methods are inherently limited by the assumption of a homogeneous space, requiring that the source and target domains share identical feature dimensions and label sets, which is often impractical in real-world applications. Therefore, effectively addressing the challenge of EEG classification under heterogeneous spaces has emerged as a crucial research topic. Methods: We present a comprehensive framework that addresses the challenges of heterogeneous spaces by implementing a cross-domain class alignment strategy. We innovatively construct a cross-encoder to effectively capture the intricate dependencies between data across domains. We also introduce a tailored class discriminator accompanied by a corresponding loss function. By optimizing the loss function, we facilitate the aggregation of features with corresponding classes between the source and target domains, while ensuring that features from non-corresponding classes are dispersed. Results: Extensive experiments were conducted on two publicly available EEG datasets. Compared to advanced methods that combine label alignment with transfer learning, our method demonstrated superior performance across five heterogeneous space scenarios. Notably, in four heterogeneous label space scenarios, our method outperformed the advanced methods by an average of 7.8%. Moreover, in complex scenarios involving both heterogeneous label spaces and heterogeneous feature spaces, our method outperformed the state-of-the-art methods by an average of 4.1%. Conclusions: This paper presents an efficient model for cross-subject EEG classification under heterogeneous spaces, which significantly addresses the challenges of EEG classification within heterogeneous spaces, thereby opening up new perspectives and avenues for research in related fields.

Ocean surface radiation measurement best practices have been developed as a first step to support the interoperability of radiation measurements across multiple ocean platforms and between land and ocean networks. This document describes the consensus by a working group of radiation measurement experts from land, ocean, and aircraft communities. The scope was limited to broadband shortwave (solar) and longwave (terrestrial infrared) surface irradiance measurements for quantification of the surface radiation budget. Best practices for spectral measurements for biological purposes like photosynthetically active radiation and ocean color are only mentioned briefly to motivate future interactions between the physical surface flux and biological radiation measurement communities. Topics discussed in these best practices include instrument selection, handling of sensors and installation, data quality monitoring, data processing, and calibration. It is recognized that platform and resource limitations may prohibit incorporating all best practices into all measurements and that spatial coverage is also an important motivator for expanding current networks. Thus, one of the key recommendations is to perform interoperability experiments that can help quantify the uncertainty of different practices and lay the groundwork for a multi-tiered global network with a mix of high-accuracy reference stations and lower-cost platforms and practices that can fill in spatial gaps.

Global horizontal irradiance (GHI) plays a vital role in estimating solar energy resources, which are used to generate sustainable green energy. In order to estimate GHI with high spatial resolution, a quantitative irradiance estimation network, named QIENet, is proposed. Specifically, the temporal and spatial characteristics of remote sensing data of the satellite Himawari-8 are extracted and fused by recurrent neural network (RNN) and convolution operation, respectively. Not only remote sensing data, but also GHI-related time information (hour, day, and month) and geographical information (altitude, longitude, and latitude), are used as the inputs of QIENet. The satellite spectral channels B07 and B11–B15 and time are recommended as model inputs for QIENet according to the spatial distributions of annual solar energy. Meanwhile, QIENet is able to capture the impact of various clouds on hourly GHI estimates. More importantly, QIENet does not overestimate ground observations and can also reduce RMSE by 27.51%/18.00%, increase R2 by 20.17%/9.42%, and increase r by 8.69%/3.54% compared with ERA5/NSRDB. Furthermore, QIENet is capable of providing a high-fidelity hourly GHI database with spatial resolution 0.02°×0.02° (approximately 2km×2km) for many applied energy fields.

Marine renewable energy is gaining prominence as a crucial component of the energy supply in coastal cities due to proximity and minimal land requirements. The synergistic potential of integrating floating photovoltaics with offshore wind turbines presents an encouraging avenue for boosting power production, amplifying spatial energy generation density, and mitigating seasonal output fluctuations. While the global promise of offshore wind-photovoltaic hybrid systems is evident, a definitive understanding of their potential remains elusive. Here, we evaluate the resource potential of the hybrid systems under geographical constraints, offering insights into sustainable and efficient offshore energy solutions. We compile a database with 11,198 offshore wind turbine locations from Sentinel-1 imagery and technical parameters from commercial project details. Our analysis reveals an underutilization of spatial resources within existing offshore wind farms, yielding a modest 26 kWh per square meter. Furthermore, employing realistic climate-driven system simulations, we find an impressive potential photovoltaic generation of 1372 ± 18 TWh annually, over seven times higher than the current offshore wind capacity. Notably, floating photovoltaics demonstrated remarkable efficiency, matching wind turbine output with a mere 17 % of the wind farm area and achieving an average 76 % increase in power generation at equivalent investment costs. Additionally, the hybrid wind and photovoltaic systems exhibit monthly-scale complementarity, reflected by a Pearson correlation coefficient of -0.78, providing a consistent and reliable power supply. These findings support the notion that hybrid offshore renewable energy could revolutionize the renewable energy industry, optimize energy structures, and contribute to a sustainable future for coastal cities.

On September 30, 2021, a saildrone uncrewed surface vehicle intercepted Hurricane Sam in the northwestern tropical Atlantic and provided continuous observations near the eyewall. Measured surface ocean temperature unexpectedly increased during the first half of the storm. Saildrone current shear and upper-ocean structure from the nearest Argo profiles show an initial trapping of wind momentum by a strong halocline in the upper 30 m, followed by deeper mixing and entrainment of warmer subsurface water into the mixed layer. The ocean initial conditions provided to operational forecast models failed to capture the observed upper-ocean structure. The forecast models failed to simulate the warming and developed a surface cold bias of ~0.5°C by the time peak winds were observed, resulting in a 12-17% underestimation of surface enthalpy flux near the eyewall. Results imply that enhanced upper-ocean observations and, critically, improved assimilation into the hurricane forecast systems, could directly benefit hurricane intensity forecasts.

Fibreoptic mammography is widely recognised as the first screening method for pathologic papillary overflow due to its significant advantages in the diagnosis of ductal dilatation, intraductal papilloma and intraductal carcinoma. The use of fibreoptic ductoscopic excisional biopsy techniques, such as biopsy needles, vacuum negative pressure aspiration, biopsy forceps and grasping baskets, has not been promoted largely due to their existing deficiencies. The imaging effect of fibreoptic ductoscopy compared with electronic ductoscopy is also one of the important factors limiting the progress of microscopic excisional biopsy techniques. Finding a more suitable operating space for electronic fibreoptic ductoscopy and the use of electrosurgical excision biopsy techniques should be the focus of research in view of achieving accurate diagnoses in electronic fibreoptic ductoscopy and microscopic excision biopsy. In this review, the development history, clinical application and existing problems of fibreoptic ductoscopy are reviewed and assessed to provide references for the clinical diagnosis and treatment of pathologic papillary overflow.

Abstract The enantioseparation of chiral molecules is a crucial and challenging task in the field of experimental chemistry, often requiring extensive trial and error with different experimental settings. To overcome this challenge, here we show a research framework that employs machine learning techniques to predict retention times of enantiomers and facilitate chromatographic enantioseparation. A documentary dataset of chiral molecular retention times in high-performance liquid chromatography (CMRT dataset) is established to handle the challenge of data acquisition. A quantile geometry-enhanced graph neural network is proposed to learn the molecular structure-retention time relationship, which shows a satisfactory predictive ability for enantiomers. The domain knowledge of chromatography is incorporated into the machine learning model to achieve multi-column prediction, which paves the way for chromatographic enantioseparation prediction by calculating the separation probability. The proposed research framework works well in retention time prediction and chromatographic enantioseparation facilitation, which sheds light on the application of machine learning techniques to the experimental scene and improves the efficiency of experimenters to speed up scientific discovery.

Abstract The western Arctic marginal seas undergo large seasonal variation, but are very challenging to observe directly due to sea ice and shallow depths. Deployments of several saildrone uncrewed surface vehicles in the summers of 2018 and 2019 provided unique opportunities to validate the satellite‐derived near surface currents, Ocean Surface Current Analysis Real‐time (OSCAR), in the western Arctic marginal seas against in situ upper ocean current measurements. Overall, OSCAR current is biased low (by 5.3 cm/s) with significant noise. Higher vector correlation estimated by the cosine similarity and speed differences often occur where stronger currents (often topography‐steered) are observed. Such differences reveal that the data set resolvability depends on spatial and temporal resolutions, smoothing, and latitudes, suggesting that OSCAR is able to depict the major current systems but significantly underestimates their strength. Poorer vector correlation occurs at weaker current regimes (

Electrical energy is essential in today's society. Accurate electrical load forecasting is beneficial for better scheduling of electricity generation and saving electrical energy. In this paper, we propose an adaptive deep-learning load forecasting framework by integrating Transformer and domain knowledge (Adaptive-TgDLF). Adaptive-TgDLF introduces the deep-learning model Transformer and adaptive learning methods (including transfer learning for different locations and online learning for different time periods), which captures the long-term dependency of the load series, and is more appropriate for realistic scenarios with scarce samples and variable data distributions. Under the theory-guided framework, the electrical load is divided into dimensionless trends and local fluctuations. The dimensionless trends are considered as the inherent pattern of the load, and the local fluctuations are considered to be determined by the external driving forces. Adaptive learning can cope with the change of load in location and time, and can make full use of load data at different locations and times to train a more efficient model. Cross-validation experiments on different districts show that Adaptive-TgDLF is approximately 16% more accurate than the previous TgDLF model and saves more than half of the training time. Adaptive-TgDLF with 50% weather noise has the same accuracy as the previous TgDLF model without noise, which proves its robustness. We also preliminarily mine the interpretability of Transformer in Adaptive-TgDLF, which may provide future potential for better theory guidance. Furthermore, experiments demonstrate that transfer learning can accelerate convergence of the model in half the number of training epochs and achieve better performance, and online learning enables the model to achieve better results on the changing load.

The working mechanisms of complex natural systems tend to abide by concise partial differential equations (PDEs). Methods that directly mine equations from data are called PDE discovery, which reveals consistent physical laws and facilitates our interactions with the natural world. In this paper, an enhanced deep reinforcement-learning framework is proposed to uncover symbolically concise open-form PDEs with little prior knowledge. Particularly, based on a symbol library of basic operators and operands, a PDE can be represented by a tree structure. A structure-aware recurrent neural network agent is designed to capture structured information, and is seamlessly combined with the sparse regression method to generate open-form PDE expressions. All of the generated PDEs are evaluated by a meticulously designed reward function by balancing fitness to data and parsimony, and updated by the model-based reinforcement learning. Customized constraints and regulations are formulated to guarantee the rationality of PDEs in terms of physics and mathematics. Numerical experiments demonstrate that our framework is capable of mining open-form governing equations of several dynamic systems, even with compound equation terms, fractional structure, and high-order derivatives. This method is also applied to a real-world problem of the oceanographic system and demonstrates great potential for knowledge discovery in more complicated circumstances with exceptional efficiency and scalability.

Based on the engineering background of weakly cemented roadways and adjacent chambers in Western China, a numerical simulation method was used to examine the changes in stress distribution and increment in the surrounding rock of weakly cemented roadways adjacent to chambers. The results show that the surrounding rock stress of the weakly cemented roadway adjacent to the chambers increases by approximately 30%. The vertical stress of the surrounding rock mainly expands along the vertical direction, and the expansion range is 7–12 times that of the chamber height. The horizontal stress of the surrounding rock mainly expands along the horizontal direction, and the expansion range is 3–6 times that of the chamber width. Based on the support idea of “allowable deformation” + “relief pressure” + “maintaining roadway shape”, the support technology of weakly cemented roadways adjacent to chambers is established with “full section U-shaped steel shed + filling flexible materials between the steel shed and surrounding rock + patching the roof bolt + laying concrete on floor”. An engineering test based on the above support technology was carried out, and it was found that the deformation of the weakly cemented roadway adjacent to the chambers was 0 in 20 days.

IntroductionGranulomatous mastitis (GM) is a chronic inflammatory breast disease. In recent years, the role of Corynebacterium in GM onset has received more and more attention. This study aims to detect the dominant bacterium in GM patients and analyze the association between clinical characteristics and infectious factors.MethodsIn this study, 88 samples from 44 GM patients, six acute lactation mastitis (ALM) patients, and 25 non-inflammatory breast disease (NIB) patients were divided into a GM pus group, a GM tissue group, an ALM pus group, and a NIB tissue group; then, 16S ribosomal DNA sequencing was used to explore their microbiota. The clinical data of all 44 GM patients were also retrospectively collected and analyzed to determine their relationship with infection.ResultsThe median age of the 44 GM patients was 33 years, and 88.6% of patients had primary-onset cases, while 11.4% were recurrences; additionally, 89.5% of patients were postpartum and 10.5% were nulliparous. The serum prolactin level was abnormal in nine patients (24.3%). Samples from 15 GM patients (34.1%) had a Corynebacterium abundance of >1% (1.08–80.08%), with eight (53.3%) displaying an abundance of >10%. Corynebacterium was the only genus with significant differences between the GM pus group and the other three groups (p < 0.05). Corynebacterium kroppenstedtii was the predominant Corynebacterium species. Among clinical characteristics, a statistical difference in breast abscess formation was observed according to Corynebacterium abundance in Corynebacterium-positive and- negative patients (p < 0.05).DiscussionThis study explored the relationship between Corynebacterium infection and GM, compared the clinical characteristics between Corynebacterium-positive and- negative patients, and provided support for the role of Corynebacterium species-in particular, C. kroppenstedtii-in the pathogenesis of GM. The detection of Corynebacterium can predict GM onset, especially in patients with high prolactin levels or a history of recent lactation.

Protecting the whole small intestine from radiation-induced intestinal injury during the radiotherapy of abdominal or pelvic solid tumors remains an unmet clinical need. Here the authors use a natural microalga to readily construct an oral delivery system to achieve effective radioprotection for the small intestine.

Data-driven discovery of partial differential equations (PDEs) has recently made tremendous progress, and many canonical PDEs have been discovered successfully for proof of concept. However, determining the most proper PDE without prior references remains challenging in terms of practical applications. In this work, a physics-informed information criterion (PIC) is proposed to measure the parsimony and precision of the discovered PDE synthetically. The proposed PIC achieves satisfactory robustness to highly noisy and sparse data on 7 canonical PDEs from different physical scenes, which confirms its ability to handle difficult situations. The PIC is also employed to discover unrevealed macroscale governing equations from microscopic simulation data in an actual physical scene. The results show that the discovered macroscale PDE is precise and parsimonious and satisfies underlying symmetries, which facilitates understanding and simulation of the physical process. The proposition of the PIC enables practical applications of PDE discovery in discovering unrevealed governing equations in broader physical scenes.

Abstract The interpretability of deep neural networks has attracted increasing attention in recent years, and several methods have been created to interpret the “black box” model. Fundamental limitations remain, however, that impede the pace of understanding the networks, especially the extraction of understandable semantic space. In this work, the framework of semantic explainable artificial intelligence (S‐XAI) is introduced, which utilizes a sample compression method based on the distinctive row‐centered principal component analysis (PCA) that is different from the conventional column‐centered PCA to obtain common traits of samples from the convolutional neural network (CNN), and extracts understandable semantic spaces on the basis of discovered semantically sensitive neurons and visualization techniques. Statistical interpretation of the semantic space is also provided, and the concept of semantic probability is proposed. The experimental results demonstrate that S‐XAI is effective in providing a semantic interpretation for the CNN, and offers broad usage, including trustworthiness assessment and semantic sample searching.

Summary: In this protocol, we detail steps for constructing a high-throughput automated platform for thin layer chromatography (TLC) analysis. We describe robotics and computer vision techniques that can handle 32 compounds under three different elution solvents in about 50 min. The established automated platform can obtain statistically standardized retardation factor (Rf) values and enhance reproducibility while reducing labor and time costs.For complete details on the use and execution of this protocol, please refer to Xu et al. (2022).1 : Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

In this study, the tail-slapping behavior of an oblique water-entry projectile is investigated through high-speed photography technology. The experimental images and data are captured, extracted and processed using a digital image processing method. The experimental repeatability is verified. By examining the formation, development and collapse process of the projectile’s cavity, this study investigates the impact of the tail-slapping motion on the cavity’s evolution. Furthermore, it examines the distinctive characteristics of both the tail-slapping cavity and the original cavity at varying initial water-entry speeds. By analyzing the formation, development and collapse process of the cavity of the projectile, the influence of the tail-slapping motion on the cavity evolution is explored. Furthermore, it examines the evolution characteristics of both the tail-slapping cavity and the original cavity under different initial water-entry speeds. The results indicate that a tail-slapping cavity is formed during the reciprocating motion of the projectile. The tail-slapping cavity fits closely with the original cavity and is finally pulled off from the surface of the original cavity to collapse. In addition, as the initial water-entry speed increases, both the maximum cross-section size of the tail-slapping cavity and the length of the original cavity gradually increase. With the increase in the number of tail-slapping motions, the speed attenuation amplitude of the projectile increases during each tail-slapping motion, the time interval between two tail-slapping motions is gradually shortened, the energy loss of the projectile correspondingly enlarges, and the speed storage capacity of the projectile decreases.

Face verification and recognition are important tasks that have made great progress in recent years. However, recognizing low-resolution faces from small images is still a difficult problem. In this paper, we advocate using diffusion models (DMs) to enhance face resolution and improve their quality for various downstream applications. Most existing DMs for super-resolution use U-Net as their backbone network, which only exploits multi-scale features along the spatial dimension. These approaches result in a slow convergence of corresponding DMs and the inability to capture complex details and fine textures. To address this issue, we propose a novel conditional generative model based on DMs called BPSR3, which replaces the U-Net in super-resolution via repeated refinement (SR3) with a multi-scale deep back-projection network structure. BPSR3 can extract richer features not only in depth but also in breadth. This helps to effectively refine the image quality at different scales. The experimental results on facial datasets show that BPSR3 significantly improved both convergence speed and reconstruction performance. BPSR3 has about 1/4 of the parameters of SR3 but achieves a 50.1% improvement in PSNR, a 19.8% improvement in SSIM, and a 15.4% reduction in FID. Our contribution lies in achieving less time and space consumption and better reconstruction results. In addition, we propose an idea of enhancing the performance of DMs by replacing the U-Net with a better network.

Image super-resolution (SR) usually synthesizes degraded low-resolution images with a predefined degradation model for training. Existing SR methods inevitably perform poorly when the true degradation does not follow the predefined degradation, especially in the case of the real world. To tackle this robustness issue, we propose a cascaded degradation-aware blind super-resolution network (CDASRN), which not only eliminates the influence of noise on blur kernel estimation but also can estimate the spatially varying blur kernel. With the addition of contrastive learning, our CDASRN can further distinguish the differences between local blur kernels, greatly improving its practicality. Experiments in various settings show that CDASRN outperforms state-of-the-art methods on both heavily degraded synthetic datasets and real-world datasets.

Human-induced earthquakes have a long history, which could be caused by impoundment of reservoirs, underground mining, geothermal operations, oil and gas production, and wastewater injection. Analyses of seismicity and other monitoring data suggest: (1) rapid increase of seismicity during 2009～2015 in Oklahoma, USA was mainly due to deep disposal of wastewater from oil and gas extraction; (2) sharp increase of seismicity during 2009～2015 in Alberta, Canada, shows strong spatial and temporal correlation with hydraulic fracturing activities, and the conclusion is that hydraulic fracturing induced earthquakes (M>3.0); (3) In the Groningen gas field in Netherlands, studies revealed strong correlation between induced seismicity and gas depletion starting from 1991 and identified a close link between induced seismicity and reservoir compaction resulting from extraction of gas. The conclusion is that the gas production induced the earthquake (ML3.6) in 2012, which significantly caused damage in the local area; (4) In Sichuan shale gas production field, China, studies suggest that both the long-term injection of wastewater and short-term shale gas hydraulic fracturing correlate with the increasing seismicity, there have been different views on the cause of sizable earthquakes (ML5.7, 5.3) occurring in late 2018 and early 2019 in the shale gas production area. Studying induced seismicity due to industry activities has been one of the major efforts and become a research field in the scientific community. Besides revealing spatial and temporal distributions of induced seismicity, this review summarizes analyses of induced earthquakes from seismology, geomechanics, and structural geology. Developing solutions to the mitigation of earthquake risk during industry activities has been another major effort. This review introduces various theoretical and practical approaches to address the subject, including induced seismicity monitoring and management system, probability estimation methods, and risk assessment using geophysical and geological data, and also presents suggestions to address the issues in China.

ABSTRACT: This study proposes a supervised learning method that does not rely on labels. We use variables associated with the label as indirect labels, and construct an indirect physics-constrained loss based on the physical mechanism to train the model. In the training process, the model prediction is mapped to the space of value that conforms to the physical mechanism through the projection matrix, and then the model is trained based on the indirect labels. The final prediction result of the model conforms to the physical mechanism between indirect label and label, and also meets the constraints of the indirect label. The present study also develops projection matrix normalization and prediction covariance analysis to ensure that the model can be fully trained. Finally, the effect of the physics-constrained indirect supervised learning is verified based on a well log generation problem.

Abstract Diverting agents that could plug a targeted location and fill different shapes of fractures have attracted an increasing attention in the field of fracturing. In this study, a series of novel diverting agents were prepared by mixing aldehyde sodium alginate (ASA) and amino gelatin (AG) through a Schiff base reaction to form a primary network, polylactic acid (PLA) fibers were added to produce a 3D network, and Ca2+ was mixed in as an ionic cross‐linking agent to strengthen the cross‐linking points. It was found that gelation time, swelling behavior, and mechanical performance could be controlled by varying the content of aldehyde groups and amino groups. Gelation time could be controlled in 1‐10 minutes. Plugging behavior and degradation tests demonstrated that the hydrogel could suit fractures with various shapes and plug effectively under high temperatures and pressures, and it could eventually become acidic fluid to help clean the wellbore. This novel ASA/AG/PLA hydrogel constitutes a promising environmentally friendly diverting agent in the application of diversion fracturing.

MicroRNAs expressed in adipocytes are involved in transcriptional regulation of target mRNAs in obesity, but miRNAs critically involved in this process is not well characterized. Here, we identified upregulation of miR-221-3p and miR-222-3p in the white adipose tissues in C57BL/6 mice fed with high fat-high sucrose (HFHS) chow by RNA sequencing. Mir221 and Mir222 are paralogous genes and share the common seed sequence and Mir221/222AdipoKO mice fed with HFHS chow demonstrated resistance to the development of obesity compared with Mir221/222flox/y. Ddit4 is a direct target of Mir221 and Mir222, and the upregulation of Ddit4 in Mir221/222AdipoKO was associated with the suppression of TSC2 (tuberous sclerosis complex 2)/mammalian target of rapamycin complex 1 (mTORC1)/S6K (ribosomal protein S6 kinase) pathway. The overexpression of miR-222-3p linked to enhanced adipogenesis, and it may be a potential candidate for miRNA-based therapy.

In obesity and type 2 diabetes, numerous genes are differentially expressed, and microRNAs are involved in transcriptional regulation of target mRNAs, but miRNAs critically involved in the appetite control are not known. Here, we identified upregulation of miR-342-3p and its host gene Evl in brain and adipose tissues in C57BL/6 mice fed with high fat-high sucrose (HFHS) chow by RNA sequencing. Mir342 (-/-) mice fed with HFHS chow were protected from obesity and diabetes. The hypothalamic arcuate nucleus neurons co-express Mir342 and EVL. The percentage of activated NPY+pSTAT3+ neurons were reduced, while POMC+pSTAT3+ neurons increased in Mir342 (-/-) mice, and they demonstrated the reduction of food intake and amelioration of metabolic phenotypes. Snap25 was identified as a major target gene of miR-342-3p and the reduced expression of Snap25 may link to functional impairment hypothalamic neurons and excess of food intake. The inhibition of miR-342-3p may be a potential candidate for miRNA-based therapy.

Partial differential equations (PDEs) are concise and understandable representations of domain knowledge, which are essential for deepening our understanding of physical processes and predicting future responses. However, the PDEs of many real-world problems are uncertain, which calls for PDE discovery. We propose the symbolic genetic algorithm to discover open-form PDEs (SGA-PDE) directly from data without prior knowledge about the equation structure. SGA-PDE focuses on the representation and optimization of PDEs. Firstly, SGA-PDE uses symbolic mathematics to realize the flexible representation of any given PDE, transforms a PDE into a forest, and converts each function term into a binary tree. Secondly, SGA-PDE adopts a specially designed genetic algorithm to efficiently optimize the binary trees by iteratively updating the tree topology and node attributes. The SGA-PDE is gradient free, which is a desirable characteristic in PDE discovery since it is difficult to obtain the gradient between the PDE loss and the PDE structure. In the experiment, SGA-PDE not only successfully discovered the nonlinear Burgers’ equation, the Korteweg–de Vries equation, and the Chafee-Infante equation but also handled PDEs with fractional structure and compound functions that cannot be solved by conventional PDE discovery methods.

More high-quality, in situ observations of essential marine variables are needed over the seasonal ice zone to better understand Arctic (or Antarctic) weather, climate, and ecosystems. To better assess the potential for arrays of uncrewed surface vehicles (USVs) to provide such observations, five wind-driven and solar-powered saildrones were sailed into the Chukchi and Beaufort Seas following the 2019 seasonal retreat of sea ice. They were equipped to observe the surface oceanic and atmospheric variables required to estimate air-sea fluxes of heat, momentum and carbon dioxide. Some of these variables were made available to weather forecast centers in real time. Our objective here is to analyze the effectiveness of existing remote ice navigation products and highlight the challenges and opportunities for improving remote ice navigation strategies with USVs. We examine the sources of navigational sea-ice distribution information based on post-mission tabulation of the sea-ice conditions encountered by the vehicles. The satellite-based ice-concentration analyses consulted during the mission exhibited large disagreements when the sea ice was retreating fastest (e.g., the 10% concentration contours differed between analyses by up to ∼175 km). Attempts to use saildrone observations to detect the ice edge revealed that in situ temperature and salinity measurements varied sufficiently in ice bands and open water that it is difficult to use these variables alone as a reliable ice-edge indicator. Devising robust strategies for remote ice zone navigation may depend on developing the capability to recognize sea ice and initiate navigational maneuvers with cameras and processing capability onboard the vehicles.

Electricity constitutes an indispensable source of secondary energy in modern society. Accurate and robust short-term electrical load forecasting is essential for more effective scheduling of load generation, minimizing the gap between generation and demand, and reducing electricity losses. This study proposes theory-guided deep-learning load forecasting (TgDLF), which is a gradient-free model that fully combines domain knowledge and machine learning algorithms. TgDLF predicts the future load through load ratio decomposition, in which dimensionless trends are obtained based on domain knowledge, and the local fluctuations are estimated via data-driven models. TgDLF simplifies the problem with the assistance of expertise, and utilizes the strong expressive power of neural networks to obtain accurate predictions. The historical load, weather forecast and calendar effect are considered in the model, and the model's robustness to inaccurate weather forecast data is improved by adding synthetic disturbance during the training process. Cross-validation experiments demonstrate that TgDLF is 23% more accurate than long short-term memory, and the TgDLF with enhanced robustness can effectively extract information from weather forecast data with up to 40% noise.

To supplement missing logging information without increasing economic cost, a machine learning method to generate synthetic well logs from the existing log data was presented, and the experimental verification and application effect analysis were carried out. Since the traditional Fully Connected Neural Network (FCNN) is incapable of preserving spatial dependency, the Long Short-Term Memory (LSTM) network, which is a kind of Recurrent Neural Network (RNN), was utilized to establish a method for log reconstruction. By this method, synthetic logs can be generated from series of input log data with consideration of variation trend and context information with depth. Besides, a cascaded LSTM was proposed by combining the standard LSTM with a cascade system. Testing through real well log data shows that: the results from the LSTM are of higher accuracy than the traditional FCNN; the cascaded LSTM is more suitable for the problem with multiple series data; the machine learning method proposed provides an accurate and cost effective way for synthetic well log generation. Key words: well log, generating method, machine learning, Fully Connected Neural Network, Recurrent Neural Network, Long Short-Term Memory, artificial intelligence

Abstract A composite one-dimensional (1D) Ag sinusoidal nanograting aiming at label-free surface enhanced Raman scattering (SERS) detection of TNT with robust and reproducible enhancements is discussed. 1D periodic sinusoidal SiO2 grating followed by Ag evaporation is proposed for the creation of reproducible and effective SERS substrate based on surface plasmon polaritons (SPPs). The optimal structure of 1D sinusoidal nanograting and its long-range SERS effect are analyzed by using the finite difference time domain (FDTD). Simulation SERS enhancement factor (EF) can be 5 orders of magnitude as possible. This SERS substrate is prepared by the interference photolithography technology, its SERS performance is tested by Rh6G detection experiments, and the actual test EF is about 104. The label-free SERS detection capacity of TNT is demonstrated in the experiment.

Data-driven discovery of partial differential equations (PDEs) has achieved considerable development in recent years. Several aspects of problems have been resolved by sparse regression-based and neural-network-based methods. However, the performance of existing methods lacks stability when dealing with complex situations, including sparse data with high noise, high-order derivatives, and shock waves, which bring obstacles to calculating derivatives accurately. Therefore, a robust PDE discovery framework, called the robust deep-learning genetic algorithm (R-DLGA), that incorporates the physics-informed neural network (PINN) is proposed in this paper. In the framework, preliminary results of potential terms provided by the DLGA are added into the loss function of the PINN as physical constraints to improve the accuracy of derivative calculation. It assists in optimizing the preliminary result and obtaining the ultimately discovered PDE by eliminating the error compensation terms. The stability and accuracy of the proposed R-DLGA in several complex situations are examined for proof and concept, and the results prove that the proposed framework can calculate derivatives accurately with the optimization of the PINN and possesses surprising robustness for complex situations, including sparse data with high noise, high-order derivatives, and shock waves.

A second species of Qingryllus Chen & Zheng is described and illustrated, Q. jiguanshanensis sp. n., from Sichuan, China. This new species is similar to Q. striofemorus Chen & Zheng, 1995, but differs from the latter by the posterior margin of pronotum being distinctly widened, the veins of tegmina yellowish-white only on lateral side of dorsal area, and the epiphallus distinctly widened posteriorly. A distribution map of the species of this genus and habitus photographs of the new species are presented.

Ocean boundary current systems are key components of the climate system, are home to highly productive ecosystems, and have numerous societal impacts. Establishment of a global network of boundary current observing systems is a critical part of ongoing development of the Global Ocean Observing System. The characteristics of boundary current systems are reviewed, focusing on scientific and societal motivations for sustained observing. Techniques currently used to observe boundary current systems are reviewed, followed by a census of the current state of boundary current observing systems globally. The next steps in the development of boundary current observing systems are considered, leading to several specific recommendations.

Partnership between the private sector and the ocean-observing community brings exciting opportunities to address observing challenges through leveraging the unique strengths of each sector. Here, we discuss a case study of a successful relationship between the National Oceanic and Atmospheric Administration (NOAA) Pacific Marine Environmental Laboratory (PMEL) and Saildrone to instrument an unmanned surface vehicle (USV) in order to serve shared goals. This case study demonstrates that a private company working with a federal laboratory has provided innovative ocean-observing solutions deployed at regional scale in only a few years, and we project that this model will be sustainable over the long term. An alignment of long-term goals with practical deliverables during the development process and integrating group cultures were key to success. To date, this effort has expanded NOAA’s interdisciplinary observing capabilities, improved public access to ocean data, and paved the way for a growing range of USV applications in every ocean. By emphasizing shared needs, complementary strengths, and a clear vision for a sustainable future observing system, we believe that this case study can serve as a blueprint for public and private partners who wish to improve observational capacity. We recommend that the international scientific community continues to foster collaborations between the private sector and regional ocean-observing networks. This effort could include regional workshops that build community confidence through independent oversight of data quality. We also recommend that an international framework should be created to organize public and private partners in the atmospheric and oceanographic fields. This body would coordinate development of observational technologies that adhere to best practices and standards for sensor integration, verification, data quality control and delivery, and provide guidance for unmanned vehicle providers. Last, we also recommend building bridges between the private sector, the ocean-observing community, and the operational forecast community to consider the future of this new private sector, with goals to determine targeted ocean-observing needs; assess the appropriateness of USVs as science platforms, sensors, and data format standards; and establish usage and data quality control and distribution protocols for ocean observing and operational forecasting.

The air–sea interface is a key gateway in the Earth system. It is where the atmosphere sets the ocean in motion, climate/weather-relevant air–sea processes occur, and pollutants (i.e., plastic, anthropogenic carbon dioxide, radioactive/chemical waste) enter the sea. Hence, accurate estimates and forecasts of physical and biogeochemical processes at this interface are critical for sustainable blue economy planning, growth, and disaster mitigation. Such estimates and forecasts rely on accurate and integrated in situ and satellite surface observations. High-impact uses of ocean surface observations of essential ocean/climate variables (EOVs/ECVs) include (1) assimilation into/validation of weather, ocean, and climate forecast models to improve their skill, impact, and value; (2) ocean physics studies (i.e., heat, momentum, freshwater, and biogeochemical air–sea fluxes) to further our understanding and parameterization of air–sea processes; and (3) calibration and validation of satellite ocean products (i.e., currents, temperature, salinity, sea level, ocean color, wind, and waves). We review strengths and limitations, impacts, and sustainability of in situ ocean surface observations of several ECVs and EOVs. We draw a 10-year vision of the global ocean surface observing network for improved synergy and integration with other observing systems (e.g., satellites), for modeling/forecast efforts, and for a better ocean observing governance. The context is both the applications listed above and the guidelines of frameworks such as the Global Ocean Observing System (GOOS) and Global Climate Observing System (GCOS) (both co-sponsored by the Intergovernmental Oceanographic Commission of UNESCO, IOC–UNESCO; the World Meteorological Organization, WMO; the United Nations Environment Programme, UNEP; and the International Science Council, ISC). Networks of multiparametric platforms, such as the global drifter array, offer opportunities for new and improved in situ observations. Advances in sensor technology (e.g., low-cost wave sensors), high-throughput communications, evolving cyberinfrastructures, and data information systems with potential to improve the scope, efficiency, integration, and sustainability of the ocean surface observing system are explored.

Turbulent and radiative exchanges of heat between the ocean and atmosphere (hereafter heat fluxes), ocean surface wind stress, and state variables used to estimate them, are Essential Ocean Variables (EOVs) and Essential Climate Variables (ECVs) influencing weather and climate. This paper describes an observational strategy for producing 3-hourly, 25-km (and an aspirational goal of hourly at 10-km) heat flux and wind stress fields over the global, ice-free ocean with breakthrough 1-day random uncertainty of 15 W m–2 and a bias of less than 5 W m–2. At present this accuracy target is met only for OceanSITES reference station moorings and research vessels (RVs) that follow best practices. To meet these targets globally, in the next decade, satellite-based observations must be optimized for boundary layer measurements of air temperature, humidity, sea surface temperature, and ocean wind stress. In order to tune and validate these satellite measurements, a complementary global in situ flux array, built around an expanded OceanSITES network of time series reference station moorings, is also needed. The array would include 500–1000 measurement platforms, including autonomous surface vehicles, moored and drifting buoys, RVs, the existing OceanSITES network of 22 flux sites, and new OceanSITES expanded in 19 key regions. This array would be globally distributed, with 1–3 measurement platforms in each nominal 10° by 10° box. These improved moisture and temperature profiles and surface data, if assimilated into Numerical Weather Prediction (NWP) models, would lead to better representation of cloud formation processes, improving state variables and surface radiative and turbulent fluxes from these models. The in situ flux array provides globally distributed measurements and metrics for satellite algorithm development, product validation, and for improving satellite-based, NWP and blended flux products. In addition, some of these flux platforms will also measure direct turbulent fluxes, which can be used to improve algorithms for computation of air-sea exchange of heat and momentum in flux products and models. With these improved air-sea fluxes, the ocean’s influence on the atmosphere will be better quantified and lead to improved long-term weather forecasts, seasonal-interannual-decadal climate predictions, and regional climate projections.

Since OceanObs’09, the Global Ocean Observing System (GOOS) has evolved from its traditional focus on the ocean’s role in global climate. GOOS now also encompasses operational services and marine ecosystem health, from the open ocean into coastal environments where much of the world’s population resides. This has opened a field of opportunity for new collaborations—across regions, communities, and technologies—facilitating enhanced engagement in the global ocean observing enterprise to benefit all nations. Enhancement of collaboration is considered from the perspectives of regional alliances, global networks, national systems, in situ observing, remote sensing, oceanography, and meteorology. Reinvigoration of GOOS Regional Alliances has been important in connecting the power of this expanded remit to the needs of coastal populations and the capabilities of regional and national marine science communities. An assessment of progress is provided, including issues/challenges with the current structure, and opportunities to increase participation and impact. Meeting the expanded requirements of GOOS will entail new system networks. The Joint Technical Commission for Oceanography and Marine Meteorology Observations Coordination Group has been working with some communities to help assess their readiness, including high frequency radars, ocean gliders, and animal tracking. Much more needs to be done, with a range of strategies considered. Other opportunities include partnering with programs such as the Global Ocean Acidification Observing Network, engaging with mature and emerging national ocean observing programs, and learning from multinational projects such as Tropical Pacific Observing System 2020 and AtlantOS, which are bringing renewed rigor to the design and operation of regional observing systems. Consideration is given to the expansion and advancement that is coming in both in situ and remote sensing ocean observation platforms over the next decade. In combination they provide the potential to measure new Essential Ocean Variables routinely at global scale. Opportunities provided by the World Meteorological Organization Integrated Global Observing System (WIGOS) in fostering a comprehensive and integrated approach across meteorology and oceanography are also considered. The focus of WIGOS on providing accurate, reliable and timely weather, climate, and related environmental observations and products sits well with the expanded requirements of GOOS, in climate, operational services, and marine ecosystem health.

Through comprehensive investigation of the environmental issues in shale gas development in the US, the environmental impacts of hydraulic fracturing in shale gas development are summarized to provide reference for the shale gas development and management in China. The environmental risks of large-scale commercial shale gas development in the United States include water consumption, water contamination, seismic inducement and air pollution. Compared to conventional oil and gas production and other energy producing industries, shale gas development is not exactly “high-water-consuming” in terms of water consuming intensity. Its water consumption, accounting for a small proportion of the total regional water consumption, will not add much more stress on water supply. In terms of water pollution, hydraulic fracturing is unlikely to cause fractures to directly connect reservoir to the shallow aquifer, the known contamination cases are most likely related to faulty well completion, therefore well integrity is the key to the prevention of contamination; the flow-back fluids in large scale shale gas development have the characteristics of large quantity, many kinds of pollutants and complex composition, thus improper treatment would lead to serious contamination, and continuous monitoring and assessment of the pollutants are necessary. Existing evidence shows that hydraulic fracturing is unlikely to trigger destructive earthquakes. Greenhouse gas emissions in the life cycle of shale gas wells were estimated differently, but no doubt more effective measures should be taken to minimize leakage. The research priorities include contamination monitoring program design, detection indicators, moving pattern of hydraulic fracturing fluid and formation fluid, the effects of shale gas development on high salinity formation water and methane migration, and treatment and re-use of flow-back fluid. Key words: shale gas, hydraulic fracturing, environmental risks, water consumption, water contamination, earthquake risk, air pollution

Fluid extraction or injection into underground reservoirs may cause ground deformation, manifested as subsidence or uplift. Excessive deformation may threaten the infrastructure of the oilfield and its surroundings and may even induce earthquakes. Therefore, the monitoring of surface deformation caused by oil production activities is important for the safe production of oilfields and safety assessments of surrounding infrastructure. Karamay oilfield is one of the major oil and gas fields in China. In this study, we take the Karamay oilfield in Xinjiang as a case study to detect surface deformation caused by subsurface fluid injection. Sentinel-1A images of 32 ascending (Path 114) and 34 descending (Path 165) tracks spanning March 2017 to August 2018, were used to derive vertical and horizontal deformation over Karamay oilfield using the MSBAS-InSAR method. The observed two-dimensional deformation exhibited significant vertical and east-west deformation in this region. The maximum uplift and horizontal velocity was approximately 160 mm/year and 65 mm/year, respectively. We also modeled one of the typical deformation zones using a dislocation model in a homogenous elastic half-space.