Computers & Geosciences最新文献_第9页

Efficient proxy for time-lapse seismic forward modeling using a U-net encoder–decoder approach

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2025-02-01 DOI: 10.1016/j.cageo.2024.105788

Michael Diniz , Masoud Maleki , Marcos Cirne , Shahram Danaei , João Oliveira , Denis José Schiozer , Alessandra Davolio , Anderson Rocha

The time-lapse seismic (4D seismic) forward modeling provides crucial data for calibrating reservoir models through different data assimilation algorithms. Unfortunately, the traditional 4D seismic forward-modeling methodology is time-expensive and entails significant computational resource consumption. To address these drawbacks, in this work, our goal is to develop a proxy model for the 4D seismic forward modeling using a class of machine learning algorithm named U-Net encoder–decoder. We applied the developed proxy model to a benchmark carbonate reservoir using an ensemble of reservoir simulation models from UNISIM IV dataset (a synthetic benchmark based on real data of a Brazilian pre-salt field). Moreover, we aim to introduce seminal strategies for interpreting the proposed proxy model operation, its outputs, and possible correlations between input and output variables. To achieve this, we trained and tested two versions of U-net-based models and applied methods for explainable artificial intelligence, such as Gradient-weighted Class Activation Mapping (Grad-CAM) and Forward Feature Selection. The experiments showed good results when applied to the test dataset. The correlation coefficient

(R^{2})

values were in the range of 0.7 to 0.9, showing the efficiency of the proxy model to replace the 4D seismic forward modeling. Through qualitative analysis, it was possible to identify which input properties and regions of the reservoir are more relevant for the model’s inference. These results are a step towards robust, explainable machine learning-based proxy forward modeling.

{"title":"Efficient proxy for time-lapse seismic forward modeling using a U-net encoder–decoder approach","authors":"Michael Diniz , Masoud Maleki , Marcos Cirne , Shahram Danaei , João Oliveira , Denis José Schiozer , Alessandra Davolio , Anderson Rocha","doi":"10.1016/j.cageo.2024.105788","DOIUrl":"10.1016/j.cageo.2024.105788","url":null,"abstract":"<div><div>The time-lapse seismic (4D seismic) forward modeling provides crucial data for calibrating reservoir models through different data assimilation algorithms. Unfortunately, the traditional 4D seismic forward-modeling methodology is time-expensive and entails significant computational resource consumption. To address these drawbacks, in this work, our goal is to develop a proxy model for the 4D seismic forward modeling using a class of machine learning algorithm named U-Net encoder–decoder. We applied the developed proxy model to a benchmark carbonate reservoir using an ensemble of reservoir simulation models from UNISIM IV dataset (a synthetic benchmark based on real data of a Brazilian pre-salt field). Moreover, we aim to introduce seminal strategies for interpreting the proposed proxy model operation, its outputs, and possible correlations between input and output variables. To achieve this, we trained and tested two versions of U-net-based models and applied methods for explainable artificial intelligence, such as Gradient-weighted Class Activation Mapping (Grad-CAM) and Forward Feature Selection. The experiments showed good results when applied to the test dataset. The correlation coefficient <span><math><mrow><mo>(</mo><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup><mo>)</mo></mrow></math></span> values were in the range of 0.7 to 0.9, showing the efficiency of the proxy model to replace the 4D seismic forward modeling. Through qualitative analysis, it was possible to identify which input properties and regions of the reservoir are more relevant for the model’s inference. These results are a step towards robust, explainable machine learning-based proxy forward modeling.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"196 ","pages":"Article 105788"},"PeriodicalIF":4.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143102327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

For any two arbitrary slices from one digital rock, its twins can be fast stably reconstructed: A novel integrated model of RVION with ADA-PGGAN

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2025-01-31 DOI: 10.1016/j.cageo.2025.105871

Yingqi Zhang , Liguo Niu , Xin Wang , Dongxing Du , Zhongwen Zhang

The amount of digital rock samples is crucial for studying pore properties. However, it is currently challenging due to equipment limitations or cost considerations. To address this issue, we propose sorts of reconstruction solutions under Data-Scarce Scenarios based on latent inversion predictions from the proposed generative model. Firstly, a novel underlying feature distribution learning model called ResNet-VGG Inversion Optimization Network (RVION) is proposed to infer the latent codes of the real rock images. During inversion, the latent codes predicted by RVION are prepared to interpolate into latent space learned by the generative model. To stably generate high-quality images, the Adaptive Data Augmentation Progressive Growing Generative Adversarial Network (ADA-PGGAN) is proposed, which includes a mechanism to supervise discriminator’s overfitting and automatically adjust levels of data augmentation. Subsequently, interpolated latent codes are input into the generator to progressively increase image resolution and reconstruct large-scale 3D digital rocks. Finally, evaluations using various metrics were conducted in both 2D and 3D on our results. The Sliced Wasserstein Distance (SWD) was used to assess our proposed data augmentation operation. The majority of SWD values remained below 0.01, and further decreased as the resolution increased. Furthermore, generated images accurately exhibited core characteristics. We also evaluated our results in 3D with corresponding metrics, structural properties to indicate consistency with given samples.

{"title":"For any two arbitrary slices from one digital rock, its twins can be fast stably reconstructed: A novel integrated model of RVION with ADA-PGGAN","authors":"Yingqi Zhang , Liguo Niu , Xin Wang , Dongxing Du , Zhongwen Zhang","doi":"10.1016/j.cageo.2025.105871","DOIUrl":"10.1016/j.cageo.2025.105871","url":null,"abstract":"<div><div>The amount of digital rock samples is crucial for studying pore properties. However, it is currently challenging due to equipment limitations or cost considerations. To address this issue, we propose sorts of reconstruction solutions under Data-Scarce Scenarios based on latent inversion predictions from the proposed generative model. Firstly, a novel underlying feature distribution learning model called ResNet-VGG Inversion Optimization Network (RVION) is proposed to infer the latent codes of the real rock images. During inversion, the latent codes predicted by RVION are prepared to interpolate into latent space learned by the generative model. To stably generate high-quality images, the Adaptive Data Augmentation Progressive Growing Generative Adversarial Network (ADA-PGGAN) is proposed, which includes a mechanism to supervise discriminator’s overfitting and automatically adjust levels of data augmentation. Subsequently, interpolated latent codes are input into the generator to progressively increase image resolution and reconstruct large-scale 3D digital rocks. Finally, evaluations using various metrics were conducted in both 2D and 3D on our results. The Sliced Wasserstein Distance (SWD) was used to assess our proposed data augmentation operation. The majority of SWD values remained below 0.01, and further decreased as the resolution increased. Furthermore, generated images accurately exhibited core characteristics. We also evaluated our results in 3D with corresponding metrics, structural properties to indicate consistency with given samples.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"197 ","pages":"Article 105871"},"PeriodicalIF":4.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143347033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Synthetic seismic data generation with pix2pix for enhanced fault detection model training

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2025-01-31 DOI: 10.1016/j.cageo.2025.105879

Byunghoon Choi , Sukjoon Pyun , Woochang Choi , Yongchae Cho

Manual fault interpretation from seismic data is time-consuming and subjective, often yielding inconsistent results. While attribute-based methods improve efficiency, they have limitations. Deep learning has emerged as a promising approach to address these challenges, but acquiring sufficient labeled data is difficult and costly. Synthetic data offers a solution, enabling easier labeling, scalability, and freedom from biases. It can be used alongside field data for pre-training or exclusively for model training. Optimizing synthetic data generation is crucial for effective fault interpretation. Previous studies have explored optimization using style transfer or generative models, which still involve numerical modeling and post-processing steps. In this study, we employ the pix2pix model to generate seismic sections for fault detection, integrating it with sketch-based modeling. Pix2pix is an image-to-image translation model within a conditional generative adversarial networks framework, tailored to the user needs by using images as conditional variables. We experiment with our proposed method using field data examples from the Netherlands Offshore F3 Block and the Thebe Gas Field. Our approach successfully replicates texture-related attributes, including noise, frequency, and amplitude, to resemble field data, thereby facilitating fault interpretation. We provide insights from variations in seismic data and fault interpretation results based on four sketch generation methods and loss function weights of pix2pix. Our approach offers notable advantages, reducing the need for extensive modeling and data processing, thereby streamlining field data analysis in generating optimal seismic sections for fault detection. It is particularly effective when the structural characteristics of reflectivity sketches closely match those of field data. Future research will focus on enhancing geological model production to capture structural characteristics of field data more effectively.

{"title":"Synthetic seismic data generation with pix2pix for enhanced fault detection model training","authors":"Byunghoon Choi , Sukjoon Pyun , Woochang Choi , Yongchae Cho","doi":"10.1016/j.cageo.2025.105879","DOIUrl":"10.1016/j.cageo.2025.105879","url":null,"abstract":"<div><div>Manual fault interpretation from seismic data is time-consuming and subjective, often yielding inconsistent results. While attribute-based methods improve efficiency, they have limitations. Deep learning has emerged as a promising approach to address these challenges, but acquiring sufficient labeled data is difficult and costly. Synthetic data offers a solution, enabling easier labeling, scalability, and freedom from biases. It can be used alongside field data for pre-training or exclusively for model training. Optimizing synthetic data generation is crucial for effective fault interpretation. Previous studies have explored optimization using style transfer or generative models, which still involve numerical modeling and post-processing steps. In this study, we employ the pix2pix model to generate seismic sections for fault detection, integrating it with sketch-based modeling. Pix2pix is an image-to-image translation model within a conditional generative adversarial networks framework, tailored to the user needs by using images as conditional variables. We experiment with our proposed method using field data examples from the Netherlands Offshore F3 Block and the Thebe Gas Field. Our approach successfully replicates texture-related attributes, including noise, frequency, and amplitude, to resemble field data, thereby facilitating fault interpretation. We provide insights from variations in seismic data and fault interpretation results based on four sketch generation methods and loss function weights of pix2pix. Our approach offers notable advantages, reducing the need for extensive modeling and data processing, thereby streamlining field data analysis in generating optimal seismic sections for fault detection. It is particularly effective when the structural characteristics of reflectivity sketches closely match those of field data. Future research will focus on enhancing geological model production to capture structural characteristics of field data more effectively.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"197 ","pages":"Article 105879"},"PeriodicalIF":4.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143347035","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Indications of abundant off-axis activity at the east Pacific rise, 9°50’ N, using a machine learning “chimney identification tool”

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2025-01-27 DOI: 10.1016/j.cageo.2025.105874

Isaac Keohane , Jyun-Nai Wu , Scott M. White , Ross Parnell-Turner

Deep-sea hydrothermal vent systems are a key mechanism for fluid and heat exchanges between the solid Earth and the ocean, but the inaccessible location, scattered occurrence, and meter scale size of vent chimneys make finding them challenging. Now that chimney-sized structures are resolved by near-bottom bathymetric maps, methods to identify potential hydrothermal chimneys in an efficient and reproducible way can be used to develop catalogs of chimney distribution and size. This study investigates the use of a previously developed machine learning Chimney Identification Tool (CIT) to identify potential chimneys in 1 m gridded bathymetric data collected by autonomous underwater vehicle Sentry in 2019–2021. The CIT uses a convolutional neural network, a deep learning model that is well suited to recognize textures and shapes in rasters, that was trained on examples from two other spreading ridge environments. This neural network is combined with a selective search to output individual point locations from input gridded bathymetric data. The CIT picked 119 chimney-like structures up to 4000 m away from the ridge axis and summit collapse trough at the East Pacific Rise between 9°43′N and 9°57′N, suggesting an abundance of off-axis hydrothermal activity that has not been previously acknowledged in estimates or models of hydrothermal activity. This machine learning approach is also compared to interpretations by two expert human analysts. We observe a wide range between the human interpretations, primarily resulting from different levels of including smaller features, with the outputs of the CIT falling within this range. These results illustrate how uncertainty is inherent to identifying seafloor chimneys from bathymetric data, whether manually or algorithmically, due to variation and ambiguity in chimney morphology. We suggest that our results underscore the promise of using an algorithmic method to produce reproducible inventories of potential chimneys with consistent criteria that can be used for broader spatial distribution insights.

{"title":"Indications of abundant off-axis activity at the east Pacific rise, 9°50’ N, using a machine learning “chimney identification tool”","authors":"Isaac Keohane , Jyun-Nai Wu , Scott M. White , Ross Parnell-Turner","doi":"10.1016/j.cageo.2025.105874","DOIUrl":"10.1016/j.cageo.2025.105874","url":null,"abstract":"<div><div>Deep-sea hydrothermal vent systems are a key mechanism for fluid and heat exchanges between the solid Earth and the ocean, but the inaccessible location, scattered occurrence, and meter scale size of vent chimneys make finding them challenging. Now that chimney-sized structures are resolved by near-bottom bathymetric maps, methods to identify potential hydrothermal chimneys in an efficient and reproducible way can be used to develop catalogs of chimney distribution and size. This study investigates the use of a previously developed machine learning Chimney Identification Tool (CIT) to identify potential chimneys in 1 m gridded bathymetric data collected by autonomous underwater vehicle <em>Sentry</em> in 2019–2021. The CIT uses a convolutional neural network, a deep learning model that is well suited to recognize textures and shapes in rasters, that was trained on examples from two other spreading ridge environments. This neural network is combined with a selective search to output individual point locations from input gridded bathymetric data. The CIT picked 119 chimney-like structures up to 4000 m away from the ridge axis and summit collapse trough at the East Pacific Rise between 9°43′N and 9°57′N, suggesting an abundance of off-axis hydrothermal activity that has not been previously acknowledged in estimates or models of hydrothermal activity. This machine learning approach is also compared to interpretations by two expert human analysts. We observe a wide range between the human interpretations, primarily resulting from different levels of including smaller features, with the outputs of the CIT falling within this range. These results illustrate how uncertainty is inherent to identifying seafloor chimneys from bathymetric data, whether manually or algorithmically, due to variation and ambiguity in chimney morphology. We suggest that our results underscore the promise of using an algorithmic method to produce reproducible inventories of potential chimneys with consistent criteria that can be used for broader spatial distribution insights.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"197 ","pages":"Article 105874"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143347034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hyperspectral image classification based on faster residual multi-branch spiking neural network

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2025-01-27 DOI: 10.1016/j.cageo.2025.105864

Yahui Li , Yang Liu , Rui Li , Liming Zhou , Lanxue Dang , Huiyu Mu , Qiang Ge

Deep convolutional neural network has strong feature extraction and fitting capabilities and perform well in hyperspectral image classification tasks. However, due to its huge parameters, complex structure and high energy consumption, it is difficult to be used in mobile edge computing. Spiking neural network (SNN) has the characteristics of event-driven and low energy consumption and has developed rapidly in image classification. But it usually requires more time steps to achieve optimal accuracy. This paper designs a faster residual multi-branch SNN (FRM-SNN) based on leaky integrate-and-fire neurons for HSI classification. The network uses the residual multi-branch module (RMM) as the basic unit for feature extraction. The RMM is composed of spiking mixed convolution and spiking point convolution, which can effectively extract spatial spectral features. Secondly, to address the problem of non-differentiability of Dirac function spiking propagation, a simple and efficient arcsine approximate derivative was designed for gradient proxy, and the classification performance, testing time, and training time of various approximate derivative algorithms were analyzed and evaluated under the same network architecture. Experimental results on six public HSI data sets show that compared with advanced SNN-based HSI classification algorithms, the time step, training time and testing time required for FRM-SNN to achieve optimal accuracy are shortened by approximately 84%, 63% and 70%. This study has important practical significance for promoting the engineering application of HSI classification algorithms in unmanned autonomous devices such as spaceborne and airborne systems.

{"title":"Hyperspectral image classification based on faster residual multi-branch spiking neural network","authors":"Yahui Li , Yang Liu , Rui Li , Liming Zhou , Lanxue Dang , Huiyu Mu , Qiang Ge","doi":"10.1016/j.cageo.2025.105864","DOIUrl":"10.1016/j.cageo.2025.105864","url":null,"abstract":"<div><div>Deep convolutional neural network has strong feature extraction and fitting capabilities and perform well in hyperspectral image classification tasks. However, due to its huge parameters, complex structure and high energy consumption, it is difficult to be used in mobile edge computing. Spiking neural network (SNN) has the characteristics of event-driven and low energy consumption and has developed rapidly in image classification. But it usually requires more time steps to achieve optimal accuracy. This paper designs a faster residual multi-branch SNN (FRM-SNN) based on leaky integrate-and-fire neurons for HSI classification. The network uses the residual multi-branch module (RMM) as the basic unit for feature extraction. The RMM is composed of spiking mixed convolution and spiking point convolution, which can effectively extract spatial spectral features. Secondly, to address the problem of non-differentiability of Dirac function spiking propagation, a simple and efficient arcsine approximate derivative was designed for gradient proxy, and the classification performance, testing time, and training time of various approximate derivative algorithms were analyzed and evaluated under the same network architecture. Experimental results on six public HSI data sets show that compared with advanced SNN-based HSI classification algorithms, the time step, training time and testing time required for FRM-SNN to achieve optimal accuracy are shortened by approximately 84%, 63% and 70%. This study has important practical significance for promoting the engineering application of HSI classification algorithms in unmanned autonomous devices such as spaceborne and airborne systems.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"197 ","pages":"Article 105864"},"PeriodicalIF":4.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fire-Image-DenseNet (FIDN) for predicting wildfire burnt area using remote sensing data Fire-Image-DenseNet (FIDN)，用于利用遥感数据预测野火燃烧面积

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-11-27 DOI: 10.1016/j.cageo.2024.105783

Bo Pang , Sibo Cheng , Yuhan Huang , Yufang Jin , Yike Guo , I. Colin Prentice , Sandy P. Harrison , Rossella Arcucci

Predicting the extent of massive wildfires once ignited is essential to reduce the subsequent socioeconomic losses and environmental damage, but challenging because of the complexity of fire behavior. Existing physics-based models are limited in predicting large or long-duration wildfire events. Here, we develop a deep-learning-based predictive model, Fire-Image-DenseNet (FIDN), that uses spatial features derived from both near real-time and reanalysis data on the environmental and meteorological drivers of wildfire. We trained and tested this model using more than 300 individual wildfires that occurred between 2012 and 2019 in the western US. In contrast to existing models, the performance of FIDN does not degrade with fire size or duration. Furthermore, it predicts final burnt area accurately even in very heterogeneous landscapes in terms of fuel density and flammability. The FIDN model showed higher accuracy, with a mean squared error (MSE) about 82% and 67% lower than those of the predictive models based on cellular automata (CA) and the minimum travel time (MTT) approaches, respectively. Its structural similarity index measure (SSIM) averages 97%, outperforming the CA and FlamMap MTT models by 6% and 2%, respectively. Additionally, FIDN is approximately three orders of magnitude faster than both CA and MTT models. The enhanced computational efficiency and accuracy advancements offer vital insights for strategic planning and resource allocation for firefighting operations.

预测大规模野火一旦点燃的范围对于减少随后的社会经济损失和环境破坏至关重要，但由于火灾行为的复杂性，这一预测具有挑战性。现有的基于物理的模型在预测大型或长时间野火事件方面是有限的。在这里，我们开发了一个基于深度学习的预测模型，Fire-Image-DenseNet (FIDN)，该模型使用了来自野火环境和气象驱动因素的近实时和再分析数据的空间特征。我们使用2012年至2019年在美国西部发生的300多起单独的野火来训练和测试这个模型。与现有模型相比，FIDN的性能不会随着火灾的大小或持续时间而下降。此外，即使在燃料密度和可燃性方面非常不均匀的景观中，它也能准确预测最终燃烧面积。与基于元胞自动机（CA）和最小旅行时间（MTT）方法的预测模型相比，FIDN模型的均方误差（MSE）分别降低了82%和67%。其结构相似指数（SSIM）平均为97%，分别比CA和FlamMap MTT模型高出6%和2%。此外，FIDN比CA和MTT模型都快大约三个数量级。计算效率和准确性的提高为消防行动的战略规划和资源分配提供了重要的见解。

{"title":"Fire-Image-DenseNet (FIDN) for predicting wildfire burnt area using remote sensing data","authors":"Bo Pang , Sibo Cheng , Yuhan Huang , Yufang Jin , Yike Guo , I. Colin Prentice , Sandy P. Harrison , Rossella Arcucci","doi":"10.1016/j.cageo.2024.105783","DOIUrl":"10.1016/j.cageo.2024.105783","url":null,"abstract":"<div><div>Predicting the extent of massive wildfires once ignited is essential to reduce the subsequent socioeconomic losses and environmental damage, but challenging because of the complexity of fire behavior. Existing physics-based models are limited in predicting large or long-duration wildfire events. Here, we develop a deep-learning-based predictive model, Fire-Image-DenseNet (FIDN), that uses spatial features derived from both near real-time and reanalysis data on the environmental and meteorological drivers of wildfire. We trained and tested this model using more than 300 individual wildfires that occurred between 2012 and 2019 in the western US. In contrast to existing models, the performance of FIDN does not degrade with fire size or duration. Furthermore, it predicts final burnt area accurately even in very heterogeneous landscapes in terms of fuel density and flammability. The FIDN model showed higher accuracy, with a mean squared error (MSE) about 82% and 67% lower than those of the predictive models based on cellular automata (CA) and the minimum travel time (MTT) approaches, respectively. Its structural similarity index measure (SSIM) averages 97%, outperforming the CA and FlamMap MTT models by 6% and 2%, respectively. Additionally, FIDN is approximately three orders of magnitude faster than both CA and MTT models. The enhanced computational efficiency and accuracy advancements offer vital insights for strategic planning and resource allocation for firefighting operations.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"195 ","pages":"Article 105783"},"PeriodicalIF":4.2,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Efficient reservoir characterization using dimensionless ensemble smoother and multiple data assimilation in damaged multilayer systems 利用无量纲系综平滑和多重数据同化在受损多层体系中有效表征储层

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-11-26 DOI: 10.1016/j.cageo.2024.105777

Adailton José do Nascimento Sousa , Malú Grave , Renan Vieira Bela , Thiago M.D. Silva , Sinesio Pesco , Abelardo Borges Barreto Junior

The ES-MDA has been extensively applied to address inverse problems related to oil reservoirs, leveraging Bayesian statistics as its cornerstone. This ensemble-based methodology utilizes historical reservoir data to infer its properties such as permeability and skin zone properties. In a recent study , the ES-MDA was utilized to estimate individual skin zone properties using well pressure responses as observed data. However, owing to insufficient reservoir information and the inherent nonlinearity of the problem, their findings lacked precision. This study presents a novel approach to efficiently characterize reservoir skin zones by employing an enhanced ES-MDA implementation and augmenting the observed data vector with flow-rate data. We introduce an analytical method for determining the pressure and flow rate observed at the well during an injectivity test, specifically tailored for multilayer reservoirs with skin zones, utilizing Laplace Transform. To convert the computed data to the real field, we use Stehfest’s algorithm. The analytical model serves a dual purpose: generating artificial data to represent a real field and predicting properties when coupled to the ES-MDA. The new analytical model enables the extraction of flow rates in each layer, which are then integrated as new data into the ES-MDA, thereby bolstering the estimation accuracy of targeted parameters. Both flow rate and pressure are employed as input data and, to alleviate the impact of orders of magnitude disparities on estimates, the ES-MDA is implemented in a dimensionless form. We tested the proposed methodology in four cases to display how adding the flow-rate data could improve results from a previous work. Moreover, the dimensionless ES-MDA offered skin zone properties with lower RMSE compared to the ones obtained in the mentioned study.

ES-MDA已广泛应用于解决与油藏相关的逆问题，以贝叶斯统计为基础。这种基于组合的方法利用历史储层数据来推断其性质，如渗透率和表皮层性质。在最近的一项研究中，ES-MDA被用于利用井压响应作为观测数据来估计单个表皮层的属性。然而，由于储层信息不足和问题固有的非线性，他们的发现缺乏精度。该研究提出了一种新的方法，通过采用增强型ES-MDA实施和用流量数据增强观测数据向量，有效地表征储层表皮层。我们介绍了一种利用拉普拉斯变换的分析方法，用于确定在注入测试期间在井中观察到的压力和流量，该方法专为具有表皮层的多层油藏量身定制。为了将计算数据转换为实际字段，我们使用Stehfest的算法。分析模型具有双重目的：生成代表真实领域的人工数据，并在与ES-MDA耦合时预测属性。新的分析模型可以提取每层的流量，然后将其作为新数据集成到ES-MDA中，从而提高目标参数的估计精度。流速和压力都被用作输入数据，为了减轻数量级差异对估计的影响，ES-MDA以无量纲形式实现。我们在四个案例中测试了所提出的方法，以显示添加流量数据如何改善先前工作的结果。此外，与上述研究中获得的结果相比，无量纲ES-MDA提供了更低RMSE的皮肤区特性。

{"title":"Efficient reservoir characterization using dimensionless ensemble smoother and multiple data assimilation in damaged multilayer systems","authors":"Adailton José do Nascimento Sousa , Malú Grave , Renan Vieira Bela , Thiago M.D. Silva , Sinesio Pesco , Abelardo Borges Barreto Junior","doi":"10.1016/j.cageo.2024.105777","DOIUrl":"10.1016/j.cageo.2024.105777","url":null,"abstract":"<div><div>The ES-MDA has been extensively applied to address inverse problems related to oil reservoirs, leveraging Bayesian statistics as its cornerstone. This ensemble-based methodology utilizes historical reservoir data to infer its properties such as permeability and skin zone properties. In a recent study , the ES-MDA was utilized to estimate individual skin zone properties using well pressure responses as observed data. However, owing to insufficient reservoir information and the inherent nonlinearity of the problem, their findings lacked precision. This study presents a novel approach to efficiently characterize reservoir skin zones by employing an enhanced ES-MDA implementation and augmenting the observed data vector with flow-rate data. We introduce an analytical method for determining the pressure and flow rate observed at the well during an injectivity test, specifically tailored for multilayer reservoirs with skin zones, utilizing Laplace Transform. To convert the computed data to the real field, we use Stehfest’s algorithm. The analytical model serves a dual purpose: generating artificial data to represent a real field and predicting properties when coupled to the ES-MDA. The new analytical model enables the extraction of flow rates in each layer, which are then integrated as new data into the ES-MDA, thereby bolstering the estimation accuracy of targeted parameters. Both flow rate and pressure are employed as input data and, to alleviate the impact of orders of magnitude disparities on estimates, the ES-MDA is implemented in a dimensionless form. We tested the proposed methodology in four cases to display how adding the flow-rate data could improve results from a previous work. Moreover, the dimensionless ES-MDA offered skin zone properties with lower RMSE compared to the ones obtained in the mentioned study.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"195 ","pages":"Article 105777"},"PeriodicalIF":4.2,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Shear wave velocity prediction based on bayesian-optimized multi-head attention mechanism and CNN-BiLSTM 基于贝叶斯优化多头注意机制和CNN-BiLSTM的横波速度预测

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-11-25 DOI: 10.1016/j.cageo.2024.105787

Wenzhi Lan , Yunhe Tao , Bin Liang , Rui Zhu , Yazhai Wei , Bo Shen

Shear wave velocity (VS) is one of the fundamental geophysical parameters essential for pre-stack seismic inversion, rock mechanics evaluation, and in-situ stress assessment. However, due to the high cost of acquiring VS log data, it is impossible to carry out this logging project in all wells. Thus, it is extremely necessary to develop an efficient and reliable VS prediction method. Deep learning methods have distinct advantages in data inversion, but different neural network have their own characteristics. A single-structured neural network has inevitable limitations in VS prediction, making it challenging to effectively capture the nonlinear mapping relationships of multiple parameters. Therefore, an integrated VS prediction model was proposed based on analyzing the applicability of classical neural networks. This new model, denoted as Bo-MA-CNN-BiLSTM, combines a Bayesian-optimized and multi-head attention mechanism (Bo-MA) with a convolutional neural network (CNN) and a bidirectional long short-term memory network (BiLSTM). It can effectively capture spatio-temporal data reflecting geophysical characteristics from log data, and the integration of the multi-head attention mechanism enhances the rational allocation of weights for log data. Bayesian optimization is utilized to determine the values of hyperparameters, overcoming the subjectivity and empiricism associated with manual selection. Actual data processing demonstrates that the new model achieves higher accuracy in predicting VS than applying CNN, LSTM, BiLSTM, and CNN-LSTM individually. The application results of well log data not involved in training indicate that, compared to other classical models, this new model exhibits optimal evaluation metrics. Especially for strongly heterogeneous formations, the predicted results demonstrate significant superiority, verifying the generalization ability and robustness of the proposed model.

横波速度（VS）是叠前地震反演、岩石力学评价和地应力评价的基本地球物理参数之一。然而，由于获取VS测井数据的成本较高，不可能在所有井中进行该测井项目。因此，开发一种高效可靠的VS预测方法是非常必要的。深度学习方法在数据反演方面具有明显的优势，但不同的神经网络各有特点。单结构神经网络在VS预测中存在不可避免的局限性，难以有效捕捉多个参数的非线性映射关系。因此，在分析经典神经网络适用性的基础上，提出了一种综合VS预测模型。该模型将贝叶斯优化的多头注意机制（Bo-MA）与卷积神经网络（CNN）和双向长短期记忆网络（BiLSTM）相结合，命名为Bo-MA-CNN-BiLSTM。该方法可以有效地从测井数据中捕获反映地球物理特征的时空数据，并且多头关注机制的集成增强了测井数据权重的合理分配。利用贝叶斯优化来确定超参数的值，克服了人工选择的主观性和经验主义。实际数据处理表明，与单独应用CNN、LSTM、BiLSTM和CNN-LSTM相比，新模型对VS的预测精度更高。非训练测井数据的应用结果表明，与其他经典模型相比，该模型具有最优的评价指标。特别是对于强非均质地层，预测结果显示出显著的优越性，验证了模型的泛化能力和鲁棒性。

{"title":"Shear wave velocity prediction based on bayesian-optimized multi-head attention mechanism and CNN-BiLSTM","authors":"Wenzhi Lan , Yunhe Tao , Bin Liang , Rui Zhu , Yazhai Wei , Bo Shen","doi":"10.1016/j.cageo.2024.105787","DOIUrl":"10.1016/j.cageo.2024.105787","url":null,"abstract":"<div><div>Shear wave velocity (VS) is one of the fundamental geophysical parameters essential for pre-stack seismic inversion, rock mechanics evaluation, and in-situ stress assessment. However, due to the high cost of acquiring VS log data, it is impossible to carry out this logging project in all wells. Thus, it is extremely necessary to develop an efficient and reliable VS prediction method. Deep learning methods have distinct advantages in data inversion, but different neural network have their own characteristics. A single-structured neural network has inevitable limitations in VS prediction, making it challenging to effectively capture the nonlinear mapping relationships of multiple parameters. Therefore, an integrated VS prediction model was proposed based on analyzing the applicability of classical neural networks. This new model, denoted as Bo-MA-CNN-BiLSTM, combines a Bayesian-optimized and multi-head attention mechanism (Bo-MA) with a convolutional neural network (CNN) and a bidirectional long short-term memory network (BiLSTM). It can effectively capture spatio-temporal data reflecting geophysical characteristics from log data, and the integration of the multi-head attention mechanism enhances the rational allocation of weights for log data. Bayesian optimization is utilized to determine the values of hyperparameters, overcoming the subjectivity and empiricism associated with manual selection. Actual data processing demonstrates that the new model achieves higher accuracy in predicting VS than applying CNN, LSTM, BiLSTM, and CNN-LSTM individually. The application results of well log data not involved in training indicate that, compared to other classical models, this new model exhibits optimal evaluation metrics. Especially for strongly heterogeneous formations, the predicted results demonstrate significant superiority, verifying the generalization ability and robustness of the proposed model.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"195 ","pages":"Article 105787"},"PeriodicalIF":4.2,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multivariate simulation using a locally varying coregionalization model 利用局部变化的核心区域化模型进行多变量模拟

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-11-24 DOI: 10.1016/j.cageo.2024.105781

Álvaro I. Riquelme, Julian M. Ortiz

Understanding the response of materials in downstream processes of mining operations relies heavily on proper multivariate spatial modeling of relevant properties of such materials. Ore recovery and the behavior of tailings and waste are examples where capturing the mineralogical composition is a key component: in the first case, to ensure reliable revenues, and in the second one, to avoid environmental risks involved in their disposal. However, multivariate spatial modeling can be difficult when variables exhibit intricate relationships, such as non-linear correlation, heteroscedastic behavior, or spatial trends. This work demonstrates that the complex multivariate behavior among variables can be reproduced by disaggregating the global non-linear behavior through the spatial domain and looking instead at the local correlations between Gaussianized variables. Local linear dependencies are first inferred from a local neighborhood and then interpolated through the domain using Riemannian geometry tools that allow us to handle correlation matrices and their spatial interpolation. By employing a non-stationary modification of the linear model of coregionalization, it is possible to independently simulate variables and then combine them as a linear mixture that locally varies according to the inferred correlation, reproducing the global multivariate behavior seen on input variables. A real case study is presented, showing the reproduction of the reference multivariate distributions, as well as direct and cross semi-variograms.

了解材料在采矿作业下游过程中的反应，在很大程度上依赖于对这些材料的相关特性进行适当的多变量空间建模。以矿石回收和尾矿及废料的行为为例，掌握矿物成分是关键的一环：前者是为了确保可靠的收益，后者是为了避免处理过程中的环境风险。然而，当变量表现出错综复杂的关系（如非线性相关性、异方差行为或空间趋势）时，多变量空间建模就会变得困难。这项研究表明，通过空间域分解全局非线性行为，转而研究高斯化变量之间的局部相关性，可以再现变量之间复杂的多变量行为。首先从局部邻域推断出局部线性相关关系，然后利用黎曼几何工具对整个域进行插值，从而处理相关矩阵及其空间插值。通过对核心区域化线性模型进行非稳态修改，可以独立模拟变量，然后将它们组合成线性混合物，该混合物根据推断的相关性在局部发生变化，从而再现输入变量的全局多元行为。本文介绍了一个实际案例研究，显示了参考多元分布以及直接和交叉半变量图的再现。

{"title":"Multivariate simulation using a locally varying coregionalization model","authors":"Álvaro I. Riquelme, Julian M. Ortiz","doi":"10.1016/j.cageo.2024.105781","DOIUrl":"10.1016/j.cageo.2024.105781","url":null,"abstract":"<div><div>Understanding the response of materials in downstream processes of mining operations relies heavily on proper multivariate spatial modeling of relevant properties of such materials. Ore recovery and the behavior of tailings and waste are examples where capturing the mineralogical composition is a key component: in the first case, to ensure reliable revenues, and in the second one, to avoid environmental risks involved in their disposal. However, multivariate spatial modeling can be difficult when variables exhibit intricate relationships, such as non-linear correlation, heteroscedastic behavior, or spatial trends. This work demonstrates that the complex multivariate behavior among variables can be reproduced by disaggregating the global non-linear behavior through the spatial domain and looking instead at the local correlations between Gaussianized variables. Local linear dependencies are first inferred from a local neighborhood and then interpolated through the domain using Riemannian geometry tools that allow us to handle correlation matrices and their spatial interpolation. By employing a non-stationary modification of the linear model of coregionalization, it is possible to independently simulate variables and then combine them as a linear mixture that locally varies according to the inferred correlation, reproducing the global multivariate behavior seen on input variables. A real case study is presented, showing the reproduction of the reference multivariate distributions, as well as direct and cross semi-variograms.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"195 ","pages":"Article 105781"},"PeriodicalIF":4.2,"publicationDate":"2024-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142722650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimized AI-MPM: Application of PSO for tuning the hyperparameters of SVM and RF algorithms 优化的 AI-MPM：应用 PSO 调整 SVM 和 RF 算法的超参数

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-11-20 DOI: 10.1016/j.cageo.2024.105785

Mehrdad Daviran , Abbas Maghsoudi , Reza Ghezelbash

Modern computational techniques, particularly Support Vector Machines (SVM) and Random Forest (RF) models, are revolutionizing predictive mineral prospectivity mapping. These advanced systems excel at identifying prime resource locations but require meticulous fine-tuning of their internal settings to achieve peak performance. Careful calibration of these configurations during the learning phase significantly enhances their ability to detect promising deposits. The main goal of this study is to introduce a hybrid model called PSO -SVM and PSO-RF, which aim to combine particle swarm optimization (PSO) with SVM (with RBF kernel) and RF models. This hybrid model automatically adjusts the optimized hyperparameters of SVM and RF, resulting in highly accurate predictions and a wide range of applicability. The PSO algorithm has been applied to fine-tune two main parameters (C and λ) for SVM-RBF and three main parameters (N_T, N_S, and d) for RF, creating efficient models for both. The proposed hybrid model as well as the traditional versions of SVM and RF models, were tested using a geo-spatial dataset related to Cu mineralization in Kerman belt, SE Iran. Forecasting algorithms were developed by integrating diverse datasets: multi-element concentrations from stream samples, bedrock and fault line maps, indicators of hot fluid interaction, aeromagnetic survey results, coordinates of previously identified copper-rich igneous intrusions, and verified ore body positions as reference points. The models' performance was evaluated using four validation methods: Multi-round data partitioning (K-fold), error classification tables (confusion matrix), true-positive vs. false-positive graphical analysis ((ROC) curve), and P-A plot were used to assess algorithms and models performance. Tests revealed that the PSO-SVM surpassed all competitors. Impressively, this fine-tuned classifier identified prime target zones in merely one-seventh of the region (14%), yet these areas encompassed nearly all verified resource sites (97%).

现代计算技术，尤其是支持向量机（SVM）和随机森林（RF）模型，正在彻底改变矿产远景预测绘图。这些先进的系统在确定主要资源位置方面表现出色，但需要对其内部设置进行细致的微调才能达到最佳性能。在学习阶段对这些配置进行仔细校准，可大大提高其探测有潜力矿藏的能力。本研究的主要目标是引入一种名为 PSO -SVM 和 PSO-RF 的混合模型，旨在将粒子群优化（PSO）与 SVM（带 RBF 内核）和 RF 模型相结合。这种混合模型可自动调整 SVM 和 RF 的优化超参数，从而实现高精度预测和广泛的适用性。PSO 算法被用于微调 SVM-RBF 的两个主要参数（C 和 λ）和 RF 的三个主要参数（NT、NS 和 d），为两者创建了高效模型。我们使用与伊朗东南部克尔曼矿带铜矿化相关的地理空间数据集对所提出的混合模型以及传统版本的 SVM 和 RF 模型进行了测试。预测算法是通过整合多种数据集而开发的：溪流样本中的多元素浓度、基岩和断层线图、热流体相互作用指标、航磁勘测结果、先前确定的富铜火成岩侵入体坐标以及作为参考点的已验证矿体位置。使用四种验证方法对模型的性能进行了评估：使用多轮数据分区（K-fold）、误差分类表（混淆矩阵）、真阳性与假阳性图形分析（（ROC）曲线）和 P-A 图来评估算法和模型的性能。测试表明，PSO-SVM 超越了所有竞争对手。令人印象深刻的是，这种经过微调的分类器仅在七分之一的区域（14%）识别出了主要目标区，但这些区域几乎涵盖了所有经过验证的资源点（97%）。

{"title":"Optimized AI-MPM: Application of PSO for tuning the hyperparameters of SVM and RF algorithms","authors":"Mehrdad Daviran , Abbas Maghsoudi , Reza Ghezelbash","doi":"10.1016/j.cageo.2024.105785","DOIUrl":"10.1016/j.cageo.2024.105785","url":null,"abstract":"<div><div>Modern computational techniques, particularly Support Vector Machines (SVM) and Random Forest (RF) models, are revolutionizing predictive mineral prospectivity mapping. These advanced systems excel at identifying prime resource locations but require meticulous fine-tuning of their internal settings to achieve peak performance. Careful calibration of these configurations during the learning phase significantly enhances their ability to detect promising deposits. The main goal of this study is to introduce a hybrid model called PSO -SVM and PSO-RF, which aim to combine particle swarm optimization (PSO) with SVM (with RBF kernel) and RF models. This hybrid model automatically adjusts the optimized hyperparameters of SVM and RF, resulting in highly accurate predictions and a wide range of applicability. The PSO algorithm has been applied to fine-tune two main parameters (<em>C</em> and <em>λ</em>) for SVM-RBF and three main parameters (<em>N</em><sub><em>T</em></sub>, <em>N</em><sub><em>S</em></sub>, and <em>d</em>) for RF, creating efficient models for both. The proposed hybrid model as well as the traditional versions of SVM and RF models, were tested using a geo-spatial dataset related to Cu mineralization in Kerman belt, SE Iran. Forecasting algorithms were developed by integrating diverse datasets: multi-element concentrations from stream samples, bedrock and fault line maps, indicators of hot fluid interaction, aeromagnetic survey results, coordinates of previously identified copper-rich igneous intrusions, and verified ore body positions as reference points. The models' performance was evaluated using four validation methods: Multi-round data partitioning (K-fold), error classification tables (confusion matrix), true-positive vs. false-positive graphical analysis ((ROC) curve), and P-A plot were used to assess algorithms and models performance. Tests revealed that the PSO-SVM surpassed all competitors. Impressively, this fine-tuned classifier identified prime target zones in merely one-seventh of the region (14%), yet these areas encompassed nearly all verified resource sites (97%).</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"195 ","pages":"Article 105785"},"PeriodicalIF":4.2,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142707103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0