Search

Co-seismic landslides can cause devastating consequences in seismically active mountainous regions. Rapid assessment of co-seismic landslides is essential in effective emergency response. A dilemma that hinders the assessment is the presence of a significant bias, either overestimating or underestimating the actual landslide hazard. To overcome this dilemma, this paper proposes a two-phase strategy that involves constructing a hierarchical Bayesian model augmenting data from a new earthquake with historical co-seismic landslide inventories. Our findings demonstrate that even with limited and evenly distributed co-seismic data from the new earthquake, it is possible to achieve an unbiased and accurate assessment. The two-phase strategy can facilitate the rapid and unbiased assessment of co-seismic landslides, thereby informing policy-making and resource allocation. © 2024 Elsevier B.V.

Vibratory penetration was successfully used to install 120 integral thin-walled steel cylinders 22 m in diameter in the Hong Kong-Zhuhai-Macao Bridge. However, the eight-hammer group was found overpowered at the western island and underpowered at the eastern site due to insufficient understanding of the mechanisms of vibratory piling. In this study, energy analysis was conducted to reveal the energy characteristics of vibratory penetration, including the periodical energy superposition along the cylinder shaft, the continuous energy consumption in the soil, and the dynamic equilibrium of the total energy. The influences of vibratory soil resistance and loading frequency on the energy distributions have been thoroughly discussed. In contrast to the notion that larger input energy leads to a faster penetration, it is revealed that the vibratory penetration velocity positively correlates to the energy stored in the cylinder, the ratio of kinetic to strain energy, and the characteristic frequencies of the soil-pile system. The actual output power of the vibratory motors is influenced by the vibratory force, the ultimate soil resistance, and the soil mobilization degree. The vibration frequency is optimized as 31.1 Hz for Cylinder E9 and 15 Hz for Cylinder W36 to ensure efficiency and safety. © 2023

Researchers view vast zeros in single-cell RNA-seq data differently: some regard zeros as biological signals representing no or low gene expression, while others regard zeros as missing data to be corrected. To help address the controversy, here we discuss the sources of biological and non-biological zeros; introduce five mechanisms of adding non-biological zeros in computational benchmarking; evaluate the impacts of non-biological zeros on data analysis; benchmark three input data types: observed counts, imputed counts, and binarized counts; discuss the open questions regarding non-biological zeros; and advocate the importance of transparent analysis.

Researchers view vast zeros in single-cell RNA-seq data differently: some regard zeros as biological signals representing no or low gene expression, while others regard zeros as missing data to be corrected. To help address the controversy, here we discuss the sources of biological and non-biological zeros; introduce five mechanisms of adding non-biological zeros in computational benchmarking; evaluate the impacts of non-biological zeros on data analysis; benchmark three input data types: observed counts, imputed counts, and binarized counts; discuss the open questions regarding non-biological zeros; and advocate the importance of transparent analysis.

Yarlung Tsangpo on the Tibetan Plateau is the highest large river in the world. Valley glaciers in the Yarlung Tsangpo Grand Canyon pose high risk to the infrastructure in the region. In particular, the Zelunglung glaciers at the entrance of the canyon are experiencing rapid movements in the warming climate. This study presents a comprehensive assessment of glacier detachment risks based on remote sensing and physically based numerical analysis. Long-term meteorological records and debris-ice rheological modeling are used to reveal the volume and thickness of the instable source mass. An erosion-deposition debris flow analysis simulation program is employed to analyze multi-scenario hazard dynamic processes to predict the run-out distance, velocity, and deposition distributions. The deposition distributions along the river course in different glacier detachment scenarios are quantified to assess the size of possible barrier dam, which lays a good foundation for a quantitative risk assessment.

Under the changing climate, extremely massive landslides, destructive avalanches, glacial debris flows, and dam-breaching outburst floods occur frequently on the Tibetan Plateau, posing severe threats to valley-crossing bridges of the railways and highways. Previous studies examined bridge-landslide interactions corresponding to separate geohazard scenarios. However, in reality, hazard type can transform from one to another. Considering generic physical mechanisms of bridge-landslide interactions under different geohazard types or evolutions of hazard chains, a unified vulnerability assessment framework is established in this study. The resultant impact forces in various geohazards are categorized into three components: boulder impact, coarse particle impact, and slurry impact. Four failure modes of the bridge are revealed under different combinations of impact forces. The integrated vulnerability of a bridge is calculated as the union failure probability of different failure modes. As an example, the proposed methodology framework is applied to assess the vulnerability of the Zhibai Bridge, which was damaged by a high mobility debris flow on September 10, 2020.

Predicting landslide dam formation is essential in mitigating landslide risks in alpine valley regions. This study assesses the landslide damming probability with the consideration of landslide characteristics, valley topography, and hydrological factors using machine learning techniques. A landslide inventory is collected, including both damming landslides and non-damming landslides in the 2008 Wenchuan earthquake region and the Bailong River basin. Three machine learning algorithms are compared, including logistic regression, random forest, and support vector machine. Results show that machine learning techniques can well predict the landslide damming probability. The random forest model achieves the best prediction performance, followed by logistic regression and support vector machine. Among six learning features, landslide area, upstream watershed area, and valley floor width are the three most important variables for landslide dam formation. An illustration example of the Tangjiashan landslide dam is used to demonstrate how the developed model can be integrated to predict landslide dam formation.

In waterway regulation engineering, submerged reefs are quite susceptible to the repetitive oblique strike from drop hammer and impact drilling, leading to a repetitive compression–shear impacting. Understanding the dynamic responses and failure mechanism of saturated rocks subjected to repetitive compression–shear impacting is thus of great significance for the efficient removal of submerged reefs. In this study, using the split Hopkinson pressure bar (SHPB) apparatus, the repetitive compression–shear impacting tests were conducted on the saturated and dry sandstone specimen from the harbor excavation project in Guoyuan Port (Chongqing, China). Our experimental results indicate that the increasing loading rate or shear component in the dynamic loading has a negative influence on the bearing capacity of saturated sandstone subjected to repetitive impacting. The deformability of saturated rocks increases as the impacting number or the shear component in the dynamic loading. The energy absorption ability as well as the energy utilization of the saturated rocks is highly improved by implementing repetitive compression–shear impacting, and such improvement yields an ascension limit with the increasing shear component in the dynamic loading. As the shear component introduces into the repetitive impacting loading, the failure patterns of rocks change from the relative tensile mode with a large deformation band along the loading direction to the combined compression–shear mode with two parallel large deformation bands inclined to the loading direction, the water content between the mineral grains promotes the shear failure in rock dynamic testing. These results can provide a better understanding of the dynamic responses and failure properties of saturated rocks subjected to repetitive compression–shear impacting. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature.

Detection of dynamic triggering of earthquakes combined with microseismic monitoring allows a better un-derstanding on rock damage process in rock engineering, which involve local stress disturbances caused by far -field earthquakes. However, rapid and efficient detection of dynamic triggering of earthquakes remain clueless due to the lack of enough investigation. This study proposed a novel method for automatic detection of dynamic triggering of earthquakes based on convolutional neural networks (CNNs). Results show that the trained model is capable of detecting earthquakes with a high accuracy rate of 98.3%, which provided strong supports for the automatic detection of dynamic triggering of earthquakes. This method is applied to the Xiaojiang fault region, and dynamically triggered earthquakes were spotted for 31 of the 47 selected earthquakes. Our method achieves 94% recognition accuracy on the detection of dynamic triggering of earthquakes. This method offers a rapid and accurate framework to detect dynamic triggering of earthquakes and investigate the damage to rock engineering structures caused by far-field earthquakes based on microseismic monitoring.

A critical challenge in microbiome data analysis is the existence of many non-biological zeros, which distort taxon abundance distributions, complicate data analysis, and jeopardize the reliability of scientific discoveries. To address this issue, we propose the first imputation method for microbiome data-mbImpute-to identify and recover likely non-biological zeros by borrowing information jointly from similar samples, similar taxa, and optional metadata including sample covariates and taxon phylogeny. We demonstrate that mbImpute improves the power of identifying disease-related taxa from microbiome data of type 2 diabetes and colorectal cancer, and mbImpute preserves non-zero distributions of taxa abundances.

Next generation sequencing (NGS) has revolutionized biomedical research and has a broad impact and applications. Since its advent around 15 years ago, this high scalable DNA sequencing technology has generated numerous biological data with new features and brought new challenges to data analysis. For example, researchers utilize RNA sequencing (RNA-seq) technology to more accurately quantify the gene expression levels. However, the NGS technology involves many processing steps and technical variations when measuring the expression values in the biological samples. In other words, the NGS data researchers observed could be biased due to the randomness and constraints in the NGS technology. This dissertation will mainly focus on microbiome sequencing data and single-cell RNA-seq (scRNA-seq) data. Both of them are highly sparse matrix-form count data. The zeros could either be biological or non-biological, and the high sparsity in the data have brought challenges to data analysis. Missing data imputation problem has been studied in statistics and social science as the survey data often experience non-response to some of the survey questions and those unresponded questions will be marked as "NA" or missing values in the data. Imputation methods are used to provide a sophisticated guess for the missing values, and the purpose is to avoid discarding the collected samples and for the ease of using the state-of-the-art statistical methods. In machine learning, the famous Netflix data challenge regarding film recommendation system also falls into the missing data imputation problem category. Netflix wants to find a way to predict users' fondness of the movies they have not watched. The potential scores these users would give to the unwatched films are regarded as missing values in the data. NGS data imputation problem is different from the previous two cases in that the missing values in the NGS data are not so well-defined. The zeros in the NGS data could either come from t

Next generation sequencing (NGS) has revolutionized biomedical research and has a broad impact and applications. Since its advent around 15 years ago, this high scalable DNA sequencing technology has generated numerous biological data with new features and brought new challenges to data analysis. For example, researchers utilize RNA sequencing (RNA-seq) technology to more accurately quantify the gene expression levels. However, the NGS technology involves many processing steps and technical variations when measuring the expression values in the biological samples. In other words, the NGS data researchers observed could be biased due to the randomness and constraints in the NGS technology. This dissertation will mainly focus on microbiome sequencing data and single-cell RNA-seq (scRNA-seq) data. Both of them are highly sparse matrix-form count data. The zeros could either be biological or non-biological, and the high sparsity in the data have brought challenges to data analysis. Missing data imputation problem has been studied in statistics and social science as the survey data often experience non-response to some of the survey questions and those unresponded questions will be marked as "NA" or missing values in the data. Imputation methods are used to provide a sophisticated guess for the missing values, and the purpose is to avoid discarding the collected samples and for the ease of using the state-of-the-art statistical methods. In machine learning, the famous Netflix data challenge regarding film recommendation system also falls into the missing data imputation problem category. Netflix wants to find a way to predict users' fondness of the movies they have not watched. The potential scores these users would give to the unwatched films are regarded as missing values in the data. NGS data imputation problem is different from the previous two cases in that the missing values in the NGS data are not so well-defined. The zeros in the NGS data could either come from t

Next generation sequencing (NGS) has revolutionized biomedical research and has a broad impact and applications. Since its advent around 15 years ago, this high scalable DNA sequencing technology has generated numerous biological data with new features and brought new challenges to data analysis. For example, researchers utilize RNA sequencing (RNA-seq) technology to more accurately quantify the gene expression levels. However, the NGS technology involves many processing steps and technical variations when measuring the expression values in the biological samples. In other words, the NGS data researchers observed could be biased due to the randomness and constraints in the NGS technology. This dissertation will mainly focus on microbiome sequencing data and single-cell RNA-seq (scRNA-seq) data. Both of them are highly sparse matrix-form count data. The zeros could either be biological or non-biological, and the high sparsity in the data have brought challenges to data analysis. Missing data imputation problem has been studied in statistics and social science as the survey data often experience non-response to some of the survey questions and those unresponded questions will be marked as "NA" or missing values in the data. Imputation methods are used to provide a sophisticated guess for the missing values, and the purpose is to avoid discarding the collected samples and for the ease of using the state-of-the-art statistical methods. In machine learning, the famous Netflix data challenge regarding film recommendation system also falls into the missing data imputation problem category. Netflix wants to find a way to predict users' fondness of the movies they have not watched. The potential scores these users would give to the unwatched films are regarded as missing values in the data. NGS data imputation problem is different from the previous two cases in that the missing values in the NGS data are not so well-defined. The zeros in the NGS data could either come from t

A critical challenge in microbiome data analysis is the existence of many non-biological zeros, which distort taxon abundance distributions, complicate data analysis, and jeopardize the reliability of scientific discoveries. To address this issue, we propose the first imputation method for microbiome data-mbImpute-to identify and recover likely non-biological zeros by borrowing information jointly from similar samples, similar taxa, and optional metadata including sample covariates and taxon phylogeny. We demonstrate that mbImpute improves the power of identifying disease-related taxa from microbiome data of type 2 diabetes and colorectal cancer, and mbImpute preserves non-zero distributions of taxa abundances.

Glacial debris flows are a common catastrophic hazard in alpine regions due to glacier recession and permafrost thawing. They are frequently triggered by heavy rainfall and outburst floods. To date, studies on debris flows originated from debris-ice mixture slope failures triggered by glacier melting in low-intensity rainfall conditions are still limited. In this study, a typical glacial debris flow that occurred in September 2020 on the west side of Namcha Barwa in southeast Tibet, China, is taken as a benchmark to reveal the initiation mechanisms and dynamics of debris-ice landslide-debris flow hazard chains. The flow discharge, erosion, entrainment, deposition, and dynamic process of the debris flow are evidenced by both field investigations and numerical simulations with a single-phase cell-based simulation program, Erosion-Deposition Debris Flow Analysis (EDDA) (Chen and Zhang, 2015) and a multi-phase flow model r.avaflow (Pudasaini and Mergili, 2019). This hazard chain was originated from a debris-ice mixture slope failure with a volume of 1.14x 106 m3 in a temperate area under little rainfall. The mobile mixture entrained saturated moraines and water in the ravine channel and rapidly transformed into a viscous debris flow, traveling about 9.78 km and entraining 50% more materials along the path. The peak discharge and frontal flow velocity at the outlet of the ravine reached 4700 m3/s and 11.4 m/s, respectively. About 3.75x 105 m3 of debris deposited on the riverbed and partially blocked the Yarlung Tsangpo, causing severe damage to local infrastructures. This study provides a scientific basis for understanding the evolution process of a debris-ice landslide-debris flow hazard chain and for minimizing its adverse impact on key infrastructures in the region.

On February 7, 2021, a catastrophic landslide occurred in Chamoli, India on the southern hills of the Himalayas (30.37 degrees N, 79.73 degrees E). About 28 x 10(6) m(3) of landslide mass detached from the mountain face, entrained deposits in the valley and riverbed, and generated a huge debris flood along the Dhauliganga River. In this study, the geomorphological and volumetric characteristics of the disaster chain were interpreted from satellite images. The full process of the disaster chain and the erosion, deposition and flow discharge processes were reproduced using a cell-based analysis program, Erosion-Deposition Debris Flow Analysis (EDDA). The results indicate a peak flow discharge of the debris flood at Tapovan Hydropower Station of about 25,000-28,000 m(3)/s. The main erosion and deposition zones are distributed along the valley floor and the Dhauliganga River, respectively. This study serves as basis for understanding the disaster chain dynamics in high mountain areas.

On February 7, 2021, a catastrophic landslide occurred in Chamoli, India on the southern hills of the Himalayas (30.37 degrees N, 79.73 degrees E). About 28 x 10(6) m(3) of landslide mass detached from the mountain face, entrained deposits in the valley and riverbed, and generated a huge debris flood along the Dhauliganga River. In this study, the geomorphological and volumetric characteristics of the disaster chain were interpreted from satellite images. The full process of the disaster chain and the erosion, deposition and flow discharge processes were reproduced using a cell-based analysis program, Erosion-Deposition Debris Flow Analysis (EDDA). The results indicate a peak flow discharge of the debris flood at Tapovan Hydropower Station of about 25,000-28,000 m(3)/s. The main erosion and deposition zones are distributed along the valley floor and the Dhauliganga River, respectively. This study serves as basis for understanding the disaster chain dynamics in high mountain areas.

Recommendation systems underpin the serving of nearly all online content in the modern age. From Youtube and Netflix recommendations, to Facebook feeds and Google searches, these systems are designed to filter content to the predicted preferences of users. Recently, these systems have faced growing criticism with respect to their impact on content diversity, social polarization, and the health of public discourse. In this work we simulate the recommendations given by collaborative filtering algorithms on users in the MovieLens data set. We find that prolonged exposure to system-generated recommendations substantially decreases content diversity, moving individual users into "echo-chambers" characterized by a narrow range of content. Furthermore, our work suggests that once these echo-chambers have been established, it is difficult for an individual user to break out by manipulating solely their own rating vector.

Live streaming platforms need to store a lot of recorded live videos on a daily basis. An important problem is how to automatically extract highlights (i.e., attractive short video clips) from these massive, long recorded live videos. One approach is to directly apply a highlight extraction algorithm to video content. However, algorithmic approaches are either domain-specific, which require experts to spend a long time to design, or resource-intensive, which require a lot of training data and/or computing resources. In this paper, we propose Lightor, a novel implicit crowdsourcing approach to overcome these limitations. The key insight is to collect users' natural interactions with a live streaming platform, and then leverage them to detect highlights. Lightor consists of two major components. Highlight Initializer collects time-stamped chat messages from a live video and then uses them to predict approximate highlight positions. Highlight Extractor keeps track of how users interact with these approximate highlight positions and then refines these positions iteratively. We find that the collected user chat and interaction data are very noisy, and propose effective techniques to deal with noise. Lightor can be easily deployed into existing live streaming platforms, or be implemented as a web browser extension. We recruit hundreds of users from Amazon Mechanical Turk, and evaluate the performance of Lightor using two popular games in Twitch. The results show that Lightor can achieve high extraction precision with a small set of training data and low computing resources.

Crowdsourcing is a multidisciplinary research area including disciplines like artificial intelligence, human-computer interaction, database, and social science. To facilitate cooperation across disciplines, reproducibility is a crucial factor, but unfortunately, it has not gotten enough attention in the HCOMP community. In this paper, we present Reprowd, a system aiming to make it easy to reproduce crowdsourced data processing research. We have open sourced Reprowd at http://sfu-db.github.io/reprowd/., Comment: HCOMP 2016 Work in Progress