Disaster Monitoring Research Articles

Large-scale integration of remotely sensed and GIS road networks: A full image-vector conflation approach based on optimization and deep learning

Road networks play an important role in the sustainable development of human society. Conventionally, there are two sources of road data acquisition: road extraction from Remote Sensing (RS) imagery and GIS based map production. Each method has its limitations. The RS road extraction methods are primarily raster-based and the extracted roads are not directly usable in GIS due to their fragmented and noisy nature, while vector-based methods cannot utilize rich raster information. Further more, the vector and raster data can have discrepancies for various reasons. Efficient road data production requires an image-vector conflation process that can match and combine raster and vector-based road data automatically.In this study, we propose a full image-vector conflation framework that directly integrates image and vector road data by appropriately transforming extracted roads from imagery and establishing a match relation between these roads and a credible target GIS road dataset. Based on analyzing these match relations, we propose new metrics for measuring the degree of agreement between the raster and vector road data. The proposed framework combines state-of-the-art deep learning methods for image segmentation and optimization-based models for object matching. We prepared a large-scale high-resolution road dataset covering two counties in Kansas, US. Using trained models from one of the two counties, we were able to extract road segments in the other county and match them to the TIGER/Line roads.Our experiments show that conventional performance metrics for road extraction (e.g. IoU) are insufficient for measuring the degree of agreement between image and vector roads as they are pixel-based and are too sensitive to spatial displacement. Instead, the newly defined vector-based agreement metrics are needed for image-vector conflation purposes. Experiments show that, by the vector-based metrics, nearly 90% of GIS road lengths in the study area were extracted and over 90% of extracted roads matched the target GIS roads. The new framework streamlines raster-vector conflation of roads and can potentially expedite relevant geospatial analyses regarding change detection, disaster monitoring and GIS data production, among others.

Combining evolutionary computation with machine learning technique for improved short-term prediction of UT1-UTC and length-of-day

Over the years, prediction techniques for the highly variable angular velocity of the Earth represented by Earth's rotation (UT1-UTC) and length-of-day (LOD) have been continuously improved. This is because many applications like navigation, astronomy, space exploration, climate studies, timekeeping, disaster monitoring, and geodynamic studies, all rely on predictions of these Earth rotation parameters. They provide early warning of changes in the Earth's rotation, allowing various industries and scientific fields to operate more precisely and efficiently. Thus, in our study, we focused on short-term prediction for UT1-UTC (dUT1) and LOD. Our prediction approach is to combine machine learning (ML) technique with efficient evolutionary computation (EC) algorithms to achieve reliable and improved predictions. Gaussian process regression (GPR) is used as the ML technique with genetic algorithm (GA) as the EC algorithm. GA is used for hyperparameter optimization of GPR model as selecting appropriate values for hyperparameter are essential to ensure that the prediction model can accurately capture the underlying patterns in the data. We conducted some experiments with our prediction approach to thoroughly test its capabilities. Moreover, two forecasting strategies were used to assess the performance in both hindcast and operational settings. In most of the experiments, the data used are the multi-technique combinations (C04) generated by International Earth Rotation and Reference Systems Service (IERS). In one of the experiments, we also investigated the performance of our prediction model on dUT1 and LOD from four different products obtained from IERS EOP 20 C04, DTRF20, JTRF20 and USNO. The prediction products are evaluated with real estimates of the EOP product with which the model is trained. The combined excitations of the atmosphere, oceans, hydrology, and sea level (AAM + OAM + HAM + SLAM) are used as predictors because they are highly correlated to the input data. The results depict the highest performance of 0.412 ms in dUT1 and 0.092 ms/day in LOD, on day 10 of predictions. It is worth noting that the later predictions were obtained by incorporating the uncertainty of the input data as weights in the prediction model, which was a novel approach tested in this study.Graphical

Risk and Response to Warm and Humidified Geological Disasters in China

Geological hazards caused by warming and humidification are particularly noticeable as the climate moves towards a warmer and more humid situation. Extreme temperatures and extreme rainfall events have led to an increase in the trend of landslides, avalanches and other geologic hazards. Strengthening the research and application of geohazard defense methods and technologies can reduce the occurrence of geohazards and the losses caused by them. This paper discusses the monitoring and prevention technology of geological disasters, analyzing the monitoring and prevention methods of geology.

Exploring the Application of Remote Sensing Technology in Natural Disaster Monitoring

Exploring the Application of Remote Sensing Technology in Natural Disaster Monitoring

CORS station for synergistic monitoring of multivariate surface parameters in expansive soils

Expansive soils cause frequent surface deformation due to their expansion and contraction, which is a serious engineering hazard, and long-term subsidence monitoring is a prerequisite for preventing and controlling expansive soil disasters. Currently, the conventional monitoring methods for the above issue include Interferometric Synthetic Aperture Radar (InSAR) technology, but InSAR is not suitable for uninterrupted monitoring of surface deformation and has low sensitivity. Meanwhile, it can’t obtain multiple surface environmental parameters around the station. The Global Navigation Satellite System (GNSS), a system that can directly acquire surface deformation, has been widely used in landslide disaster monitoring, and in recent years, this technology has also been applied to the field of expansive soil disaster monitoring. At the same time, GNSS can also provide a constant stream of L-band microwave signals to obtain ground environmental information such as precipitable rainfall and soil moisture around the station. In previous studies of expansive soil hazards, GNSS technology has been mainly used to provide surface deformation information without exploring its potential to invert ground environmental information around stations. This paper proposes a ground-based GNSS remote sensing integrated monitoring system that integrates expanding land surface parameters such as “precipitable rainfall, soil moisture, and three-dimensional deformation” and analyses the ability of ground-based GNSS to be used for integrated monitoring of expanding soil hazards by combining ten years of consecutive observational data from GNSS stations along the coastal area of Houston. The experimental results show that the GNSS is capable of providing highly accurate time-series characterization of deformation, and inelastic subsidence in recent years has resulted in a cumulative permanent elevation loss of 2 cm along the Houston coast. The correlation coefficient between soil moisture extracted by the fifth-generation European reanalysis data (ERA5) and soil moisture inverted by ground-based GNSS is 0.514. At the same time, the GNSS was also able to monitor the zenithal precipitable water vapor (PWV) and soil moisture changes around the GNSS station and further analyze the response relationship among the three parameters, which could comprehensively evaluate the stability of expansive soils, avoiding the unreliability of relying on a single piece of monitoring information to assess the stability of expansive soils. We hope to construct a more comprehensive ground-based GNSS remote sensing monitoring system to better monitor expansive soil hazards.

Automated Disaster Monitoring From Social Media Posts Using AI-Based Location Intelligence and Sentiment Analysis

Worldwide disasters like bushfires, earthquakes, floods, cyclones, and heatwaves have affected the lives of social media users in an unprecedented manner. They are constantly posting their level of negativity over the disaster situations at their location of interest. Understanding location-oriented sentiments about disaster situation is of prime importance for political leaders, and strategic decision-makers. To this end, we present a new fully automated algorithm based on artificial intelligence (AI) and natural language processing (NLP), for extraction of location-oriented public sentiments on global disaster situation. We designed the proposed system to obtain exhaustive knowledge and insights on social media feeds related to disaster in 110 languages through AI- and NLP-based sentiment analysis, named entity recognition (NER), anomaly detection, regression, and Getis Ord Gi* algorithms. We deployed and tested this algorithm on live Twitter feeds from 28 September to 6 October 2021. Tweets with 67 515 entities in 39 different languages were processed during this period. Our novel algorithm extracted 9727 location entities with greater than 70% confidence from live Twitter feed and displayed the locations of possible disasters with disaster intelligence. The rates of average precision, recall, and F₁-Score were measured to be 0.93, 0.88, and 0.90, respectively. Overall, the fully automated disaster monitoring solution demonstrated 97% accuracy. To the best of our knowledge, this study is the first to report location intelligence with NER, sentiment analysis, regression and anomaly detection on social media messages related to disasters and has covered the largest set of languages.

An investigation and future prospects of artificial intelligence applications in natural disaster prediction

The monitoring of natural hazards, such as those affecting the oceans, forests and geological formations, is important for safeguarding human life, protecting property, preserving the ecological balance and promoting technological progress. Research on the prediction and prevention of natural disasters has been conducted for decades in a wide range of areas, and this paper summarise recent research on the use of emerging computer technologies for such prediction, with the expectation that it will serve as a recommendation for future developments in the field. Effective prediction of such disasters depends on the ability to detect anomalous signals in a timely manner. Current methods that rely heavily on manual observation and data analysis are inefficient and prone to human error. Rapid advances in computer science and artificial intelligence technologies offer a promising solution, such as the use of deep learning algorithms or the Spark framework to improve model prediction accuracy and response timeliness. In recent years, computer technology has played a key role in improving these capabilities, contributing to better management of the nations natural resources, and enhancing emergency response strategies. By leveraging the capabilities of deep learning, machine learning and others in data processing and analysis, the field of natural disaster monitoring will continue to evolve towards more efficient and reliable methods.

Research on Flood Disaster Monitoring and Early Warning Technology Based on UAV Remote Sensing and AI Algorithm

This paper, based on UAV far-flung sensing science and synthetic brain algorithm, studies the monitoring and early warning technological know-how of flood catastrophes. Firstly, the far-flung sensing monitoring science of flood catastrophes is summarized. Then, blended with geographic records machine (GIS) technology, the UAV far-flung sensing science software in flood catastrophe monitoring is analyzed. Finally, combining the micro-UAV technological know-how and AI technology, the key applied sciences of the micro-UAV flood catastrophe monitoring and early warning device primarily based on AI imaginative and prescient are studied, along with the lookup of the micro-UAV key gadget and the lookup of AI imaginative and prescient key technology. The key device lookup of micro-UAV is the definition of mindset. Mindset A mindset lookup on key applied sciences of AI, imaginative and prescient broadly speaking, expounds the precept of CNN and introduces Faster RCNN. It lays the groundwork for the later lookup on flood catastrophe monitoring and early warning technological know-how primarily based on UAV far-off sensing and AI algorithms.

Landslide Hazard Prediction Based on Small Baseline Subset–Interferometric Synthetic-Aperture Radar Technology Combined with Land-Use Dynamic Change and Hydrological Conditions (Sichuan, China)

Le’an Town, located in the southwest of Qingchuan County, Guangyuan City, Sichuan Province, boasts a unique geographical position. The town’s terrain is complex, and its geological environment is fragile. Multiple phases of tectonic movements have resulted in numerous cracks and faults, making the area prone to landslides, debris flows, and other disasters. Additionally, heavy rainfall and fluctuating groundwater levels further exacerbate the instability of the mountains. Human activities, such as overdevelopment and deforestation, have significantly increased the risk of geological disasters. Currently, the methods for landslide prediction in Le’an Town are limited; traditional techniques cannot provide precise forecasts, and the study area is largely covered by tall vegetation. Therefore, this paper proposes a method that combines SBAS-InSAR technology with dynamic changes in land use and hydrological conditions. SBAS-InSAR technology is used to obtain surface deformation information, while land-use changes and hydrological condition data are incorporated to analyze the dynamic characteristics and potential influencing factors of landslide areas. The innovation of this method lies in its high-precision surface deformation monitoring capability and the integration of multi-source data, which can more comprehensively reveal the geological environmental characteristics of the study area, thereby achieving accurate predictions of landslide development. The study results indicate that the annual subsidence rate in most deformation areas of Le’an Town ranges from −10 to 0 mm, indicating slow subsidence. In some areas, the subsidence rate exceeds −50 mm per year, showing significant slope aspect differences, reflecting the combined effects of geological structures, climatic conditions, and human activities. It is evident that land-use changes and hydrological conditions have a significant impact on the occurrence and development of landslides. Therefore, by utilizing SBAS-InSAR technology and cross-verifying it with other techniques, the consistency of identified landslide deformation areas can be enhanced, thereby improving results. This method provides a scientific basis for the monitoring and early warning of landslide disasters and has important practical application value.

Development of an Algorithm for Assessing the Scope of Large Forest Fire Using VIIRS-Based Data and Machine Learning

Forest fires pose a multifaceted threat, encompassing human lives and property loss, forest resource destruction, and toxic gas release. This crucial disaster’s global occurrence and impact have risen in recent years, primarily driven by climate change. Hence, the scope and frequency of forest fires must be collected to establish disaster prevention policies and conduct relevant research projects. However, some countries do not share details, including the location of forest fires, which can make research problematic when it is necessary to know the exact location or shape of a forest fire. This non-disclosure warrants remote surveys of forest fire sites using satellites, which sidestep national information disclosure policies. Meanwhile, original data from satellites have a great advantage in terms of data acquisition in that they are independent of national information disclosure policies, making them the most effective method that can be used for environmental monitoring and disaster monitoring. The Visible Infrared Imaging Radiometer Suite (VIIRS) aboard the Suomi National Polar-Orbiting Partnership (NPP) satellite has worldwide coverage at a daily temporal resolution and spatial resolution of 375 m. It is widely used for detecting hotspots worldwide, enabling the recognition of forest fires and affected areas. However, information collection on affected regions and durations based on raw data necessitates identifying and filtering hotspots caused by industrial activities. Therefore, this study used VIIRS hotspot data collected over long periods and the Spatio-Temporal Density-Based Spatial Clustering of Applications with Noise (ST-DBSCAN) algorithm to develop ST-MASK, which masks said hotspots. By targeting the concentrated and fixed nature of these hotspots, ST-MASK is developed and used to distinguish forest fires from other hotspots, even in mountainous areas, and through an outlier detection algorithm, it generates identified forest fire areas, which will ultimately allow for the creation of a global forest fire watch system.

Study on the failure modes of water-bearing soft rock under different low-frequency disturbance

To investigate the evolution of failure modes in soft rock under various low-frequency disturbances and varying moisture content, a creep impact dynamic disturbance loading system was employed. Experiments were conducted on soft rock samples with different moisture contents, while the loading process was monitored using an acoustic emission system. The fracture development and propagation characteristics were reconstructed using a CT scanning system. The experimental results reveal that: (1) The failure mode of dry soft rock transitions from “shear-tension-shear” with increasing initial disturbance, whereas water-bearing soft rock primarily undergoes tensile failure. (2) During disturbance, saturated soft rock exhibits significant fluctuations in acoustic emission characteristic parameters, with the initiation of larger fractures. As the initial disturbance increases, failure transitions from plastic to brittle, resembling the behavior of dry samples. (3) Microcracks in both dry and saturated soft rock predominantly range from 500 to 1000 µm, while those in naturally moist soft rock are less than 500 µm. Although the volume of primary fractures varies slightly among different states, the overall degree of damage increases with higher moisture content. These findings elucidate the evolution of failure modes and fracture development in soft rock under low-frequency disturbances, providing valuable insights for monitoring, early warning, and mitigation of dynamic disasters in deep mines.

Early Detection and Stability Assessment of Hazardous Rock Masses in Steep Slopes

The assessment of slope stability plays a critical role in the prevention and management of slope disasters. Evaluating the condition and stability of hazardous rock masses is essential for predicting potential collapses and assessing treatment effectiveness. However, conventional measurement techniques are inadequate in high slope areas, which lack sufficient spatial data to support subsequent calculations and analyses. Therefore, this paper presents a method for the early identification and evaluation of unstable rock masses in high slopes using Unmanned Aerial Vehicle (UAV) digital photogrammetry and geographic information technology. By considering nine evaluation indices including geology, topography, and induced conditions within the study area, weights for each index are determined through an analytic hierarchy process. A semi-automatic approach is then utilized to extract and analyze rock mass stability. The reliability of this early identification method is confirmed by applying the limit equilibrium principle. The findings reveal that 17.6% of dangerous rock masses in the study area fall into the unstable category (W4, W6, W10). This method effectively assesses slope rock mass stability while providing technical support for disaster monitoring systems, warning mechanisms, and railway infrastructure safety defense capability to ensure safe mountain railway operations.

Research on GNSS-IR soil moisture retrieval based on random forest algorithm

Abstract Soil moisture (SM) retrieval is of great significance in climate, agriculture, ecology, hydrology, and natural disaster monitoring, and it is one of the essential hydrometeorological parameters studied in the world at present. With the continuous development of the GNSS, a technique called GNSS-IR became widely used in ground SM inversion. Therefore, based on the frequency, amplitude and phase of signal-to-noise ratio residuals (δSNR), this study takes P037 and P043 stations set by UNAVCO in the United States as examples and develops the research of SM inversion from Random Forest Regression (RFR) prediction. The experimental results show that the retrieval accuracy of SM under different practical schemes can be in descending order: L1 + L2 dual frequency combination > L2 single frequency > L1 single frequency. It is confirmed that the experimental scheme based on the L1+L2 dual-frequency combination is beneficial to the inversion of SM. In the L1+L2 dual-frequency combination, the prediction set accuracy of the P037 station is as follows: R is 0.796, RMSE is 0.032 cm3cm-3, ME is 0.002 cm3cm-3. The prediction accuracy of the P043 station is as follows: R is 0.858, RMSE is 0.039 cm3cm-3, ME is -0.009 cm3cm-3. Among them, the RMSE of the L1+L2 dual-frequency combination of the two stations has an improvement effect of 13%-37% compared with their single-frequency, which has a noticeable improvement effect. The difference between the SM retrieved by GNSS-IR and the reference value of PBO-H2O is concentrated around 0, further showing the accuracy of SM retrieved by GNSS-IR technology. To sum up, this study considers that SM retrieval based on the RFR model has good reliability and accuracy, which makes GNSS-IR technology an efficient means for SM retrieval. With the continuous improvement of the GNSS system and technology, the application of GNSS-IR technology in SM will become broader.

Innovative landslide disaster monitoring: unmanned aerial vehicle-deployed GNSS technology

Real-time monitoring technologies of surface deformation represented by the global navigation satellite system (GNSS) are essential for landslides early warning. Monitoring methods such as GNSS require on-site manual installation. Therefore, it becomes nearly impossible to deploy surface monitoring equipment when landslides are located in high mountain valleys that are challenging for personnel to access, and/or in hazardous situations. We propose an intelligent real-time monitoring and early-warning technology that employs an unmanned aerial vehicle (UAV) as a carrier to deploy GNSS equipment. A system was initially designed with five components: an adaptive sampling GNSS receiver, an intelligent cooperative network transmission module, UAV-dropped GNSS equipment, special delivery UAV, and an intelligent monitoring and early warning cloud platform. This new technology was applied to a real landslide in Gansu Province, China, in 2020. The mean absolute error of 30 days in the east, north, and upward directions were 1.2, 1.3, and 2.9 mm/d, respectively, comparing the deformation velocity between the traditional monitoring station and the UAV-dropped monitoring station. This observation facilitated the successful prediction of a landslide hazard on 27 January 2021, allowing for a timely alert to be issued. Collectively, the results of this study established that UAV-deployed GNSS technology can accomplish unmanned deployment of GNSS equipment for landslide monitoring to ensure early warning issuance in inaccessible and/or high-risk areas.

A dataset of annual TCI, VCI, VHI and TVDI with the resolution of 1 km in the Yellow River Basin (2003–2022)

Primarily situated in arid and semi-arid climatic zones, characterized by naturally limited water resources, the Yellow River Basin is the most drought-affected among China's major basins. As global environmental and climate change continue to evolve, droughts in the Yellow River Basin have become increasingly frequent, making drought monitoring research a current focal point. Based on MODIS vegetation data and surface temperature products, this dataset undergoes quality control procedures such as cloud removal and data reconstruction to collect the data of annual Temperature Condition Index (TCI), Vegetation Condition Index (VCI), Vegetation Health Index (VHI) and Temperature Vegetation Dryness Index (TVDI) in the Yellow River Basin from 2003 to 2022. The spatial scope of this dataset ranges from 32°10′N–41°50′N，95°53′E–119°05′E. The dataset is in GeoTiff format with a spatial resolution of 1km. It can provide foundational data and scientific support for drought disaster monitoring in the Yellow River Basin.

SSA-LHCD: A Singular Spectrum Analysis-Driven Lightweight Network with 2-D Self-Attention for Hyperspectral Change Detection

As an emerging research hotspot in contemporary remote sensing, hyperspectral change detection (HCD) has attracted increasing attention in remote sensing Earth observation, covering land mapping changes and anomaly detection. This is primarily attributable to the unique capacity of hyperspectral imagery (HSI) to amalgamate both the spectral and spatial information in the scene, facilitating a more exhaustive analysis and change detection on the Earth’s surface, proving to be successful across diverse domains, such as disaster monitoring and geological surveys. Although numerous HCD algorithms have been developed, most of them face three major challenges: (i) susceptibility to inherent data noise, (ii) inconsistent accuracy of detection, especially when dealing with multi-scale changes, and (iii) extensive hyperparameters and high computational costs. As such, we propose a singular spectrum analysis-driven-lightweight network for HCD, where three crucial components are incorporated to tackle these challenges. Firstly, singular spectrum analysis (SSA) is applied to alleviate the effect of noise. Next, a 2-D self-attention-based spatial–spectral feature-extraction module is employed to effectively handle multi-scale changes. Finally, a residual block-based module is designed to effectively extract the spectral features for efficiency. Comprehensive experiments on three publicly available datasets have fully validated the superiority of the proposed SSA-LHCD model over eight state-of-the-art HCD approaches, including four deep learning models.

Experimental study on joint sensing and early warning method of landslide disaster based on NPR-OFST

In order to explore more effective methods of landslide disaster monitoring and controlling, NPR anchor cable and optical fiber grating strain sensor are physically combined to form a slope reinforcement-monitoring integration system, with PVC pipe an intermediary. Physical model test is carried out according to timely warning of landslide disasters of the Newtonian force monitoring system. At the same time, the optical fiber sensing technology has the condition of continuous perception of time and space. The feasibility of monitoring the whole life cycle of slope instability and positioning the potential sliding surface is discussed. Through the analysis of the test results, it is concluded that the fiber grating strain sensor can effectively monitor the deep displacement of the slope. PVC pipe and the soil have deformation coordination, which is able to effectively reinforce the slope body together with the combined anchor cable sensing device, and can continuously sense the potential sliding surface in the whole life cycle of the slope. By comparing the results of Newton force monitoring curve with the fiber grating strain monitoring, the mechanical law of “Newton force sudden drops, immediately catastrophe happens” is further verified. The feasibility of the combined sensing and early warning method of optical fiber sensing technology and Newton force monitoring is verified. This method realizes the further optimization of Newton force monitoring system.

Assessing the informativeness of a coupled surface–subsurface watershed model for understanding debris flow: a hydrological perspective

ABSTRACT Characterization of debris flow is critical to both risk assessment and hazard mitigation. Recent technologies enable onsite environmental monitoring sensors for geological disaster monitoring. However, the spatiotemporal understanding of debris flow in remote mountainous areas is still limited due to difficulties in observation networks and its complex driving conditions. Here we apply a coupled surface–subsurface hydrological model to examine the characteristics of the water movements near debris flow sites in southwest China. Our approach captured the temporal dynamics of infiltration, redistribution of soil moisture, groundwater storage, and lateral groundwater fluxes. The lateral groundwater flux and groundwater storage were informative indicators in identifying debris flow location. Such informativeness was only effective when hourly dynamics were analysed. Our findings provide new insight into quantifying debris flow susceptibility. This study suggests that the coupled surface–subsurface watershed modelling approach can be informative for preliminary monitoring, planning and management of debris flow.

A study on the monitoring of landslide deformation disasters in Wenxian County, Longnan City based on different time-series InSAR techniques

A study on the monitoring of landslide deformation disasters in Wenxian County, Longnan City based on different time-series InSAR techniques

Rectilinear Building Footprint Regularization Using Deep Learning

Abstract. Nowadays, deep learning allows to automatically learn features from data. Buildings are one of the most important objects in urban environments. They are used in applications such as inputs to building reconstruction, disaster monitoring, city planing and environment modelling for autonomous driving. However, it is not enough to represent them in raster format, since applications require buildings as polygons. We use an existing, learning based approach to extract building footprints from ortho imagery and digital surface model (DSM) and propose a pipeline for building polygon extraction, which we call primary orientation learning (POL). The first step is to extract initial polygons, that contain a vertex for each pixel in the boundary of the footprint. Afterwards, the two primary orientation angles are regressed continuously. Using these orientation, we insert vertices such that all consecutive edges are perpendicular. To the best of our knowledge, our approach is the first to predict a continuous orientation angle for building boundary regularization. Furthermore, the proposed method is highly efficient with an average processing time of 2.879 ms for a single building.