Spatio-temporal Clustering Algorithm Research Articles

A joint spatio-temporal characterization of the major meteorological droughts in Europe

Abstract Drought can be considered one of the most severe and complex weather-related natural hazards. It is a relevant stressor for ecosystems, affecting vegetation, ecosystem productivity, and water and carbon cycles, with a complex web of related impacts. Despite the interlink between the spatial and temporal scales of droughts, these two aspects are often studied separately. In addition, studies generally focus on detecting the events, without trying to investigate similarities among them. In this work, we introduce a set of tools used to summarize the main properties of major droughts in Europe, with the goal of subdividing the events in groups characterized by similar properties. We used a European dataset of meteorological droughts (from 1981 to 2020) that detects events based on the Standardized Precipitation Index using an event-oriented spatio-temporal clustering algorithm. From the analysis, we identified three groups of major meteorological droughts: a first group that is comprised by warm-season events, characterized by a longer duration, a shorter early growing phase, and a longer exhaustion phase; a second group, less numerous, comprised by droughts occurring during the cold season, that tend to have a shorter duration, a longer early growing phase and a shorter exhaustion phase; and a third group comprised of droughts occurring across the two periods. This last class is characterized by a longer duration and a high variability in most of the other characteristics, suggesting that these events may be associated with a large range of driving mechanisms. The proposed procedure allows for a drought classification that can be used for better understanding the mechanisms behind spatio-temporal evolution of these events.

Dynamic relationships and propagation characteristics between meteorological drought and vegetation drought based on a three-dimensional identification algorithm

Drought is a regional phenomenon and progressive in specific dimensions of time and space, with significant continuity and dynamic characteristics on a spatiotemporal scale. The meteorological drought acts as a driving factor for vegetation drought, and studying the response characteristics of vegetation drought to meteorological drought is crucial for comprehending the mechanisms of drought evolution. In this study, based on a three-dimensional spatiotemporal clustering algorithm, meteorological and vegetation drought events in the Yellow River Basin (YRB) from 1982 to 2020 were identified, and the dynamic characteristics of typical drought events were depicted, elucidating the inherent correlations and propagation features between meteorological and vegetation drought events. Additionally, according to the spatiotemporal matching criteria of drought events, the meteorological-vegetation drought event pairs were successfully matched, which can reveal the lag time between meteorological drought and vegetation drought. From a three-dimensional perspective, we revealed the dynamic propagation characteristics of meteorological and vegetation drought events, which could provide an effective way for monitoring and mitigating regional vegetation drought, restoring ecosystem functionality, and promoting sustainable development of the ecological environment.

Presenting the AI models in predicting the settlement of earth dams using the results of spatiotemporal clustering and k-means algorithm

In this work, the results of instrumentation over 8 years, including the phases of construction, first impounding, and operation, have been used to analyze the location of the Eyvashan Dam settlement. Mohr–Coulomb behavioral model and numerical model of Plaxis 2D software were used to verify the monitoring results. The results demonstrated that settlement of the dam has increased in the dam's core since the beginning of construction, and they eventually stabilized during the operation phase. After the completion of the construction phase, the maximum settlement of the dam core was recorded as 809 mm, which is equivalent to 1.2% of the height of the dam at the middle level. Also, an approach to interpreting the settlement behavior of earth dams has been presented that is based on spatiotemporal clustering. Also, RF, MARS, and GMDH models were created based on a proposed scenario to predict settlement using points located in a cluster. Therefore, the settlement location of the studied dam was determined using the results of the k-means clustering algorithm in the aforementioned AI models. The high accuracy of the results of the proposed method confirms the proper performance of using AI models in predicting and diagnosing the settlement of earthen dams using the results of k-means spatiotemporal clustering algorithm. The evaluation of the models shows that the ENN model is a more suitable and efficient tool in this field and can be useful in monitoring the settlement of earth dams.

Intelligent Caching for Vehicular Dew Computing in Poor Network Connectivity Environments

In vehicular networks, some edge servers may not function properly due to the time-varying load condition and the uneven computing resource distribution, resulting in a low quality of caching services. To overcome this challenge, we develop a Vehicular dew computing (VDC) architecture for the first time by combining dew computing with vehicular networks, which can achieve wireless communication between vehicles in a resource-constrained environment. Consequently, it is crucial to develop an adaptive caching scheme that empowers vehicles to form efficient cooperation in VDC. In this paper, we propose an intelligent caching scheme based on VDC architecture, which includes two parts. First, to meet the dynamic nature of VDC, a spatiotemporal vehicle clustering algorithm is proposed to establish adaptive cooperation to assist content caching for vehicles. Second, the multi-armed bandit algorithm is employed to select suitable content for caching in vehicles based on real-time file popularity, and a model is established to dynamically update each vehicle’s request preferences. Extensive experiments are conducted to demonstrate that the proposed scheme has excellent performance in terms of cluster head stability and cache hit rate.

Observed links between heatwaves and wildfires across Northern high latitudes

Data on Arctic and Sub-Arctic summer heat events are limited due to the sparse network of surface observation stations. Here, we analyze large heat events within 60°–80°N using land surface temperature (LST) observations from the moderate resolution imaging spectroradiometer (MODIS) sensor aboard the Terra satellite. Our heatwave (HW) detection method uses exceedances of the climatological 90th percentile of LST across summer months, and a spatio-temporal density-based clustering algorithm to distinguish space-time coherent events across Northern Hemispheric high latitudes for the summers of 2000–2022. We find a close link between HW duration, spatial extent and amplitude across the study region (correlations ranging from 0.63 to 0.73). MODIS-derived burned area data show that wildfire seasons are significantly correlated to summer HW activity, particularly in Siberia (r = 0.87 at p < 0.05) and Alaska and NW Canada (r = 0.45 at p < 0.05), and are also spatially co-located. MODIS active fire data also show substantial increase during larger heat events. For the strongest HWs in Siberia, the peak in daily fire count (from the MODIS active fire archive) lags behind peak HW activity. We conclude that there is a close link between intense fire summers and extensive HWs over boreal and shrub tundra regions (Interior Alaska, the Canadian Prairies and Eastern Siberia).

Deformation prediction of large-scale civil structures using spatiotemporal clustering and empirical mode decomposition-based long short-term memory network

Structural deformation prediction is important for maintaining the serviceability and safety of civil infrastructure. However, current deep learning-based single prediction models face difficulties in accurately tracking deformation evolution of large-scale civil structures with numerous diverse monitoring data. In this study, a novel deformation prediction framework is proposed based on a spatiotemporal clustering algorithm and an empirical mode decomposition (EMD)-based long short-term memory (LSTM) network. Experiments on a large excavated foundation pit show that the proposed framework has a prediction error of RMSE less than 0.4 mm, outperforming LSTM models without using spatiotemporal clustering or EMD. Such high-precision predictions at many different monitoring locations demonstrate its promising application on large-scale structures. More deformation prediction experiments are required in the future to further validate the capability and performance of the proposed framework.

Clustering Based on Eye Tracking Data for Depression Recognition

Attention-based approach would be a good way of detecting depression, assisting medical diagnosis and treating the patients at risk earlier. In this paper, a new approach of recognizing depression is proposed, which avoids eye movement event identification and directly performs clustering based on eye tracking data to obtain Regions of Interesting (ROIs), and then conducts depression recognition modelling. Based on these, a novel spatio-temporal clustering algorithm was proposed, i.e. ROI Clustering with Deflection Elimination, which takes the noisy data into consideration to betterly describe attention patterns. On the dataset with 45 depression patients and 44 healthy controls, the proposed algorithm achieved the best classification accuracy of 76.25%, which has the potential to provide methodological reference on the assessment of mental disorders based on eye movements.

A Novel and Efficient Spatio-Temporal Clustering Algorithm to Discover Patterns in Indian Earthquake Data

The seismic activity in India, including the Himalayas, the North-East region, and the Andaman-Nicobar Islands, is prominently displayed on the seismic map. It is important to analyze the specific characteristics of seismic events on the Indian sub-continent. This research paper introduces a new algorithm for data analysis using a partitioning technique that mines clustering patterns. The algorithm generates clusters of spatial and spatio-temporal data by distributing the data into spatial bins or buckets, identifying neighboring bins, and minimizing distance calculations. Furthermore, the selection of centroids is based on the density of data in the spatial region, utilizing a random selection method. It requires minimal parameter settings, enhancing its efficiency and practicality for analyzing large-scale spatio-temporal datasets. We validate the algorithm's efficacy through experiments conducted on the Indian earthquake spatio-temporal dataset, demonstrating its ability to detect spatio-temporal patterns and identify earthquake-prone regions.The conducted experiments reveal a correlation between the frequency of earthquake events and the number of clusters formed, indicating that regions with a higher occurrence of earthquakes exhibit a greater clustering tendency, signifying their susceptibility to seismic activity. The results imply promising clustering quality, with Silhouette index in the range of 0.88 to 0.93.

Identifying spatial heterogeneity of COVID-19 transmission clusters and their built-environment features at the neighbourhood scale

The identification of high-risk areas for infectious disease transmission and its built-environment features are crucial for targeted surveillance and early prevention efforts. While previous research has explored the association between infectious disease incidence and urban built environment, the investigation of spatial heterogeneity of built-environment features in high-risk areas has been insufficient. This paper aims to address this gap by analysing the spatial heterogeneity of COVID-19 clusters in Shanghai at the neighbourhood scale and examining associated built-environment features. Using a spatiotemporal clustering algorithm, the study analysed 1395 reported cases in Shanghai from March 6 to March 17, 2022. Both global Poisson regression (GPR) and geographically weighted Poisson regression (GWPR) models were applied to examine the association between built-environment variables and the size of COVID-19 clusters. Our findings suggest that larger COVID-19 clusters emerging in the suburbs compared with the downtown and multiple built-environment features are significantly associated with this pattern. Specifically, neighbourhoods with a higher proportion of commercial, public service and industrial land, higher centrality of metro stations, and proximity to hospitals are positively associated with larger COVID-19 clusters, while neighbourhoods with higher land use mix and green/open spaces density are associated with smaller COVID-19 clusters. Moreover, we identified that metro stations with high centrality present the highest risk in the downtown, while commercial and public service places exhibit the highest risk in the suburbs. By highlighting the overlooked spatial heterogeneity of built-environment features for high-risk areas, this study aims to provide valuable guidance for public health departments in implementing place-based interventions to effectively prevent the spread of potential epidemics.

Exploring year-to-year spatiotemporal changes in cycling patterns for bike-sharing system in the pre-, during and post-pandemic periods

The bike-sharing system plays an important role during the COVID-19 pandemic due to its benefit of reducing the risk of infection. Previous studies have employed bike trip datasets to analyze the impact of COVID-19 on spatiotemporal cycling mobility, which was limited to a short pandemic period. An understanding of the long-term impact of the pandemic on human travel patterns is urgently needed. This paper presents a framework for investigating the long-term spatiotemporal change in cycling patterns in the pre-, during and post-pandemic periods. A spatiotemporal clustering algorithm is combined with spatiotemporal visualization, network analysis, and important location identification to explore latent mobility patterns. The strength of the proposed framework is that it can model a unified representation of cycling networks across different years, and provide insightful spatiotemporal patterns that allow easy interpretation. The experiments are conducted on five years of data collected from the ‘Citibike’ system in New York City. The results demonstrate the efficiency of the proposed approach in exploring the evolutionary cycling patterns of how people respond to the COVID-19 pandemic and are beneficial for various stakeholders.

A novel spatio-temporal clustering algorithm with applications on COVID-19 data from the United States

A new clustering algorithm for spatio-temporal data is developed. The proposed method leverages a weighted combination of a spatial haversine distance matrix and a spectral-density based temporal distance matrix between the locations. Concepts of partition around medoids algorithm and the gap statistic are utilized to develop the algorithm and to determine the optimal number of clusters. Such a non-parametric algorithm is novel as it incorporates both spatial and temporal distances of the units and it can work for time-series of possibly different lengths. Theoretical guarantee of consistency of the proposed method is provided. An elaborate simulation study is also given to demonstrate the efficacy of the algorithm. As an interesting real life application, the proposed algorithm is implemented to analyze the spatio-temporal dynamics of the time series of coronavirus (COVID-19) incidence rates observed at county-level in the United States of America. The results are demonstrated on datasets of different sizes: the entire country, the Midwest region and the state of California. Special emphasis is given on the last two cases to display how the clustering results offer interesting insights into the epidemic progression in these areas. Particularly, it sheds light on whether state-mandated restrictions impacted the entire state similarly or if there are interesting local behaviors in terms of the COVID-19 spread.

A Travel Demand Response Model in MaaS Based on Spatiotemporal Preference Clustering

To respond to travel demand in the MaaS system, improve transport efficiency, and optimize the framework of MaaS, we propose a travel demand response model based on a spatiotemporal preference clustering algorithm that considers the impact of travel preferences and features of the MaaS system to improve travel demand response and achieve full coverage of travel demands. Specifically, in the MaaS system, the time preference hierarchical clustering algorithm is optimized with travel preference as the perception factor and preference priority order as the iteration index. Then, we cluster the departure and arrival times of reservation demand points and iteratively analyze the discrete points to obtain the set of reservation demand points with convergent time dimensions under similar preferences. Then, the spatial preference DBSCAN clustering algorithm is improved with travel preference and preference priority order as the iteration indices, and spatial clustering of the time-dense points are updated by the silhouette coefficient to obtain reservation demand points with similar spatiotemporal preference and respond to the demands. Meanwhile, traffic resources are coordinated by the MaaS system and the flexible means of transport are deployed to spatiotemporal discrete points to achieve full coverage of travel demand. Simulation shows that when the neighborhood range is 0.5 km and the least number of reservation demand sites is 3, our spatiotemporal model achieves a response rate of reservation demand points at 95%, and a demand coverage rate of 100%, which is 15% and 6.7% higher than the hierarchical clustering model and the DBSCAN clustering model, respectively. The demand response rate is also improved compared to the spatiotemporal clustering model in the customized bus model. The model and algorithm have some applicability and can be applied to areas with fixed, semifixed and flexible route transport, thereby considerably improving the travel demand response efficiency and transport service quality.

Optimal capacity configuration model of power-to-gas equipment in wind-solar sustainable energy systems based on a novel spatiotemporal clustering algorithm: A pathway towards sustainable development

For wind-solar sustainable energy systems with a large amount of abandoned wind and solar energy and carbon dioxide emissions, a power-to-gas equipment can be introduced to synthesize methane together, which is an effective way to alleviate the greenhouse effect, improve the utilization rate of new energy, and promote sustainable development. Focusing on how to perform the optimal capacity configuration of the newly introduced power-to-gas equipment more accurately and simply, this paper make progress through a more similar scenario reduction and a more accurate power-to-gas capacity configuration model. First, in view of the problem that the existing scenario reduction methods are difficult to take into account the coherent temporal characteristics and spatial amplitude characteristics, a novel multi-source-load double-layer spatiotemporal clustering algorithm is proposed. Second, in view of the problem that the power-to-gas equipment can only capture the actual carbon dioxide, a more comprehensive model considering the coupling relationship of multi-energy flows such as electricity-natural gas-hydrogen-oxygen-actual carbon dioxide-virtual carbon dioxide is proposed. The example analysis shows that the proposed strategies can reduce the error rate of the power-to-gas capacity configuration with a similar calculation amount.

Unsupervised change detection of man-made objects using coherent and incoherent features of multi-temporal SAR images

Constrained by complex imaging mechanism and extraordinary visual appearance, change detection with synthetic aperture radar (SAR) images has been a difficult research topic, especially in urban areas. Although existing studies have extended from bi-temporal data pair to multi-temporal datasets to derive more plentiful information, there are still two problems to be solved in practical applications. First, change indicators constructed from incoherent feature only cannot characterize the change objects accurately. Second, the results of pixel-level methods are usually presented in the form of the noisy binary map, making the spatial change not intuitive and the temporal change of a single pixel meaningless. In this study, we propose an unsupervised man-made objects change detection framework using both coherent and incoherent features derived from multi-temporal SAR images. The coefficients of variation in time-series incoherent features and the man-made object index (MOI) defined with coherent features are first combined to identify the initial change pixels. Afterwards, an improved spatiotemporal clustering algorithm is developed based on density-based spatial clustering of applications with noise (DBSCAN) and dynamic time warping (DTW), which can transform the initial results into noiseless object-level patches, and take the cluster center as a representative of the man-made object to determine the change pattern of each patch. An experiment with a stack of 10 TerraSAR-X images in Stripmap mode demonstrated that this method is effective in urban scenes and has the potential applicability to wide area change detection.

A Spatio-Temporal Feature Trajectory Clustering Algorithm Based on Deep Learning

The trajectory data of aircraft, ships, and so on, can be analyzed to obtain valuable information. Clustering is the basic technology of trajectory analysis, and the feature extraction process is one of the decisive factors for clustering performance. Trajectory features can be divided into two categories: spatial features and temporal features. In mainstream algorithms, spatial features are represented by latitude and longitude coordinates. However, such algorithms are only suitable for trajectories where spatial features are tightly coupled with latitude and longitude. When the same types of trajectories are in different latitude and longitude ranges or there are transformations such as rotation, scaling, and so on, this kind of algorithm is infeasible. Therefore, this paper proposes a spatio-temporal feature trajectory clustering algorithm based on deep learning. In this algorithm, the extraction process of the trajectory spatial shape feature is designed based on image matching technology, and the extracted spatial features are combined with the trajectory temporal features to improve the clustering performance. The experimental results on simulated and real datasets show that the algorithm can effectively extract the trajectory spatial shape features and that the clustering effect of the fused spatio-temporal feature is better than that of a single feature.

Modelling the Risk of Imported COVID-19 Infections at Maritime Ports Based on the Mobility of International-Going Ships

Maritime ports are critical logistics hubs that play an important role when preventing the transmission of COVID-19-imported infections from incoming international-going ships. This study introduces a data-driven method to dynamically model infection risks of international ports from imported COVID-19 cases. The approach is based on global Automatic Identification System (AIS) data and a spatio-temporal clustering algorithm that both automatically identifies ports and countries approached by ships and correlates them with country COVID-19 statistics and stopover dates. The infection risk of an individual ship is firstly modeled by considering the current number of COVID-19 cases of the approached countries, increase rate of the new cases, and ship capacity. The infection risk of a maritime port is mainly calculated as the aggregation of the risks of all of the ships stopovering at a specific date. This method is applied to track the risk of the imported COVID-19 of the main cruise ports worldwide. The results show that the proposed method dynamically estimates the risk level of the overseas imported COVID-19 of cruise ports and has the potential to provide valuable support to improve prevention measures and reduce the risk of imported COVID-19 cases in seaports.

Short-Term 4D Trajectory Prediction for UAV Based on Spatio-Temporal Trajectory Clustering

To improve the trajectory prediction accuracy of unmanned aerial vehicles (UAVs) with random behavior intentions, this paper presents a short-term four-dimensional (4D) trajectory prediction method based on spatio-temporal trajectory clustering. A spatio-temporal trajectory clustering algorithm is first designed to cluster the UAV trajectory segments divided by a fixed time window. Each trajectory segment is given a category label that represents some certain type of behavior characteristics, such as climbing, turning, descending, etc. The convolutional neural network (CNN) is used to identify the category label of a given trajectory segment by learning the behavior characteristics of different trajectory segments. Based on the long-short-term memory network (LSTM), a short-term trajectory prediction model for different categories of label trajectory segments is established. The global trajectory prediction includes several steps adopting the corresponding prediction models. Historical trajectory data of UAVs are used to validate the proposed prediction method. Experiment results indicate that the method can obtain obviously better prediction accuracy in a short prediction time range (0-3s) with acceptable efficiency compared to LSTM, GRU and velocity trend extrapolation.

Evaluation of Four Satellite-Derived Fire Products in the Fire-Prone, Cloudy, and Mountainous Area Over Subtropical China

In the absence of historical fire records, end-users intend to adopt free satellite-derived fire products, including global burn area (BA) and active fire (AF) products, to understand the historical fire dynamics for better forest management. Previous literature evaluated the accuracy of these fire products in regions with different environments, but no study evaluated the performance of these fire products in fire-prone, cloudy and mountainous areas. This study contributed to filling this gap, through the first evaluation of four broadly used fire products: MODIS-based MCD64A1, MCD14ML, VIIRS-based VNP14DLIMGTDL&#x005F;NRT (hereafter simplified as VNP14DL), and ESA Fire&#x005F;CCI51 over subtropical China. Two methods were applied to this end, the spatio-temporal clustering algorithm based on official historical fire records and the density-based random sampling and estimation method from the Landsat 8 fire scenes. The results show that both the AF and BA products show poor fire detection ability in this area (with all the F1-Score &#x003C; 0.5). Among them, the VNP14DL performed best, followed by MCD14ML, Fire&#x005F;CCI51, and MCD64A1. The MCD14ML had the best detection capability for small fires (&#x003C; 50 hectares). AF products have an overall higher fire detection ability than BA products. These findings provide insights for the improvement of fire detection algorithms of these fire products in the fire-prone, cloudy and mountainous area over subtropical China.

An Angle Recognition Algorithm for Tracking Moving Targets Using WiFi Signals with Adaptive Spatiotemporal Clustering.

An angle estimation algorithm for tracking indoor moving targets with WiFi is proposed. First, phase calibration and static path elimination are proposed and performed on the collected channel state information signals from different antennas. Then, the angle of arrival information is obtained with the joint estimation algorithm of the angle of arrival (AOA) and time of flight (TOF). To deal with the multipath effects, we adopt the DBscan spatiotemporal clustering algorithm with adaptive parameters. In addition, the time-continuous angle of arrival information is obtained by interpolating and supplementing points to extract the dynamic signal paths better. Finally, the least-squares method is used for linear fitting to obtain the final angle information of a moving target. Experiments are conducted with the tracking data set presented with Tsinghua’s Widar 2.0. The results show that the average angle estimation error with the proposed algorithm is smaller than Widar2.0. The average angle error is about 7.18° in the classroom environment, 3.62° in the corridor environment, and 12.16° in the office environment; they are smaller than the errors of the existing system.

Roads constrain movement across behavioural processes in a partially migratory ungulate

BackgroundHuman disturbance alters animal movement globally and infrastructure, such as roads, can act as physical barriers that impact behaviour across multiple spatial scales. In ungulates, roads can particularly hamper key ecological processes such as dispersal and migration, which ensure functional connectivity among populations, and may be particularly important for population performance in highly human-dominated landscapes. The impact of roads on some aspects of ungulate behaviour has already been studied. However, potential differences in response to roads during migration, dispersal and home range movements have never been evaluated. Addressing these issues is particularly important to assess the resistance of European landscapes to the range of wildlife movement processes, and to evaluate how animals adjust to anthropogenic constraints.MethodsWe analysed 95 GPS trajectories from 6 populations of European roe deer (Capreolus capreolus) across the Alps and central Europe. We investigated how roe deer movements were affected by landscape characteristics, including roads, and we evaluated potential differences in road avoidance among resident, migratory and dispersing animals (hereafter, movement modes). First, using Net Squared Displacement and a spatio-temporal clustering algorithm, we classified individuals as residents, migrants or dispersers. We then identified the start and end dates of the migration and dispersal trajectories, and retained only the GPS locations that fell between those dates (i.e., during transience). Finally, we used the resulting trajectories to perform an integrated step selection analysis.ResultsWe found that roe deer moved through more forested areas during the day and visited less forested areas at night. They also minimised elevation gains and losses along their movement trajectories. Road crossings were strongly avoided at all times of day, but when they occurred, they were more likely to occur during longer steps and in more forested areas. Road avoidance did not vary among movement modes and, during dispersal and migration, it remained high and consistent with that expressed during home range movements.ConclusionsRoads can represent a major constraint to movement across modes and populations, potentially limiting functional connectivity at multiple ecological scales. In particular, they can affect migrating individuals that track seasonal resources, and dispersing animals searching for novel ranges.