Spatial Flow Patterns Research Articles

Spatial evolution pattern of tourism flow in China: case study of the May Day Holiday based on Baidu migration data

ABSTRACT Tourism flows are an elementary part of mobility and an important topic in tourism research. Utilising Baidu migration data, we model tourism flows during China’s May Day holiday from 2019 to 2023. Employing hotspot analysis and the differential index for tourism flow methods, we scrutinise the spatial and temporal dynamics of tourism flows in China before, during, and after the COVID-19 pandemic. The findings are as follows: (1) Regarding temporal variations, the COVID-19 pandemic has profoundly impacted China’s tourism industry, albeit with a slow recovery currently underway. (2) Spatially, tourism flows primarily concentrated within four major city clusters, with emerging tourism destinations like Zibo gradually gaining prominence. (3) Analysing changes in tourism flow during holidays reveals that cities can be categorised into four major types and six minor types, with a predominant trend of continuous decrease in tourism flow. These findings shed light on the intricate dynamics of tourism flows in China, offering valuable insights for stakeholders in the tourism sector.

Highly clustered mating networks in naturally fragmented riparian tree populations.

Understanding how spatial patterns of mating and gene flow respond to habitat loss and geographical isolation is a crucial aspect of forest fragmentation genetics. Naturally fragmented riparian tree populations exhibit unique characteristics that significantly influence these patterns. In this study, we investigate mating patterns, pollen-mediated gene flow, and genetic diversity in relict populations of Frangula alnus in southern Spain by testing specific hypotheses related to the riparian habitat. We employ a novel approach that combines paternity analysis, particularly suited for small and isolated populations, with complex network theory and Bayesian models to predict mating likelihood among tree pairs. Our findings reveal a prevalence of short-distance pollination, resulting in spatially driven local mating clusters with a distinct subset of trees being highly connected in the mating network. Additionally, we observe numerous pollination events over distances of hundreds of metres and considerable pollen immigration. Local neighbourhood density is the primary factor influencing within-population mating patterns and pollen dispersal; moreover, mating network properties reflect the population's size and spatial configuration. Conversely, among-population pollen dispersal is mainly determined by tree size, which influences floral display. Our results do not support a major role of directional pollen dispersal in longitudinal trends of genetic diversity. We provide evidence that long-term fragmented tree populations persist in unique environments that shape mating patterns and impose constraints to pollen-mediated gene flow. Nevertheless, even seemingly strongly isolated populations can maintain functional connectivity over extended periods, especially when animal-mediated mating networks promote genetic diversity, as in this riparian tree species.

Пространственная картина центро-периферийных миграционных взаимодействий в России

The migration appeal of major cities and their suburbs (centers) and the outflow of the population from the periphery in Russia is a well-documented phenomenon. Utilizing detailed spatial data, this study examines the spatial diversity of the migration balance of 137 identified centers, represented by cities with a population of 100,000 or more and their suburbs, and 82 peripheral areas for the 33years 2011-2020. The analysis encompasses not only the flow between centers and the periphery but also among individual centers, as well as migration involving the centers and peripheries within their regions and in interregional exchanges. Out of the variety of centers, 21 main ones are distinguished, concentrating the primary flow of population from peripheral territories and attracting populations from other centers. Meanwhile, half of the centers experienced migration loss. The centers leading the migration flow in each federal district are identified, with particular attention given to the migration balance characteristics of the largest centers. The data used allowed for analysis with and without considering auto-return, which impacts the scale of migration gain/loss, but, as it turned out, virtually does not alter the spatial pattern of population flow between centers and periphery.

Quantifying Intermittent Flow Regimes in Ungauged Basins: Optimization of Remote Sensing Techniques for Ephemeral Channels Using a Flexible Statistical Classification

Intermittent and ephemeral channels are a critical component of the global hydrologic network. The dominant feature in dryland environments, ephemeral channel transmission loss facilitates aquifer recharge. Characterizing flow intermittency improves groundwater storage estimates; however, limited gauging of intermittent systems impedes this understanding. This research develops an improved classification for surface flow, optimized for ephemeral systems using linear discriminant function analysis and remotely sensed imagery. It further applies this methodology to assess temporal and spatial flow patterns across the Souss channel, an ungauged, ephemeral system in central Morocco. Linear discriminant function analysis demonstrates high predictive accuracy for Landsat imagery, with significantly improved classification success as compared to the Modified Normalized Difference Water Index. Application to the Souss channel from 1984 to 2022 points to a decreasing trend in flow frequency. Despite this change, flow events remain concentrated within the wet season, critical for regional aquifer recharge. Spatial flow characteristics further support sustained infiltration, with the majority of events focused within the upstream channel section during both dry and wet seasons. Decreased occurrence moving downstream highlights the likely impact of additional factors such as transmission loss, evapotranspiration, and agricultural abstraction contributing to channel intermittency.

Detailed turbulent structures investigation around piers group induced flow

This paper aims to investigate the mean velocities and turbulence characteristics, which include Reynolds shear stresses (RSS) and turbulent intensities, in order to study temporal and spatial flow patterns around a complex three square-piers group. Measurements have been performed by acoustic Doppler velocimeter (ADV) to identify turbulence flow structures alteration and vortices formation in longitudinal, transverse, and vertical directions. In total, 164 locations around the bridge piers have been measured, while each location consists of no lesser than 15 vertical depth ADV data point measurements. To investigate the flow turbulence, critical locations have been observed from high vortex-influenced regions. The impacted mean velocity profiles have been compared to that at the unaffected region to study the flow alteration caused by piers group. The result reveals that the log-law layer near wall has been enhanced by the installation of pier. Furthermore, as identified from the most critical locations of turbulent structures across different measurement rows, the accelerated flow will result in lower observed turbulent structures. Although flow after each pier has been influenced by strong reversal velocities, vortices have managed to form at near wake. Analysis has been divided into inner and outer flow regions (IFR and OFR) based on flow depth to enhance the understanding of how bridge piers contribute to the development of flow turbulence.

Спектр сверхзвукового обтекания вокруг летательного аппарата с управляющими тормозными щитками

The article presents the results of the study of the spatial flow pattern of the aircraft model, the study of the flow structure near the surface of the aircraft by experimental and numerical methods. Comparative materials for visualizing flows around a conical model with a stern control body – a flat end shield during supersonic flight in atmospheric pressure air are presented. The difference from the qualitative flow pattern according to experimental spectra at M≈4 and M≈5.5 lies in the presence, in addition to the formation of a breakaway flow zone in front of the shield, an oblique jump of the seal (from the point of separation of the flow) and a more intense direct jump of the seal, the possibility of the appearance of a flow separation zone induced by the shield on the side of the model surface opposite the shield, with the formation of an oblique jump of the seal. The efficiency of increasing the number of brake shields installed on the cylindrical-conical model is considered. The results of measurements of the dependence of the increments of the drag coefficient of the tested models on the area of the shields are presented.

Heat exposure assessment based on high-resolution spatio-temporal data of population dynamics and temperature variations

Urban heat waves pose a significant risk to the health and safety of city dwellers, with urbanization potentially amplifying the health impact of extreme heat. Accurate assessments of population heat exposure hinge on the interplay between temperature, population spatial dynamics, and the epidemiological effects of temperature on health. Yet, many past studies have over-simplified the matter by assuming static populations, leading to substantial inaccuracies in heat exposure assessments. To address these issues, this study integrates dynamic population data, fluctuating temperature, and the exposure-response relationship between temperature and health to construct an advanced heat exposure assessment framework predicated on a population dynamic model. We analyzed urban heat island characteristics, population dynamics, and heat exposure during heat wave conditions in Beijing, a major city in China. Our findings highlight significant intra-day population movement between urban and suburban areas during heat wave conditions, with spatial population flow patterns showing clear scale-dependent characteristics. These population flow dynamics intensify heat exposure levels, and the disparity between dynamic population-weighted temperature and average temperature is most pronounced at night. Our research provides a more comprehensive understanding of real urban population heat exposure levels and can furnish city administrators with more scientifically rigorous evidence.

Lagrangian investigation of the interface dynamics in single-mode Rayleigh–Taylor instability

We apply Lagrangian particle tracking to the two-dimensional single-mode Rayleigh–Taylor (RT) instability to study the dynamical evolution of fluid interface. At the onset of the nonlinear RT stage, we select three ensembles of tracer particles located at the bubble tip, at the spike tip, and inside the spiral of the mushroom structure, which cover most of the interfacial region as the instability develops. Conditional statistics performed on the three sets of particles and over different RT evolution stages, such as the trajectory curvature, velocity, and acceleration, reveals the temporal and spatial flow patterns characterizing the single-mode RT growth. The probability density functions of tracer particle velocity and trajectory curvature exhibit scalings compatible with local flow topology, such as the swirling motion of the spiral particles. Large-scale anisotropy of RT interfacial flows, measured by the ratio of horizontal to vertical kinetic energy, also varies for different particle ensembles arising from the differing evolution patterns of the particle acceleration. In addition, we provide direct evidence to connect the RT bubble re-acceleration to its interaction with the transported fluid from the spike side, due to the shear driven Kelvin–Helmholtz instability. Furthermore, we reveal that the secondary RT instability inside the spiral, which destabilizes the spiraling motion and induces complex flow structures, is generated by the centrifugal acceleration.

Mapping the intrinsic photocurrent streamlines through micromagnetic heterostructure devices

Photocurrent in quantum materials is often collected at global contacts far away from the initial photoexcitation. This collection process is highly nonlocal. It involves an intricate spatial pattern of photocurrent flow (streamlines) away from its primary photoexcitation that depends sensitively on the configuration of current collecting contacts as well as the spatial nonuniformity and tensor structure of conductivity. Direct imaging to track photocurrent streamlines is challenging. Here, we demonstrate a microscopy method to image photocurrent streamlines through ultrathin heterostructure devices comprising platinum on yttrium iron garnet (YIG). We accomplish this by combining scanning photovoltage microscopy with a uniform rotating magnetic field. Here, local photocurrent is generated through a photo-Nernst type effect with its direction controlled by the external magnetic field. This enables the mapping of photocurrent streamlines in a variety of geometries that include conventional Hall bar-type devices, but also unconventional wing-shaped devices called electrofoils. In these, we find that photocurrent streamlines display contortion, compression, and expansion behavior depending on the shape and angle of attack of the electrofoil devices, much in the same way as tracers in a wind tunnel map the flow of air around an aerodynamic airfoil. This affords a powerful tool to visualize and characterize charge flow in optoelectronic devices.

Supply-demand relationship and spatial flow of urban cultural ecosystem services: The case of Shenzhen, China

Cultural ecosystem services (CES) have garnered significant attention due to their capacity to comprehensively evaluate the influence of ecosystem degradation on human well-being. It is very important to evaluate the spatial coordination relationship between the supply and demand of urban CES and explore its spatial flow mechanism for guiding the sustainable development of urban regions. However, research on the relationship between supply and demand of urban CES and their spatial flows is still in the conceptual exploration stage, and its quantitative assessment methods are not yet mature. Taking Shenzhen as an example, this study uses the Social Values for Ecosystem Services model to evaluate the supply of CES based on environmental background and survey statistics; Uses the topic model of natural language processing, quantify the demand of CES based on social media data. The Bivariate Local Indicators of Spatial Association index is utilized to examine the correlation between supply and demand, while a crowd activity extraction method based on mobile phone signaling data is developed to simulate the spatial flow pattern of CES. The results show that: (1) The supply of urban CES shows obvious spatial differentiation; There are differences in the perception of the needs of various CES. Among them, the beneficiaries have higher needs for Aesthetic and Recreation. (2) The relationship between supply and demand of urban CES shows different degrees of spatial mismatch, with the proportion of the supply-demand matching area to the total area being less than 20%, and the main type of supply-demand mismatch being low supply-high demand. (3) All kinds of CES flow from the center of the study area to the east, among which the flow of recreation and Aesthetic is more active, with the highest flow of 189.7 and 135.3, respectively. Overall, the coordination between various types of CES in the region is still insufficient. We should encourage the establishment and improvement of a diversified transportation system connecting the central urban area and other regions while enhancing the condition of the local ecological conditions, to improve the capacity of urban ring roads and rail transit, to promote the sustainable development of the regional ecological environment and social economy, and to enhance human well-being.

Ocean connectivity and habitat characteristics predict population genetic structure of seagrass in an extreme tropical setting

AbstractUnderstanding patterns of gene flow and processes driving genetic differentiation is important for a broad range of conservation practices. In marine organisms, genetic differentiation among populations is influenced by a range of spatial, oceanographic, and environmental factors that are attributed to the seascape. The relative influences of these factors may vary in different locations and can be measured using seascape genetic approaches. Here, we applied a seascape genetic approach to populations of the seagrass, Thalassia hemprichii, at a fine spatial scale (~80 km) in the Kimberley coast, western Australia, a complex seascape with strong, multidirectional currents greatly influenced by extreme tidal ranges (up to 11 m, the world's largest tropical tides). We incorporated genetic data from a panel of 16 microsatellite markers, overwater distance, oceanographic data derived from predicted passive dispersal on a 2 km‐resolution hydrodynamic model, and habitat characteristics from each meadow sampled. We detected significant spatial genetic structure and asymmetric gene flow, in which meadows 12–14 km apart were less connected than ones 30–50 km apart. This pattern was explained by oceanographic connectivity and differences in habitat characteristics, suggesting a combined scenario of dispersal limitation and facilitation by ocean current with local adaptation. Our findings add to the growing evidence for the key role of seascape attributes in driving spatial patterns of gene flow. Despite the potential for long‐distance dispersal, there was significant genetic structuring over small spatial scales implicating dispersal and recruitment bottlenecks and highlighting the importance of implementing local‐scale conservation and management measures.

Assessment of ecosystem service flow and optimization of spatial pattern of supply and demand matching in Pearl River Delta, China

Exploring the spatial flow pattern of ecosystem services and clarifying the transmission path between the supply area and the demand area will help to formulate more scientific and reasonable ecological protection policies. This paper takes the rapid urbanization area representing the Pearl River Delta urban agglomeration as an example. Using Invest model to quantitative the assessment of the supply and demand of carbon sequestration services and water ecosystem services from 2000 to 2020, a spatial flow model of ecosystem services is constructed to clarify the spatial flow of regional ecosystem services pattern. In addition, the Geo-detector model is used to explore the driving factors of the supply and demand relationship of ecosystem services, and the spatiotemporal geographic weighted regression model is further used to analyze the temporal and spatial differentiation of the impact degree. Finally, based on the Bayesian belief network, the optimal state factor configuration is selected to optimize the spatial pattern, and the corresponding optimization strategy is given. The main conclusions are as follows: (1) The supply of carbon sequestration services in the Pearl River Delta decreased gradually during 2000–2020, while the demand increased gradually. The supply of water ecosystem services increased first and then decreased, while the demand showed a downward trend. (2) In terms of spatial heterogeneity of supply and demand matching, the main driving factors of carbon sequestration services included night light brightness value, temperature and vegetation index; The main driving factors of water ecosystem services include night light brightness value, land use type and vegetation index. (3) In terms of optimal areas for supply and demand matching, carbon sequestration services were mainly distributed in Zhaoqing, Huizhou and Jiangmen; The optimal matching areas of water ecosystem service supply and demand were mainly distributed in Zhaoqing, Guangzhou, Huizhou and other cities.

Verification of Geographic Laws Hidden in Textual Space and Analysis of Spatial Interaction Patterns of Information Flow

The rapid development of Internet technology has formed a huge virtual information space. In the information space, information flow has become a link of communication between objects. Information flow is an alternative or supplement to the traditional physical flow for the study of the spatial interaction of geographical entities. The research uses toponym co-occurrence and search index as information flow data, verifies the geographical laws hidden in the information space by spatial autocorrelation analysis and gravity model fitting, and analyzes the spatial interaction patterns of provinces in China in the information space by complex network analysis methods. The results show that: (1) information flow in the information space obeys Tobler’s first law of geography and Goodchild’s second law of geography. The spatial interaction represented by information flow has a distance decay effect. The best distance decay coefficients for toponym co-occurrence and the search index are 0.189 and 0.186, respectively. (2) The inter-provincial spatial interaction network of China shows a hierarchical pattern of the triangular primary network and diamond secondary network, and the ranking of provinces in the centrality analysis is basically stable, but the network hierarchy is deepening. The gravity center of spatial interaction is located in the east-central region of China. (3) The information flow-based interaction network is of higher asymmetry than the population mobility network, and its spatial structure is also obvious. This research provides a new idea for studying the spatial interaction of geographical entities in the physical world from the perspective of information flow.

Using jointly calibrated fine-scale drain models across Denmark to assess the influence of physical variables on spatial drain flow patterns

Study regionDenmark Study focusTile drainage, widespread across agricultural areas in Denmark, significantly impacts the hydrological cycle. Tile drain flow dynamics and their spatial patterns are crucial for water managers to address water quality and quantity issues. However, these processes are challenging to simulate accurately. In this study, we developed 10 m resolution drain models in MIKE SHE for 26 drain sites across Denmark to predict drain flow spatial and temporal dynamics. Joint calibration of all drain models was conducted by evaluating PBIAS and KGE of simulated and observed drain flow. Subsequently, we performed a correlation analysis between physical parameters and spatial patterns of simulated drain fraction (DF, ratio of drain flow to recharge per grid cell). New hydrological insightsThe jointly calibrated models achieved average PBIAS and KGE of − 6.7% and 0.53 respectively, for drain flow predictions across 26 drain sites. The correlation of DF with topographical variables was highest on a national scale and on most of the drain sites, including Gyldenholm1–4, Norsminde1–11, and other Jylland drain sites. Lillebæk drain sites showed a high correlation with average clay content in 0–30 cm layer of soil. Thus, in addition to developing scalable fine-resolution drain models for Denmark, this study also identifies the control parameters for spatial patterns of drain flows across Denmark.

Weak range-wide population structure in the blackfin tuna (Thunnus atlanticus) revealed by analysis of genome-wide SNPs

Abstract Blackfin tuna (Thunnus atlanticus) is a small tuna distributed in the western Atlantic Ocean where it is exploited by growing recreational and commercial regional fisheries. In this work, genome-wide genetic variation was analysed to investigate the occurrence of stock subdivision. A de novo assembly of the blackfin tuna genome was generated using Illumina paired-end sequencing data and applied as a reference for population genomic analysis of specimens from nine localities (average sample size per locality n = 72) spanning most of the blackfin tuna distribution range. A total of 2139 single-nucleotide polymorphisms were discovered and genotyped using the double-digest restriction associated DNA sequencing. Pairwise exact homogeneity tests were significant in 24 out of 36 population pairs and significant spatial autocorrelation of genotypes was observed for specimens collected within 2250 km of each other. However, divergence among locality samples was very low (pairwise FST range 0.0002–0.0025) and significant temporal variations were detected in localities sampled multiple times. Approaches to detect cryptic groups de novo were unsuccessful. Additional sampling is warranted to determine if multiple stocks need to be defined for management and assess temporal and spatial patterns of gene flow connecting them.

Design optimization of prismatic rib turbulators in a rectangular channel based on multi-objective criterion

A multi-objective shape optimization of truncated prismatic ribs in a rectangular channel has been performed to optimize simultaneously, heat transfer and pressure loss. The numerical analysis was performed using Reynolds-averaged Navier-Stokes equations with eddy viscosity-based models for turbulence closure. The optimization was performed using multi-objective genetic algorithm technique, and the fitness functions for the objectives were obtained using surrogate modeling. Using literature survey and preliminary parametric analysis, three design variables related to prismatic ribs viz. ratio of rib-height (e) to hydraulic-diameter (Dh) of a rectangular channel, ratio of rib-pitch (p) to rib-height (e), and ratio of rib-height at end (ep) to rib-height at center (e) were selected. The augmentation Nusselt number (FNu) and friction factor (Ff) ratios were chosen as the multi-objective functions for heat transfer and pressure drop, respectively in the rectangular channel. Numerical analysis and multi-objective optimization have been done at fixed Reynolds number, Re = 42,500. Using the proposed optimization framework, Pareto-optimal front design (POD) signifies the trade-off between the obtained objective functions, and among them, five PODs have been chosen using K-means clustering technique. In the paper, results indicated that the POD which offer high blockage effect to the incoming flow lead to high heat transfer performance, but accompanied by greater pressure losses. The blockage effect is discussed using spatial flow and thermal patterns. An optimum design with a relative thermal performance enhancement of 168.60% compared with the reference design is reported.

Geographically Weighted Flow Cross K-Function for Network-Constrained Flow Data

Network-constrained spatial flows are usually used to describe movements between two spatial places on a road network. The analysis of the spatial associations between different types of network-constrained spatial flows plays a key role in understanding the spatial relationships among different movements. However, existing studies usually do not consider the effect of distance decay, which may reduce the effectiveness of the detected bivariate spatial flow patterns. Moreover, most existing studies are based on the planar space assumption, which is not suitable for network-constrained spatial flows. To overcome these problems, this study proposed a new statistical method, the Geographically Weighted Network Flow Cross K-function, which improves the Flow Cross K-Function method by taking the distance decay effect and the constraints of road networks into account. Both global and local versions are extended in this study: the global version measures the overall spatial association and the local version identifies the exact locations where a spatial association occurs. The experiments on simulated datasets show that the proposed method can identify predefined bivariate flow patterns. In a case study, the proposed method is also applied to flow data comprising Xiamen taxi and ride-hailing datasets. The results demonstrate that the proposed method effectively identifies the competitive relationships between taxi and ride-hailing services.

Exploring Spatial Patterns of Interurban Passenger Flows Using Dual Gravity Models.

Geographical gravity models can be employed to quantitatively describe and predict spatial flows, including migration flows, passenger flows, daily commuting flows, etc. However, how to model spatial flows and reveal the structure of urban traffic networks in the case of missing partial data is still a problem to be solved. This paper is devoted to characterizing the interurban passenger flows in the Beijing-Tianjin-Hebei region of China using dual gravity models and Tencent location big data. The method of parameter estimation is the least squares regression. The main results are as follows. First, both the railway and highway passenger flows can be effectively described by dual gravity models. A small part of missing spatial data can be compensated for by predicted values. Second, the fractal properties of traffic flows can be revealed. The railway passenger flows follow the gravity scaling law better than the highway passenger flows. Third, the prediction residuals indicate the changing trend of interurban connections in the study area in recent years. The center of gravity of the spatial dynamics has shifted from the Beijing-Tianjin-Tangshan triangle to the Beijing-Baoding-Shijiazhuang axis. A conclusion can be reached that the dual gravity model is an effective tool for analyzing spatial structures and dynamics of traffic networks and flows. Moreover, the model provides a new approach to estimating the fractal dimensions of traffic networks and spatial flow patterns.

Fine-Grained Vessel Traffic Flow Prediction With a Spatio-Temporal Multigraph Convolutional Network

The accurate and robust prediction of vessel traffic flow is gaining importance in maritime intelligent transportation system (ITS), such as vessel traffic services, maritime spatial planning, and traffic safety management, etc. To achieve fine-grained vessel traffic flow prediction, we will first generate the maritime traffic network (which is essentially a graph), and then propose a graph-driven neural network. In particular, to represent various correlations among spatio-temporal vessel traffic flow, we tend to extract the feature points (i.e., starting, way and ending points) by utilizing the knowledge of vessel positioning data. These feature points are essentially related to the geometrical structures of massive vessel trajectories collected from massive automatic identification system (AIS) records, contributing to the generation of maritime traffic network. We then propose a spatio-temporal multi-graph convolutional network (STMGCN)-based vessel traffic flow prediction method by exploiting multiple types of inherent correlations in the generated maritime graph. The proposed STMGCN mainly contains one spatial multi-graph convolutional layer and two temporal gated convolutional layers, beneficial for extracting spatial and temporal traffic flow patterns. The main benefit of our graph-driven prediction method is that it takes full advantage of the maritime graph and multi-graph learning. Comprehensive experiments have been implemented on realistic AIS dataset to compare our method with several state-of-the-art prediction methods. The fine-grained prediction results have demonstrated our superior performance in terms of both accuracy and robustness.

Learning Dynamic and Hierarchical Traffic Spatiotemporal Features With Transformer

Traffic forecasting has attracted considerable attention due to its importance in proactive urban traffic control and management. Scholars and engineers have exerted considerable efforts in improving the performance of traffic forecasting algorithms in terms of accuracy, reliability, and efficiency. Spatial feature representation of traffic flow is a core component that greatly influences traffic forecasting performance. In previous studies, several spatial attributes of traffic flow are ignored due to the following issues: a) traffic flow propagation does not comply with the road network, b) the spatial pattern of traffic flow varies over time, and c) single adjacent matrix cannot handle the complex and hierarchical urban traffic flow. To address the abovementioned issues, this study proposes a novel traffic forecasting algorithm called traffic transformer, which achieves great success in natural language processing. The multihead attention mechanism and stacking layers enable the transformer to learn dynamic and hierarchical features in sequential data. Two components, namely, global encoder and global–local decoder, are proposed to extract and fuse the spatial patterns globally and locally. Experimental results indicate that the proposed traffic transformer outperforms state-of-the-art methods. The learned dynamic and hierarchical features of traffic flow can help achieve a better understanding of spatial dependency of traffic flow for effective and efficient traffic control and management strategies.