Spatio-temporal Process Research Articles

Inhibition of cardiomyocyte neddylation impairs embryonic cardiac morphogenesis

Heart development is a complex spatiotemporal process involving a series of orchestrated morphogenic events that result in the formation of an efficient pumping organ. How posttranslational mechanisms regulate heart development remains poorly understood. Therefore, we investigate how neddylation, the attachment of NEDD8 to target proteins, coordinates cardiogenesis. Abrogation of neddylation by deleting Nae1 in the heart via Sm22αCre led to early embryonic lethality. Mutant hearts exhibited deficits in trabeculation and expansion of the compact layer due to reduced cardiomyocyte proliferation, which was linked to abnormal Notch signaling in the developing heart. Overall, our findings demonstrate an essential role for neddylation in cardiogenesis.

Unraveling the developmental heterogeneity within the human retina to reconstruct the continuity of retinal ganglion cell maturation and stage-specific intrinsic and extrinsic factors.

Tissue development is a complex spatiotemporal process with multiple interdependent components. Anatomical, histological, sequencing, and evolutional strategies can be used to profile and explain tissue development from different perspectives. The introduction of scRNAseq methods and the computational tools allows to deconvolute developmental heterogeneity and draw a decomposed uniform map. In this manuscript, we decomposed the development of a human retina with a focus on the retinal ganglion cells (RGC). To increase the temporal resolution of retinal cell classes maturation state we assumed the working hypothesis that that maturation of retinal ganglion cells is a continuous, non-discrete process. We have assembled the scRNAseq atlas of human fetal retina from fetal week 8 to week 27 and applied the computational methods to unravel maturation heterogeneity into a uniform maturation track. We align RGC transcriptomes in pseudotime to map RGC developmental fate trajectories against the broader timeline of retinal development. Through this analysis, we identified the continuous maturation track of RGC and described the cell-intrinsic (DEGs, maturation gene profiles, regulons, transcriptional motifs) and -extrinsic profiles (neurotrophic receptors across maturation, cell-cell interactions) of different RGC maturation states. We described the genes involved in the retina and RGC maturation, including de novo RGC maturation drivers. We demonstrate the application of the human fetal retina atlas as a reference tool, allowing automated annotation and universal embedding of scRNAseq data. Altogether, our findings deepen the current knowledge of the retina and RGC maturation by bringing in the maturation dimension for the cell class vs. state analysis. We show how the pseudotime application contributes to developmental-oriented analyses, allowing to order the cells by their maturation state. This approach not only improves the downstream computational analysis but also provides a true maturation track transcriptomics profile.

A doubly stochastic point process approach for spatio-temporal dynamics of crime data

Crime, in general, is at the base of crucial issues for many societies living in large cities worldwide. Indeed, crime and neighbourhood disorder may negatively impact the health of urban residents. Thus, the crime rate reduction is at the core of many local policies driven by active plans supported by police action and local authorities. Considering crime reports as a spatio-temporal point pattern, we propose spatio-temporal log-Gaussian Cox processes as a modelling framework for crimes in space and time. We model the spatial and temporal variation through generalized parametric additive and linear models, and a Gaussian space-time process approximates the residual variation. The inference is performed via Markov chain Monte Carlo through MALA algorithms. We provide short-term forecasts of future crimes and suggest a surveillance system that operates by reporting predictive probabilities. Our data come from the reported crimes in the locality of Kennedy (Bogota) over several years and several types of crimes. The police department may use our method to help allocate police resources and design crime prevention strategies and policies, such as surveillance, in specific zonal planning units.

Análisis espacio-temporal del proceso de urbanización en la ciudad de Santa Tecla, El Salvador

This study aims to analyze the urbanization process in the municipality of Santa Tecla and propose a generic methodology with free tools so that the population can monitor the spatio-temporal process of urbanization in the municipalities of the metropolitan area of San Salvador using Sentinel satellite images and the current regulatory planning instrument as a reference. The evolution of the 30-year temporal space (1991-2021) of the five-year urban sprawl of the municipality of Santa Tecla was analyzed using Landsat satellite data and contrasted with current regulatory instruments or reference techniques. NDVI and NDBI indices calculated from Sentinel-2 satellite data and applying a hybrid image classification methodology, conflicts in real land use were identified with high resolution. The study concluded that the process in the municipality of Santa Tecla has been disorderly in the last 30 years, which has suspected the risk for the population, because although there is a legal framework for land use planning, its application has been flexible. A hybrid method of classification based on NDVI and NDBI from satellite images has been used to simplify the manual inspection of the area in a given period of time with an acceptable margin of error. ILIA: Investigaciones Latinoamericanas en Ingeniería y Arquitectura, No. 01, 2024: 53-60.

A Method for Constructing an Urban Waterlogging Emergency Knowledge Graph Based on Spatiotemporal Processes

Urban waterlogging is one of the major “diseases” faced by cities, posing a great challenge to the healthy and sustainable development of cities. The traditional geographic knowledge graph struggles to capture dynamic changes in urban waterlogging over time. Therefore, the objective of this study is to analyze the time, events, properties, geographic objects, and activities associated with urban waterlogging emergency responses from the geographic spatial and temporal processes perspective and to construct an urban waterlogging emergency knowledge graph by combining top-down and bottom-up approaches. We propose a conceptual model of urban waterlogging emergency response ontology based on spatiotemporal processes by analyzing the basic laws and influencing factors of urban waterlogging occurrence and development. Secondly, we describe the construction process of the urban waterlogging emergency response knowledge graph from knowledge extraction, knowledge fusion, and knowledge storage. Finally, the knowledge graph was visualized using 159 urban waterlogging events in China from 2020–2022, with a quality assessment indicating 81% correctness, 65.5% completeness, and 95% data conciseness. The results show that this method can effectively express the spatiotemporal process of an urban waterlogging emergency response and can provide a reference for the spatiotemporal modeling of the knowledge graph.

Time-guided convolutional neural networks for spatiotemporal urban flood modelling

Urban flood modelling is key to understand flood risks and develop effective interventions in flood management. Deep learning (DL), known for its robust and automatic feature extraction capabilities, has been applied for urban flood predictions. However, the hybrid spatiotemporal structure of conventional DL-enabled urban flood models is limited in terms of accuracy and efficiency. To address this gap, this study develops a new DL model guided by time information. This model uses a classic CNN (Convolution Neural Network) architecture, Unet, as its backbone. Time information is integrated into inputs via an extra channel to specify the desired prediction time, facilitating the simulation of the spatiotemporal flood process. Additionally, a modified loss function is formulated to tackle the sample imbalance problem between flooded and non-flooded sites. The model performance is assessed in an urban area in Dalian, China with a total of 18 rainfall events of varying return periods. The model attains average precision and recall values of 0.90 and 0.81, respectively, across different time steps during various events. Furthermore, the model exhibits transferability in ungauged regions where a high influence of surrounding environments on local flood processes is identified by Grad-CAM (Gradient-weighted Class Activation Mapping) analysis. The results show that the new Unet model has great promise in efficiently providing accurate spatiotemporal flood simulations. The time-guided Unet model can serve as practical tools for rapid flood simulation in urban areas.

National burden and risk factors of diabetes mellitus in China from 1990 to 2021: Results from the Global Burden of Disease study 2021.

In recent years, the prevalence and mortality rates of diabetes have been rising continuously, posing a significant threat to public health and placing a heavy burden on the population. This study was conducted to describe and analyze the burden of diabetes in China from 1990 to 2021 and its attributable risk factors. Utilizing data from the Global Burden of Disease Study 2021, we analyzed the incidence, prevalence, and disability-adjusted life years (DALYs) of type 1 diabetes (T1DM) and type 2 diabetes (T2DM) in China from 1990 to 2021. We extracted sex- and age-specific data on diabetes, focusing on DALYs, years lived with disability, and years of life lost. Bayesian meta-regression and spatiotemporal Gaussian process regression were used to estimate disease parameters. Age-standardized rates (ASRs) and estimated annual percentage changes (EAPC) were calculated using direct standardization and log-linear regression. The population-attributable fractions were also determined for each risk factor. In 2021, the absolute number of incident diabetes mellitus (DM) cases was estimated at 4003543.82, including 32 000 T1DM and 3971486.24 T2DM cases. The ASRs were 244.57 for DM, 2.67 for T1DM, and 241.9 for T2DM (per 100 000 population). The absolute number of prevalent DM cases was 117288553.93, including 1442775.09 T1DM and 115845778.84 T2DM cases. The ASRs were 6142.29 for DM, 86.78 for T1DM, and 6055.51 for T2DM (per 100 000 population). In 2021, there were 178475.73 deaths caused by DM, with an ASR of mortality of 8.98 per 100 000 population. The DALYs due to DM in 2021 were 11713613.86, with an ASR of 585.43 per 100 000 population and an EAPC of 0.57. This increase can be attributed to several factors, including high body mass index, air pollution, and dietary habits. The burden of diabetes is considerable, with high prevalence and incidence rates, highlighting the urgent need for public health interventions. Addressing factors like high fasting plasma glucose, body mass index, air pollution, and dietary risks through effective interventions is critical.

A longitudinal MRI analysis reveals altered brain connectivity and microstructural changes in a transgenic mouse model of Alzheimer's disease

Alzheimer's disease (AD) is characterized by progressive cognitive decline and neuropathological changes, yet the underlying neurobiological mechanisms remain elusive. Here, we employed a multimodal longitudinal neuroimaging approach, using anatomical and functional sequences on a high field magnetic resonance imaging (MRI) preclinical scanner, to investigate alterations in brain connectivity and white matter microstructure in a transgenic mouse model of AD (J20) when compared to wild-type (WT) littermates. Functional connectivity analysis revealed distinct network disruptions in J20 mice, primarily involving connections between posterior and anterior brain regions; importantly, a significant interaction between group and age highlighted an exacerbation of these connectivity changes with advancing age in J20 mice. In addition, significant reductions in fractional anisotropy (FA) were observed in the corpus callosum of J20 mice compared to WT, indicative of microstructural alterations consistent with white matter pathology. The observed alterations in brain connectivity and microstructure provide valuable insights into the spatiotemporal processes underlying AD-related decline and underscore the utility of multimodal neuroimaging in elucidating the neurobiological substrates of AD pathology in animal models.

Spatially explicit simulation and forecasting of urban growth using weights of evidence based cellular automata model in a millennium city of India

The present study focuses on quantifying and simulating the future urban growth based on the land use/land cover (LULC) data created from the Landsat images of the year 1999, 2011, and 2022. These LULC maps help in analysing the expansion of urban areas over the years and forecast their potential growth in the future. The spatio-temporal processes of urban growth are quantified, and future patterns are simulated and forecasted using Weights of Evidence based Cellular Automata model built in Dinamica EGO (Environment for Geoprocessing Objects) platform. The process of urban growth was manifested through prominent contributing factors of infill expansion namely, distance to built-up areas, distance to main roads, population density, and public services etc. The model's performance was evaluated using Kappa statistics and the percentage of correct prediction (PCP) based two-way comparison method. For this purpose, the simulated map was first compared with the observed information of year 2022 using Kappa indices followed by the PCP value (90.40%) exhibiting high predictive ability of the model. These findings corroborate that the model can forecast the future urban growth scenarios effectively with reasonable accuracy. Based on the outcomes, the forecasting of future urban growth scenarios for years 2033 and 2044 was accomplished. Analysis of the LULC changes displays that urban land use will experience the highest increase. Growth in the study area is predicted to increase by 23.5% and 26.7% in year 2033 and 2044 respectively where new urban settlements can appear. The results demonstrated that an integrated geospatial model provides essential information about the pattern, simulation, and prediction of urban growth associated with various driving variables.

Research on the Unsteady Flow Characteristics of Solid–Liquid Two-Phase Flow in a Deep-Sea Mining Lift Pump and Model Experimental Verification

The deep-sea mining lift pump is one of the pivotal components in deep-sea mineral transportation systems and its internal flow is very complex; consequently, unraveling its unsteady flow behavior pattern holds immense practical value. This study adopts numerical methods to analyze the time-averaged distribution characteristics of the internal flow field in mining lift pumps, as well as the flow field’s pulsation intensity distribution characteristics, the vortex’s spatiotemporal evolution process in both moving and static cascades, and the time- and frequency-domain pulsation characteristics of internal pressure in each flow passage component under four different flow conditions are also investigated. The hydraulic properties of mining lift pumps under these four different conditions are also evaluated, and the outcomes are benchmarked against those of numerical predictions. Our findings reveal that the interplay between impeller blades and guide vanes significantly influences the pump’s flow characteristics, with the pump’s unsteady flow influencing its hydraulic properties. Experimental validation of this system confirms that the pump under study is in line with design specifications in terms of hydraulic properties. The method validation test on the prototype pump shows that the SST k-ω model is capable of successfully forecasting instability in the flow features of deep-sea mining lift pumps. These results will serve as a theoretical reference for regulating the flow state inside deep-sea mining lift pumps.

Analysis of Global and Key PM2.5 Dynamic Mode Decomposition Based on the Koopman Method

Understanding the spatiotemporal dynamics of atmospheric PM2.5 concentration is highly challenging due to its evolution processes have complex and nonlinear patterns. Traditional mode decomposition methods struggle to accurately capture the mode features of PM2.5 concentrations. In this study, we utilized the global linearization capabilities of the Koopman method to analyze the hourly and daily spatiotemporal processes of PM2.5 concentration in the Beijing–Tianjin–Hebei (BTH) region from 2019 to 2021. This approach decomposes the data into the superposition of different spatial modes, revealing their hierarchical spatiotemporal structure and reconstructing the dynamic processes. The results show that PM2.5 concentrations exhibit high-frequency cycles of 12 and 24 h, as well as low-frequency cycles of 124 and 353 days, while also revealing spatiotemporal modes of growth, recession, and oscillation. The superposition of these modes enables the reconstruction of spatiotemporal dynamics with a mean absolute percentage error (MAPE) of only 0.6%. Unlike empirical mode decomposition (EMD), Koopman mode decomposition (KMD) method avoids mode aliasing and provides a clearer identification of global and key modes compared to wavelet analysis. These findings underscore the effectiveness of KMD method in analyzing and reconstructing the spatiotemporal dynamics of PM2.5 concentration, offering new insights into the understanding and reconstruction of other complex spatiotemporal phenomena.

Code and Data Repository for Clustering Then Estimation of Spatio-Temporal Self-Exciting Processes

This directory contains the code for the Clustering then Estimation of Spatio-Temporal Self-Exciting Processes (CTE) algorithm.

Studies of dynamic spatio-temporal processes of electrons in a 50 kHz/5 MHz dual-frequency dielectric barrier discharge plasma at atmospheric pressure

Dual-frequency excitation of dielectric barrier discharge (DBD) plasma reactors, where the plasma is generated by a low-frequency (LF) source and modulated by a radio-frequency (RF) source, have been widely adopted in the low-pressure regime. However, the impacts of the RF voltage and LF voltage amplitudes on the plasma parameters, including the spatiotemporal distributions of ion and electron densities, electron dynamics, and gas temperatures, remain poorly understood in the atmospheric pressure regime. The present work addresses this issue by conducting joint experimental and simulation studies for an atmospheric-pressure 50 kHz/5 MHz dual-frequency driven DBD plasma reactor based on phase-resolved optical emission spectra and a drift–diffusion model. The results demonstrate that the dynamic spatiotemporal behaviors of electrons change dramatically as the voltage of the RF component increases from 50 V to 300 V with the voltage of the LF component fixed at 1 kV. Moreover, the RF component is further demonstrated to modulate other plasma characteristics, such as particle densities, gas temperature, and argon emissions. These results contribute toward the tailoring of the non-equilibrium and nonlinear plasma parameters obtained under atmospheric pressure conditions.

Societal Costs of Dementia: 204 Countries, 2000-2019.

Dementia prevalence is expected to increase as populations grow and age. Therefore, additional resources will be needed to meet the global demand for care for Alzheimer's disease and related dementias (ADRD). Estimate global and country-level health care spending attributable to ADRD and the cost of informal care for people living with ADRD. We gathered data from three systematic literature reviews and the Global Burden of Disease 2019 study. We used spatiotemporal Gaussian process regression to impute estimates for the many countries without underlying data. We projected future costs to 2050 based on past trends in costs, diagnosis rates, and institutionalization rate. We estimated that in 2019, the direct health care spending attributable to ADRD across 204 countries reached $260.6 billion (95% uncertainty interval [UI] 131.6-420.4) and the cost of informal ADRD care was $354.1 billion (95% UI 190.0-544.1). On average, informal care represents 57% (95% UI 38-75%) of the total cost of care. We estimated that direct health care spending attributable to ADRD will reach $1.6 trillion (95% UI 0.6-3.3) in 2050, or 9.4% (95% UI 3.9-19.6%) of projected health spending worldwide. We estimated the cost of informal care will reach $0.9 trillion (95% UI 0.3-1.7) in 2050. These cost estimates underscore the magnitude of resources needed to ensure sufficient resources for people living with ADRD and highlight the role that informal care plays in provision of their care. Incorporating informal care cost estimates is critical to capture the social cost of ADRD.

Diffusion distribution model for damage mitigation in scanning transmission electron microscopy.

Despite the widespread use of Scanning Transmission Electron Microscopy (STEM) for observing the structure of materials at the atomic scale, a detailed understanding of some relevant electron beam damage mechanisms is limited. Recent reports suggest that certain types of damage can be modelled as a diffusion process and that the accumulation effects of this process must be kept low in order to reduce damage. We therefore develop an explicit mathematical formulation of spatiotemporal diffusion processes in STEM that take into account both instrument and sample parameters. Furthermore, our framework can aid the design of Diffusion Controlled Sampling (DCS) strategies using optimally selected probe positions in STEM, that constrain the cumulative diffusion distribution. Numerical simulations highlight the variability of the cumulative diffusion distribution for different experimental STEM configurations. These analytical and numerical frameworks can subsequently be used for careful design of 2- and 4-dimensional STEM experiments where beam damage isminimised.

Nonequilibrium Antigen Recognition during Infections and Vaccinations

The initial immune response to an acute primary infection is a coupled process of antigen proliferation, molecular recognition by naive B cells, and their subsequent clonal expansion. This process contains a fundamental problem: the recognition of an exponentially time-dependent antigen signal. Here, we show that an efficient immune response must be stringently constrained to B-cell lineages with high antigen binding affinity. We propose a tuned proofreading mechanism for primary recognition of new antigens, where the molecular recognition machinery is adapted to the complexity of the immune repertoire. We show that this process produces potent, specific, and fast recognition of antigens, maintaining a spectrum of genetically distinct B-cell lineages as input for affinity maturation. Our analysis maps the proliferation-recognition dynamics of a primary infection to a generalized Luria-Delbrück process, akin to the dynamics of the classic fluctuation experiment. This map establishes a link between signal recognition dynamics and evolution. We derive the resulting statistics of the activated immune repertoire: Antigen binding affinity, expected size, and frequency of active B-cell clones are related by power laws, which define the class of generalized Luria-Delbrück processes. Their exponents depend on the antigen and B-cell proliferation rate, the number of proofreading steps, and the lineage density of the naive repertoire. We extend the model to include spatiotemporal processes, including the diffusion-recognition dynamics of a vaccination. Empirical data of activated mouse immune repertoires are found to be consistent with activation involving about three proofreading steps. The model predicts key clinical characteristics of acute infections and vaccinations, including the emergence of elite neutralizers and the effects of immune aging. More broadly, our results establish infections and vaccinations as a new probe into the global architecture and functional principles of immune repertoires. Published by the American Physical Society 2024

Irregular surface temperature monitoring in liver procurement via time-vertex signal processing

The liver viability monitoring during its procurement is critical to guarantee the safe liver transportation. Traditionally, the viability is assessed by taking invasive biopsy on the liver surface. Recently, the noninvasive thermal images of liver surface have been used as an alternative assessment way. Researchers have proposed monitoring and classification approaches based on the thermal images. Existing works have demonstrated the importance of the temporal variation or the spatial variation of the thermal images to the viability monitoring. However, there is no prior work to leverage the graph structure of the spatio-temporal processes for the viability monitoring. In this paper, we propose a time-vertex signal processing framework for the irregular thermal data of the pure liver region monitoring. In particular, we extract features of irregular thermal data based on the joint Fourier transform (JFT), which is an integration of graph Fourier transform (GFT) and discrete Fourier transform (DFT). The extracted JFT features can accurately reconstruct the thermal images with a limited number of features. Then, we use a non-parametric online change-point estimation method, based on online scan B statistics, to monitor the JFT features without a parametric distribution. Our proposed framework is applied to both simulation and the liver procurement data, and achieves good monitoring performance.

The spatio-temporal evolution of spatial structure and supply-demand relationships of the ecological network in the Yellow River Delta region of China

Rapid urbanization has caused unregulated transformation of land use/land cover, further exacerbating the imbalance in ecological supply-demand relationships. Constructing the ecological supply-demand network by integrating natural ecological processes and human ecological demands into a unified spatio-temporal framework and considering their internal dynamic flow characteristics is an effective approach to address this issue. However, there is currently a lack of evaluation methods for the coordination of ecological network supply-demand relationships, which could provide essential guidance and reference. This article takes Yellow River Delta region (YRDR) as an example and utilizes the integrated traditional ecological network paradigm and ecosystem services theory to simulate the ecological supply-demand network for the years 2000, 2010, and 2020 in the study area. Then, a novel evaluation index system for the ecological network supply-demand relationships was proposed from the perspectives of intensity, capacity, and equilibrium. Based on this, combined with a graph-based network structure evaluation system, it reveals the spatio-temporal evolution process of the spatial structure and supply-demand relationship of the ecological network in the study area. The study results showed that: (1) The quantity of elements in the ecological supply network and the demand network show opposite trends, leading to an exacerbation of the spatial imbalance in the arrangement of ecological supply-demand network elements. (2) Although the structural changes in the ecological supply network were significantly greater than those in the ecological demand network, the overall structure of the ecological supply-demand network gradually improved. (3) However, under the coupled influence of imbalanced change in the spatial distribution and connectivity of ecological elements, the overall coordination of the ecological supply-demand relationships within the YRDR's ecological network is gradually deteriorating. (4) The proposed optimization strategy, "Two zones, Three belts, Four corridors", for the ecological security pattern in YRDR can accommodate the ecological process cycle and human ecological demand. This study can provide scientific references for "harmonious coexistence between humans and nature" ecological protection and sustainable development planning under the background of rapid urbanization.

Microplastic clouds in rivers: spatiotemporal dynamics of microplastic pollution in a fluvial system

BackgroundThe microplastic transport of rivers is a complex spatiotemporal process; however, only limited knowledge exists on it, making its monitoring complicated. The study aimed to analyze the spatial and temporal dynamics of suspended sediments and microplastics based on measurements (1) every five days for 2 years at one site and (2) annual repetition at 29 sites along the 750-km-long Tisza River for 3 years. Water samples were taken by pumping (1 m3). Machine learning algorithms were applied to Sentinel images to analyze the spatiality of sediment transport.ResultsIn the Tisza River (Central Europe), the microplastic concentration (MPCmean: 35 ± 27 item/m3) and the suspended sediment concentration (SSCmean: 60 ± 57 g/m3) showed high temporal variations. During low stages, the concentrations dropped as most transported sediments were deposited on the bottom. These sediments, including microplastics, were remobilized during flood waves, thus, higher MPC and SSC were measured. The first flood wave after a low-stage period had the highest concentrations. The increased transport capacity of the river during floods created large-scale suspended sediment and microplastic waves with increased concentrations. The mean MPC gradually increased between 2021 (19 ± 13.6 item/m3) and 2022 (23.7 ± 15.8 item/m3), and then it more than doubled (2023: 57 ± 44.8 item/m3). The tributaries acted as suspended sediment and microplastic conveyors.On the Sentinel images, medium-scale clouds were identified, with the suspended sediment clouds being more pronounced than microplastic clouds. Fewer and longer clouds appeared during low stages, separated by clearer water bodies. During flood waves, shorter clouds were detected. The tributaries with increased suspended sediment and microplastic transport created well-distinguishable clouds in the main river.ConclusionsIdentifying suspended sediment and microplastic clouds in a river could support more precise monitoring. The hydrological background of the monitoring and the existence of these clouds should be considered, as sampling from clouds with increased SSC and MPC provides different data than sampling from the clearer water bodies between two clouds.Graphical

Probing the capacity of a spatiotemporal deep learning model for short-term PM2.5 forecasts in a coastal urban area

Accurate forecast of fine particulate matter (PM2.5) is crucial for city air pollution control, yet remains challenging due to the complex urban atmospheric chemical and physical processes. Recently deep learning has been routinely applied for better urban PM2.5 forecasts. However, their capacity to represent the spatiotemporal urban atmospheric processes remains underexplored, especially compared with traditional approaches such as chemistry-transport models (CTMs) and shallow statistical methods other than deep learning. Here we probe such urban-scale representation capacity of a spatiotemporal deep learning (STDL) model for 24-hour short-term PM2.5 forecasts at six urban stations in Rizhao, a coastal city in China. Compared with two operational CTMs and three statistical models, the STDL model shows its superiority with improvements in all five evaluation metrics, notably in root mean square error (RMSE) for forecasts at lead times within 12 h with reductions of 49.8 % and 47.8 % respectively. This demonstrates the STDL model's capacity to represent nonlinear small-scale phenomena such as street-level emissions and urban meteorology that are in general not well represented in either CTMs or shallow statistical models. This gain of small-scale representation in forecast performance decreases at increasing lead times, leading to similar RMSEs to the statistical methods (linear shallow representations) at about 12 h and to the CTMs (mesoscale representations) at 24 h. The STDL model performs especially well in winter, when complex urban physical and chemical processes dominate the frequent severe air pollution, and in moisture conditions fostering hygroscopic growth of particles. The DL-based PM2.5 forecasts align with observed trends under various humidity and wind conditions. Such investigation into the potential and limitations of deep learning representation for urban PM2.5 forecasting could hopefully inspire further fusion of distinct representations from CTMs and deep networks to break the conventional limits of short-term PM2.5 forecasts.