Spatial-temporal Factors Research Articles

Study on Spatial-Temporal Disparities and Factors Influencing Electricity Consumption Carbon Emissions in China

Investigating the factors influencing the spatial-temporal disparities in China’s electricity consumption carbon emissions (ECCEs) will be of great help to advancing the reduction in carbon emissions on the consumption side of electricity. Based on the measurement of the ECCEs in 30 Chinese provinces between 2005 and 2021, we utilized the natural breakpoint method and the Dagum Gini coefficient to analyze the spatial-temporal disparities in ECCEs at the provincial and regional levels, and then we used Geodetector to explore the factors influencing the spatial-temporal disparities in ECCEs. The results revealed the following: (1) There were obvious inter-provincial spatial disparities in ECCEs, with coastal provinces such as Jiangsu and Guangdong consistently ranking at the top of the country and inland provinces such as Qinghai and Yunnan having relatively low carbon emission values. (2) The overall disparities in China’s ECCEs fluctuated and rose, with inter-regional disparities being the primary source of the overall disparities. (3) Economic development, industrialization level, population density, and foreign direct investment all had strong explanations for the spatial-temporal disparities in China’s ECCEs. When all these influencing factors were spatially superimposed, their effects were enhanced.

Study on Spatiotemporal Behavior of Self-driving Tourists in Inner Mongolia

Against the backdrop of normalized tourism and upgraded consumption, self-driving tours have become one of the main choices for tourists. Faced with the booming self-driving tour market, it is of great significance to explore the spatial behavior characteristics and temporal constraint factors of self-driving tourists within a geographical range using GPS trajectory data. This exploration aids in the planning and design of self-driving tour routes, tourism products, and promotes the healthy and sustainable development of the self-driving tour industry. This paper selects Inner Mongolia Autonomous Region as the self-driving tour hotspot area, collects GPS data of self-driving tourists from 2017 to 2022 on the online platform "Foooooot" using Python, and conducts in-depth exploration and comparative analysis of the spatiotemporal behavior trajectories of self-driving tourists, aiming to explore the spatiotemporal behavior characteristics of self-driving tourists and their constraining factors.

Research on carbon emission efficiency and spatial-temporal factors in the transportation industry: Evidence from the Yangtze River Economic Belt

<p>Severe climate problems have forced the Chinese government to put forward the goal of "carbon peak" and "carbon neutrality". The transport sector is a key area for carbon reduction. Based on this background, this paper constructs the Super-SBM model and the Malmquist-Luenberger index model containing the undesirable output. This paper measures the carbon emission efficiency of transport industry in China's Yangtze River Economic Belt from static and dynamic perspectives during 2014-2020. Finally, the GTWR model is constructed to analyze the factors affecting carbon emission efficiency. The results show that: (1) the carbon emission efficiency of provinces in the Yangtze River Economic Belt is characterized by heterogeneity. Hubei was the highest, while Shanghai, Hunan, Sichuan and Yunnan were below average. (2) According to the Malmquist-Luenberger index analysis, the transportation carbon emission efficiency of the Yangtze River Economic Belt has a downward trend. Over the 2014-2020 period, carbon efficiency will decline by an average of 7% per year. (3) Energy consumption structure, industrial upgrading, economic development and population agglomeration have significant effects on carbon emission efficiency.</p>

MVSTT: A Multiview Spatial-Temporal Transformer Network for Traffic-Flow Forecasting.

Accurate traffic-flow prediction remains a critical challenge due to complicated spatial dependencies, temporal factors, and unpredictable events. Most existing approaches focus on single-or dual-view learning and thus face limitations in systematically learning complex spatial-temporal features. In this work, we propose a novel multiview spatial-temporal transformer (MVSTT) network that can effectively learn complex spatial-temporal domain correlations and potential patterns from multiple views. First, we examine a temporal view and design a short-range gated convolution component from a short-term subview, and a long-range gated convolution component from a long-term subview. These two components effectively aggregate knowledge of the temporal domain at multiple granularities and mine patterns of node evolution across time steps. Meanwhile, in the spatial view, we design a dual-graph spatial learning module that captures fixed and dynamic spatial dependencies of nodes, as well as the evolution patterns of edges, from the static and dynamic graph subviews, respectively. In addition, we further design a spatial-temporal transformer to mine different levels of spatial-temporal features through multiview knowledge fusion. Extensive experiments on four real-world traffic datasets show that our method consistently outperforms the state-of-the-art baseline. The code of MVSTT is available at https://github.com/JianSoL/MVSTT.

City-level livestock methane emissions in China from 2010 to 2020

Livestock constitute the world’s largest anthropogenic source of methane (CH4), providing high-protein food to humans but also causing notable climate risks. With rapid urbanization and increasing income levels in China, the livestock sector will face even higher emission pressures, which could jeopardize China’s carbon neutrality target. To formulate targeted methane reduction measures, it is crucial to estimate historical and current emissions on fine geographical scales, considering the high spatial heterogeneity and temporal variability of livestock emissions. However, there is currently a lack of time-series data on city-level livestock methane emissions in China, despite the flourishing livestock industry and large amount of meat consumed. In this study, we constructed a city-level livestock methane emission inventory with dynamic spatial-temporal emission factors considering biological, management, and environmental factors from 2010 to 2020 in China. This inventory could serve as a basic database for related research and future methane mitigation policy formulation, given the population boom and dietary changes.

Spatial-temporal risk factors in the occurrence of rabies in Mexico.

Rabies is a zoonotic disease that affects livestock worldwide. The distribution of rabies is highly correlated with the distribution of the vampire bat Desmodus rotundus, the main vector of the disease. In this study, climatic, topographic, livestock population, vampire distribution and urban and rural zones were used to estimate the risk for presentation of cases of rabies in Mexico by co- Kriging interpolation. The highest risk for the presentation of cases is in the endemic areas of the disease, i.e. the States of Yucatán, Chiapas, Campeche, Quintana Roo, Tabasco, Veracruz, San Luis Potosí, Nayarit and Baja California Sur. A transition zone for cases was identified across northern Mexico, involving the States of Sonora, Sinaloa, Chihuahua, and Durango. The variables topography, vampire distribution, bovine population and rural zones are the most important to explain the risk of cases in livestock. This study provides robust estimates of risk and spread of rabies based on geostatistical methods. The information presented should be useful for authorities responsible of public and animal health when they plan and establish strategies preventing the spread of rabies into rabies-free regions of México.

AI-Oriented Two-Phase Multifactor Authentication in SAGINs: Prospects and Challenges

Space-air-ground integrated networks (SAGINs), which have emerged as an expansion of terrestrial networks, provide flexible access, ubiquitous coverage, high-capacity backhaul, and emergency/disaster recovery for mobile users (MUs). While the massive benefits brought by SAGIN may improve the quality of service, unauthorized access to SAGIN entities is potentially dangerous. At present, conventional crypto-based authentication is facing challenges, such as the inability to provide continuous and transparent protection for MUs. In this article, we propose an AI-oriented two-phase multi-factor authentication scheme (ATMAS) by introducing intelligence to authentication. The satellite and network control center collaborate on continuous authentication, while unique spatial-temporal features, including service features and geographic features, are utilized to enhance the system security. Our further security analysis and performance evaluations show that ATMAS has proper security characteristics which can meet various security requirements. Moreover, we shed light on lightweight and efficient authentication mechanism design through a proper combination of spatial-temporal factors.

Exploring the utilization of Electric-vehicles’ free public charging facilities in small towns. A case of the town of Cary, North Carolina

This study explores the influence of spatial–temporal factors on the utilization of publicly available electric vehicle (EV) charging stations. It utilized the EV energy consumption data collected between 2019 and 2022 from six charging locations in the town of Cary, North Carolina. Descriptive analysis, multiple linear regression, and survival models were applied to explore energy consumption and the charging duration. Study results revealed a great variation in energy consumption and charging duration. Sundays have the highest frequency of charging activities, the largest amount of energy consumed, and the longest total duration, while Mondays have the lowest frequency of charging activities. Most of the charging activities occurred between 6 a.m. and 9p.m. The relationship between charging duration and energy consumed revealed that, for each charging minute, the EV consumed between 87 and 111.3 Wh of energy. The observed variations were mainly associated with the location of a charging facility and charging model type. The regression model results revealed that, unlike nighttime, energy consumption is statistically significantly lower in the daytime. Exploration of the day of the week suggested that weekdays, especially Mondays and Thursdays, have higher energy consumption than weekends. Further, the energy consumption increased over time, as indicated by the higher energy consumed in 2022 compared to 2019. The pattern for charging duration is similar to that of energy consumption except for the days of the week. The findings are crucial to planners for better planning and deployment of charging facilities based on the expected utilization.

Mortality from neglected tropical ciseases in Brazil in the 21st Century: Analysis of spatial and temporal trends and associated factorsMortalidad por enfermedades tropicales desatendidas en Brasil en el siglo XXI: análisis de tendencias espaciales y temporales y factores asociados

To analyze the spatial-temporal distribution and factors associated with mortality from neglected tropical diseases (NTDs) in Brazil from 2000 to 2019. We performed an ecological study to analyze NTD-related deaths recorded in the Ministry of Health Mortality Information System (SIM). For the temporal analysis, the joinpoint method was used. Spatial dependence was analyzed using global Moran and local Getis-Ord Gi* indices. Four non-spatial and spatial regression models were used to identify factors associated with mortality. The mean mortality rate from NTDs in Brazil during the study period was 3.32 deaths per 100 000 inhabitants, with the highest rate (8.68 deaths per 100 000 inhabitants) recorded in the Midwest. The most prevalent causes of death were Chagas disease (n = 94 781; 74.9%) and schistosomiasis (n = 10 271; 8.1%). There was a 1.24% reduction (95%CI = -1.6; -0.9; P < 0.001) in NTD-related mortality in Brazil per year. A high/high spatial distribution pattern and hotspots were observed in municipalities in the states of Goiás, Minas Gerais, Bahia, Tocantins, and Piauí. The indicators "population in households with density > 2 people per bedroom" (β = -0.07; P = 0.00) and "municipal human development index" (β = -3.36; P = 0.08) were negatively associated with the outcome, while the "index of social vulnerability" (β = 2.74; P = 0.05) was positively associated with the outcome. Lower human development and higher social vulnerability are associated with higher mortality from NTDs, which should guide NTD prevention and control efforts.

What is affecting the improvement of green innovation efficiency in the old industrial base: evidence from Northeast China.

Green innovation is an important driving force for high-quality development and is vital for reinvigorating the old industrial bases in Northeast China. As such, this study investigates the spatial-temporal evolution characteristics and factors influencing green innovation efficiency (GIE) in Northeast China from 2005 to 2020, using the super-efficient EBM-Malmquist model, kernel density estimation, and random forest model. The results show the following. (1) The "growth effect" of technological change is the main force driving GIE improvement; the "horizontal effect" of pure technical efficiency change has started to play an important role; and the club convergence characteristics of GIE in Northeast China have started to be optimized. (2) GIE in Northeast China shows significant spatial differentiation. The urban agglomeration of Mid-southern Liaoning and Harbin-Changchun has had high values for GIE, indicating that green innovation has a cyclic cumulative effect and the spatial pattern of green innovation needs to be further optimized. (3) The random forest model is more accurate and provides more trustworthy results compared with the traditional multiple linear regression model. The results of random forest model measurement illustrate that the number of digital economy enterprises, public finance expenditure, GDP per capita, and vegetation coverage play a positive role in promoting GIE. The proportion of the non-farm population and industrial agglomeration plays a negative role in GIE. In the same period, the contribution of the number of digital economy enterprises≥0.41, public expenditure ≥0.47, GDP per capita≥0.39, and vegetation coverage≥0.36 to GIE reach maximum values and then remain unchanged.

Using Street View Images to Examine the Impact of Built Environment on Street Property Crimes in the Old District of CA City, China

Property crimes on the street are common in cities, posing a certain threat to people’s daily life safety and social stability. Therefore, it is essential to analyze the characteristics and spatial patterns of street property crimes in the built environment to make cities safe. Based on environmental criminological theories, this study takes the MC old district in CA City as a case study and uses a negative binomial regression model to analyze the influencing factors of street property crimes in different periods. The results show the temporal and spatial differentiation in street property crimes. In terms of time, the number of crime cases presents the features of “three peaks and two troughs.” In terms of space, crime cases show spatial clustering patterns, mainly concentrated in the commercial and prosperous areas where the main roads of the city are located. During the whole day, openness, banks, bars, and restaurants have a significant positive effect on crime occurrence; closeness, police cameras, grocery stores, and distance to the nearest police patrol station had a significant negative effect on crime occurrence. There are two explanations for the positive and negative correlations of some environmental variables with a crime before dawn, daytime, and nighttime. This study explored the spatial-temporal distribution and factors that influence the old district street property crimes by extracting physical environmental characteristics from street view images using deep learning algorithms and providing a reference base for police departments to prevent and combat crime.

Adaptive Spatiotemporal Transformer Graph Network for Traffic Flow Forecasting by IoT Loop Detectors

Extensive traffic flow data are received from the loop detector networks every second, which requires us to develop an effective and efficient algorithm to predict future traffic flow. However, dynamic traffic conditions on a road are not just influenced by sequential patterns in the temporal dimension, but also by other roadways in the spatial dimension. Although many successful models have been developed in previous studies to forecast future traffic flows, most of them have shortcomings in modeling spatial and temporal dependencies. In this article, we focus on spatial-temporal factors and propose a new adaptive spatial-temporal transformer graph network (ASTTGN) to improve the accuracy of traffic forecasting by jointly modeling the spatial-temporal information of road networks. Specifically, we propose an adaptive spatial-temporal transformer module, which contains two developed adaptive transformer modules for capturing dynamic spatial dependence and temporal dependence across multiple time steps, respectively. Finally, feature fusion is performed through a gated feature aggregation layer to simulate the effect of complex spatial-temporal factors on traffic conditions. In particular, the multihead attention mechanism employed by the transformer can effectively explore the potential spatial-temporal dependence patterns in different subspaces. Experimental results on two real-world traffic data sets demonstrate the superiority of the proposed model compared to existing techniques.

DISTRIBUIÇÃO ESPAÇO-TEMPORAL E FATORES ASSOCIADOS À MORTALIDADE POR HIV/AIDS ENTRE JOVENS NO NORDESTE BRASILEIRO

ABSTRACT Objective to analyze the spatial-temporal distribution and factors associated with HIV/AIDS mortality among young people in the Northeast from 2001 to 2020. Method ecological study with 2,509 deaths from HIV/AIDS from northeastern residents aged between 10 and 24 years of age, reported in the Mortality Information System. Temporal analysis techniques (Joinpoint) and detection of spatial clusters (Spatial Autocorrelation, Gets-Ord Gi* and Scan were used. Three spatial error and spatial lag (Spatial Error and Spatial Lag) and non-spatial regression models (Ordinary Least Squares-OLS) were used to identify the factors associated with mortality in northeastern municipalities, considering p&lt;0.05. Results the HIV/AIDS mortality rate among young people in the northeast was 0.4 deaths per 100,000 inhabitants. Maranhão (APC:7.1; CI95%:2.3-12.1), Sergipe (PcA:6.9; CI95%:1.8-12.2), Rio Grande do Norte (PcA:6.4; CI95%:1.8-11.2), Ceará (PcA:4.2; CI95%:1.5-7.0) and Alagoas (APC:3.2; CI95%:0.1-6.4) showed a significant increasing trend of deaths. Cluster detection techniques indicated clusters of deaths mainly on the coast of Pernambuco and north-central Maranhão. The indicators proportion of the population in households with density &gt;2 (β=0.012; p&lt;0.001) and per capita transfer of the continued benefit (β=0.000; p&lt;0.001) showed a positive relationship with the outcome. On the other hand, the proportion of extremely poor people (β=-0.011; p=0.029) and the Brazilian Deprivation Index (β=-0.195; p=0.009) were negatively associated with mortality. Conclusion there was a trend of increased mortality in five of the nine northeastern states. The spatial clusters were located mainly in Pernambuco and North-Central Maranhão. Interventions aimed at socioeconomic factors should be established to prevent HIV infections and deaths among young people.

Agricultural Price Prediction Based on Combined Forecasting Model under Spatial-Temporal Influencing Factors

Grain product price fluctuations affect the input of production factors and impact national food security. Under the influence of complex factors, such as spatial-temporal influencing factors, price correlation, and market diversity, it is increasingly important to improve the accuracy of grain product price prediction for agricultural sustainable development. Therefore, successful prediction of the agricultural product plays a vital role in the government’s market regulation and the stability of national food security. In this paper, the price of corn in Sichuan Province is taken as an example. Firstly, the apriori algorithm was used to search for the spatial-temporal influencing factors of price changes. Secondly, the Attention Mechanism Algorithm, Long Short-term Memory (LSTM), Autoregressive Integrated Moving Average (ARIMA), and Back Propagation (BP) Neural Network models were combined into the AttLSTM-ARIMA-BP model to predict the accurate price. Compared with the other seven models, the AttLSTM-ARIMA-BP model achieves the best prediction effect and possesses the strongest robustness, which improves the accuracy of price forecasting in complex environments and makes the application to other fields possible.

Multi-Semantic Path Representation Learning for Travel Time Estimation

Travel time estimation of a given path is a crucial task of Intelligent Transportation Systems (ITS). Accurate travel time estimation can benefit multiple downstream applications such as route planning, real-time navigation, and urban construction. However, it is a challenging problem since the travel time is largely affected by multiple complicated factors including spatial factors, temporal factors and external factors, and obtaining informative representations of a given path is not trivial. Most previous works solved this problem in either Euclidean space or non-Euclidean space, which was unilateral to represent the actual traveling path and led to relatively poor performance. To address this, this paper proposes a multi-semantic path representation method to exploit information in Euclidean space and non-Euclidean space simultaneously. First, since the path is composed of several segments, we generate semantic representations of segments in non-Euclidean space by taking both the time information and the historical co-occurrence into consideration. Second, as the path could be equally represented as several travelled intersections, semantic representations of intersection sequences are also extracted to improve the capability of the method by considering information in Euclidean space. Meanwhile, semantic representations from properties, including the length and the type of segments, are also incorporated into the model. Finally, a sequence learning component is added on the top to aggregate the information along the entire path and provides the final estimation. Extensive experiments were conducted on two real-world taxi trajectories datasets, and the experimental results demonstrate the superiority of the proposed method.

A Novel Vision-Based Framework for Identifying Dynamic Vehicle Loads on Long-Span Bridges: A Case Study of Jiangyin Bridge, China

Since the high randomness feature of traffic flows, it is challenging to be aware of the spatial-temporal characteristics of dynamic vehicle loads on a long-span bridge. Compared with conventional measuring approaches, computer vision-based techniques have competitive advantages in the measurement of factual vehicle loads parameters, including the passing time, velocity, vehicle type, and corresponding locations on the bridge at each timestamp. In this paper, a vision-based framework for the identifications of dynamic vehicle loads on long-span bridges is proposed. To be specific, the framework consists of two major parts. The first part is the extraction of spatial-temporal factors of vehicles, including the detection of vehicles and the estimation of tracklets. Here, Single Shot Multi-Box Detection (SSD) architecture with MobileNets is utilized for detecting vehicles, while the data association technique with Kalman filter is applied on detection results for generating tracklets. Then, in the second part, based on the detection and tracking results, a Multi-layer Monte Carlo (MLMC) model is developed for axle-load intensities inference and estimation. The learned model is utilized to infer the dynamic load of vehicles from the detection results and generated tracklets. The proposed methodology is validated via an experiment of a case study on the Jiangyin bridge, Jiangsu Province, China, and several spatial-temporal distributions of vehicle loads are specifically analyzed and discussed. Experimental results indicate our proposed framework is a feasible and reliable solution to predict and identify dynamic vehicle loads on long-span bridges, without the assistance of the deck-embedded Weigh-in-Motion (WIM) system.

Intelligent cognition of traffic loads on road bridges: From measurement to simulation – A review

Traffic load is a crucial but complicated factor in determining the in-service performance and deterioration behavior of bridges. A better understanding of traffic loads in different traffic densities has become increasingly important in structure health monitoring. As a result, for the traffic load measurement, the relevant technologies had great progress in the past decades. Therefore, we focus on introducing the state-of-the-art approaches most relevant to the traffic load cognition on road bridges, including in-site measurement and data-driven simulation. General principles of the traffic load cognition are firstly presented by reviewing different statistical analysis techniques for determining the spatial-temporal factors of vehicles. Then, this paper reviews various measurement methods carried out for the essential data of traffic loads. The methods are roughly grouped into mechanical, optical and microwave sensor-based methods. Within each category, technical descriptions of the sensor types, properties and applications are discussed in terms of theoretical formulas and feasible scenarios. This paper also implements qualitative and comprehensive comparisons between multiple measurement sensors to show the efficiency of each method or technique. Base on in-site measurement, several kinds of simulation models can be established for traffic loads on road bridges, including the modelling of single vehicles and the overall traffic flow. Considering the significant contribution of statistics-based deterministic, direct probabilistic methods, and artificial intelligence to traffic load cognition, we carried out the investigation on them in vehicle modelling. For on-bridge traffic flow simulation, three representative microscopic models are reviewed, involving the car-following, hydrodynamic, and cellular automatic models. Overall, this study highlights the application of intelligent cognition methods in identifying and simulating traffic loads on road bridges, potentially providing support for digitalised design, operation, and maintenance.

Processes of preferential flow in a eurasian steppe under different scenarios

The control mechanisms of temporal and small-scale spatial characteristics on different types of preferential flow (PF) processes during infiltration are still lacking. Two-year high-resolution in situ soil moisture profiles were used to investigate the controls of PF processes and its interaction with moisture and vegetation dynamics, which covers three hillslope positions (upslope–14˚, midslope–9˚, and downslope–4˚) and three vegetation treatments (bare ground, stubble grassland and natural grassland) in a typical semi-arid grassland of Eurasia. Two types of PF were identified. Non-sequential response (NSR) was taken as one indication of PF using the disordered response of soil moisture profiles to rainfall. Fast preferential flow in sequence response (PF-rate) was taken as another indicator when a measured maximum pore water rate higher than saturated hydraulic conductivity. The conditions for the occurrence of the two PFs were inconsistent and different from the studies which showed the higher probability of PFs under dryer conditions and with higher rainfall intensities. Small rainfall events with low rainfall intensity and relatively low soil moisture at most locations showed higher NSR frequency, which further highlight the strong dependence of NSR on hydrophobicity and macropore structure of the topsoil. Whereas a higher rainfall intensity increased PF-rate under wet conditions. Meanwhile, the occurrence of PF-rate was also significantly controlled by vegetation coverage, soil texture and saturated hydraulic conductivity of 5–10 cm soil layer. Furthermore, the frequency of PF was 15.9%–44.5%, and its contribution to the total infiltration was >50%. The water recession rate of NSR and PF was slower and faster than that of matrix flow, respectively. This demonstrates the importance of PF to water replenishment in scarce rain areas and the influence of combinations of spatial–temporal factors on different flow processes.

Spatial-Temporal Driving Factors of Urban Landscape Changes in the Karst Mountainous Regions of Southwest China: A Case Study in Central Urban Area of Guiyang City

Rapid urban expansion has significantly altered the regional landscape pattern, posing a serious threat to the sustainable development of natural and social ecosystems. By using landscape patterns indices and an area transfer matrix, this study analyzed the spatial-temporal changes of landscape patterns in the karst mountainous cities of southwest China from 2000 to 2020, by taking the central urban area of Guiyang City (CUAG) as the study area. This study explored the spatial and temporal driving factors of landscape pattern changes by using stepwise multiple linear regression and geographic detector methods. The results show: (1) CUAG’s landscape types altered changed drastically, with the area of forestland and construction land rapid increment and cultivated land decrement significantly. (2) The patches of construction land and forestland tended to be aggregated, the degree of fragmentation was reduced, and the shape was complex; cultivated land fragmentation was intensified. The connectivity of the landscape was improved, while the level of landscape diversity declined, the trend of landscape homogenization was obvious. (3) Socioeconomic and geographical endowment drivers have determined landscape pattern changes. The findings of this study may be used to interpret other similar landscapes worldwide and may imply the protection of urban ecosystem and sustainable development.

Fine-Grained Individual Air Quality Index (IAQI) Prediction Based on Spatial-Temporal Causal Convolution Network: A Case Study of Shanghai

Accurate and fine-grained individual air quality index (IAQI) prediction is the basis of air quality index (AQI), which is of great significance for air quality control and human health. Traditional approaches, such as time series, recurrent neural network or graph convolutional network, cannot effectively integrate spatial-temporal and meteorological factors and manage the dynamic edge relationship among scattered monitoring stations. In this paper, a ST-CCN-IAQI model is proposed based on spatial-temporal causal convolution networks. Both the spatial effects of multi-source air pollutants and meteorological factors were considered via spatial attention mechanism. Time-dependent features in the causal convolution network were extracted by stacked dilated convolution and time attention. All the hyper-parameters in ST-CCN-IAQI were tuned by Bayesian optimization. Shanghai air monitoring station data were employed with a series of baselines (AR, MA, ARMA, ANN, SVR, GRU, LSTM and ST-GCN). Final results showed that: (1) For a single station, the RMSE and MAE values of ST-CCN-IAQI were 9.873 and 7.469, decreasing by 24.95% and 16.87% on average, respectively. R2 was 0.917, with an average 5.69% improvement; (2) For all nine stations, the mean RMSE and MAE of ST-CCN-IAQI were 9.849 and 7.527, respectively, and the R2 value was 0.906. (3) Shapley analysis showed PM10, humidity and NO2 were the most influencing factors in ST-CCN-IAQI. The Friedman test, under different resampling, further confirmed the advantage of ST-CCN-IAQI. The ST-CCN-IAQI provides a promising direction for fine-grained IAQI prediction.