Space-time Interaction Research Articles

Spatiotemporally Comparative Analysis of HIV, Pulmonary Tuberculosis, HIV-Pulmonary Tuberculosis Coinfection in Jiangsu Province, China.

Pulmonary tuberculosis (PTB) is a severe chronic communicable disease that causes a heavy disease burden in China. Human Immunodeficiency Virus (HIV) and PTB coinfection dramatically increases the risk of death. This study analyzes the spatiotemporal dynamics of HIV, PTB and HIV-PTB coinfection in Jiangsu Province, China, and explores the impact of socioeconomic determinants. The data on all notified HIV, PTB and HIV-PTB coinfection cases were extracted from Jiangsu Provincial Center for Disease Control and Prevention. We applied the seasonal index to identify high-risk periods of the disease. Time trend, spatial autocorrelation and SaTScan were used to analyze temporal trends, hotspots and spatiotemporal clusters of diseases. The Bayesian space-time model was conducted to examine the socioeconomic determinants. The case notification rate (CNR) of PTB decreased from 2011 to 2019 in Jiangsu Province, but the CNR of HIV and HIV-PTB coinfection had an upward trend. The seasonal index of PTB was the highest in March, and its hotspots were mainly distributed in the central and northern parts, such as Xuzhou, Suqian, Lianyungang and Taizhou. HIV had the highest seasonal index in July and HIV-PTB coinfection had the highest seasonal index in June, with their hotspots mainly distributed in southern Jiangsu, involving Nanjing, Suzhou, Wuxi and Changzhou. The Bayesian space-time interaction model showed that socioeconomic factor and population density were negatively correlated with the CNR of PTB, and positively associated with the CNR of HIV and HIV-PTB coinfection. The spatial heterogeneity andspatiotemporal clusters of PTB, HIV and HIV-PTB coinfection are exhibited obviously in Jiangsu. More comprehensive interventions should be applied to target TB in the northern part. While in southern Jiangsu, where the economic level is well-developed and the population density is high, we should strengthen the prevention and control of HIV and HIV-PTB coinfection.

Space–time interactions between free-ranging red deer (Cervus elaphus) and grazing livestock on a fenced pasture

Livestock farming, which produces milk and meat, is practised by nearly all cultures. But in addition to this, grazing livestock fulfils other fundamentally important roles, one of which is the maintenance of open landscapes, which in turn provides habitat for many rare plant and animal species. Whether wild large herbivores, such as red deer (Cervus elaphus), benefit from the management of the open landscapes is unknown, and there are conflicting views on the nature of the interaction between livestock and wildlife. To investigate this question, we studied interactions in space and time between red deer and livestock in an 787 ha fenced pasture in eastern Germany. First, we used faecal pellet counts to determine whether red deer were using the fenced pasture and whether livestock were using the entire available pasture. Second, we fitted ten red deer, as well as eight Konik horses (Equus ferus caballus) and eight Heck cattle (Bos taurus) with GPS-collars. These data allowed us to perform analyses of space–time interactions. The faecal pellet data showed that red deer did indeed use the fenced pasture and that livestock used the entire extent of the pasture. Further, we verified a neutral space–time interaction between red deer and livestock within the pasture. Thus, there was no indication of a negative effect of the presence of livestock on red deer inside the pasture. For these reasons, we recommend that low-intensity grazing by cattle and horses can maintain open landscapes without detrimental effects on the use of the area by red deer.

Spatiotemporal analysis of the effect of global development indicators on child mortality

BackgroundChild mortality continue to be a major public health issue in most developing countries; albeit there has been a decline in global under-five deaths. The differences in child mortality can best be explained by socioeconomic and environmental inequalities among countries. In this study, we explore the effect of country-level development indicators on under-five mortality rates. Specifically, we examine potential spatio-temporal heterogeneity in the association between major world development indicators on under-five mortality, as well as, visualize the global differential time trend of under-five mortality rates.MethodsThe data from 195 countries were curated from the World Bank’s World Development Indicators (WDI) spanning from 2000 to 2017 and national estimates for under-five mortality from the UN Inter-agency Group for Child Mortality Estimation (UN IGME).We built parametric and non-parametric Bayesian space-time interaction models to examine the effect of development indicators on under-five mortality rates. We also used employed Bayesian spatio-temporal varying coefficient models to assess the spatial and temporal variations in the effect of development indicators on under-five mortality rates.ResultsIn both parametric and non-parametric models, the results show indicators of good socioeconomic development were associated with a reduction in under-five mortality rates while poor indicators were associated with an increase in under-five mortality rates. For instance, the parametric model shows that gross domestic product (GDP) (β = − 1.26, [CI − 1.51; − 1.01]), current healthcare expenditure (β = − 0.40, [CI − 0.55; − 0.26]) and access to basic sanitation (β = − 0.03, [CI − 0.05; − 0.01]) were associated with a reduction under-five mortality. An increase in the proportion practising open defecation (β = 0.14, [CI 0.08; 0.20]) an increase under-five mortality rate. The result of the spatial components spatial variation in the effect of the development indicators on under-five mortality rates. The spatial patterns of the effect also change over time for some indicators, such as PM2.5.ConclusionThe findings show that the burden of under-five mortality rates was considerably higher among sub-Saharan African countries and some southern Asian countries. The findings also reveal the trend in reduction in the sub-Saharan African region has been slower than the global trend.

Predicting of vertical displacements of structures of engineering buildings and facilities

Purpose. The research is aimed at solving the tasks of monitoring and assessing the state of an engineering object and determining predictive characteristics in the form of construction of various models of settlement and deformation of a structure based on the results of geodetic measurements, taking into account the space-time interaction of engineering objects with the geological environment in a seismic zone with the earthquake energy class values of K = 6–7.5, in the area of possible underground fluctuations with an intensity of 3–4 points. Methodology. An analytical methodology has been developed that takes into account the geological conditions of the foundation, changes in the groundwater level, soil compaction under significant static load, as well as the accumulation of damage in the supporting structures caused by the impact of numerous and insignificant underground seismic shocks, for predicting the vertical displacements of the engineering facility supporting structures. To assess the durability and stability of design components, the least squares method is used, which makes it possible to display the patterns of deformation process development. Obtaining new results during the period of further operation in order to draw up a technical conclusion on the state of the facility and its stability is possible by monitoring and predicting deformation deviations of individual components of an engineering object in the vertical plane. Findings. With the help of mathematical modeling of strength calculations of supporting structures based on geodetic measurement data, it is possible to determine the quantitative characteristics and patterns of deformation process development during the engineering facility operation. The performed calculations on mathematical modeling make it possible to reveal the distribution of the parameters of amplitude-frequency characteristics of the linearly deformable monolithic plate upper layer along the perimeter and to predict possible deformation changes over a certain period of time during the engineering facility operation. According to the executive survey data, vertical deviations of structures along the facility perimeter AB, BC, CD, DA have been determined in the range from 1 to 27 mm, which is the basis for predicting deformation deviations in the vertical plane. In addition, as a result of engineering-geological surveys conducted on the building construction site, the geological-lithological structure of the site has been determined. Originality. A methodology for predicting the deformation processes of individual sections of vertical structures and monolithic walls of an engineering facility, which are associated with the complex lithological structure of the foundation section and the location of the object in a seismically hazardous zone, is proposed. Practical value. The obtained results of studying the deformation processes of structures and individual facilities make it possible to take into account the form of complex interaction of individual sections and, in general, to predict deformation deviations in the vertical plane.

Seasonal and spatial variations of malaria transmissions in northwest Ethiopia: Evaluating climate and environmental effects using generalized additive model

The impacts of climate change and environmental predictors on malaria epidemiology remain unclear and not well investigated in the Sub-Sahara African region. This study was aimed to investigate the nonlinear effects of climate and environmental factors on monthly malaria cases in northwest Ethiopia, considering space-time interaction effects. The monthly malaria cases and populations sizes of the 152 districts were obtained from the Amhara public health institute and the central statistical agency of Ethiopia. The climate and environmental data were retrieved from US National Oceanic and Atmospheric Administration. The data were analyzed using a spatiotemporal generalized additive model. The spatial, temporal, and space-time interaction effects had higher contributions in explaining the spatiotemporal distribution of malaria transmissions. Malaria transmission was seasonal, in which a higher number of cases occurred from September to November. The long-term trend of malaria incidence has decreased between 2012 and 2018 and has turned to an increased pattern since 2019. Areas neighborhood to the Abay gorge and Benshangul-Gumuz, South Sudan, and Sudan border have higher spatial effects. Climate and environmental predictors had significant nonlinear effects, in which their effects are not stationary through the ranges of values of variables, and they had a smaller contributions in explaining the variabilities of malaria incidence compared to seasonal, spatial and temporal effects. Effects of climate and environmental predictors were nonlinear and varied across areas, ecology, and landscape of the study sites, which had little contribution to explaining malaria transmission variabilities with an account of space and time dimensions. Hence, exploring and developing an early warning system that predicts the outbreak of malaria transmission would have an essential role in controlling, preventing, and eliminating malaria in areas with lower and higher transmission levels and ultimately lead to the achievement of malaria GTS milestones.

Flexible spatio-temporal Hawkes process models for earthquake occurrences

Hawkes process is one of the most commonly used models for investigating the self-exciting nature of earthquake occurrences. However, seismicity patterns have complicated characteristics due to heterogeneous geology and stresses, for which existing methods with Hawkes process cannot fully capture. This study introduces novel nonparametric Hawkes process models that are flexible in three distinct ways. First, we incorporate the spatial inhomogeneity of the self-excitation earthquake productivity. Second, we consider the anisotropy in aftershock occurrences. Third, we reflect the space–time interactions between aftershocks with a non-separable spatio-temporal triggering structure. For model estimation, we extend the model-independent stochastic declustering (MISD) algorithm and suggest substituting its histogram-based estimators with kernel methods. We demonstrate the utility of the proposed methods by applying them to the seismicity data in regions with active seismic activities.

Bivariate spatial clustering in differential time trends of related tropical diseases: Application to diarrhea and intestinal parasite infections

There has been a rapid development of space–time multivariate disease mapping methods that focus on space–time variation of risks. Examining the posterior estimates of the space–time random effects can provide compelling epidemiological information that is necessary for public health monitoring. In this study, we propose and evaluate the posterior estimates of the random effects to examine spatial and temporal trends of two tropical diseases. Our model is a multivariate Bayesian space–time model with common spatial and temporal trends, and a space–time interaction term that allows different time trends for different areas. When applied to diarrhea and intestinal parasites data from Ghana, the model that consolidates all random effects as a multivariate conditional autoregressive prior was the best fit. The implementation is based on the notion that diarrheal and intestinal parasite infections share common risk factors Our novel contribution concerns the posterior joint evaluations of the spatial and temporal random effects into a 5 × 5 cross-tabulation table. The method presented is useful for developing and implementing joint epidemiological control strategies, especially in countries where resources are scarce.

A decade of child pedestrian safety in England: a Bayesian spatio-temporal analysis

BackgroundChild pedestrian injury is a public health and health equality challenge worldwide, including in high-income countries. However, child pedestrian safety is less-understood, especially over long time spans. The intent of this study is to understand factors affecting child pedestrian safety in England over the period 2011–2020.MethodsWe conducted an area-level study using a Bayesian space-time interaction model to understand the association between the number of road crashes involving child pedestrians in English Local Authorities and a host of socio-economic, transport-related and built-environment variables. We investigated spatio-temporal trends in child pedestrian safety in England over the study period and identified high-crash local authorities.ResultsWe found that child pedestrian crash frequencies increase as child population, unemployment-related claimants, road density, and the number of schools increase. Nevertheless, as the number of licensed vehicles per capita and zonal-level walking/cycling increase, child pedestrian safety increases. Generally, child pedestrian safety has improved in England since 2011. However, the socio-economic inequality gap in child pedestrian safety has not narrowed down. In addition, we found that after adjusting for the effect of covariates, the rate of decline in crashes varies between local authorities. The presence of localised risk factors/mitigation measures contributes to variation in the spatio-temporal patterns of child pedestrian safety.ConclusionsOverall, southern England has experienced more improvement in child pedestrian safety over the last decade than the northern regions. Our study revealed socio-economic inequality in child pedestrian safety in England. To better inform safety and public health policy, our findings support the importance of a targeted system approach, considering the identification of high-crash areas while keeping track of how child pedestrian safety evolves over time.

Human Behavior Recognition Based on Multiscale Convolutional Neural Network

The key problem in human behavior recognition is how to build a spatiotemporal feature extraction and classification network. Aiming at the problem that the existing channel attention mechanism directly pools the global average information of each channel and ignores its local spatial information, this paper proposes two improved channel attention modules, namely the space-time (ST) interaction module of matrix operation and the depth separable convolution module, combined with the research of human behavior recognition. Combined with the superior performance of convolutional neural network (CNN) in image and video processing, a multi-scale convolutional neural network method for human behavior recognition is proposed. Firstly, the behavior video is segmented, and low rank learning is performed on each video segment to extract the corresponding Low rank behavior information, and then these Low rank behavior information are connected on the time axis to obtain the Low rank behavior information of the whole video, so as to effectively capture the behavior information in the video, avoiding tedious extraction steps and various assumptions. The ability of neural network to model human behavior can be transferred and reused in networks with different structures. According to the different characteristics of data features at different network levels, two effective feature difference measurement functions are introduced to reduce the difference between features extracted from different network structures. Experiments on several public datasets show that the proposed method has a good classification effect. The experimental results show that the method has a good accuracy in human behavior recognition. It is proved that the proposed model not only improves the recognition accuracy, but also effectively reduces the computational complexity of output weights and improves the compactness of the model structure.

Bayesian hierarchical spatiotemporal modeling for forecasting diarrhea risk among children under 5 in Bandung city, Indonesia

The main objectives of this research are to identify significant spatial and temporal compo-nents associated with diarrhea and provide an accurate forecast. Using data from the Ban-dung city health surveillance system, the analysis reveals a decreasing trend in both the number of incidences and the estimated relative risks of diarrhea in most districts. Key fac-tors contributing to diarrhea variation include temporally structured, spatially structured, and unstructured effects of space-time interaction Type I. No clear seasonal pattern is observed in diarrhea incidence among children under five, emphasizing the need for consistent vigilance and preventive measures. Spatial clustering was observed in the eastern and western parts of Bandung city. The forecasting model predicts a continued decline in diarrhea incidence and relative risk throughout 2022.

The Folkloric-Festive continuum of an Emerging Elite in La Paz

This article presents the results of the ethnographic research conducted in the period 2015-2017 with the Fraternidad Verdaderos Intocables for the purpose of analyzing reciprocal exchanges through drinking and their impact on the collective sense of community in an urban context. It is proposed that social interactions mediated by alcohol at the space-time of the fiesta signal/impulse chains of reciprocity in present/future overlapping space-time interactions resulting in a folkloric-festive continuum that reinforces the bonds of participants as compadres. This article intends to contribute to the emerging scholarly interest in calendric folkloric festivities in Bolivia by linking that research with the study of alternatives ways to conceive citizenship through public ritual interactions. Contrary to a common assumption that ostentation and big quantities of drunkenness in folkloric events related to a new folklorist elite are “irrational,” ethnographic research found that the dynamics of alcohol distribution/consumption is very carefully structured ritually and has direct implications in social reproduction and collective identity among the participants.

Relational Analysis of Dirac Equation in Momentum Representation

In terms of the relational approach to space-time geometry and physical interactions, we show that the Dirac equation for a free fermion in the momentum representation can be obtained starting from a binary system of complex relations (BSCR) between elements of two abstract sets. With the derivation performed we show that the 4-dimensional pseudo-Euclidean momentum space is not needed a priori but naturally emerges from considerations of rather general character (2-spinor algebra). A bispinor wave function is constructed for a fermion with positive energy and an arbitrary distribution of momenta. Special attention is paid to physical assumptions that should be made to enable the construction.

Exploring spatiotemporal interactions: On the superiority of time over space.

Space and time mutually influence each other such that space affects time estimation (space-on-time effect), and conversely (time-on-space effect). These reciprocal interferences suggest that space and time are intrinsically linked in the human mind. Yet, recent evidence for an asymmetrical advantage for space over time challenges the classical theoretical interpretation. In the present study, we tested whether the superiority of space over time in magnitude interference depends on the cognitive resources engaged in the spatial task. We conducted three experiments in which participants performed judgments on temporal intervals and spatial distances in separate blocks. In each trial, two dots were successively flashed at various locations, and participants were to judge whether the duration or distance between the dots was short or long. To manipulate cognitive demands in the spatial task, distances varied across experiments (highly discriminable for the non-demanding spatial task in Experiment 1 and scarcely discriminable for the demanding spatial task in Experiment 2). Importantly, this manipulation tended to enhance perceptual sensitivity (as indexed by Weber Ratios) but slowed down the decision process (as indexed by response times) in the demanding experiment. Our results provide evidence for robust space-on-time and time-on-space effects (Experiments 1 and 2). More crucially, the involvement of cognitive resources in a demanding spatial task causes a massive time-on-space effect: Spatial judgments are indeed more influenced by irrelevant temporal information than the reverse (Experiments 2 and 3). Overall, the flexibility of spatiotemporal interferences has direct theoretical implications and questions the origins of space-time interaction.

Spatiotemporal analysis of African swine fever outbreaks on South African smallholder farms, 1993-2018.

African swine fever (ASF) is a contagious viral disease of swine worldwide. ASF in South Africa has for many years been confined to a controlled area in the northeast of the country that was proclaimed in 1935. Since 2012, outbreaks are more likely to occur in the historically ASF-free area. This study aimed to analyse the spatial and spatiotemporal structure of ASF outbreaks in South Africa between 1993 and 2018. Global space-time clustering of ASF outbreaks was investigated by the Diggle space-time K-function while Kulldorff's spatial scan statistic was applied to detect local cluster of ASF outbreaks. Globally, ASF outbreaks exhibit statistically significant spatial clustering. They have shown a significant negative space-time interaction at month scale (p = 0.003) but no significant space-time interaction at year scale (p = 0.577), revealing strong evidence that ASF cases that are close in space occur in months which are close and vice versa. In studying local area space-time clustering at both month and year scale, three significant local clusters associated with high-rate were detected. These clusters are localised in both the ASF-controlled area and outside the controlled area with radius varying from 60.84 km up to 271.43 km and risk ratio varying from 6.61 up to 17.70. At month scale, clusters with more outbreaks were observed between June 2017 and August 2017 and involved 22 outbreaks followed by the cluster that involved 13 outbreaks in January 2012. These results show the need to maintain high biosecurity standards on pig farms in both inside and outside the ASF-controlled areas.

Spatial functional data modeling of plant reflectances

Plant reflectance spectra, the profile of light reflected by leaves across different wavelengths, supply the spectral signature for a species at a spatial location to enable estimation of functional and taxonomic diversity for plants. We consider leaf spectra as “responses” to be explained spatially. These reflectance spectra are also functions over wavelength that respond to the environment. Our motivating data are gathered for several plant families from the Greater Cape Floristic Region (GCFR) in South Africa and lead us to develop rich novel spatial models that can explain spectra for genera within families. Wavelength responses for an individual leaf are viewed as a function of wavelength, leading to functional data modeling. Local environmental features become covariates. We introduce a wavelength, covariate interaction, since the response to environmental regressors may vary with wavelength, as may variance. Formal spatial modeling enables prediction of reflectances for genera at unobserved locations with known environmental features. We incorporate spatial dependence, wavelength dependence, and space–wavelength interaction (in the spirit of space–time interaction). We implement out-of-sample validation for model selection, finding that the model features above are informative for the functional data analysis. We supply ecological interpretation of the results under the selected model.

RELATIONAL SUBSTANTIATION OF THE HUBBLE LAW

In recent years, within the framework of the relational approach to the description of space-time and physical interactions, a number of studies have been carried out to substantiate cosmological effects. It has been shown that the cosmological redshift and the cosmic microwave background may be the result of contributions from emitted but not absorbed radiation. However, until now, Hubble’s law has not been fully derived on the basis of only relational ideas. In this paper, this conclusion is presented and the Hubble parameter is calculated in the modern era. For this, the contributions of the pulses of the emitted but not absorbed radiation to the pulse of a distant astronomical object (cluster of galaxies) are considered. It is shown that taking these contributions into account leads to the linear Hubble law.

Netizens’ risk perception in new coronary pneumonia public health events: an analysis of spatiotemporal distribution and influencing factors

BackgroundInternet search volume reflects the level of Internet users’ risk perception during public health events. The Internet search volume index model, an algorithm of concentration of Internet users, and statistical analysis of popular topics on Weibo are used to analyze the effects of time, space, and space-time interaction. We conducted in-depth research on the characteristics of the spatial and temporal distribution of Internet users’ risk perceptions of public health events and the associated influential factors.MethodsWe analyzed the spatiotemporal distribution characteristics of Internet users’ risk perception after the Wuhan “city closing” order during the coronavirus disease 2019 (COVID-19) pandemic. We established five linear regression models according to different time periods and analyzed factors influencing Internet users’ risk perception by employing a Poisson and spatial distribution and topic modeling analysis.ResultsEconomy, education, health, and the degree of information disclosure affect Internet users’ risk perception significantly. Internet users’ risk perception conforms to the exponential distribution law in time and has periodic characteristics and stability trends. Additionally, Internet users’ average arrival rate dropped from week 1 to week 8 after the “city closing.” Internet users’ risk perception has a uniform distribution in space, economic and social development level distribution consistency, spatial agglomeration, and other characteristics. The results of the time-space interaction show that after 8 weeks of COVID-19, Internet search hot topics have become more stable, and Internet users’ information demand structure has become more rational.ConclusionsThe Internet search cycle of the COVID-19 event is synchronized with the evolution cycle of the epidemic. The physical risk of Internet users is at the top of the risk structure, focusing on the strong concern about the government’s ability to control COVID-19 and its future trend. The government should strengthen network management; seize the risk control focus of key time nodes, regional locations, and information content of online communication; actively adjust the information content supply; effectively control the rebound of Internet users’ risk perception; establish a data-driven, risk-aware intelligence system for internet users; and guide people to actively face and overcome the potential risks and threats of COVID-19.

Amplification in Time and Dilution in Space: Partitioning Spatiotemporal Processes to Assess the Role of Avian-Host Phylodiversity in Shaping Eastern Equine Encephalitis Virus Distribution

Eastern equine encephalitis virus (EEEv) is an arthropod-borne virus and the causative agent of neurologic disease in humans, horses, poultry, and wildlife. Although EEEv is known to be transmitted in cycles involving avian hosts and ornithophilic mosquitoes, there is ongoing debate about the role avian-host phylodiversity plays in diluting or amplifying virus prevalence across geographic space and through time. This study leveraged seventeen years of non-human EEEv detections to quantify possible EEEv dilution and amplification effects in response to avian-host phylodiversity. In assessing EEEv and avian-host diversity relationships, comparisons were performed to illustrate how modeling decisions aimed at capturing spatial patterns, temporal trends, and space–time interactions impacted results and the interpretations drawn from those results. Principal findings indicated that increased avian phylodiversity promotes EEEv dilution across geographic space, but this dilution effect is scale-dependent and masked by amplification effects that occur through time. Findings further demonstrated that the decisions made when modeling complex spatiotemporal dynamics can readily contribute to contrasting statistical outcomes and results misinterpretation, even when arithmetic and mathematics are strictly correct.

Monitoring Land-Use Efficiency in China’s Yangtze River Economic Belt from 2000 to 2018

Monitoring of the indicator Sustainable Development Goal (SDG) 11.3.1 is important for understanding the coordination between land consumption rate (LCR) and population growth rate (PGR). However, the spatiotemporal indicator SDG 11.3.1 changes at the urban agglomeration (UA) level, and the relationship between LCR and PGR in the prefecture-level cities from different UAs remains unclear. In this study, we monitored the spatiotemporal indicator SDG 11.3.1 in the Yangtze River Economic Belt (YREB) and its three major UAs (i.e., Chengdu–Chongqing (CC), the Middle Reaches of the Yangtze River (MRYR), and the Yangtze River Delta (YRD)) for the periods 2000–2010, 2010–2015, and 2015–2018, using the space–time interaction (STI) method and Pearson’s method. Our major findings were as follows: (1) Compared with the world average of 1.28 for LCRPGR (i.e., ratio of LCR to PGR), except for the LCRPGR of the YRD (2000–2018) and CC (2000–2010), the LCRPGR of CC, the MRYR, and the YREB was lower than 1.28 during 2000–2018. (2) The gaps in both population and built-up area between the YREB and the three UAs did not narrow, but widened. (3) Compared with the LCRPGR in China, except for the LCRPGR of the YRD (2000–2018) and CC (2000–2010), the LCRPGR of the YREB increased from 1.21 to 1.23 between 2000–2010 and 2010–2015, and then decreased to 1.16 in 2015–2018, indicating that the relationship between LCR and PGR in the YREB is relatively stable. (4) A significant positive relationship (p < 0.001) was found between LCR and PGR in CC, the MRYR, the YRD, and the YREB. We conclude that the indicator SDG 11.3.1 is a helpful tool for evaluating land-use efficiency caused by the LCR and PGR at the UA level. Our results provide information support for promoting sustainable and coordinative development between LCR and PGR.

SPATE-GAN: Improved Generative Modeling of Dynamic Spatio-Temporal Patterns with an Autoregressive Embedding Loss

From ecology to atmospheric sciences, many academic disciplines deal with data characterized by intricate spatio-temporal complexities, the modeling of which often requires specialized approaches. Generative models of these data are of particular interest, as they enable a range of impactful downstream applications like simulation or creating synthetic training data. Recently, COT-GAN, a new GAN algorithm inspired by the theory of causal optimal transport (COT), was proposed in an attempt to improve generation of sequential data. However, the task of learning complex patterns over time and space requires additional knowledge of the specific data structures. In this study, we propose a novel loss objective combined with COT-GAN based on an autoregressive embedding to reinforce the learning of spatio-temporal dynamics. We devise SPATE (spatio-temporal association), a new metric measuring spatio-temporal autocorrelation. We compute SPATE for real and synthetic data samples and use it to compute an embedding loss that considers space-time interactions, nudging the GAN to learn outputs that are faithful to the observed dynamics. We test our new SPATE-GAN on a diverse set of spatio-temporal patterns: turbulent flows, log-Gaussian Cox processes and global weather data. We show that our novel embedding loss improves performance without any changes to the architecture of the GAN backbone, highlighting our model's increased capacity for capturing autoregressive structures.