Spatial-temporal characters of Antarctic sea ice variation

Using sea ice concentration dataset covering the period of 1968-2002 obtained from the Hadley Center of UK, this paper investigates characters of Antarctic sea ice variations .The finding demonstrates that the change of mean sea-ice extent is almost consistent with that of sea-ice area, so sea-ice extent can be chosen to go on this research. The maximum and the minimum of Antarctic sea ice appear in September and February respectively. The maximum and the maximal variation of sea ice appear in Weddell Sea and Ross Sea, while the minimum and the minimal variation of sea-ice appear in Antarctic Peninsula. In recent 35 years, as a whole, Antarctic sea ice decreased distinctly. Moreover, there are 5 subdivision characteristic regions considering their different variations. Hereinto, the sea-ice extent of Weddell Sea and Ross Sea regions extends and area increases, while the sea-ice extent of the other three regions contracts and area decreases. They are all of obvious 2-4 years and 5-7 years significant oscillation periods. It is of significance for further understanding the sea-ice-air interaction in Antarctica region and discussing the relationship between sea-ice variation and atmospheric circulation.

Comprehensive evaluation and spatial-temporal evolution characteristics of urban resilience in Chengdu-Chongqing Economic Circle

To clarify the connotations and extensions of urban resilience,this study focuses on the Chengdu-Chongqing Economic Circle with 16 cities as research subjects.A comprehensive evaluation index system was constructed to measure the resilience of each city from 2003 to 2020.The spatial-temporal evolution characteristics were analyzed using Kernel density estimation,standard deviation ellipse,and spatial Markov chain analysis,and the spatial Tobit model was introduced to discover the influencing factors.The results indicate the following:①Urban resilience in the Chengdu-Chongqing Economic Circle displays an upward trend,with the center of gravity moving to the southwest,and the polarization phenomenon intensifying.②The urban resilience level in a region has certain spatial and geographical dependence,while the probability of urban resilience transfer differs in adjacent cities with different resilience levels.③Urban centrality,economic scale,openness level,and financial development promote urban resilience,whereas government scale significantly inhibits it.Finally,this paper proposes countermeasures and suggestions to improve the urban resilience of the Chengdu-Chongqing Economic Circle.

Spatial-temporal Divergence Characteristics and Driving Factors of Green Economic Efficiency in the Yangtze River Economic Belt of China

The spatial and temporal variation of green economic efficiency and its driving factors are of great significance for the construction of high-efficiency and low-consumption green development model and sustainable socio-economic development.The research focused on the Yangtze River Economic Belt(YREB)and employed the miniumum distance to strong efficient frontier DEA(MinDs)model to measure the green economic efficiency of the municipalities in the region between 2008 and 2020.Then,the spatial autocorrelation model was used to analyze the evolution characteristics of its spatial pattern.Finally,Geodetector was applied to reveal the drivers and their interactions on green economic efficiency.It is found that:1)the overall green economic efficiency of the YREB from 2008 to 2020 shows a W-shaped fluctuating upward trend,green economic efficiency is greater in the downstream and smallest in the upstream;2)the spatial distribution of green economic efficiency shows clustering characteristics,with multi-core clustering based on‘city clusters-central cities'becoming more obvious over time;the High-High agglomeration type is mainly clustered in Jiangsu and Zheji-ang,while the Low-Low agglomeration type is clustered in the western Sichuan Plateau area and southwestern Yunnan;3)from input-output factors,whether it is the YREB as a whole or the upper,middle and lower reaches regions,the economic development level,labor input,and capital investment are the leading factors in the spatial-temporal evolution of green economic efficiency,with the com-prehensive influence of economic development level and pollution index being the most important interactive driving factor;4)from so-cio-economic factors,information technology drivers such as government intervention,transportation accessibility,information infra-structure,and Internet penetration are always high impact influencers and dominant interaction factors for green economic efficiency in the YREB and the three major regions in the upper,middle and lower reaches.Accordingly,the article puts forward relevant policy re-commendations in terms of formulating differentiated green transformation strategies,strengthening network leadership and informa-tion technology construction and coordinating multi-factor integrated development,which could provide useful reference for promoting synergistic green economic efficiency in the YREB.

Spatial-temporal Variation Characteristics of Water Quality in the Lower Reaches of the Nenjiang River

As an important river in the western part of Jilin Province,the lower reach of the Nenjiang River is an important wetland water source conservation area in Jilin Province.Within the watershed,it governs the Momoge Wetland,the Xianghai Wetland,and the Danjiang Wetland in Jilin Province.The main problem in the lower reaches of the Nenjiang River is the uneven distribution of water resources in time and space,and the intensification of land salinization.Zhenlai County and Da an City in the Nenjiang River Basin have sufficient surface water resources,with surface water as the drinking water source.Baicheng City and Tongyu County have scarce surface water resources,and both use groundwater as their domestic water source.The main polluted section in the basin is the Xianghai Reservoir,and the annual water quality evaluation is Class V.However,the water quality of the Tao er River,the main stream of the Nenjiang River,is significantly better than that of the Xianghai Reservoir.In order to better study the water environmental pollution situation in the Nenjiang River basin,monitoring data from five sections of non seasonal rivers in the basin from 2012 to 2021 were selected for studying water quality.This in-depth exploration of the water pollution status and river water quality change trends in the Nenjiang River basin is of great significance for future rural development,agricultural pattern transformation,and the promotion of water ecological civilization construction.

Spatial-temporal Evolvement Characteristics of Climate Productivity for the Plants on Inner Mongolia Desert Steppe

Thornthwaite Memorial model and other statistic methods were used to calculate the climate-productivity of plants with the meteorological data from 1961 to 2007 at 9 stations distributed on Inner Mongolia desert steppe.The spatial and temporal variation characteristics of climate-productivity were analyzed by using the methods of the tendency rate of the climate trend,accumulative anomaly,and spatial difference and so on.The results showed that the climate-productivity kept linear increased trend over Inner Mongolia desert steppe in recent 47 years,but not significant.In spatial distribution,the climate-productivity reduced with the increased latitude.The climate-productivity in southwest part of Inner Mongolia desert steppe was growing while that in the southeast was reducing.The variation rate of the climate-productivity increased from the northwest part to the southeast part of Inner Mongolia desert steppe.In recent 47 years,the climate-productivity in southeast Jurh underwent the greatest decreasing extent,and the region was the sensitive area of the climate-productivity variation.

Spatial-Temporal Distribution Characteristics and Limiting Factors of Medium-low Yield Farmland in Tianjin

[Objective] This paper aimed to understand the area change and distribu- tion of medium-low yield farmland, and offered basis to the improvement of mediumlow farmland and its increase of grain production in Tianjin. [Method] Based on the statistical date of Tianjin and its relevant counties and districts, the yield standard was set up to classify high-yield, medium-yield and low-yield farmland in Tianjin. The author analyzed area change of medium-low yield farmland in six agricultural counties and districts （including Jixian County, Wuqing District, Baodi District, Ninghe County, Jinghai County and Dagang district of Binghai New Area） from 1980 to 2010. [Result] The results showed that the average yield of grain rose from 2 445 kg/hm^2 in 1980 to 5 130 kg/hm^2 in 2010, increasing 109.82%. The area of mediumlow yield farmland was reduced from 291 250.13 hm^2 in 1985 to 76 489.87 hm^2 in 2010, coming down 74%. In Tianjin, the area of medium-low yield farmland of 2010 accounted for 19% of the total farmland, of which the ratios of medium-low yield farmland of Jinghai County, Jixian County, Dagang district of Binghai New Area, Wuqing District, Baodi District and Ninghe County were 43.12%, 18.59%, 17.23%, 14.01%, 7.05% and 0, respectively. Low soil nutrient content, drought and water shortage, as well as soil salinization were the main yield limiting factors to mediumlow yield farmland in Tianjin in 2010. [Conclusion] The countermeasures to improve the medium-low yield farmland were proposed, involving enhancing the investment of the government, strengthening the construction of water conservancy infrastructure, further improving the soil fertility, as well as saline and alkaline land, optimizing the farming system and planting drought and salt tolerance crops, etc.

Research on Spatial-Temporal Characteristics and Driving Factor of Agricultural Carbon Emissions in China

Macroscopic grasp of agricultural carbon emissions status, spatial-temporal characteristics as well as driving factors are the basic premise in further research on China’s agricultural carbon emissions. Based on 23 kinds of major carbon emission sources including agricultural materials inputs, paddy ifeld, soil and livestock breeding, this paper ifrstly calculated agricultural carbon emissions from 1995 to 2010, as well as 31 provinces and cities in 2010 in China. We then made a decomposed analysis to the driving factors of carbon emissions with logarithmic mean Divisia index （LMDI） model. The results show：（1） The amount of agricultural carbon emissions is 291.1691 million t in 2010. Compared with 249.5239 million t in 1995, it increased by 16.69%, in which, agricultural materials inputs, paddy ifeld, soil, enteric fermentation, and manure management accounted for 33.59, 22.03, 7.46, 17.53 and 19.39%of total agricultural carbon emissions, respectively. Although the amount exist ups and downs, it shows an overall trend of cyclical rise; （2） There is an obvious difference among regions：the amount of agricultural carbon emissions from top ten zones account for 56.68%, while 9.84%from last 10 zones. The traditional agricultural provinces, especially the major crop production areas are the main source regions. Based on the differences of carbon emission rations, 31 provinces and cities are divided into ifve types, namely agricultural materials dominant type, paddy ifeld dominant type, enteric fermentation dominant type, composite factors dominant type and balanced type. The agricultural carbon emissions intensity in west of China is the highest, followed by the central region, and the east zone is the lowest; （3） Compared with 1995, efifciency, labor and structure factors cut down carbon emissions by 65.78, 27.51 and 3.19%, respectively;while economy factor increase carbon emissions by 113.16%.

A SPATIAL-TEMPORAL DISTRIBUTION CHARACTERISTICS STUDY ON THE ATMOSPHERIC CARBON DIOXIDE OBSERVED BY GOSAT SATELLITE REMOTE SENSING

The variation of the atmospheric Carbon Dioxide （CO2） concentration plays an important role in global cli- mate and agriculture. We analyzed the spatial-temporal characteristics of CO2 in the China region and around the globe with the CO2 column mixing ratios observed by the Japanese GOSAT satellite （Greenhouse Gases Observing Satellite）. In order to make sure that the accuracy of the CO2 data retrieved by the satellite meets the needs of the climate charac- teristics analyses, we ran a validation on the CO2 column mixing ratios retrieved by the satellite against the ground-based TCCON （Total Carbon Column Observing Network） observation data. The result shows that the two sets of data have a correlation coefficient of higher than 0.7, and a bias of within 2.2 ppmv. Therefore, the GOSAT CO2 da- ta can be used for the climate characteristics analysis of global CO2. Our analysis on the spatial-temporal characteristics of the CO2 column mixing ratios observed during the period of June 2009 through January 2014 proved that, with the impact of the natural emission of near ground CO2 and human activities, the global CO2 concentration has a significant latitudinal characteristics with its highest level averaging 390 oomv in the 0-40？N latitudinal zone in the Northern Hemisphere, and 387 ppmv in the Southern Hemisphere. China has a relatively higher CO2 concentration with the highest level exceeding 398 ppmv, and the eastern area higher than the western area. The variation of global CO2 concentration shows a seasonal pattern, i.e. the CO2 concen- tration reaches its highest in spring in the Northern Hemisphere averaging more than 392 ppmv, second highest in win- ter, and lowest in summer averaging less than 387 ppmv. It fluctuates the most in the Northern Hemisphere with an av- erage concentration of 392.5 ppmv in April, and 385.5 ppmv in July. While in the Southern Hemisphere, the seasonal fluctuation is smaller with the highest concentration occurring in July. Over the recent years, the global CO2 concentra- tion has shown an elevating trend with an average annual increase rate of 1.58 ppmv per year. It is a challenge that the human kind has to face to slow down the increase of the CO2 concentration.

Characterizing urban expansion of Korla City and its spatial-temporal patterns using remote sensing and GIS methods

Cities provide spatial contexts for populations and economic activities. Determining the spatial-temporal patterns of urban expansion is of particular significance for regional sustainable development. To achieve a better understanding of the spatial-temporal patterns of urban expansion of Korla City, we explore the urban expansion characteristics of Korla City over the period 1995-2015 by employing Landsat TM/ETM＋ images of 1995, 2000, 2005, 2010, and 2015. Urban land use types were classified using the supervised classification method in ENVI 4.5. Urban expansion indices, such as expansion area, expansion proportion, expansion speed, expansion intensity, compactness, and fractal dimension, were calculated. The spatial-temporal patterns and evolution process of the urban expansion （e.g., urban gravity center and its direction of movement） were then quantitatively analyzed. The results indicated that, over the past 25 years, the area and proportion of urban land increased substantially with an average annual growth rate of 15.18%. Farmland and unused land were lost greatly due to the urban expansion. This result might be attributable to the rapid population growth and the dramatic economic development in this area. The city extended to the southeast, and the urban gravity center shifted to the southeast as well by about 2118 m. The degree of urban compactness tended to decrease and the fractal dimension index tended to increase, indicating that the spatial pattern of Korla City was becoming loose, complex, and unstable. This study could provide a scientific reference for the studies on urban expansion of oasis cities in arid land.

Multivariate Time Series Anomaly Detection Based on Spatial-Temporal Network and Transformer in Industrial Internet of Things

In the Industrial Internet of Things(IIoT),sensors generate time series data to reflect the working state.When the systems are attacked,timely identification of outliers in time series is critical to ensure security.Although many anomaly detection methods have been proposed,the temporal correlation of the time series over the same sensor and the state(spatial)correlation between different sensors are rarely considered simultaneously in these methods.Owing to the superior capability of Transformer in learning time series features.This paper proposes a time series anomaly detection method based on a spatial-temporal network and an improved Transformer.Additionally,the methods based on graph neural networks typically include a graph structure learning module and an anomaly detection module,which are interdependent.However,in the initial phase of training,since neither of the modules has reached an optimal state,their performance may influence each other.This scenario makes the end-to-end training approach hard to effectively direct the learning trajectory of each module.This interdependence between the modules,coupled with the initial instability,may cause the model to find it hard to find the optimal solution during the training process,resulting in unsatisfactory results.We introduce an adaptive graph structure learning method to obtain the optimal model parameters and graph structure.Experiments on two publicly available datasets demonstrate that the proposed method attains higher anomaly detection results than other methods.

Spatial-temporal Distribution Characteristics of Global Seismic Clusters and Associated Spatial Factors

Earthquakes exhibit clear clustering on the earth. It is important to explore the spatial-temporal characteristics of seismicity clusters and their spatial heterogeneity. We analyze effects of plate space, tectonic style, and their interaction on characteristic of cluster.Based on data of earthquakes not less than moment magnitude(M_w) 5.6 from 1960 to 2014, this study used the spatial-temporal scan method to identify earthquake clusters. The results indicate that seismic spatial-temporal clusters can be classified into two types based on duration: persistent clusters and burst clusters. Finally, we analysed the spatial heterogeneity of the two types. The main conclusions are as follows: 1) Ninety percent of the persistent clusters last for 22-38 yr and show a high clustering likelihood;ninety percent of the burst clusters last for 1-1.78 yr and show a high relative risk. 2) The persistent clusters are mainly distributed in interplate zones, especially along the western margin of the Pacific Ocean. The burst clusters are distributed in both intraplate and interplate zones, slightly concentrated in the India-Eurasia interaction zone. 3) For the persistent type, plate interaction plays an important role in the distribution of the clusters’ likelihood and relative risk. In addition, the tectonic style further enhances the spatial heterogeneity. 4) For the burst type,neither plate activity nor tectonic style has an obvious effect on the distribution of the clusters’ likelihood and relative risk. Nevertheless,interaction between these two spatial factors enhances the spatial heterogeneity, especially in terms of relative risk.

AFSTGCN:Prediction for multivariate time series using an adaptive fused spatial-temporal graph convolutional network

The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other fields.Furthermore,it is important to construct a digital twin system.However,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted accuracy.In this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)layer.Specifically,we fuse the spatial-temporal graph based on the interrelationship of spatial graphs.Simultaneously,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding methods.Subsequently,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)module.The module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like data.AFSTGCN dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted accuracy.Experiments on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.

Spatial-temporal variance of reclamation soil physical and chemical character in opencast mine region

In order to study the effects of soil compaction, and soil physical and chemicalcharacteristics after land reclamation, selected lands that were reclaimed after 1, 2, 3, 4,and 5 a, respectively, in the Majiata Mine of the Shendong Open Pit; tested the effects ofsoil compaction; and collected soil samples from 5 different depths, which are 0-7.62,7.62-15.24, 15.24-22.86, 22.86-30.48, and 30.48-38.10 cm, respectively. The resultsshow that: Land reclamation leads to soil compaction. The lowest effect of soil compaction is in the top layer and the highest one at the depth of 20-30 cm; The bulk density of reclaimed soil is higher than that of undisturbed soil; this declines with the reclamation and nearly reaches the level of undisturbed soil after 5-year reclamation;The content of reclaimed soil nutrients is lower than that of undisturbed soil. The lowest one is inthe soil dumping site, which reaches the level of undisturbed soil after 5-year reclamation;The pH value of reclaimed soil is lower than that of undisturbed soil. The highest one isin the soil dumping site; this declines with the reclamation.

Instance Segmentation of Characters Recognized in Palmyrene Aramaic Inscriptions

This study presents a single-class and multi-class instance segmentation approach applied to ancient Palmyrene inscriptions,employing two state-of-the-art deep learning algorithms,namely YOLOv8 and Roboflow 3.0.The goal is to contribute to the preservation and understanding of historical texts,showcasing the potential of modern deep learning methods in archaeological research.Our research culminates in several key findings and scientific contributions.We comprehensively compare the performance of YOLOv8 and Roboflow 3.0 in the context of Palmyrene character segmentation—this comparative analysis mainly focuses on the strengths and weaknesses of each algorithm in this context.We also created and annotated an extensive dataset of Palmyrene inscriptions,a crucial resource for further research in the field.The dataset serves for training and evaluating the segmentation models.We employ comparative evaluation metrics to quantitatively assess the segmentation results,ensuring the reliability and reproducibility of our findings and we present custom visualization tools for predicted segmentation masks.Our study advances the state of the art in semi-automatic reading of Palmyrene inscriptions and establishes a benchmark for future research.The availability of the Palmyrene dataset and the insights into algorithm performance contribute to the broader understanding of historical text analysis.

Fireworks Optimization with Deep Learning-Based Arabic Handwritten Characters Recognition Model

Handwritten character recognition becomes one of the challenging research matters.More studies were presented for recognizing letters of various languages.The availability of Arabic handwritten characters databases was confined.Almost a quarter of a billion people worldwide write and speak Arabic.More historical books and files indicate a vital data set for many Arab nationswritten in Arabic.Recently,Arabic handwritten character recognition(AHCR)has grabbed the attention and has become a difficult topic for pattern recognition and computer vision(CV).Therefore,this study develops fireworks optimizationwith the deep learning-based AHCR(FWODL-AHCR)technique.Themajor intention of the FWODL-AHCR technique is to recognize the distinct handwritten characters in the Arabic language.It initially pre-processes the handwritten images to improve their quality of them.Then,the RetinaNet-based deep convolutional neural network is applied as a feature extractor to produce feature vectors.Next,the deep echo state network(DESN)model is utilized to classify handwritten characters.Finally,the FWO algorithm is exploited as a hyperparameter tuning strategy to boost recognition performance.Various simulations in series were performed to exhibit the enhanced performance of the FWODL-AHCR technique.The comparison study portrayed the supremacy of the FWODL-AHCR technique over other approaches,with 99.91%and 98.94%on Hijja and AHCD datasets,respectively.

Spatial-temporal characteristics and influencing factors of relative humidity in arid region of Northwest China during 1966–2017

Playing an important role in global warming and plant growth,relative humidity(RH)has profound impacts on production and living,and can be used as an integrated indicator for evaluating the wet-dry conditions in the arid and semi-arid area.However,information on the spatial-temporal variation and the influencing factors of RH in these regions is still limited.This study attempted to use daily meteorological data during 1966–2017 to reveal the spatial-temporal characteristics of RH in the arid region of Northwest China through rotated empirical orthogonal function and statistical analysis method,and the path analysis was used to clarify the impact of temperature(T),precipitation(P),actual evapotranspiration(ETa),wind speed(W)and sunshine duration(S)on RH.The results demonstrated that climatic conditions in North Xinjiang(NXJ)was more humid than those in Hexi Corridor(HXC)and South Xinjiang(SXJ).RH had a less significant downtrend in NXJ than that in HXC,but an increasingly rising trend was observed in SXJ during the last five decades,implying that HXC and NXJ were under the process of droughts,while SXJ was getting wetter.There was a turning point for the trend of RH in Xinjiang,which occurred in 2000.Path analysis indicated that RH was negatively correlated to T,ETa,W and S,but it increased with increase of P.S,T and W had the greatest direct effects on RH in HXC,NXJ and SXJ,respectively.ETa was the factor which had the greatest indirect effect on RH in HXC and NXJ,while T was the dominant factor in SXJ.

Spatial-temporal distribution and geochemistry of highly evolved Mesozoic granites in Great Xing’an Range,NE China:Discriminant criteria and geological significance

Highly evolved granite is an important sign of the mature continent crust and closely associated with deposits of rare metals.In this work,the authors undertake systematically zircon U-Pb ages and whole rock elemental data for highly evolved granitic intrusions from the Great Xing’an Range(GXR),NE China,to elucidate their discriminant criteria,spatial-temporal distribution,differentiation and geodynamic mecha-nism.Geochemical data of these highly evolved granites suggest that high w(SiO_(2))(>70%)and differentiation index(DI>88)could be quantified indicators,while strong Eu depletion,high TE_(1,3),lowΣREE and low Zr/Hf,Nb/Ta,K/Rb could only be qualitative indicators.Zircon U-Pb ages suggest that the highly evolved gran-ites in the GXR were mainly formed in Late Mesozoic,which can be divided into two major stages:Late Ju-rassic-early Early Cretaceous(162-136 Ma,peak at 138 Ma),and late Early Cretaceous(136-106 Ma,peak at 126 Ma).The highly evolved granites are mainly distributed in the central-southern GXR,and display a weakly trend of getting younger from northwest to southeast,meanwhile indicating the metallogenic potential of rare metals within the central GXR.The spatial-temporal distribution,combined with regional geological data,indicates the highly evolved Mesozoic granites in the GXR were emplaced in an extensional environ-ment,of which the Late Jurassic-early Early Cretaceous extension was related to the closure of the Mongol-Okhotsk Ocean and roll-back of the Paleo-Pacific Plate,while the late Early Cretaceous extension was mainly related to the roll-back of the Paleo-Pacific Plate.

Spatial-Temporal Characteristics of Regional Extreme Low Temperature Events in China during 1960-2009

An objective identification technique is used to detect regional extreme low temperature events （RELTE） in China during 1960-2009. Their spatial-temporal characteristics are analyzed. The results indicate that the lowest temperatures of RELTE, together with the frequency distribution of the geometric latitude center, exhibit a double-peak feature. The RELTE frequently happen near the geometric area of 30°N and 42°N before the mid-1980s, but shifted afterwards to 30°N. During 1960-2009, the frequency~ intensity, and the maximum impacted area of RELTE show overall decreasing trends. Due to the contribution of RELTE, with long duratioh and large spatial range, which account for 10% of the total RELTE, there is a significant turning point in the late 1980s. A change to a much more steady state after the late 1990s is identified. In addition, the integrated indices of RELTE are classified and analyzed.

Adaptive spatial-temporal graph attention network for traffic speed prediction

Considering the nonlinear structure and spatial-temporal correlation of traffic network,and the influence of potential correlation between nodes of traffic network on the spatial features,this paper proposes a traffic speed prediction model based on the combination of graph attention network with self-adaptive adjacency matrix(SAdpGAT)and bidirectional gated recurrent unit(BiGRU).First-ly,the model introduces graph attention network(GAT)to extract the spatial features of real road network and potential road network respectively in spatial dimension.Secondly,the spatial features are input into BiGRU to extract the time series features.Finally,the prediction results of the real road network and the potential road network are connected to generate the final prediction results of the model.The experimental results show that the prediction accuracy of the proposed model is im-proved obviously on METR-LA and PEMS-BAY datasets,which proves the advantages of the pro-posed spatial-temporal model in traffic speed prediction.

Spatial-temporal Variation Characteristics of Sunshine Duration in China and Its Influence Factors

Sunshine duration has an important impact on agriculture.In order to study the temporal and spatial variation of sunshine duration in China under the background of climate change,based on the observation data of 2089 national meteorological stations during 1961-2017,trend analysis,mutation analysis,partial correlation analysis,and Mann-Kendall mutation analysis were used to analyze the temporal and spatial variation characteristics of sunshine duration in different regions of China,and the influence of main climatic factors and human activities.The results showed that:the sunshine duration in China showed a significant decreasing trend with a rate of-45.8 h/10 a,and the sunshine duration in 7 regions of Northwestern China,Northern China,Northeastern China,Center China,Eastern China,Southern China,and Southwestern China also showed a significant decrease.The spatial distribution of sunshine duration in China was characterized by"less in the south and more in the north",and the sunshine duration in the northern region was significantly higher than those in the southern region.The sunshine duration in Tibet,Qinghai,Gansu and west Inner Mongolia was higher,while it was obviously lower in Sichuan Basin.Through M-K mutation test analysis,it was found that sunshine duration in the whole country decreased significantly,but there was no mutation station,while mutation occurred at 1989 in Southwestern China,1983 in Northwestern China,1985 in Northeastern China.Seasonally,there was the highest sunshine duration in summer,followed by spring,autumn,and winter;the decline rate was also the highest in summer,followed by autumn,winter,and spring.Sunshine duration had a highly negative correlation with total cloud amount,visibility,haze,and precipitation,while had a strongly positive correlation with wind speed,and had no significant correlation with fog.