In the view of pan-spatial information system, we apply the complex network analysis method to analyze and visualize the relationship among the major countries in the complex network of the global pipeline natural gas from its attribute relations, temporal relations, spatial relations and comprehensive relations. The results indicate that the path dependence of the trade of pipeline natural gas is prominent and the global trade of pipeline natural gas is evolutionarily stable, indicating the "small world" characteristics of the regional market. Due to the differences in regional resource endowments, the trade of pipeline natural gas forms "core-edge" network structure. The trade of pipeline natural gas is mainly concentrated in Eurasia. In 2009, the demand center of global pipeline natural gas trade was located in European countries. In 2015, the dual demand core network relationship between Western Europe and China was formed. At the same time, we expounds the characteristics of the pan-spatial information system and prospects of the application of the platform based on the excavation analysis and visual expression of the complex network of global natural gas pipeline.

Currently route maps are often displayed on mobile devices for the route selection. However, for such devices, a major problem is their small display sizes. In this study, a new method on adaptive generation of variable scale maps for small display sizes was proposed and aimed to improve the clarity of the whole map. This method consists of two steps, i.e. estimating the density of global distribution on maps and conducting variable scale transformation based on the estimated density values. Compared with current variable scale methods, this method considered the space for representation of the landmarks and text symbols, and the reduction of overall distortions arising from map deformation. The experimental results indicated that the new method is able to adaptively generate the variable scale maps for different small display sizes and provide a map overview with an improved clarity as well as well-preserved map recognition compared with original shapes.

With the rapid development of spatial information technology, the concept of Pan-spatial Information System has been proposed. It extends the scope of spatial information system from the traditional mapping space to the space, interior space, microscopic space and other measurable space. Location data is one of the important research objects of Pan-spatial Information System and it has become a way of studying people's social life and urban dynamics. In this paper, we propose a new crowd classification method based on check-in data which is different from the traditional method based on socioeconomic attributes. Firstly, using the time series of check-in data, we build a matrix model. Then, we analyze the temporal characteristics of residents’ check-in activities. The analytical process starts from spatial-temporal profiles, learns the different behaviors, and returns annotated profiles. In the analytical process, we use the K-means clustering algorithm and K-NN algorithm to learn how to annotate profiles with a city user category (resident, dynamic resident, commuter, or visitor). Finally, according to the classification results of the population, we analyze the temporal and spatial behavior of different city user category and find their differences and potential regularity of spatial behavior. Our method can be applied to a new research perspective for characterizing the composition and characteristics of the urban population and studying urban spatiotemporal structure.

The pan-spatial information system is a world of data that abstracts the real world into spatio-temporal objects of multi-granularity. Various types of spatio-temporal entities and objects in the dynamic and complex real world can be expressed, analyzed, queried and so on. Spatio-temporal data model is the key of temporal GIS. Moreover, it has achieved better application in some special fields and expressed the temporal information and temporal object at certain degree. However, it still cannot completely describe the relationship and diversification of spatio-temporal object. We took the modeling of pan-spatio-temporal objects of multi-granularity as guiding ideology based on the analysis of the current status and existing problems of spatio-temporal data model. We took Beijing district as an example and analyzed the characteristics of the district and took it as a multi-granularity spatial-temporal object. The modeling method of spatio-temporal object of multi-granularity is explored, and the spatio-temporal object of multi-granularity is expressed in the political area and verified by visual expression method. In conclusion, the modeling of spatio-temporal object of multi-granularity can better reflect the spatial, temporal and attribute characteristics of the district . It also reflects its evolution in time and space, and facilitate the query, analysis and visualization of spatial-temporal objects.

The new generation of parallel spatial analysis in pan-spatial information system is challenged by analysis of spatial big data and real-time spatial service. As one of the most important part of GIS, vector spatial analysis has some performance bottlenecks such as load unbalance, less ability of parallel expansion and low I/O efficiency. First, we review the history of the developing process of vector spatial analysis from application requirement and technical progress. Then, we expound the research findings of spatial analysis of parallel vector, summarize the algorithm features and technical bottlenecks, compare the different parallel programming model and present the parallel spatial algorithm of R&D processing. Finally, we predict the spatial data model in the future and the computing method based on spatio-temporal objects of multi-granularity in pan-spatial information system. Also, we present the new techniques which use memory computing to realize the storage-computing integration in vector spatial analysis.

The dynamic variation rules of spatial position of a geographical entity could vary with the granularity of space and time. Scholars have tried to analyze the spatio-temporal position of geographical entities at multiple levels during the process of spatio-temporal reasoning, track data mining, and so on. Therefore, it has been a hotspot in GIS that how to effectively organize and express the spatio-temporal position of geographical entities under different spatio-temporal granularities. To solve this problem, two strategies have been applied for one geographical entity based on object-oriented thinking: "a three-level space" and "0-1 variation series of positions". Based on that two strategies, a method has been proposed to support the multi-granularity representation of spatio-temporal position of a geographical entity. Firstly, a three-level space including global space, relative space and object space has been constructed to guarantee the multi-granularity of the space, the transformation from different time period or moment to a series of discrete time points with different temporal granularity helping break the limit of temporal granularity. Then, for one geographical entity, its change process of spatio-temporal position could be divided into a series of stages according to the "0-1 variation series of positions". Based on this, different storage schemes for its spatial position information under each time point have been designed to reduce redundancy. Furthermore, the progressive recognition aroused from the three-level space, and the transitions between time points and periods could help to obtain the spatio-temporal positions of one geographical entity at more spatio-temporal granularities through its positions at existing spatiotemporal granularities. By the loose coupling between the space and time in describing the positions of geographical entities, the method could efficiently represent the spatio-temporal positions of geographical entities under variable granularity of space and time, which could help provide a reference for temporal GIS or multi-granularity spatio-temporal database.

Along with the increasing expansion of GIS in diverse application fields, spatial information systems face a variety of challenges, such as full-scale, full-category and full dynamics. Pan-spatial information system is an effective way to deal with the challenges mentioned above, in which spatial-temporal reference framework (STRF) is the core for the quantitative representation for human’s view of time and space. When handling objects of study with different scales and in different domains, the STRF in pan-spatial information system has various difficulties including descriptive scale, datum transformation, as well as integration of space and time. Thus, there is a huge demand for the development of a generic and unitary approach for the description and transformation of STRF. Through comparison and analysis on current STRFs in 1-, 2-, 3- or even higher dimensions, a new ubiq-uitous STRF is put forward in this study. In fact, the proposed ubiquitous representation of STRS in pan-space is applicable to the corresponding mathematical quantitative description and the related representa-tion of semantic vagueness. Thus, it provides reference for spatial-temporal representation in a broad sense. Meanwhile, based on fractal dimension characteristics extensively available in the real world and the proposed ubiquitous representation method, a unitary representation of STRF is investigated by making use of recursive computation. A full-scale and full-dimensional transformation of STRF is then realized in the pan-space.

Traditional behavior-driven spatio-temporal data models mainly focus on the spatial movement property of an spatio-temporal object. However, they ignore the changes of attributive characteristics and relational properties caused by the behavior. As a result, the behavior-driven mechanism—an important feature in the pan-spatial information—has not been adequately investigated. In this paper, the spatio-temporal behavior was comprehensively investigated from the following aspects: firstly, the definition of the behavior was provided and the necessity of taking the behavior as one of the tuples of spatiotemporal ontology was demonstrated; secondly, the behavior was partitioned as 4 categories according to the changed features, including spatial behavior, attributive behavior, relational behavior and composite behavior; thirdly, definitions of the four behavior types and their corresponding formalized descriptive methods were proposed; finally, the behavior-driven mechanism in the spatio-temporal ontology was studied. This research lays a theoretical foundation for the study of spatio-temporal object model under the behavior-driven mechanism.

As the development of Smart City and other GIS applications, we need to collect many types of attribute data of spatio-temporal objects derived from a variety of geographical entities or geographical phenomena. Generally, these attribute data have features of multi-scale, multi-dimension and dynamic, which pose challenges to effective expression and management of attribute characteristics of spatio-temporal objects. In view of the unclear structure, storage redundancy and semantic heterogeneity in the current expression method of attribute for spatio-temporal object, we present a set of expression and operation methods of attribute characteristics that take semantic and dynamic features into account. Based on the concept of classification theory, this method classifies different types of attribute information, introduce the time stamps that are independent of the spatial information of spatio-temporal object. It was coordinated with original sets and update sets to record and manage attribute information at different time nodes. Furthermore, to attribute with expression requirement of different semantic dimensions, we add semantic scale identifier to point out the semantic dimension and set knowledge reference to describe the transformation relation between different semantic scales. Finally, based on the design of dependent time stamp, the knowledge reference and the semantic scale identifier, we formulate the operation methods of transformation between different time scales of attribute information as well as semantic scales and give an example of both operation methods. The expression and operation methods of attribute information we proposed, can help to reduce the redundancy of the expression model for spatio-temporal object realize the cognitive mode of attribute with multi-time scale and multi-semantic scale, and provide a novel method to fine management of attributes of spatio-temporal object.

Spatial relationships play an important role in spatial query language, data retrieval and spatial analysis. However, the current research of spatial relations are hard to realize the unified expression and calculation of spatio-temporal objects of multi-granularity. In this paper, the spatial relationships computing operators of the simple objects are designed based on the type-independence and dimension-independence characteristics of GA operators. The operators are then generalized to spatio-temporal objects of multi-granularity by the union operator. Lastly, we realized the unified expression and calculation of three kinds of spatial relationships for the spatio-temporal objects of multi-granularity under the framework of pan-spatial GIS. The triangulation intersection algorithm is raised as an example to prove the reliability of our methods. Our research also provides the reference for expression and calculation of spatial relationships in pan-spatial GIS.

The modern expression and modeling of spatial objects is more related to the description of spatial and temporal data of multi-granularity than the correlation of spatio-temporal objects of multi-granularity. Multi-granularity expression of spatio-temporal object is a new method of expressing the temporal and spatial objects. The evolution of spatio-temporal objects is abstracted as a complex network. In this paper, based on the representation of spatio-temporal objects of multi granularity, the evolution process is formally defined. We present an initial model for constructing the process of relationship evolution with time-dependent network. In this paper, through the description and expression of the relation with the evolution process of returning farmland to forest based on time slices, we construct a dynamic and real-time network model and abstract the evolution process of object relationship of returning farmland to forests. We applied time-dependent network to clarify the evolutionary process of spatio-temporal object of multi-granularity relations, and initial expression and modeling of the evolution process. This method can make the object relation change more clearly, improve its hierarchy and efficiency, and lay the foundation for the study of the relationship of spatio-temporal objects of multi-granularity.

The relationship of geographic objects is the interaction between geographic objects. Any geographical objects are not independent, and have a certain relationship with other geographical objects. Besides, these associations are complex and diverse. The description of relationship of geographical objects in traditional GIS data model focuses on spatial relations. Other relationships often need complicated calculation and inference to obtain. So it's hard to completely express the complex relationships of geographical objects, and it does not completely meet human cognition of geographical phenomena. However, in pan-spatial system, according to the needs of the common geographic analysis, we can classify and express the relationship of spatiotemporal objects of multi-granularity, which is convenient for the pan-position abstraction and expression of the objective entity. The relationship of the objects can be divided into five categories: spatial relation, temporal relation, attribute relation, causal relation and cognitive relation. Finally, we use the geographic entity as a case to analyze the expression model of the relation.

The associative relation of spatial-temporal objects of multi-granularity describes the relations and mutual effects among objective entities in real world. It is the expression of entity relationship and mutual effects. According to its different describing subject, the associative relation can be divided into temporal relation, spatial relation, attribute relation and comprehensive relation. This paper begins with the background of pan-spatial information system, elaborates the basic conception of the associative relation of spatial-temporal objects of multi-granularity , and defines its formalized expression and classification. We also illustrate and analyze the construction and administration of associative relation, the applicative scene of static construction method and dynamic construction method. The associative relation maintains the consistency and linkage of the digital world through constraining effects and behavior conduction. Thus, this paper gives a brief illustration of constraining types and definition of associative relation and relation-behavior conduction mechanism.

With the accelerated and deepened exploration of the space, it’s difficult for existing spatial represen-tation forms to meet the demands of human cognition of the objective world. Thus, pan-spatial information system has been proposed. In order to explore its regulation of spatial cognition and make the presentation of pan-spatial information more in line with cognitive characteristics of human, we firstly study the development of geographic spatial cognition and presentation, and analyze the problems of representation forms in paper maps, electronic maps and GIS. Then, we discuss the cognitive characteristics of pan-spatial information system, and find that it extends traditional representation forms from the breadth and depth of spatial cognition and cognitive subject. It is the expansion and extension of the map and GIS in the era of big data. Finally, based on spatial cognition, we make a preliminary explanation for the representation of pan-spatial information system, pointing out that it not only needs to represent spatiotemporal objects of multi-granularity on display, but also needs analytical and exploring expression. We also discuss its cognitive connotation of the representation levels.

The pan-spatial information system (PSIS) is a spatial information system that describes dynamic and complicated world from micro to macro perspective. Its theory is based on data model of spatio-temporal objects of multi-granularity. In order to determine the specific content of spatio-temporal object model of multi-granularity, the basic framework for illustrating the characteristic of spatio-temporal object of multi-granularity is required. In order to model spatio-temporal object of multi-granularity practically, the modelling process for the data model of spatio-temporal object of multi-granularity is required. Firstly, we analyze the relation and difference between the pan-spatial information system and the traditional GIS from five aspects: data model, data management, visualization, spatial analysis and practical application. Then, we analyze the shortcomings of spatial data model of traditional GIS from six aspects: space category, dynamic change, complex relation, cognition and behavior, visualization technology and analysis of spatio-temporal big data. We also put forward seven characters of spatio-temporal object of multi-granularity: multi-granularity, multi-type, multi-modality, multi-reference system, multi-association, multi-dimensional dynamic and multi-autonomous. We define the descriptive framework of data model of spatio-temporal object of multi-granularity consisting of spatio-temporal reference, spatial location, spatial morphology, composition structure, association relation, cognitive ability, behavioral ability and attribute characteristics. Finally, we put forward the modeling process and the idea of data model of spatio-temporal object of multi-granularity on the basis of the analysis of the modeling process of spatial data model of traditional GIS.