【文章內(nèi)容簡介】 tics. In order to address this issue the GIS munity has introduced a number of ways for modeling real world data. In a vector model, reality is perceived as an aggregation of discrete entities defined by their geometry, topology and thematic attributes. Another popular way to describe real world phenomena is by using raster structures that can sufficiently describe continuous/field data. The raster model in its basic form is considered as a rectangular array of equally spaced pixels. Each pixel can be defined uniquely inside the array through i,j coordinates. The values stored in the pixels of the array depend on the phenomenon and can be from elevation or temperature to the reflectance of light for part of thespectrum. . The XML approach XML enables domain experts to create properly structured formats, which can serve the storage and exchange of a wide variety of data types. Apart from that XML itself can store efficiently various types of data and it can easily describe continuous data using a tablebased mapping. The tablebased mapping is used to model XML documents as a single table or set of tables (Fig. 1). tableholder table1 row column1.../column1 column2.../column2 ... /row row ... /row ... /table1 table2 ... /table2 ... /tableholder Fig. 1. Tablebased mapping of field data. This general form can be modified accordingly in order to acmodate the description of different phenomena. For instance the description of a threeband raster image could be achieved in the way shown in Fig. 2. A problem easily solved through XML is whether the auxiliary data should be stored as child elements or as attributes, as well as the names to be used for each element or attribute. In addition, developers who use tablebased mappings often include table and column metadata either at the beginning of the document or as attributes of each table or column element. For image metadata one could store pixel size, date and conditions of capture, details of the camera used, etc. Using this method efficient description, storage and exchange of fieldtype geographic data can be achieved utilizing open standards and promoting interoperability (Bourret, 2021). image band1 pixel id=0 row=0 column=0 r201/r g171/g b81/b /pixel pixel id=1 row=0 column=1 r203/r g175/g b78/b /pixel pixel ............ /pixel ............ /band1 band2 ............ /band2 band3 ............ /band3 /image Fig. 2. Description of a threeband raster image. Apart from continuous data, XML can easily encode data modeled as vectors. An XML schema is developed describing the definitions of geometric primitives. In fact that is what OGC provides through GML, which is an XMLbased specification designed to describe vector data. . The GML approach GML has been a turning point in Geomatics to the extent that many national and private anizations have already adopted this format. GML specification introduced a number of new ponents improving vector data encoding (Cox et al., 2021). GML provides only three core schemas (, geometry. xsd and ) whereas in GML there are 28 core schemas. This version supports new geometry types including Arc, Circle, CubicSpline, Ring, OrientableCurve, OrientableSuface, Solid, and the aggregates CompositeCurve, CompositeSurface and CompositeSolid. GML addresses many limitations of the previous version like: topology, temporal ponents, dynamic features, coverages and coordinate reference systems. These ponents enable the encoding of sophisticated vector models, which are more accurate representations of the real world. Furthermore, GML provides XLink and XPointer mechanisms to make geospatial data interoperable. Through XLink and Xpointer, different features and feature collections, which may be located remotely, can be associated at the feature level (Peng and Zhang, 2021). Apart from the improvements in vector encoding, GML introduced a new way for the description of continuous data with the use of ‘‘coverages’’ (Fig. 3). The specification defines the GML encoding for coverages and is pliant with the conceptual model of ISO 19123: Coverages support mapping from a spatiotemporal domain to attribute values where attribute types are mon to all geographic positions within the spatiotemporal domain. A spatiotemporal domain consists of a collection of direct positions in a coordinate space. Examples of coverages include rasters, triangulated irregular works, point coverages, and polygon coverages (Cox et al., 2021). There are basically two methods to describe information in a coverage. The first is by creating a set of discrete location–value pairs. The domain set of the coverage is formed from discrete and homogeneous feature collections and the range set is formed by a set of values. The properties of each feature are assigned to the location of the feature. The second method describes the spatiotemporal domain set and the set of values from the range set. It also describes the rule that assigns a value from the range set to each position within the domain (Cox et al., 2021). Two schemas, and support coverages in GML and the values that can be assigned in the domain set are Grid and RectifiedGrid as shown in Figs. 4 and 5. The gml:Grid element defines an nonrectified grid, which is a work posed of two or more sets of equally spaced parallel lines intersecting each other. The gml:RectifiedGrid defines another kind of grid. The points of a rectified grid refer to geographic locations. The element describes the position of the origin of the grid along with the position of the offset vectors that determine the spacing of the points of the grid. In the above example the RectifiedGrid st