The basic classes in the conceptual model


See also:
Conceptual Universe - Summary

Conceptual Universe Classes


Geo-Fields

“A geo-field represents a spatial distribution of a variable that has values in all points belonging to a geographical region.”

The geo-fields can be specialized as:

  • THEMATIC: Given a geographical region R, a thematic geo-field associates a theme to each point in the space (for instance, a pedology map is characterized by the set {Dusky-Red latosol, litosol, cambisol, ...}). The figure below shows an example of a thematic geo-field (soil map);

  • NUMERICAL: Given a geographical region, a numerical geo-field associates, to each point in space, a real value (e.g. a magnetic field map);

    mnt_amo1.gif - 11880 Bytes

  • IMAGE: Given a geographical region, it is obtained by the discretization of the signal received by a sensor (passive or active) for all the points in the region. The figure below shows a Landsat satellite image from Manaus (AM).

    Conceptual Universe Classes


Geo-Objects

“A geo-object is a unique element that has non-spatial attributes and is associated to multiple geographical locations. The location is precise and the object is distinguishable from its contour.”

Each geo-object in a GIS is associated to one or more regions (or locations) in space. As the Geoprocessing applications usually do not store or manipulate isolated elements in space, it is convenient to store a graphical representation of an geo-object together with its neighbors, keeping the topological relations. For example, in a cadastral urban map, the lots in the same neighborhood are stored and presented all together.

These basic characteristics lead to an introduction to the idea of geo-object maps, which groups geo-objects for a certain cartographic projection and geographical region, as instances of the GEO-OBJECT MAP class.

The relation between a geo-object and a geo-object map is described by IS_LOCATED_AT. This situation is typical in large geographical data bases, which include maps in different scales and projections, covering several UTM zones. The figure below shows the geo-object “Yang Tse Kiang River” represented in three distinct maps.

Objects and Object Maps.


Conceptual Universe Classes


Geo-Object Maps

Consider the GEO-OBJECT MAP specialized into two classes:

  • CADASTRAL MAP: they describe an elements location from an urban or rural cadastral information system, which uses the arc-node-polygon topology for storage.
  • NETWORK MAPS: these are maps that use the arc-node topology, they store the location and symbology associated to structures linearly connected. Additional information in this type of map includes flow directions and dynamic segmentations.

Geo-object map applications

The geo-object map concept has several practical and important applications. It allows the same geo-object to be represented in several maps, which can be in different scales or projections (see the figure below). This is very useful in network maps. In this case, it allows that more than one object share the same graphical representation.

As an example consider the electrical network case, which is usually represented by the arc-node topology. In this case, a transmission line can be geometrically represented by a set of arcs and poles, substations and large consumers can have their geometries associated to the network nodes. It happens that in the same pole it is possible to have other elements (like keys and transformers), as illustrated in the figure below.


Different geo-objects associated to the same geometry.


Conceptual Universe Classes



Information Layer

As the geo-fields and geo-object maps concepts are both related to the location in a geographical region in space, it is useful to define the idea of information layer.

An information layer is the basis for the geographical representation of different types of geographic data. This is related to the generalization of the geo-object maps and geo-fields concepts. A class instance INFORMATION LAYER represents, for a given geographical region, a geometric place of a set of geographic data (a geo-field or a geo-object map).


Conceptual Universe Classes



Non-spatial objects

In several situations it is convenient to allow the association of non-spatial information to a georeferenced data base. For instance, consider an application of urban register in a city hall, which already has a system for tax computation based on an alpha-numeric properties registers. In this case, it is possible to associate the alpha-numerical list to a geo-referenced map which has the geographical location of these properties.

The notion of non-spatial object covers any type of information which is not geo-referenced and which we want to aggregate to a GIS.

The example given shows the case of an application of rural registration. In this case, we have the geo-object class “farm” (which are located in a map) and we want to establish a connection among these geo-objects and the existing alpha-numeric information (the registers). In this case we only have to consider the registers as non-spatial objects.

 

A connection between a geo-object and non-spatial object example.


Conceptual Universe Classes




Geographical Data Bases

“A geographical data base is composed by a set of information layers, a set of geo-objects and a set of non-spatial objects.”

This definition is particularly interesting because there is no restriction about data scale and there is no spatial continuity of the information layers composing the data base. As an example take a geographical data base about the Legal Amazon Forest, with the following data:

  • Vegetation (phytophysiognomy), specialization of the Thematic class, with 780 vegetation themes defined by the IBGE (Brazilian Institute of Geography and Statistics - RADAM project): 26 digitalized charts from Lambert project maps with 1:1.000.000 scale.
  • Satellite Images, specialization of the DATA_SENSOR_REMOTE class, obtained by the LANDSAT TM satellite, in three years (1986, 1990 and 1994): 228 images, UTM projection.
  • Geo-fields of soil usage, Thematic class specializations, with the classes: primary forest, secondary forest, floodplain forest, savannah, regeneration areas, rivers and clouds. The final maps will be compatible with the 1:250.000 scale, UTM projection.
  • Economic-ecological Zoning in the Roraima state, specialization of the Cadastral class, with 15 information layers generated from studies using 1:100.000 scale.

Conceptual Universe Classes



Conceptual Universe Summary

As a summary, the conceptual universe of the model has as basic entities geo-objects, object maps and geo-fields (these last two aggregated by the idea of an information layer) and non-spatial objects.

The geo-fields can be specialized in the thematic fields, numerical models of a terrain and remote sensing images. The geo-object maps can be specialized in cadastral maps and network maps.

In this perspective, a geographical data base is a collection of information layers, geo-objects, and non-spatial objects. The following figure summarizes the conceptual universe model.

The SPRING data model of the conceptual Universe.


Conceptual Universe Classes

Conceptual Universe