Chapter 7 - Geographic Data Modeling

Representation

Typically used in conceptual and scientific discussions

Model

Typically used in practical and database circles

Representation vs Model

General use focuses on how geographic reality is modeled (abstracted or simplified)

Specifically, the generation of a simulation used as an experiment basis for studying interactions in the real world and making projections of future conditions based on specific alterations of known variables.

Data Model

Set of constructs for describing and representing parts of the real world in a digital computers system

Levels of Data Model Abstraction

Generalization

Reality → Conceptual Model → Logical Model → Physical Model

Increasing abstraction →

Human oriented ——————> Computer-Oriented

Reality

Composed of all real-world phenomena (pr

Conceptual Model

Human-oriented, partially structured model of selected objects and process that are believed to be relevant to a problem domain.

Logical Model

Implementation-oriented representation of reality often expressed in the form of diagrams and lists

Physical Model

Portrays the actual implementation in a GIS and comprises tables stored as files or databases

Metadata

Data about data; includes geographic coordinates of upper-left corner, cell-size, number of rows and columns, and projection (as well as sensor or bands utilized)

Lossless Compression

Types of image compressions that results in no loss or degradation of information with successive compressions or decompressions

Run-length Encoding

Encoding adjacent row cells that have the same value with a pair of values indicating the number of cells with the same value and the actual value

Block (Quadtree) Encoding

Two-dimensional version of run-length encoding, in which areas of common cells (instead of rows) are represented with a single value

Lossy Compression

Types of image compressions that results in some information being irrevocably lost (discarded) during compression

Wavelet

Remove information by recursively examining patterns in datasets at different scales, always trying to reproduce a "faithful” representation of the original

Vector Data Model

Uses points, lines (polylines) and polygons to represent real-world objects (typically discrete object type data)

Features

Geographic entities encoded using vector data model

Feature Class

Features of the same geometric type (point, line, or polygon)

Feature Table

Stored in physical (database) representation (features as row and each property as the column)

Topology

The science and mathematics of relationships used to validate the geometry of vector entities, and for operations such as network tracing and test of polygon adjacency (Longley et al. 2011)

Importance of Employing Topological Relationships Within a Vector Data Model

• Data validation

• Modeling integrated feature behavior

• Editing productivity

• Optimizing query

Topological Features: Data Validation

Geographic data collection resulting in simple features limits its structural intelligence

Topological integrity can validate the geometric quality of the data and its suitability for geographic analysis

Topological Features: Modeling Integrated Feature Behavior

Integrated feature behavior refers to cases where multiple objects share common locations and partial identities

• Ex., Electric distribution corresponding and linked to subdivision areas or common conduits shared by multiple fiber and cable lines

GIS Database Approaches

• Single object with separate geometry representations

• Multiple objects with separate geometry integrated for editing, analysis, and representation

Topological Features: Editing Productivity

Simplifies the editing process and provides additional capabilities to manipulate feature geometries (requires topological data and tools)

Topological Features: Optimizing Queries

GIS queries can be optimized by precomputing and storing information about topological relationships

• Network Tracing (e.g., finding all connected water pipes and fittings)

• Polygon Adjacency (e.g., determining who owns the parcels adjoining those owned by a specific owner)

• Containment (e.g., finding out which manholes lie within the pavement area of a given street)

• Intersection (e.g., examines which census tracts intersect with a set of health areas)

TIN (triangulated irregular network)

Represents a surface with nonoverlapping triangular elements created from a set of points with x, y, and z coordinates

Topological data structure that stores information about the nodes and adjacent triangles

Advantages of TIN

Store actual point data for accuracy analysis

Efficient (more so than a DEM) as point density can vary

Data structure convenient for calculating slope, aspect, and line-of-sight between points

TIN Uses

Common use for volumetric calculations (road construction) and visualization of urban forms

• Can’t handle slope discontinuities well and must sufficiently sample elevation extents to create reliable TINs

Principles of Object Data Model

Integrates geometry, properties, and methods

Geometry is no longer defining but only one attribute of the object (still in an applied way of central concern)

Object

Self-contained package of information describing the characteristics and capabilities of an entity under study

(Real world is modeled as a collection of objects and the relationship between objects)

Class

A collection of objects of the same type

Encapsulation

Packing of description of state (properties/attributes) with the behavior (operations can be performed)

Inheritance

Ability to reuse some or all of the characteristics of one object in another

Polymorphism

Process where objects has its own specific implementation for operations and therefore does not need to be rewritten if additional object classes are added (they just have their own rule)