Week 7: Data models, databases Flashcards
(36 cards)
The object model is implemented with
Vector data where we use points, lines and polygons to represent geographic phenomena
Phenomena can be represented with either
Vector or raster data
Geographic information links a place (and often a time) with
Some properties of that place
Two conceptual spatial data models that describe how aspects of the real world are represented in a GIS
- Field models
- Object models
Field models can be associated with
Raster
Object models can be associated with
Vector
Field models have properties that
Vary continuously over geographic space
Characteristics of field models
- Every point on the earths surface can be recorded as a single value (single x y coordinate, 0 dimensional)
- Value at any point is a function of its location
- The property of the points can be of any attribute type (nominal, ordinal, interval, ratio, cyclic e.g wind direction)
Examples of field models
- Elevation
- Noise levels
- Temperature
Object models contain geographic space that is populated by
Discrete spatial features / objects
Characteristics of object models
- Each object has precisely defined spatial boundaries
- Each object is a generic element of the real world (e.g physical - lakes; human - roads, buildings)
- Each object is represented in the data base through generic feature types (points, lines, areas) and single or multiple attributes for all feature types
Object models contain features that
We can see clearly and count on the landscape
Difficult cases in determining between field and object models
- Natural phenomena such as lakes, habitats
- often conceived as objects, but difficult to define or count precisely (e.g vegetative pontoons) - Socio-cultural distance based concepts
- Spatial: near, far, about, close to, etc.
- Spatial and/or behavioral: neighbourhood, community etc. - Phenomena that cross the field-object classification
- weather forecasting
- Forecasts originate in models of fields, but are presented in terms of discrete objects (highs, lows, fronts)
Raster data models contain regular tessellation AKA
Division of space
Characteristics of raster - an implementation of field representation
- Square cells
- Register the grid corners in map coordinates
- Features are represented as collections of one or more cells
- Represent fields by assigning attribute values to cells
Characteristics of vector - an implementation of the object model of spatial data
- Discrete point, line, and area objects
- Location of features is represented explicitly
- Model does not store the space between objects / features
- Vector model is more compact than its counterpart - Many different formats (CAD, GIS / surveying vendors)
Data refer to
Physical and/or abstract facts about real world phenomena
Data is derived through
Cognitive processes: selection, generalization, synthesis
Data is the result of
Representing, organizing, labeling, encoding and relating real world phenomena
Data is dependent on
Observers perceptions
Forms that data may take
Numeric / descriptive; digital / analogue
What is special about spatial data
Include a locational dimension to its description of real-world entities, processes, events
In spatial data, location may be specified with
Absolute (e.g x,y), relative (e.g topological) or nominal referencing (e.g addresses)
Spatial data are / can be
- Multi-dimensional linking place, time and attributes
- Voluminous in nature (raster and imagery)
- Expensive and time consuming to update
- Often compiled from multiple sources
- Require special analsis methods
- Spatial data are scale dependent
- Spatial data are typically sampled data
- need to interpolate among sample points - Sample data are often interrelated across space (spatial autocorrelation) and time (temporal autocorrelation)
- Spatial data are subject to a host of uncertainties
- As a result we need special formats and databases to store and retrieve spatial data
