Notes on Spatial Data Structures for GIS
Overview of Spatial Data and Structures
- Spatial Data: Represents features and phenomena related to geographic locations.
- GIS Function: Simplifies the complex real world using data models and structures.
Levels of Abstraction in GIS
- Real World: Physical entities and phenomena.
- Data Model: Simplified representation of objects from the real world.
- Data Structure: Implementation and organization of the data model.
- File Structure: How data is stored on physical media.
Spatial Data Models
- Definition: A spatial data model is a simplified representation of spatial features on the Earth's surface.
- Types of Models:
- Vector Model: Includes Points, Lines, and Polygons.
- Raster Model: Comprised of Grid cells and Pixels.
- Attributes: Additional information associated with spatial features.
Vector Data and Topology
- Vector Data can have: Topology (relationships between features) or No Topology.
- Topology Definitions:
- Connection between lines.
- Shared boundaries between areas.
- Inclusion of points in areas.
- Topology Function: Validates spatial relationships of neighboring and overlapping features.
Data Models for Vector Objects
- Georelational Model:
- Geometry and attributes stored separately.
- Uses Feature Identification number to link geometry to attributes.
- Object-Based Data Model (OODM):
- Geometry and attributes stored in a unified system.
- Features are treated as objects with properties and methods.
Georelational Data Model: Shapefile
- Shapefiles include multiple files with the same base name but different extensions (e.g.,
country.shp, country.shx, country.dbf). - Advantages of Shapefile:
- Faster display and non-proprietary.
- Disadvantages:
- Shapes can overlay, duplicate boundary storage, limited error checking, and potential data quality issues.
Georelational Data Model: Coverage
- Coverage: Contains both spatial and descriptive attribute data.
- Introduced in the early '80s with ArcInfo, made of multiple physical files in a folder.
- Topology: May include feature relationships.
Object-Oriented Data Model (OODM)
- Definition: Stores geometry and attributes as objects.
- Key Components:
- Feature Class: Storage for spatial features with the same geometry.
- Feature Dataset: Storage for features sharing extent and coordinate systems.
- OODM differs from Georelational Model in that both geometry and attributes are stored within the same system, often in BLOB format.
Geodatabase in ArcGIS
- Geodatabase: Centralized system for storing feature layers.
- Types of Geodatabases:
- Personal Geodatabase: Uses Microsoft Access, 2GB limit.
- File Geodatabase: Stores data in multiple small files, no size limit.
Topology in Geodatabase
- Defined as rules ensuring the integrity of spatial relationships within vector layers.
- Examples include no overlapping boundaries for countries or no intersecting contour lines.
Raster Data Model
- Definition: Represents continuous surface features using a grid of pixels.
- Storage Methods:
- Cell by Cell Coding: Simplest method, ideal for rapidly changing surfaces.
- Run Length Coding: Efficient for repetitive values, stores runs of similar values.
- Quadtree Coding: Recursively divides raster into quadrants based on pixel similarity.
Raster Data Compression
- Types of Compression:
- Lossless: Original data can be fully reconstructed.
- Lossy: Allows for compression but results in loss of data quality, generally used for background images.
Conversions and Integration
- Conversion Types:
- Rasterization: Converts vector data to raster format.
- Vectorization: Converts raster data to vector format.
- Hybrid Models: Utilize both formats simultaneously.
- Integration in GIS: Crucial for analysis, combining various data types.