Unit 3 : Spatial Data Structure

and Database Design

Model
A model is a simplified representation of a phenomenon or a system. A map is
a model.

A model help us better understand a phenomena or a system by retaining the

significant features and relationships of reality

Models are the simplification of reality. A map is a symbolic model, because

itis a simplified representation of part of the real world. The components of the
model are spatial objects, they are represented on the map by graphical
symbols.

Raster Data Model

Definition:
Represents spatial data as a grid of cells (pixels), where each cell has a value
representing information such as temperature, elevation, or land cover.

Characteristics:
Grid-based (rows and columns).

Each pixel holds a single value.

Suitable for continuous data.

Examples:
Satellite images

Digital Elevation Models (DEM)

Land use/land cover maps

Weather maps

Unit 3 : Spatial Data Structure and Database Design 1

 Advantages:
Efficient for representing continuous data.

Simple data structure.

Ideal for image processing and modeling.

Disadvantages:
Large file sizes.

Lower spatial accuracy compared to vector.

Difficult to represent discrete features (e.g., roads, buildings).

Vector Data Model

Definition:
Represents spatial data using points, lines, and polygons to depict real-world
features.

Characteristics:
Precise geometric shapes.

Attribute data stored in tables.

Suitable for discrete data.

Examples:
Road networks (lines)

City locations (points)

Land parcels (polygons)

Advantages:
High spatial precision.

More compact than raster for certain datasets.

Suitable for topology-based analysis (e.g., network analysis).

Unit 3 : Spatial Data Structure and Database Design 2

 Disadvantages:
Complex structure.

Not ideal for continuous data.

Spatial analysis may be computationally intensive.

Summary Table
Feature Raster Model Vector Model

Data Structure Grid (cells) Geometric shapes (points/lines/polygons)

Best For Continuous data Discrete features

File Size Large Smaller (usually)

Accuracy Depends on resolution High spatial accuracy

Examples Satellite image, elevation Roads, boundaries, buildings

Geographic phenomena
Geographic phenomena are observable events or processes that occur on the
Earth's surface. These phenomena can be natural or human-
made and can include phenomena such as land cover, land use, population
distribution, transportation networks, and environmental
changes.
Characteristics of Geographic phenomena may include location,

Unit 3 : Spatial Data Structure and Database Design 3

 attributes, relationships with other phenomena, and how these
characteristics change over time.
Discrete phenomena have well-defined boundaries, such as
people, cars, houses, hospitals, and roads.
Continuous phenomena have ill-defined or irrelevant boundaries,
such as temperature, air quality, and elevation.
Data modeling of geographic phenomena:
Data modeling in GIS involves representing real-world geographic features and
phenomena within a computer-based system.
Data modeling include:
•
Entity-Relationship Modeling: Identifying the entities (geographic features) and
their relationships in the real world and representing them in the GIS database.
•
Attribute Data: Capturing attributes or characteristics associated with
geographic features. For example, attributes of a city might include
population, area, and elevation.
•
Topology: Representing the spatial relationships between geographic
features, such as adjacency, connectivity, and containment.
•
Metadata: Describing the characteristics and properties of geographic data,
including its source, accuracy, and projection information.

Spatial Database Design

A spatial database is a database optimized to store and query spatial data such
as maps, geographic features, coordinates, and topology. Spatial database design
refers to the structured process of modeling, organizing, and implementing
spatial data efficiently in a database system.

Steps in Spatial Database Design

1. Requirement Analysis

Understand the type of spatial data needed.

Unit 3 : Spatial Data Structure and Database Design 4

 Identify user requirements: what kind of queries and analysis will be
performed?

Define the spatial and non-spatial attributes.

2. Conceptual Design (Using UML or ER Diagrams)

Design the overall structure using an Entity-Relationship (ER) model.

Identify:

Entities (e.g., City, River, Road)

Relationships (e.g., River crosses City)

Attributes (e.g., Population, Length, Coordinates)

3. Spatial Data Types Selection

Choose appropriate data types:

Point – e.g., city location

Line– e.g., roads, rivers

Polygon – e.g., country boundaries, lakes

Raster – e.g., satellite images, elevation

4. Logical Design

Convert the conceptual model into tables and relationships.

Define spatial indexes for performance (like R-Tree).

Normalize the database (apply 1NF, 2NF, 3NF where applicable).

5. Physical Design

Implement the database in a Spatial DBMS like:

PostgreSQL

Oracle Spatial

MySQL Spatial

SQLite

Optimize storage, indexing, and performance.

Unit 3 : Spatial Data Structure and Database Design 5

 6. Data Loading and Conversion

Import spatial data from GIS formats (e.g., shapefiles, GeoJSON, KML).

Clean and validate the data before storage.

7. Query Design

Create SQL queries for spatial analysis:

Distance, area, intersection, containment, etc.

Use spatial functions like:

ST_Within(), ST_Intersects(), ST_Distance(), ST_Area()

8. Security & Maintenance

Set up access control for spatial data.

Backup regularly and monitor performance.

🔹Management)
Example: Mini Spatial Database Design (Land

Entities:
LandParcel (Polygon): id, area, land_type, geometry

Owner (Point): owner_id, name, address, location

Road (Line): road_id, name, length, geometry

Relationships:
One owner can have many land parcels.

Roads can pass through multiple parcels.

What is a Geo-Database?

Unit 3 : Spatial Data Structure and Database Design 6

 A Geo-database (Geographic Database) is a database designed to store,
manage, analyze, and visualize spatial data (data that has a geographic or
location-based component).

It integrates spatial data (like maps, coordinates, shapes) with attribute data (like
names, IDs, population, etc.) in a structured format, enabling powerful GIS
(Geographic Information System) applications.

Types of Geo-Databases
1. File Geo-database
Stored as files in a folder.

Developed by ESRI (used in ArcGIS).

Can store large datasets (up to 1 TB per dataset).

Good performance and multi-user read access.

2. Personal Geo-database
Stored in a Microsoft Access (.mdb) file.

Size limit ~2 GB.

Suitable for smaller projects and single users.

3. Enterprise Geo-database
Stored in high-end relational DBMS (like PostgreSQL, Oracle, SQL Server).

Supports multi-user editing, versioning, and large-scale operations.

Used in large organizations and government systems.

Advantages of Geo-Databases
Efficient spatial data storage

Centralized data management

Supports complex queries and analysis
Integration with GIS software

Unit 3 : Spatial Data Structure and Database Design 7

 Better data validation and integrity

Versioning and multi-user editing (Enterprise)

Common Uses of Geo-Databases

Urban planning (roads, zoning, buildings)

Environmental monitoring (forests, rivers, pollution)

Disaster management (earthquake zones, flood-prone areas)

Transportation (route mapping, logistics)

Agriculture (crop zones, irrigation maps)

Steps for Designing a Geodatabase

1. Define the Purpose and Scope
Identify the goals of the geodatabase.

Understand who will use it, what data it will store, and how it will be used.

2. Identify and List the Required Datasets

Determine what feature classes, raster datasets, and tables are needed.

Example: Roads, buildings, rivers, land use, elevation.

3. Determine the Spatial Data Model

Choose between vector (points, lines, polygons) and raster (grids) formats.

Decide if data is best represented as objects or fields.

4. Design Feature Classes and Tables

Define the feature classes (e.g., Roads, Parcels, Lakes).

Design related attribute tables with appropriate fields (e.g., Name, Type,
Area).

Unit 3 : Spatial Data Structure and Database Design 8

 5. Define Relationships and Topology Rules
Establish relationships between feature classes and tables (e.g., one-to-
many).

Apply topology rules to maintain spatial integrity (e.g., no overlaps, must

connect).

6. Set Coordinate System and Projection

Choose an appropriate coordinate system (e.g., UTM, WGS84).

Ensures all spatial data aligns correctly.

7. Apply Data Integrity Rules

Use domains to restrict field values (e.g., land type = [Residential,
Commercial]).

Use subtypes to categorize features within a feature class.

8. Organize into Feature Datasets (Optional)

Group related feature classes into feature datasets for better organization.

9. Load and Validate the Data

Import or digitize spatial and attribute data into the geodatabase.

Validate against topology and domain rules.

10. Document the Design

Create metadata for datasets.

Document field definitions, domains, relationships, and design decisions.

Spatial Relationship and Topology in GIS

Both spatial relationships and topology are fundamental concepts in Geographic
Information Systems (GIS) used to understand how geographic features interact
with each other in space.

Unit 3 : Spatial Data Structure and Database Design 9

 Spatial Relationships
Spatial relationships describe how geographic objects (points, lines, polygons)
relate to each other in location, distance, and direction.

Topology
Topology refers to the rules and relationships that define how geographic
features share geometry. It focuses on connectivity, adjacency, and
containment, rather than coordinates.

Importance of Topology:
Ensures data integrity in spatial databases

Helps in network analysis (e.g., finding shortest paths)

Prevents errors (e.g., overlapping polygons, dangling lines)

Supports editing rules (e.g., if you move a shared boundary, both features
update)

Aspect Spatial Relationships Topology

Based on Geometry/location Rules/connections

How features are located relative How features are connected and
Focus
to others behave

Find all rivers within 5 km of a

Example Query Ensure roads connect without gaps
school

Topology rules in GIS editors (e.g.,

Tools Used Spatial SQL, GIS queries
ArcGIS, QGIS)

Geographic Object and Field – Short Notes

Geographic Object (Object-Based Model)

Represents discrete geographic features with clear boundaries.

Examples: Buildings, roads, lakes, trees.

Stored using vector data (point, line, polygon).

Unit 3 : Spatial Data Structure and Database Design 10

 Each object has its own geometry and attributes.

Ideal for man-made or distinguishable natural features.

Geographic Field (Field-Based Model)

Represents continuous geographic phenomena.

Examples: Elevation, temperature, rainfall, soil type.

Stored using raster data (grids or pixels).

Each location (cell) has a measured value.

Ideal for natural phenomena that vary smoothly over space.

Unit 3 : Spatial Data Structure and Database Design 11