Geodatabase A container & a

 A Geodatabase is a database designed specifically comprehensive data model

to store, manage, and organize various types of
geographic information (spatial data) along with A central location for all
its related attributes. organizational GIS data
 Unlike a normal database (which stores only text,
numbers, or tables), a geodatabase can also Domains &
Feature Raster Tables Topology Relationships subtypes
handle: Datasets Networks Attribute
Classes Rules
o Spatial data (points, lines, polygons, rasters). Rules
o Attributes (information about each spatial No Automated connection Domains
Continuous geometry checks
feature). data Control
spatial Logical
o Relationships (connections btn datasets). integrity Constraint allowable
(Vector Spatial values for
constraints rules
Components of a Geodatabase Data) (Building attributes
1. Feature Datasets: groups of related feature Arial Images Parcel height > 0)One-to-one
classes with the same coordinate system.
DEM owner one-to-many Land Use =
datasets to
2. Feature Classes: collections of features of the
Land Cover database many-to-many {Residential,
model Commercial,
same type (e.g., roads, rivers, schools). Points Calculation
connected Agricultural}.
Lines rules (area field
3. Raster Datasets: images, DEMs, satellite Polygons Geometrically systems
auto-calculates
[Link] → attribute information (e.g., valid
Subtypes
features from geometry)
population, land ownership).
Utility networks
4. Relationships: define connections (e.g., (water, electricity, Classify features
schools linked to districts). sewer). into categories
5. Topology Rules: maintain spatial integrity Validation
Transportation rules (identify
(e.g., roads must connect, parcels must not networks Roads = {Highway,
overlap). incorrect
(roads, Local Road,
railways) entries)
 Simple Relationships
Types of Relationships in a Geodatabase

[Link] Relationships
 Link between a feature class and a table or between
two tables.
 Related objects can exist independently of each
other
 Changes in one dataset do not cascade to the other.
 Example: A table of landowners linked to a parcel
feature class by owner ID. Deleting a landowner does
not delete the parcel.
[Link] Relationships
 Parent–child relationship, where the child is
dependent on the parent.
 Changes cascade: deleting the parent deletes the
child.
 Example: A building feature (parent) and its rooms
(child). If the building is deleted, the rooms are
deleted too.
[Link] (how many records relate to each other)
 One-to-one (1:1): One parcel → One title deed.
 One-to-many (1:M): One landowner → Many parcels.
 Many-to-many (M:N): Many landowners ↔ Many
parcels (joint ownership)
4. Spatial Relationships
 Defined by location/topology
 A spatial relationship describes how two or more geographic features are related
to each other in space. It answers questions such as: Where is it? How close is it?
Does it touch? Does it overlap?
 Example: "A road crosses a river" or "A house must be contained within a parcel."
Containment
•One feature completely contains another.
•Example: A parcel contains a building.
Adjacency (Touches/Neighbors)
•Features share a common boundary but do not overlap.
•Example: Two parcels sharing a border.
Intersection (Overlap)
•Features partly cover the same area.
•Example: A flood zone overlapping with land parcels.
Proximity (Distance-based)
•Measures how close one feature is to another.
•Example: Schools located within 500 m of a road.
Connectivity
•Features connected through a network.
•Example: Road segments connected at intersections.
Directionality
 Relative orientation of one feature to another
•One feature is north, south, east, west of another.
•Example: A river flowing east of a village.
Rules in a Geodatabase Sample Topology rules
Rules ensure data integrity by enforcing constraints on how 1. Features must not overlap
features are stored, related, or interact. 2. Features must not have gaps (slivers)
a) Attribute Rules 3. Polygons must be closed.
• Constraints on attribute values. 4. Undershoots or overshoots of the border lines are not
• Eg. The "Building Height" field must be > 0. allowed.
b) Topology Rules 5. Contour lines in a vector line layer must not intersect
• Control spatial relationships among features. (cross each other)
6. ………………………etc.
o Parcels must not overlap.
o Roads must connect at endpoints.
o Rivers must not intersect themselves.
c) Domain Rules
• Restrict allowable attribute values to a valid set.
• Eg. A “Land Use” field may only contain: {Residential,
Commercial, Agricultural}.
d) Subtype Rules
• Define feature subcategories and apply rules per
subtype.
• E.g.. A road network may have subtypes {Highway,
Local Road, Footpath}, each with different speed limits.
e) Relationship Class Rules
• Rules applied within relationships. e.g..
• Every parcel must have at least one owner.
• A well must belong to only one water supply zone.
 Rationale (Why Use a Geodatabase?)
1. Centralized Data Management
o Stores all GIS layers, attributes, and imagery in one place, avoiding
scattered files.
o Makes data sharing and updating easier.
2. Data Integrity & Consistency
o Enforces rules and topology (e.g., parcels should not overlap, roads must
connect).
o Prevents data errors that can mislead analysis.
3. Scalability
o Can handle large volumes of data (from small local projects to national
spatial data infrastructures).
o Supports file-based, desktop, or enterprise-level (multi-user)
environments
4. Advanced Analysis
o Enables spatial queries (e.g., “Which schools are within 1 km of a
road?”).
o Supports network analysis, 3D modelling, and surface modelling.
5. Multi-User Collaboration
o In enterprise geodatabases, multiple users can edit and access data
simultaneously. Supports diverse
o Useful for government, organizations, and collaborative projects.
data types used in
6. Integration with Other Systems
o Can link with statistical databases, remote sensing imagery, and CAD
GIS
data.
o Makes it easier to integrate GIS into broader decision-making workflows.
Geodatabase & the Geo-information Cycle
The cycle has five main stages:
1. Data Acquisition:
o Collecting raw spatial data (GPS surveys, remote
sensing, field mapping, existing maps,
sensors).Example: satellite images of a city or land
parcel survey.
2. Data Pre-processing
o Cleaning, converting, projecting, and integrating
data into a usable form.
o Example: correcting satellite images, converting
coordinates into a common system.
3. Data Storage & Management
o Organizing data into a structured system.
o This is where the Geodatabase comes in. It provides
a standardized way to store spatial + attribute data
in a consistent, queryable environment
4. Data Analysis & Modelling
o Running queries, spatial analysis, overlays,
simulations
o Example: analyzing flood-risk zones, land-use
change, or transport networks.
5. Information Dissemination & Use The geodatabase sits at the center of
o Producing maps, reports, dashboards, or decision- the cycle, enabling smooth transition
support tools. Information is delivered to planners, from raw data to actionable
policymakers, or the public. geographic knowledge.
 Stages of Geodatabase Design
Geodatabase Design & Development
Requirement Analysis
Geodatabase design is the process of planning, structuring, and • Identify users, stakeholders, and applications.
implementing a geodatabase to support spatial data storage, • Define data needs: feature types, attributes, relationships, scale,
management, analysis, and dissemination. A well-designed precision.
geodatabase ensures efficiency, scalability, accuracy, and usability. • Determine expected operations: queries, analysis, updates,
reporting.
Conceptual Design
• High-level description of what the database should contain.
• Use Entity-Relationship (ER) diagrams or UML class diagrams.
Spatial Data • Define entities (e.g., parcels, roads, rivers, buildings) and thei
Non-Spatial Data relationships.
(Shspefiles
(Access, Excel…files Logical Design
• Translate conceptual design into a database schema.
• Decide on data models:
o Vector (point, line, polygon) for discrete features.
Geodatabase o Raster (grids, images) for continuous data.
• Define tables, attributes, keys, domains, & topology rules.
Physical Design
Space Required Data Retrival , • Implement schema in a specific DBMS (e.g.,
& Attributes Edit & Update PostgreSQL/PostGIS, Oracle Spatial, ESRI Geodatabase).
• Optimize storage: indexing (R-trees, quadtrees), partitioning.
• Establish rules, constraints, and validation checks.
• Configure user roles, permissions, and security policies
Data visualization Implementation and Testing
& Map editing • The geodatabase moves from plan to practice
 Geodatabase Design & Development…Cont
b) Data Requirements
Stages of Geodatabase Design  Thematic data: parcels, roads, utilities, administrative
boundaries, DEM, imagery.
1. Requirement Analysis  Data types: vector, raster, tabular.
 Scale and resolution: level of detail (1:1000 cadastral vs.
Requirement Analysis of a Geodatabase is the first step 1:50,000 national maps).
in designing and implementing a geodatabase. It ensures  Data sources: existing shapefiles, CAD drawings, surveys,
the database will meet user needs, support workflows, satellite imagery.
 Data volume: estimate storage capacity.
and integrate with existing systems.
c) Attribute and Metadata Requirements
a) Stakeholder and User Needs Assessment  Define what attributes must be stored for each dataset
 Identify end-users (planners, surveyors, engineers, public). (e.g., Parcel ID, Owner, Area, Land Use).
 Determine use cases (land administration, urban planning,  Identify domains and subtypes (to control field values).
disaster monitoring, utilities management).  Define metadata standards (ISO 19115, FGDC).
•Collect functional requirements: d) Relationship and Rule Requirements
• What queries and analyses need to be supported? •Identify relationships among datasets:
• What kind of maps and reports are required? • Parcel ↔ Owner (relationship class).
• What rules and constraints must be enforced? • Road ↔ Maintenance records.
•Collect non-functional requirements: • Building ↔ Utility connections.
• Performance (speed, scalability). •Define rules:
• Security and access levels. • Topology
• Interoperability with other systems (CAD, NSDI) • Relationship (a parcel must have at least one owner).
• Attribute (land use must match zoning domain).
e) System and Technical Requirements Key Steps in Conceptual Design
 Software: ArcGIS Desktop/Pro, ArcGIS Enterprise, or open-
source (QGIS + PostGIS). a) Identify Entities (Feature Classes & Tables)
 Hardware: servers, storage, backup systems. •Real-world objects to be stored.
 Database type: File geodatabase, Enterprise geodatabase (SQL • Parcels, Buildings, Roads, Rivers, Owners,
Server, Oracle, PostgreSQL). Utilities.
 Network requirements: multi-user access, web services, b) Define Attributes
synchronization. •Describe the properties of each entity.
f) Data Management and Maintenance • Parcel: ParcelID, Area, LandUse, OwnerID.
• Road: RoadID, Name, Type, Surface.
 Data update workflows (surveys, crowdsourcing,
satellite updates). c) Define Relationships
 Versioning and edit tracking (especially in multi-user •Show how entities are connected.
enterprise environments). Multiple users to edit the • Parcel ↔ Owner (1:M relationship class).
• Building ↔ Parcel (M:1).
same geodatabase at the same time without
• Roads ↔ Maintenance Records (1:M).
overwriting each other’s work.
 Backup and recovery strategy. d) Define Rules and Constraints
• Topology rules:
 Access control and security policies (who can edit vs. • Relationship rules: a parcel must have an owner.
who can view). • Domains: LandUse must be from a coded list
2 . Conceptual Design (Residential, Commercial, etc.).
Purpose e) Define Domains & Subtypes
 This comes right after requirement analysis and •Sub-categories within a feature class.
before logical/physical design. •Example:
• Roads → {Highway, Local, Footpath}.
 The goal is to create a high-level model of the
• Land Parcels → {Residential, Agricultural,
geodatabase that captures what information will be Industrial}.
stored and how it relates
Conceptual Data Model Diagram 3 . Logical Design
 Entities (feature classes/tables) as boxes. The Logical Design of a Geodatabase translates the abstract
 Attributes inside the boxes. conceptual model (entities, attributes, rules) into a formal
 Relationships as connecting lines with cardinality (1:1, 1:M, database schema that can be implemented in ArcGIS
M:N).
(ArcMap/ArcGIS Pro) or another GIS database system. The
 Rules and subtypes noted.
Schema Design includes the following:
Parcel
 Identify entities (parcels, roads, rivers).
ParcelID
1:1 Title Deed  Define attributes (owner name, width, flow rate)
Area
Name  Establish relationships (road connects to city, river flows into lake).
Landuse
Area  Decide the structure, data types, and rules within the GIS software
OwnerID
environment.
1:M A parcel can  Normalize Attributes: reduce attribute redundancy e.g in parcel
M:1 Every parcel have many buildings table, there should be no owner name. Its already in owner table
must have an owner
Key Components
Bulding a) Datasets and Feature Classes
Owner BuildingID  Group spatial layers into feature datasets (e.g., “Cadastral)
OwnerID ParcelID
MM Road Must  Define feature classes (points, lines, polygons).
Name Type
Connect to parcels • Parcels (Polygon FC)
Contact Height • Buildings (Polygon FC)
• Roads (Line FC)
Road • Owners (Non-spatial Table)
RoadID b) Attributes → Field Definitions
Name  Assign field names, data types, lengths, and nullability.
Type • ParcelID → Text (10), Not Null, Unique.
• Area → Double.
• LandUse → Text (20), with domain.
c) Domains and Subtypes Logical schema diagram
 Domains: Parcel (Polygon FC)
• LandUse_Domain = {Residential, Commercial, ParcelID” Text(10) (PK)
Agricultural, Industrial}. Area: Double 1:1 Title Deed (Table)
Landuse: Text(10) (Domain: LU_Domain
• RoadType_Domain = {Primary, Secondary, Tertiary}. OwnerID: Text(10) (FK) Name: Text(50)
 Subtypes: Area: Double
• Parcel → {Public Land, Private Land}.
• Road → {Highway, Local}. 1:M Buildings must
M:1 Every parcel
d) Topology Rules be inside parcels
must have an owner
 Defined at the feature dataset level.
• Parcels must not overlap. Bulding (Polygon FC)
• Buildings must be inside parcels. Owner (Table) BuildingIDText(10) (PK)
• Roads must not overlap parcels except at OwnerID: Text(10) (PK) ParcelID: Text(10) (FK)
Name: Text(50) MM Road Must Type: Text(20) (Domain: Bldtype_Domain
easements. Height: Double
Contact: Text(50)
e) Relationship Classes Connect to parcels
 Implement relationships defined in conceptual design.
• Parcel ↔ Owner (1:M, simple relationship).
• Parcel ↔ Building (1:M, composite relationship). Road(Line FC)
RoadID: Text(10) (PK)
• Parcel ↔ Road (M:N with intersection table).
Name: Text(50)
f) Identifiers & Keys Type: Text(20) (Domain: Roadtype_Domain)
•Primary Keys:
• OwnerID (Owner table).
• ParcelID (Parcel FC).
•Foreign Keys:
• OwnerID in Parcel FC (links to Owner).
• ParcelID in Building FC (links to Parcel).
 4. Physical Design b) Physical Storage Structure
•Feature Datasets → hold related feature classes that share a
Logical schema is translated into an actual implementation in coordinate system and topology.
ArcGIS (or another GIS DBMS), with specific storage structures, • Example: Cadastral_Dataset (Parcels, Buildings,
indexing, performance, and user access considerations. Roads).
•Feature Classes & Tables → created with specified
 Define how and where the data will physically be stored. geometry types.
 Decide on geodatabase type, storage location, indexing, and • Parcel (Polygon), Road (Line), Owner (Non-spatial
access control. table).
 Ensure performance, scalability, and security. c) Field Implementation
•Data types implemented as per logical design:
Key Components • ParcelID → Text(10).
• Area → Double.
a) Choice of Geodatabase Type
• OwnerID → Text(10), indexed for fast joins.
•File Geodatabase (.gdb)
• Best for single-user or small teams. d) Rules & Constraints Implementation
• Stores datasets as binary files in folders.  Domains: Created in geodatabase properties and
assigned to fields.
• Capacity ~1 TB per dataset.
 Subtypes: Defined for feature classes (e.g., Road =
•Personal Geodatabase (.mdb) {Primary, Secondary, Tertiary}).
• Based on Microsoft Access. Limited (2 GB)  Topology Rules: Implemented in a topology within the
•Enterprise Geodatabase (SDE): feature dataset.
• Multi-user, built on RDBMS (SQL Server, Oracle,  Relationship Classes:
PostgreSQL, etc.).
 For open Source GIS software e.g. QGIS
• GeoPackage
• PostgreSQL database with PostGIS extension
 e) Performance Optimization 5. Implementation and Testing
 Indexes:
o Attribute indexes: Create attribute indexes on ParcelID,  Final phases of geodatabase development
OwnerID, RoadID) . It speed up queries, joins, and  The geodatabase moves from plan to practice
selection.  Testing ensures the geodatabase works as
o Spatial indexes: Automatically generated for feature expected and meets requirements.
classes to speed up spatial queries (intersections,
buffers, proximity searches). 1. Schema Validation
 Storage Settings:  Check that all feature classes, attributes, and domains
o Compress feature classes, enable archiving/versioning were created correctly.
for enterprise use.  Confirm data types and field lengths match logical design.
2. Data Integrity Testing
f) Access Control and Security  Run topology validation to detect overlaps, slivers, and
 File GDB: Access control handled by OS file permissions. disconnections.
 Enterprise GDB: User roles, privileges (read-only, editor,  Test relationship rules (e.g., cannot add a Parcel without
admin). an Owner).
 Versioning: Allows multiple editors to work simultaneously.  Validate domains (e.g., LandUse only accepts
Residential/Commercial/etc.).
g) Backup and Maintenance 3. Functionality Testing
 Backup strategy: scheduled backups or .gdb folder backups.  Perform test queries:
 Data compression to reduce storage size.  Find Parcels owned by “John Doe”.
 Periodic integrity checks and topology validation.  Identify Buildings taller than 20m within
Commercial LandUse.
 Generate test maps and reports.
4. Performance Testing Final Geodatabase Structure
 Test query speed on large datasets.
 Test concurrent edits (multi-user environment).
 Check indexing effectiveness.
5. User Acceptance Testing (UAT)
 Let actual users (surveyors, planners, managers) test
workflows.
 Validate that the system supports decision-making needs
(e.g., land allocation, infrastructure planning).
6. Error Handling and Correction
 Document errors (e.g., missing attributes, topology
violations).
 Correct or adjust schema/data.
 Re-test until validation passes.

Geodatabase

Rater Dataset
Geodatabase design & Development Implications of Poor Design
Integration with the Geo- Choices
information Cycle  Redundancy and Inconsistency: Poor
schema design can lead to duplication of
 The geo-information cycle involves data data, conflicting records, and inefficient
acquisition, data pre-processing, storage & storage.
management, analysis & modelling,  Loss of Spatial Integrity: Failure to
visualization, information dissemination & use. enforce topological rules results in errors
 Geodatabase design and development Provided a structured such as gaps, overlaps, and dangling
underpins this cycle by providing a structured framework to manage nodes.
framework to manage spatial data consistently spatial data  Reduced Analytical Reliability:
at each stage. consistently at each Inaccurate or incomplete attribute
 In the capture stage, well-designed schemas stage. definitions compromise spatial analyses
ensure data is stored in appropriate formats with like overlays or network analysis.
proper attribute domains.  Poor Decision-making: Erroneous outpu
 During management and analysis, rules, from a poorly designed geodatabase can
topologies, and relationships in a geodatabase misinform planning, resource allocation, o
ensure that spatial and attribute integrity is risk assessment.
maintained.  Scalability and Maintenance Issues:
 In the visualization and dissemination stage, Inefficient structures hinder updates, mult
geodatabases allow efficient querying and map user collaboration, and integration with
production, enabling reliable communication of other datasets. (5 marks)
spatial information. (5 marks)
 Geodatabase Approaches
a) Relational Approach How it Works in GIS
 Based on Relational Database Management Systems
(RDBMS), first proposed by Edgar Codd (1970).
In a relational geodatabase, spatial data and attribute data are
 In this model, all data is stored in tables (relations) usually stored separately but linked
consisting of rows (records/tuples) and columns  Geometry Table (or spatial column): stores the shape
(fields/attributes). of the feature (points, lines, polygons).
 In GIS:  Attribute Table: stores descriptive data (names,
o Rows represent geographic features (e.g., a road, categories, IDs).
parcel, or school).  Joins & Relates: used to connect different tables for
o Columns represent the attributes (e.g., road analysis.
name, parcel ID, school type).
 Relationships between tables are established via keys:
 This makes the relational approach very powerful for
 Primary Key (PK): a unique identifier in a table (e.g., querying and managing spatial + non-spatial data.
ParcelID).
Tables
 Foreign Key (FK): a field in another table that links back
to the PK (e.g., OwnerID in the Parcels table that links to Attribute Table
the Owners table). Geometry Table
 Example: Storing parcels in one table, owners in
another, linked by parcel ID.

SELECT ParcelID, Location, Area

Sample query: FROM Parcels
WHERE Area > 500;
 Geodatabase Approaches….Cont.
b) Object-Relational Approach Sample cadastral system:
The object–relational approach is a hybrid model  Parcel Feature Class (geometry + ParcelNo, Area, LandUse).
that combines:  Owners Table (OwnerID, Name, Contact).
 The relational database model (tables, rows,  A relationship class links ParcelNo (Table1) to ParcelNo (Table2)
columns, keys) with object-oriented concepts (e.g o Selecting a parcel automatically highlights the related
rules, relationships, and behavior). owner(s)
 Traditional relational databases can store tabular Relationship
data very well, but spatial data (points, lines, Class
polygons) and their behaviors (connectivity,
topology, rules) are not naturally supported.
 Therefore it has direct support for POINT,
LINESTRING, POLYGON.
When
records are
 The object–relational model brings the two
selected in
worlds together; tables + spatial objects + rules
one table,
+ behavior.
their
related
Why Its Object-Relational Approach
records in
o Data is in relational tables
the other
o Features are objects with spatial types.
table are
o The geodatabase enforces relationship and
automatical
rules beyond what a pure relational DB can
ly selected
do.
 Geodatabase Approaches Example
c) Object-Oriented Approach  Class: Parcel
 Built entirely on the object-oriented programming  Parcel 101: Object
model.  Attributes: ParcelID, Area, LandUse
 Data is stored as objects (just like in Java, Python, or  Methods: CalculateArea()-auto-calculates from polygon geometry
C++). CheckEncroachment() - verifies if parcel overlaps another.
 Data is stored as objects with attributes and  Class: Building
behaviors.  Building 10: Object
 Attributes: BuildingID, Floors, Height
 Each object has:
 Methods: CalculateFloorAreaRatio() – building area ÷ parcel area
o Attributes (data values) CheckZoningCompliance() –ensures height is allowed in
o Methods (behaviors or functions) zoning law.
 Instead of only using tables and relationships (like in  Class: Owner
the relational approach), it organizes geographic data  Owner 30: Object
as objects that combine both attributes (data) and  Attributes: OwnerID, Name, Contact
behavior (methods/functions) into one unified  Methods: ValidateDocuments()-ValidateTitleDeed() – checks
model. ownership documents.

In the parcel Example:

 Class: Parcel = Template (general design)
 Object: Parcel 101 = a specific parcel created from that class
 Attributes: ParcelID, Area, LandUse = data that describes the
parcel.
 Methods:
o CalculateArea() → behavior to compute area.
o CheckEncroachment() → behavior to check overlaps.
 Object-Oriented Approach……… Cont.
NB:
Pure Object-Oriented GIS (OOGIS) is more of a programming
concept than something you can fully set up inside a
geodatabase.

Sample case
[Link] draws/updates the building geometry in the spatial layer. building Not well
[Link] geodatabase method CheckEncroachment() is triggered. Digitized
[Link] compares the new building polygon against:
•Parcel boundaries
•Easements, road reserves, buffer zones
4. If no encroachment (True)
•The building is accepted into the spatial layer.
•The layer updates normally.
5. If encroachment is detected (False)
Depending on the system configuration:
•Reject: The system blocks saving the feature (common in
cadastral systems).
•Flag: The building is saved but flagged in the attribute table
(e.g., Status = “Encroachment”).
•Auto-correct: The geodatabase automatically clips or
adjusts the geometry to fit inside the parcel.
•Trigger workflow: A notification or task is sent to
surveyors/planners to resolve the encroachment.
 Data Structures & Spatial ADTs in a Geodatabase
Data Structures in Databases: Representation of ADTs
Ways of organizing and storing data so it can be efficiently
retrieved and manipulated. Two main types in a geodatabase: 1. In geodatabases, the primary way Spatial ADTs are
• Non-spatial: Numbers, text, dates eg Owner table represented is through the geometry field (Shape
• Spatial: Points, lines, polygons, rasters. Eg A Parcel field) of feature classes
feature class
2. But apart from pure geometry storage, there are other
Spatial Abstract Data Types (ADTs) supporting mechanisms that represent or extend
Spatial ADTs:
 Spatial ADTs are specialized data types in a geodatabases that
represent spatial objects (geometry) and provide operations
Spatial ADTs in Geodatabases
to analyze them
 They extend the database with geometry awareness, so the
database can understand and process spatial data.
 They represent geometry (location + shape) and allow spatial
operations Primary
 Operations = actions like calculate area, find length, check Representation
Supporting
overlap, buffer, intersect. Geometry Field (Shape
mechanisms
 They allow spatial objects to be stored, queried, and field) with feature
 Raster datasets
manipulated classes
 Network datasets
 Point
 Topology rules
 Line
 Polygon
Spatial Abstract Data Types (ADTs)
2. Raster ADTs
 Used for continuous surfaces stored as grids of
1. Geometry Representation cells.
 Represent vector objects such as points, lines, and polygons.  Each cell holds a value (e.g., elevation,
 Allow both the storage of shape and methods to analyze it. temperature, rainfall).
 Enable spatial queries like distance, area, buffer, intersection.  Support operations across cells (map algebra,
 Examples: statistics). eg
 POINT → location of a borehole or school. o Elevation model (DEM): used for slope and
 LINE → a road or river. contour generation.
 POLYGON → a land parcel or administrative [Link] o Land cover raster: forest, grassland, urban
o Temperature raster: average cell values for
climate zones.
o reclassify(raster):group temperature values
into classes.
3. Topology ADTs 4. Network ADTs
 Store and enforce spatial relationships like adjacency,  Represent spatial features as graphs with nodes and
connectivity, containment. edges.
 Ensure data quality by applying rules.  Store connectivity and support network-based analysis.
 Provide methods for validating geometry eg  Useful for transport, utilities, and communication
o Parcels must not overlap (rule in cadastral mapping) networks.
 Examples:
o Road network dataset : used for shortest path
routing.
o Water supply pipes: flow capacity analysis.
o Power grid: connectivity modeling.
o shortestPath (Hospital, FireStation) :fastest
emergency route.
o flowCapacity(PipeSegment): max water flow.
Spatial Abstract Data Types (ADTs)……Cont.

Spatial
ADTs

Summary: Spatial ADTs extend geodatabases with:

 Geometry ADTs → vector objects + geometric operations.
 Raster ADTs → grids for continuous data + cell operations.
 Topology ADTs → rules and relationships for data quality.
 Network ADTs → connectivity models for routing and
flow analysis.
 Topology Encoding in Geospatial Databases
 In geospatial databases, topology encoding means storing not only
the geometry of features (points, lines, polygons) but also their
spatial relationships such as adjacency, connectivity, and Topological Elements and
containment in a structured way. relationships
 It transforms raw spatial data into a network of relationships that
can be efficiently queried, validated, and used for analysis.
Key Concepts
Nodes and Edges
 Points (nodes) represent intersections or endpoints.
 Lines (edges) connect nodes, forming networks (roads, rivers).
Faces (Polygons)
 Enclosed areas are stored as faces bounded by edges. For example
a land parcel polygon is defined by boundary lines.
Topological Rules
 No gaps between adjacent polygons (e.g., cadastral parcels).
 Lines connect only at endpoints (e.g., road networks).
 Polygons must close properly.
Encoding Methods
1. Arc-node model: Stores lines (arcs) with start and end nodes. 4
2. Polygon-arc model: Polygons defined by the arcs forming their
boundary.
3. Adjacency lists / matrices: Store which features are next to or
connected to others.
Standard Query Language (SQL) in Geodatabases
 Structured Query Language (SQL) is a standard language for b) Data Querying/Query Generation
managing, manipulating, and retrieving data in relational
databases  SQL enables users to extract meaningful
 In geodatabases, SQL is extended to handle spatial data information from a geodatabase using queries.
(geometries like points, lines, polygons) in addition to  Retrieve records (attributes and/or geometries).
traditional attribute data.  These queries can be simple, such as selecting all
 This extended SQL is sometimes called Spatial SQL. features in a specific region (SELECT * FROM Parcels
Functions/Roles of SQL in Geodatabases WHERE Zone='A'), or complex, involving joins,
a) Schema Specification aggregations, and subqueries to analyze spatial and
 SQL is used to define the structure of a geodatabase, attribute data simultaneously.
including tables, fields (attributes), data types, and  In geodatabases, SQL supports both attribute
relationships. queries (filtering based on field values) and spatial
 It uses commands such as CREATE TABLE, ALTER TABLE, queries when combined with spatial functions (e.g.,
and DROP TABLE, ST_Within, ST_Intersects).
 GIS professionals can specify how spatial and non-  Example:
spatial data are organized and linked. For example, a
table for parcels can include fields for parcel ID, owner,
and geometry type, ensuring consistent data storage.
 Example: create tables, feature classes, or
relationships.
o CREATE TABLE Parcels (ParcelID INT PRIMARY KEY,
OwnerName TEXT, Shape GEOMETRY);
Standard Query Language (SQL) in Geodatabases…..Cont.
c) Data Manipulation e) Analysis & Derived Data
 SQL allows users to insert, update, and delete records  SQL allows users to create views, generate reports,
in a geodatabase (INSERT, UPDATE, DELETE). or export subsets of data for further analysis or
 This supports maintaining accurate, up-to-date spatial mapping.
and attribute datasets, which is crucial for GIS  Derived data, such as calculated fields, summaries,
analyses. or spatial statistics, can be produced directly from
 For example: SQL queries to support GIS workflows.
 Overlay, buffering, intersections, etc. for example:
d) Spatial Queries
 SQL enables the extraction of data beyond attribute
queries and include spatial queries (when combined
with spatial functions like ST_Within or ST_Intersects).
(Create a 500m buffer around highways.)
 Users can filter data, perform aggregations, or join
tables to derive new datasets, supporting spatial  Tabular reports
analyses and decision-making. Output:  Thematic maps
 Derived datasets
o Containment:
(overlay, buffer etc.)
(Find schools within each district.) SQL is fundamental in geodatabases for schema definition,
data manipulation, querying (including spatial queries),
o Adjacency:
analysis, and generating outputs. It enables GIS
professionals to manage, analyze, and derive meaningful
(Find parcels that share boundaries.) insights from geospatial data efficiently.
 Sample R-tree diagram
Algorithms and Data Structures in Geodatabases
A geodatabase is not just about storing features (points, lines, o Roots - Holds largest Region
polygons, raster). It also needs efficient algorithms and (T,U)
specialized data structures to: o Internal Nodes - Store data
 Store large spatial datasets. bounded within parent area
 Support spatial queries (e.g., “Which parcels intersect Road X?”). (NPQRS)
 Perform spatial analysis (buffers, overlays, shortest path). o Leaf Nodes- Store actual
 Ensure integrity (topology rules, indexing). data within respective
minimum Bounding Region
Core Data Structures in Geodatabases (MBR)-A to M
 Efficient storage and retrieval of spatial records
 Includes key data indexing structures
Sample Quadtree diagram with points A, B, C, E, and F
a) Spatial Index Structures: organize spatial data for fast access. NW NE B ACFEB
 R-tree / R*-tree: Hierarchical bounding boxes for efficient A
0

spatial searches (e.g., “find features within this rectangle”). F Root = whole space
0
 Quadtrees: Divide space recursively into quadrants, efficient C E
for rasters or point clouds. A C E NE = Internal Nodes
SW SE
 Grid Indexes: Divide space into fixed grid cells, useful for dense
data. B 0 = Leaf Nodes
b) Topology Structures: Encodes relationships: F
Algorithms in Geodatabases: 0
 Nodes = points at intersections.
 Edges = lines connecting nodes.  Spatial Joins: Find objects that intersect/contain each other.
 Faces = polygons enclosed by edges.  Nearest Neighbor Search: E.g. find the closest hospital to a house.
 Overlay Operations: Union, intersection, difference of polygons.
c) Network Data Structures  Buffering: Creating zones around objects.
 Graphs = nodes + edges for modeling transportation or utilities.
 Shortest Path: Dijkstra’s algorithm for network routing.
.d) Raster Data Structures: Regular grids, pyramids etc  Voronoi Diagrams: Partition space into regions closest to given points
 GIS Analyses Supported by Geodatabases
Spatial Queries & Retrieval
 Function: Uses SQL and spatial operators to retrieve
features based on location or attribute conditions (e.g.,
ST_Intersects, ST_Within). eg “Find all parcels within 500 m
of a river.”
 Best suited for: Identifying and selecting features that
meet specific criteria, such as locating schools within 2
km of a highway.
Overlay Analysis
 Function: Combines two or more spatial layers to derive
new information For example:
o Intersecting land use with soil types
o Intersect parcels with flood hazard zones to
identify at-risk land.
 Best suited for: Suitability studies, environmental impact
assessments, or identifying areas meeting multiple
conditions.
Buffering & Proximity Analysis Parcels within a 30 m buffer around river
distance of 30 m from
 Function: Creates zones around spatial features at a rivers
centerline
specified distance (e.g., 500 m buffer around rivers).
 Best suited for: Planning and regulatory compliance,
such as defining protected areas or service catchments.
GIS Analyses Supported by Geodatabases
Network Analysis
 Function: Uses network data models to analyze
connectivity, shortest paths, and service areas.
 Best suited for: Transportation planning, logistics,
and emergency response (e.g., finding the fastest A
B
route for ambulances)

Surface & Terrain Analysis

 Function: Uses elevation data stored in the
geodatabase. (DEM, DTM, TIN) to analyze slope, Finding Shortest Path
aspect, hillshade, visibility, and hydrological flow. From Point A to Point B
 Best suited for: Engineering projects, watershed
management, hazard mapping, and infrastructure
planning in hilly terrains.

Topological Analysis
 Function: Enforce and examine spatial
relationships such as adjacency, connectivity, and
containment between features.
 Best suited for: Network analysis (e.g., road
connectivity), parcel boundary validation, and
Surface/Terrain Anlysis
detecting overlaps or gaps in cadastral mapping.
 Geo-visualisation and Information Dissemination
 Geo-visualization (method of presentation - how you show
Role of Geo-visualization
spatial info) refers to the interactive visual representation of
o To convert raw geospatial data into
geospatial data for:
meaningful visual insights.
o Analysis, Exploration & Decision-making
o To enable spatial pattern recognition,
 When integrated within a geodatabase, it enhances how spatial
trend analysis, and scenario modeling.
information is managed, displayed, and shared
 Information dissemination then ensures that the insights derived
from spatial data reach various stakeholders in usable formats.
 Geo-visualization plays a central role in
Geovisualization cube
transforming raw spatial data into clear,
Geovisualization: interpretable visual products that support
From exploration of communication and decision-making.
unknown data set,  It allows complex spatial patterns, relationships,
validation of results and trends to be represented in intuitive forms
, communication of such as maps, dashboards, and 3D models.
known information  Enhances understanding by making technical
data accessible to both experts and non-
specialists.
 Supports evidence-based decision-making by
presenting spatial insights in a visual context.
Three main dimensions  Facilitates stakeholder engagement through
o User or audience (Public vs Expert) interactive and dynamic map products.
o Interaction (High vs. Low)
o Information content (Unknown vs. Known),
 Geo-visualisation and Information Dissemination
Geovisualization techniques refer to the set of methods and tools
Thematic Mapping:
used to visually explore, analyze, and communicate spatial data. Focuses on visualizing a specific geographic
They combine cartography, geographic information systems (GIS), pattern or dataset, such as population
and interactive visualization to support spatial reasoning and density, temperature, or language, across a
decision-making particular area etc..
Map-Based Visualization Techniques

Thematic Mapping: Representing variables such as population

density, land use, or rainfall.
 Proportional Symbol Maps: Represent data using varying
symbol sizes at point locations. Visualizing city populations,
number of facilities.
 Map Symbolization: Use of colors, patterns, and symbols to
represent attributes.
 Dot Density Maps: Use dots to show distribution and
density of a feature. Population or disease case distribution.
 Flow Maps: Show movement between locations using
arrows or lines. Migration patterns, trade routes.
 Choropleth Maps: Use color gradients or patterns to
represent quantitative data aggregated by region.
Population density, literacy rates.
Proportional Symbol Maps
Geo-visualisation and Information Dissemination……………. Cont.

Flow Map

Sample population density

Dot Map Sample Health Coverage

Map Symbolization: 3D Visualization Techniques
Continuous digital
Tools and Platforms
It is the fundamental language of maps,  ArcGIS Pro / ArcMap (Map
using shapes, sizes, and colors to encode representation of the Earth’s
surface from discrete elevation documents connected to
information about location, quantity, and geodatabases)
quality in a simplified, visual format or attribute data.
 QGIS with PostGIS or
GeoPackage
Sample Rates Payment Status  ArcGIS Online, Mapbox or
Leaflet for web
dissemination
 Dashboards and Story Maps
for interactive storytelling

Paid
Not Paid
Not Known
Temporal (Time-based) Visualization Interactive Visualization:
The process of browsing, exploring, and analyzing spatial data within a real
Focus on how spatial data changes over time
GIS environment, where users can dynamically interact with the map rather
 Detect trends, patterns, and cycles over time than just view it passively:
 Compare conditions between different periods
 Control what you see (choose themes, layers, or time periods).
 Forecast or simulate future conditions based on
 Interact with the map (click, hover, or query features).
temporal trends.
 Integrate additional content (charts, media, dashboards, or tables).
 Support decision-making through time-aware analysis
 Receive real-time feedback based on their selections.
e.g, understanding how urban growth
Sample Land use Visualization from a satellite imagery

Same Place 2016

Same Place 2024

Information Dissemination from a Geodatabase
Geo-visualization - Complex spatial data
Information dissemination (delivery - how you share it with stakeholders) is understandable)
 To make spatial data and insights accessible to various users
 From technical analysts to policymakers and the public

 Information dissemination is the process of sharing geographic

information with users, decision-makers, or the public.
 Can be through maps, web portals, reports, mobile apps, dashboards.
 The goal: Make GIS outputs accessible and actionable
Dissemination Type Example Tools Description
Static Maps PDF, JPG exports Suitable for reports and print
ArcGIS Online, Allow users to pan, zoom,
Interactive Web Maps
Leaflet, Mapbox and query data
ArcGIS Dashboards, Real-time monitoring of
Dashboards
Power BI spatial indicators
Combine maps, text, images,
Story Maps ArcGIS StoryMaps
videos, and data visualization
World Bank Open
Provide downloadable
Open Data Portals Data Portal, UN GIS
datasets
Portal …..etc
Disseminate and collect Information dissemination = sharing those
Mobile Apps Field Maps, QField
spatial data in the field insights with decision-makers and the public
Information Dissemination from a Geodatabase……Cont.
Data Access Models Public Access (Open Data Model)
 Data access models define:
 Who can access spatial data  Public access means spatial data and maps are freely
 How they can access it available for anyone to view, download, or reuse
 Under what conditions without restrictive permissions
 These models are critical in ensuring the right balance  Datasets are hosted on open platforms (e.g., ArcGIS
between data openness, security, and usability. In a Hub, GeoNode, CKAN, [Link], OpenStreetMap)
geodatabase-driven environment, they determine how  Access requires no login credentials
information is shared for analysis, visualization, or decision-  Data is usually shared in open formats (e.g.,
making. shapefiles, GeoJSON, CSV, KML).
 Metadata describing data sources, accuracy, and
update frequency is provided.
Examples of Data Access Models
1. Public Access: Open data for general Example:
public use A National Open Data Portal provides free access to base
2. Restricted Access: For authorized users maps, administrative boundaries, and land cover data for
(internal use, government, partners) planning and research.
3. API-based Access: Sharing through web
services (WMS, WFS, REST APIs)
4. Hybrid Access (Combined Model)
combination of all the three above
Information Dissemination from a Geodatabase……Cont.
API-based Access (Web Services Model)
Restricted Access (Controlled Data Model)
 This model provides data access through standardized
Application Programming Interfaces (APIs) and web
 This model limits access to specific users or
services, allowing systems and users to connect directly to
organizations, such as government departments,
the geodatabase for visualization, analysis, or integration.
technical partners, or licensed entities, through
 Types of Services:
authentication or role-based permissions.
o WMS (Web Map Service): Delivers map images
 Requires login credentials or institutional
rendered by the server, users can view maps but not
authorization.
modify or download raw data - NASA Global Imagery
 Users are assigned roles with defined permissions
o WFS (Web Feature Service): Allows querying and
(e.g., read-only, edit, publish).
downloading of actual vector features for spatial
 Access is controlled through security protocols, VPNs,
analysis Natural Earth via GeoServer, OSM basemaps
or enterprise authentication (Active Directory, SSO).
o WCS (Web Coverage Service): Provides access to raster
 Data may include sensitive or proprietary
datasets (e.g., DEMs, satellite imagery) - European
information.
Space Agency
o REST API (Representational State Transfer): Enables
Example:
interaction with ArcGIS Server, GeoServer, or other
A Ministry of Lands manages cadastral and ownership
platforms through URLs and JSON/XML responses.-
data in a centralized geodatabase. Only authorized
ArcGIS Online
surveyors and registry staff can access or update parcel
Example:
records.
A disaster management portal accesses flood risk layers via
REST APIs from the national geodatabase to update its web
dashboard every hour.
Information Dissemination from a Geodatabase……Cont.

Hybrid Access (Combined Model) Key Considerations in Designing Access Models

 In practice, many organizations adopt a hybrid model,
combining all three types:  Data sensitivity: Ownership, security, and
o Public layers for base maps and non-sensitive privacy concerns.
data.  User roles: Who needs access and for what
o Restricted layers for sensitive or internal datasets. purpose.
o APIs for cross-platform access and automated  Technical infrastructure: Server capacity,
updates bandwidth, and interoperability standards.
 Legal framework: Data licensing, intellectual
Example: property, and open data policies.
A National Spatial Data Infrastructure (NSDI) platform  Sustainability: Maintenance, versioning, and
might publicly share base maps and demographic data long-term availability.
while restricting cadastral information to licensed users,
all served through APIs for real-time integration.
Errors and Uncertainty in Geodatabases  Topological Errors: Violations of spatial rules (e.g.,
overlaps, gaps, unconnected line segments, dangling
In a geodatabase, errors and uncertainty refer to imperfections
nodes etc).
in the spatial or attribute information stored, and the degree to
o e.g. Two land parcels overlap even though parcels
which data deviates from its true real-world position or
should share boundaries exactly
characteristics.
 Temporal Errors: Outdated data that no longer reflects
real-world conditions.
Why It is Important to understand Errors and Uncertainty
o A road network layer from 2014 is used for
 Data quality
planning, but new roads built in 2020 are missing.
 Spatial analysis
o A land use map from 2016 still shows farmland
 Decision-making
where housing estates were constructed in 2021.
o A population dataset from the 2009 census is used
Errors and Uncertainties in Geodatabases
to plan school locations in 2025.
 Positional Errors: Inaccuracies in the location of spatial
o Forest cover dataset from 2010 does not show
features. A geodatabase can contain mistakes such as wrong
areas that were reforested in the last decade.
locations, misplaced or shifted features). This means that the
o A parcel ownership attribute table is outdated
feature is in the wrong position on the earth’s surface.
because recent land transfers or subdivisions were
e.g.
not updated in the geodatabase……..etc
o A school point is 20 meters from the real school
 Measurement/Resolution Errors: Limitations in scale or
location.
sensor precision leading to generalized representations.
o A road is shifted because of a wrong coordinate system.
o These errors happen when data cannot perfectly
o A parcel boundary is slightly misplaced.
capture reality due to the precision of instruments
o Attribute Errors: Incorrect or incomplete descriptive
or scale of mapping
information linked to spatial features (e.g., wrong name, ID,
o These errors are not mistakes, but limitations of
Type, Area etc of a parcel)
data collection tools.
 Errors and Uncertainty in Geodatabases…..Cont.
o Topological errors, such as gaps in a road network, can
 Uncertainty: The level of doubt or confidence in the accuracy, distort shortest path or network analyses
position, or classification of a feature in a geodatabase o A school point collected with a GPS accuracy of ±10 m
o While errors are mistakes, uncertainty acknowledges is buffered by 500 m; The buffer location is also
limitations in measurement, classification, or uncertain by ±10 m, affecting distance-based analysis
interpretation. like:
o It helps users decide how much trust to place in the data  Eg which houses are within 500 m
for analysis or decision-making.  As multiple layers are combined, each containing
 A GPS point of a well is recorded with ±8 meters accuracy inherent uncertainty, the overall accuracy of the final
 A river digitized from a low-resolution satellite image may output decreases unless error handling methods (e.g.,
not follow the true meanders precisely data validation or error tracking) are applied.
 A land cover map classifies a pixel as “forest” with 70%
confidence; it might actually be shrub-land.
Error Propagation in GIS:
 When geoprocessing operations such as overlay,
buffering, or spatial analysis are performed, existing
errors can accumulate and magnify. Examples:
o In an overlay analysis of land parcels and flood zones,
positional errors may shift boundaries, leading to
misclassification of parcels as flood-prone.
o Digitizing error in a parcel boundary causes error in
area calculation , misrepresentation in land taxation
Error in a parcel boundary = Error in area calculation
 Errors and Uncertainty in Geodatabases…Cont.
Managing Errors and Uncertainty in GIS Documentation:
 Metadata (data about data) helps track the accuracy, source, a. Documenting errors and uncertainty
and reliability of each dataset. o Even with QC, some errors or uncertainties remains
o Metadata = “data about data.” o Documenting them helps users make informed
o It describes how, when, and by whom the data was decisions e.g.
collected, its accuracy, coordinate system, lineage  A flood map may indicate areas of high,
among others. medium, and low confidence based on source
o It helps users assess reliability before using data for data accuracy.
analysis  A land cover dataset could include accuracy
o Ensures traceability if errors are discovered later percentages for each class (forest, urban,
o Supports data sharing and interoperability farmland).
 Good quality control: Ensures that datasets are consistent, b. Documenting standard operating procedures (SOPs)
accurate, and ready for analysis Ensure consistency, accuracy and reliability in how data
o Topology checks - Polygons should not overlap, lines is collected, edited, and analyzed
should connect properly, and points should fall inside o Consistency: Everyone follows the same steps for
polygons. data collection, editing, and analysis.
o Snapping - Ensures digitized features align perfectly with o Accuracy: Define how to check and validate data
existing features (e.g., roads connecting at intersections). (e.g. digitizing rules, attribute domains)
o Attribute domains - Restrict values to valid entries (e.g., o Traceability: SOPs describe the workflow so errors
land use = “residential,” “commercial,” “agricultural”) to can be tracked back to their source.
prevent typos. o Training: New staff can quickly understand proper
o GPS/GNSS accuracy checks-Compare field data to known methods, reducing mistakes.
control points to verify precision o Documentation: SOPs themselves become part of
the metadata and quality control records.
 Data Management in GIS
Data Management in GIS refers to the overall set of strategies,
policies, and practices that govern how spatial and attribute  Integrate and transform: Ensuring compatibility
data are collected, stored, maintained, validated, secured, and between heterogeneous datasets by applying
shared. standardized formats (e.g., OGC standards like WMS,
How data management works WFS) and harmonization methods.
Effective data management in a GIS environment ensures that  Secure and Govern: Implementing role-based
spatial and attribute data remain accurate, consistent, and permissions, backups, and data encryption to
accessible for analysis and decision-making. Here’s how it works: safeguard sensitive spatial data while allowing
 Collect: Careful selection of reliable data sources, proper controlled access.
digitization, georeferencing, and metadata  Access and share: Using portals, web services, or
documentation to ensure that data entering the GIS is dashboards to distribute data products to
accurate and traceable. stakeholders, ensuring accessibility while respecting
 Store and organize: Use of geodatabases to structure licensing and policy requirements.
spatial and attribute data, enforce integrity through rules  Data Maintenance and Updating: Establishing
and relationships, and support scalability for multi-user protocols for routine updates and versioning to
environments. ensure datasets remain current and relevant for
 Catalogue and document: Comprehensive decision-making.
documentation of metadata; data lineage, accuracy,
resolution, and update cycles so that users understand
the reliability and applicability of datasets.
 Data management models and tools
This is how GIS data is :
 Organized 2. Decentralized (Distributed) Data Management Model:
 Maintained Each department or unit maintains its own GIS data and
 Shared………… across users and departments systems independently.
1. Centralized Data Management Model: All GIS data is  Data resides in multiple locations or departmental
stored and managed in a single central database or databases.
repository (e.g., a centralized geodatabase on a main  Each group is responsible for its own updates and
server). accuracy.
 One authoritative source of data.
 Centralized control over data editing, updating, and Advantages
security.  Greater flexibility for departments to tailor data to their
 Usually managed by a dedicated GIS team or data needs
administrator.  Reduces dependency on a central GIS team
 Advantages  Easier for large or multi-location organizations
 Ensures data consistency and standardization.
 Easier to enforce data quality and version control. Disadvantages
 Reduces duplication of data.  Data redundancy and inconsistency across departments.
 Difficult to integrate or share data organization-wide.
 Disadvantages  Requires strong coordination for enterprise-level
 Can become a bottleneck if the central system fails. analysis.
 Users depend heavily on the central GIS team. Sample Tool: Local GIS systems: Each city department
 May limit flexibility for departments with unique (planning, utilities, environment) maintains its own GIS
data needs. data independently
Sample Tool: ArcGIS Enterprise Geodatabase,
 Data management models ……Cont.
3. Federated (Hybrid) Data Management Model:
4. Cloud-Based GIS Data Management Model
Combines centralized standards with decentralized
Data is stored and managed on cloud platforms rather
data ownership.
than on-premises servers
Departments maintain their own datasets but follow
 Uses web-based GIS platforms and cloud storage.
common standards and share data through an
 Accessible from anywhere with proper
integrated framework.
authentication.
 Central GIS unit sets data standards and provides
infrastructure(Hardware, software, data, services,
Advantages:
security, governance, support etc).
 Scalable and cost-effective (no need for local
 Departments manage & update their data locally
servers).
 Data is accessible organization-wide through web
 Supports collaboration and real-time access.
services or portals.
 Easy data backup and disaster recovery.
Advantages
Disadvantages:
 Balances data consistency
 Dependent on internet connectivity.
 Promotes data sharing and interoperability.
 Potential security and privacy concerns.
 Reduces redundancy while allowing flexibility.
 Subscription or usage-based costs.
Disadvantages:
 Requires strong coordination and communication.
 Needs good network infrastructure and metadata
management.
Sample Tool: ArcGIS Online
Sample Tool: NSDI
 Data management models ……Cont.
Benefits of effective data management 5. Cost Savings
o Minimizes redundant data collection and storage.
1. Improved Data Accuracy and Consistency o Optimizes infrastructure and workforce utilization.
o Ensures all spatial data is up-to-date, complete, and 6. Stronger Data Security and Control
standardized. o Implements access permissions, backups, and
o Reduces errors, duplication, and discrepancies across version control to protect data integrity.
datasets. o Prevents unauthorized edits and data loss.
2. Enhanced Decision-Making 7. Supports Spatial Analysis and Modelling
o Provides reliable geographic insights for planning, o Clean and consistent data enables more accurate
analysis, and policy-making. spatial modeling, mapping, and scenario
o Supports evidence-based decisions in areas like urban simulations.
planning, resource management, and disaster response. 8. Long-Term Sustainability
3. Increased Efficiency and Productivity o Ensures data remains usable and relevant over time
o Centralized, well-organized data saves time spent through proper metadata and documentation.
searching for or re-creating datasets. 9. Compliance and Standardization
o Automates repetitive tasks like data validation and o Meets national/international standards (e.g., ISO,
updates. OGC), improving interoperability and legal
4. Better Data Sharing and Collaboration compliance.
o Facilitates seamless data exchange between 10. Informed Public Services and Policy
departments, organizations, and the public. o Helps governments and agencies provide better
o Encourages cross-sector cooperation using interoperable services e.g., emergency response, land
GIS platforms and web services (e.g., WMS, WFS). management, and infrastructure planning.
 Setting Up a GIS
To effectively use spatial data for planning, analysis, and A systematic process ensures successful GIS
decision-making, the first step is setting up a functional implementation.
Geographic Information System (GIS).
 GIS can support any decision-making Steps:
 A proper GIS infrastructure must exist  Awareness Creation: Educating stakeholders about
GIS benefits. This helps to gain support, funding, and
Components of Setting Up a GIS cooperation from stakeholders. E.g
 Hardware (computers, servers, GPS equipment): To  Introducing GIS concepts to management and staff
 Demonstrating how GIS improves decision-making
provide the computing power and field tools needed
 Holding awareness workshops or presentations
to create and manage geospatial information.  Showing examples from similar organizations
 Software (ArcGIS, QGIS, databases like PostGIS): These  System Requirements Analysis: This identifies what
are the applications used to store, analyze, visualize, the organization actually needs in order to build a
and manage spatial data. functioning GIS e.g.
 Data (base maps, satellite imagery, boundaries, o Determine the types of maps and analyses
geodatabases): Data is the core of any GIS. A GIS needed
without data cannot function o Identify hardware requirements (computers,
 Standards & Metadata: Standards ensure that data is servers, GPS units)
consistent and interoperable. Metadata describes the o Identify software needs (ArcGIS, QGIS, PostGIS,
quality and source of the data WebGIS)
 People/ Skills: Skilled professionals who can manage o List required datasets (boundaries, roads,
data, run analyses, and maintain systems imagery, survey data)
 Policies & Workflows: These define how GIS o Identify user skills and training needs
operations should run.
 System Development
Setting Up a GIS
 System Acquisition and Start-Up
This is where the GIS begins to take shape.
This is where actual investment begins
o Designing the geodatabase structure
o Buying hardware (servers, GPS, computers)
o Setting coordinate systems and data standards
o Purchasing software licenses
o Collecting or importing data (GPS, digitizing, satellite
o Hiring or assigning GIS staff
imagery)
o Setting up network access and user permissions
o Cleaning, validating, and integrating datasets
o Installing GIS applications & database configuration
o Developing basic GIS tools, maps, and layers
 System Operation and Maintenance
 System Evaluation/Testing
A GIS must be continuously updated and maintained
Entails testing after development, before full implementation
o Updating datasets (roads, boundaries, imagery, etc.)
o Testing data accuracy (topology checks, positional accuracy)
o Training new staff and refreshing skills
o Testing system performance (speed, stability)
o Monitoring system performance
o Checking user workflows
o Backups and data security
o Identifying missing or incorrect data
o Version control(track, manage, coordinate changes to
o Running pilot tests with real users spatial data
 Implementation Plan o Producing maps, reports, and analysis for users
Once satisfied with the system, a detailed plan is created for
Once the GIS system is properly set up and populated with
launching it
accurate spatial data, it becomes a powerful engine for
o Timeline for full rollout
analysis and decision-making
o Resource allocation (who does what)
 Run spatial analysis (buffers, overlays, suitability etc.)
o Training schedule for users
 Produce maps, models, and scenarios
o Data update cycle  Becomes the foundation for informed decisions
o Maintenance strategy Spatial analysis forms the core input for higher-level tools
o Risk management plan i.e. Decision Support Systems (DSS) & Expert Systems
Introduction to Decision Support Systems (DSS) and Expert Systems
 A Decision Support System (DSS) is a computer-based An Expert System (ES) is a type of Artificial Intelligence (AI)
information system that helps decision-makers analyze data, program that mimics the decision-making ability of a
solve problems, and make informed decisions human expert.

 GIS-based DSS are widely used in fields requiring spatial Key characteristics:
decision-making  Contains a knowledge base
o A applies rules to interpret or classify that data e.g.
NB: Doesn’t attempt to replace human judgment; It supports it if–then logic
 Uses an inference engine to “reason” about spatial
Key characteristics: conditions.
 Integrates spatial datasets, maps, and attribute information.  Can provide explanations for conclusions (“why” a
 Allows scenario testing (“what if” analysis) decision was reached)
 Perform spatial analysis (buffering, overlay, proximity)  Results are mapped back in GIS for visualization and
 Visualize spatial problems (maps, patterns, trends) decision-making.
 Model geographic scenarios: multi-criteria evaluation (MCE)  Aims to imitate expert judgment rather than support it.
(e.g., flood risk zones, optimal facility locations-suitability
models, network optimization).
 Focuses on decision-awareness, not expert reasoning.
Introduction to Decision Support Systems (DSS) and Expert Systems
A Decision Support System (DSS) is a computer-based An Expert System (ES) is a type of Artificial Intelligence (AI)
information system that helps decision-makers analyze data, program that mimics the decision-making ability of a
solve problems, and make informed decisions human expert.
Role of Expert Systems in GIS
NB: Doesn’t attempt to replace human judgment; It supports it Expert System = Human Expertise + Computer Intelligence
 Expert Systems use rules, inference engines (the
Role of DSS in GIS component that applies logical rules to the knowledge
GIS-DSS = DSS + Spatial (location-based) data base ), and knowledge bases to simulate the decision-
 DSS in GIS combines spatial data, analytical models, and making ability of human experts
visualization tools to support decision-making processes.  They are particularly useful in cases where decisions
 It allows decision-makers to test different scenarios (e.g., flood require specialized domain knowledge (e.g., mineral
risk mapping, urban planning, land allocation) using what-if exploration, environmental monitoring, medical GIS).
analysis.  By embedding expert knowledge, these systems help
 Helps in multi-criteria evaluation (MCE) where various spatial users with limited expertise make complex spatial
factors (e.g., soil, slope, land use) are analyzed together for site decisions accurately.
selection or resource allocation.
 Enhances policy planning and resource management, making
GIS more than just a mapping tool, but a decision-support Expert system (applies the logic and outputs ):
environment.  A applies rules to interpret or classify that data.
 Evaluates conditions (e.g., thresholds, patterns).
 Visualize spatial problems (maps, patterns, trends)  Results are mapped back in GIS for visualization and
 Perform spatial analysis (buffering, overlay, proximity) decision-making.
 Model geographic scenarios: multi-criteria evaluation Example: If soil type = clay AND slope > 25° AND rainfall
(MCE) (e.g., flood risk zones, optimal facility locations) > 1000 mm/year = High landslide susceptibility zone
 Core Difference Between DSS & ES
Aspect Decision Support System (DSS) Expert System (ES)
Helps decision-makers analyze data and choose Emulates a human expert’s reasoning to give
Purpose
among alternatives advice or a solution
Knowledge-driven (rules, experience, expert
Knowledge Type Data-driven (quantitative models, analysis)
knowledge)
Decision Role Supports the user’s decision Advises or recommends a decision
User Involvement User interprets and decides System provides a decision or explanation
Main Component Models + databases (analytical tools) Knowledge base + Logical reasoning
“Where should the hospital be built based on
Example Question “What are the possible suitable sites for a hospital?”
expert rules?”
Example Approach “What if” analysis, scenario simulation Rule-based reasoning (“IF... THEN...”)

Example in a GIS Context

Step GIS-DSS GIS-ES
Input Data Land use, slope, roads, population density Same data
Apply expert rules (e.g., “If slope < 10° & near road, then
Approach Overlay, weight factors, run suitability model
suitable”)
Output Suitability map with scores (user interprets) Recommendation map with classifications (system explains why)
User’s Role Interprets results and makes final choice Reviews recommendation and reasoning provided by system
DSS Real World GIS Applications of DSS & ES
Example A: Flood Evacuation Planning Tool
A GIS platform that shows:
ES
Example A: Soil Classification Based on Rules
 where floods are likely,  A system that says: “If soil type = clay AND slope > 25°
 where people live, AND rainfall > 1000 mm/year = High landslide
 which roads are safe, susceptibility zone”
 Proposes evacuation routes.  soil texture = clay AND slope < 5% AND drainage = poor,
It supports decision-making but does not make the final THEN classify as Vertisol.”
decision.  This is expert knowledge turned into rules.
Example B: Land Suitability Analysis for New Schools  It performs expert-like classification automatically
A GIS tool that:
Example B: Health Risk Diagnosis Map:
 Overlays population density, distance to roads, land
 A rule-based tool: “IF rainfall > threshold AND
availability, and slope,
temperature is high AND stagnant water
 Gives a suitability map,
present THEN malaria risk = very high.”
 Allows planners to compare different sites.
 It is giving expert-like conclusions, not scenario
The user makes the final choice, the system only supports.
options.
Example C: Agricultural DSS for Fertilizer Needs DSS and Expert Systems can be designed and
A map-based system showing: implemented directly inside a GIS environment
 Soil fertility
 Rainfall trends  QGIS (with plugins, Python scripts, SAGA, GRASS)
 Crop type  ArcGIS Pro / ArcGIS Online
 Outputs the recommended fertilizer rate  WebGIS platforms (GeoServer + Leaflet/OpenLayers)
The user decides the final action.  Spatial databases (PostGIS)
DSS Example: Flood DSS in Bangladesh and Netherlands Expert System Example: FAO’s CROPWAT system
Decision Support Systems (DSS) and Expert Systems (ES)……Cont.
 Together, DSS and Expert Systems in GIS transform Real-world GIS applications of DSS & ES
geospatial technologies into powerful tools for Sample Flood DSS (Decision Support System)
decision-making.
 DSS provides flexible analytical frameworks for
evaluating scenarios
 Expert Systems offer domain-specific knowledge and
reasoning.
 Their integration enhances the ability of GIS to
o Support planning
o Support management
o Problem-solving across sectors ( environment,
health, disaster risk, urban development……etc)

Sample Agriculture Expert System

 Setting up a GIS
Setting up a GIS can mean different things depending on 2. GIS Architecture (Recommended Setup)
whether you’re doing it for a small office, a corporate company, a) Desktop GIS (eg QGIS, Arcmap… etc)
or a national-level program. A successful GIS integrates:  Data creation and editing
o Clear objectives  Spatial analysis
o Appropriate technology  Cartographic map production
o High-quality data  Data validation and quality control
o Standards and governance b) Spatial Database (eg. PostgreSQL with
o Skilled personnel PostGIS extension)
Every GIS must be established to meet clearly defined  Stores vector and raster data
objectives.  Centralized spatial data management  Supports multi-user access
 Accurate mapping and visualization  Maintains data integrity
 Spatial analysis and modeling  Enables scalability
 Decision support for planning and policy c) GIS Server / Web GIS ( eg. WMS, WFS. WCS)
 Data sharing through web services  GIS servers allow spatial data to be published and
1. Define the Purpose of the GIS accessed by multiple users.
 What problems will the GIS solve?  Serve maps and spatial data
 Who are the users (technicians, planners, management)?  Enable system-to-system data sharing
 What decisions will the system support?  Support dashboards and web applications
 Which sectors will rely on the GIS?
o Land & cadastral d) Web Mapping and Visualization (for non-technical
o Urban planning users to interact with spatial data)
o Climate risk & early warning  Management dashboards
o Natural resources / mining  Interactive maps
o Infrastructure & utilities  Public data portals
3. GIS Data Requirements 5. Hardware and Infrastructure (Supports GIS performance)
Data is the most valuable component of a GIS.  High-performance workstations
 Central server
a) Base (Reference) Data (Most valuable component of a GIS)
 Stable power supply and backups
 Administrative boundaries
 Network connectivity
 Road networks
 Cloud-based deployment
 Rivers and hydrology
 Automated backup systems
 Settlements and buildings
 High-availability configurations
 Digital Elevation Models (DEM)
b) Thematic (Sectoral) Data (Supports specific applications) 6. GIS Workflows and Processes - Data Lifecycle
 Land parcels and cadastral records  Data acquisition
 Urban zoning and land use  Data validation and quality control
 Climate and environmental data  Metadata documentation
 Infrastructure networks  Database storage
 Natural resource inventories  Analysis and modeling
c) Coordinate Reference Systems (Use of consistent spatial  Map and service publication
reference systems e.g. UTM, WGS84, Cassini etc)  Archiving and updates
7. Quality Control Procedures (Trust in GIS outputs)
 Topology rules
4. Data Standards and Governance (Ensure interoperability,
 Attribute validation
consistency, and quality) e.g.
 Accuracy assessments
 ISO 19115 – Metadata
8. Human Resources and Skills
 ISO 19157 – Data Quality
9. Security and Access Control
 OGC standards – WMS, WFS, WCS
 User authentication and authorization
 Etc….
 Role-based access control
 Audit trails, Backup and disaster recovery plans
 Case Study Examples Agriculture & Food Security GIS
 Crop monitoring and land suitability
Land Administration & Cadastre GIS
analysis
 Digitization of parcel maps and land records
 Food security and agricultural planning
 Secure land tenure and boundary management
Transport & Road Management GIS
Urban Planning & Municipal GIS
 Road inventory and maintenance planning
 Centralized GIS for zoning, infrastructure, and urban growth
 Route optimization and infrastructure
 Support development control and city planning
investment
Disaster Risk Management GIS
National Spatial Data Infrastructure (NSDI)
 Flood, drought, and hazard mapping
 Coordinating spatial data across
 Emergency preparedness and response planning
government agencies
Climate Change & Environmental Monitoring GIS
 Data sharing, standards, and national
 Monitoring land cover change and climate impacts
planning
 Environmental assessment and reporting
Natural Resources & Mining GIS
 Mapping mineral licenses and extraction areas
 Resource management and compliance monitoring
Infrastructure & Utilities GIS and many others………………………………
 Mapping and managing water, power, and telecom
networks
 Asset management and service delivery improvement
Health GIS
•Mapping health facilities and disease patterns
•Healthcare planning and outbreak monitoring
 GIS Developments
1. Web GIS 3. Cloud GIS
 Delivery and visualization of maps over the internet through  GIS platforms hosted on the cloud, offering
browsers and apps. spatial data storage, analysis, visualization,
Impacts: and sharing.
o Accessibility: Easy access to maps without requiring GIS Impacts:
software. o Accessibility: Users can log in from
o Integration: Supports embedding of maps into websites, anywhere without installing heavy
mobile apps, and dashboards. desktop GIS.
o Utilization: Enhances public participation and navigation in o Integration: Allows multiple users to
everyday life (tourism, disaster response, logistics). collaborate on the same datasets and
o Examples: Google Maps, OpenStreetMap, ArcGIS Online analyses in real time.
interactive maps. o Utilization: Reduces cost barriers;
2. Location-Based Services (LBS) organizations can scale operations and
 Applications that provide real-time services to users based on their perform advanced analytics with less
geographic location, often via GPS and mobile networks. infrastructure.
Impacts: o Examples: ESRI ArcGIS Online, Mapbox,
o Accessibility: Makes GIS functions available to the general public CARTO, Google Earth Engine.
through smartphones.
o Integration: Combines spatial data with mobile technology, IoT,
and real-time tracking.
o Utilization: Widely used in transportation, emergency response,
marketing, and tourism.
o Examples: Uber, Bolt, Glovo/Food delivery apps etc
 GIS Developments……Cont.
5. High-resolution remote sensing, UAVs and LiDAR
4. GeoAI
 The availability of frequent high-resolution imagery (commercial
 It is the integration of Artificial Intelligence (AI),
satellites, drone/UAV platforms) and LiDAR point clouds
Machine Learning (ML), and Deep Learning
,together with automated processing chains has made frequent,
techniques with geospatial data and GIS
fine-scale mapping feasible
systems. It enhances spatial analysis by
 Impact:
automating pattern recognition, prediction, and
 Accessibility: more timely and local data
decision-making. e.g.
 Integration: sensor data ingested directly into enterprise
o Automated feature extraction from
geodatabases
imagery
 Utilization: operational monitoring (e.g., vegetation change,
o Semantic segmentation (roads, buildings,
road condition) and targeted interventions.
crops)
o Geospatial reasoning that combines text
and maps. 6. Open-source ecosystem maturity (QGIS, PostGIS, GeoServer,
 GeoAI reduces manual processing, speeds up etc.)
data production  Open tools (QGIS, PostGIS, GeoServer, GDAL/OGR and others)
Impact: have continued to improve in capability, performance and
 Accessibility: subject-matter experts can get usability, making robust GIS accessible at low cost.
automated outputs without deep ML expertise Impact:
 Integration: unstructured sources (reports,  Accessibility: low-cost, community-driven tools available to
social media, imagery) can be turned into many
structured spatial data;  Integration: open formats and libraries ease linking systems
 Utilization: faster, scaled analytics for  Utilization: broader experimentation, local solutions and
humanitarian, urban, and environmental tasks. capacity building.
 Trends and Future Prospects in GIS 3. Real-Time GIS
1. Big Data GIS GIS systems capable of ingesting, processing, and
The integration of GIS with massive, high-velocity, and diverse visualizing live data feeds from sensors, GPS trackers, and
datasets generated from IoT devices, remote sensors, satellite streaming services.
imagery, mobile apps, and social media streams. Why it matters:
Why it matters: • Provides instant insights for emergency response
• Enhances situational awareness (e.g., disaster response, (wildfires, floods, pandemics).
epidemiology, urban mobility). • Enables dynamic traffic management, logistics
• Enables predictive modeling by analyzing spatial patterns routing, and smart city operations.
at unprecedented scales. • Creates interactive dashboards for monitoring
• Supports multi-source data fusion (climate, health, traffic, (e.g., power outages, weather updates).
economic data in one framework). 4. 3D/4D GIS and Spatial Digital Twins
2. Artificial Intelligence (AI) & Machine Learning (ML) in GIS Advanced GIS models that represent the world in 3D and
(GeoAI) incorporate time (4D), enabling simulations and predictive
The application of AI/ML models to automate spatial data modeling.
processing, pattern recognition, and predictive analytics. Why it matters:
Why it matters:  Supports urban planning with realistic 3D city
 Reduces human effort in tasks like feature extraction from models and infrastructure simulations.
satellite imagery, land-cover classification, and anomaly  Enhances environmental management (e.g.,
detection. coastal erosion, groundwater modeling).
 Enhances forecasting (e.g., crop yields, flood risk zones,  Powers digital twins — virtual replicas of real-
urban expansion). world systems that allow testing, prediction, and
 Supports intelligent decision-making in planning, optimization.
environment, and resource allocation.
Trends and Future Prospects in GIS…. Cont.
5. Cloud GIS
Delivery of GIS software, data storage, and analytical tools via
cloud platforms, reducing the need for local hardware.
Why it matters:
• Democratizes GIS by lowering costs and making it
accessible anywhere via the web.
• Enables collaborative projects across organizations and
geographies.
• Offers elastic scalability for processing massive datasets
(e.g., climate models, satellite archives).
•Example: ESRI ArcGIS Online, Google Earth Engine, and
Mapbox, which provide global-scale GIS capabilities over the
cloud.
Summary

 Big Data GIS: Integration with IoT, sensors, social media.

GIS evolution and future trends.
 AI and Machine Learning: Automated feature extraction
from imagery.
 Real-time GIS: Live traffic updates, disaster monitoring.
 3D/4D GIS: Time-enabled spatial analysis and
visualization.
 Cloud GIS: Cost-effective and scalable data processing.
Technological, Institutional, and Legal Challenges
Technological Challenges
Institutional Challenges
 Limited Infrastructure & Hardware Capacity
 Many developing countries lack high-performance
 Weak Institutional Coordination
computing infrastructure, stable internet connectivity,
 Multiple ministries/agencies collect spatial data,
and reliable power supply.
but with little harmonization.
 This limits the ability to run advanced GIS applications
 Leads to duplication, conflicting datasets, and
(real-time GIS, big data analytics, 3D/4D).
“data silos.”
 Data Availability and Quality
 Capacity and Human Resources
 Spatial datasets are often incomplete, outdated, or
 Shortage of trained GIS professionals to
inconsistent.
manage, analyze, and maintain systems.
 Lack of standardized formats leads to duplication and
 Limited investment in training and retention of
poor interoperability.
skilled personnel.
 Critical sectors (land tenure, health, environment) may
 Sustainability & Maintenance
suffer from data scarcity.
 Many GIS projects are donor-driven and
 High Costs of Technology & Licensing
collapse after initial funding ends.
 Proprietary GIS software (e.g., ArcGIS) is costly, making
 Lack of long-term institutional strategy and
adoption difficult.
budgetary support.
 Even when open-source options exist, they often
require skilled personnel that are scarce.
 Technological, Institutional, and Legal Challenges….Cont.
Legal and Policy Challenges
Critical Examination of these
 Lack of Data-Sharing Policies and Standards Challenges
 Spatial data often treated as “institutional property,”
limiting access.  These challenges are interconnected:
 No clear frameworks for interoperability (e.g., weak or o Weak technological gaps = worse
absent SDI policies). institutional inefficiencies
 Data Privacy and Security Concerns o Weak legal frameworks = weak
 Absence of legal safeguards for geospatial data use technological advancements + institutional
(e.g., location privacy, health or land records). inefficiencies.
 Risk of misuse, especially in politically sensitive  For example: A government may have advanced
contexts. GIS tools (tech) but without data-sharing laws
 Intellectual Property and Licensing Issues (legal) and trained staff (institutional), the
 Unclear copyright/licensing rules for datasets (e.g., can system fails to deliver benefits.
local agencies use satellite data freely?).  In developing countries, sustainability is the
 Dependence on foreign vendors raises concerns about biggest issue: projects thrive under donor
data sovereignty. support but collapse without local ownership
and legal/institutional backing.