Introduction to Geographic Information Systems
Open in the Padhaiko app1.1 GIS: definition, components and functions
A Geographic Information System (GIS) is a computer-based system designed to capture, store, manipulate, analyze, manage, and present geographically referenced (spatial) data along with its descriptive attributes to solve real-world spatial problems. Unlike ordinary information systems that handle only tabular records or documents, every feature in a GIS is tied to a location on or near the Earth's surface through coordinates, addresses, or administrative zones.
Formal Definition of GIS
A Geographic Information System (GIS) is an organized integration of hardware, software, geographic data, people, and methods designed to efficiently capture, store, update, manipulate, analyze, and display all forms of geographically referenced information.
Five Components of a GIS
A working GIS operates through five tightly coupled components:
- Hardware: The physical computing infrastructure on which the GIS runs and interacts with the physical world.
- Processing & Storage: Workstations, laptops, high-capacity enterprise servers, cloud compute instances, and spatial storage arrays.
- Data Capture Peripherals: GNSS receivers, total stations, drones (UAVs), flatbed/drum scanners, and digitizing tablets.
- Output Devices: High-resolution monitors, large-format plotters, and printers.
- Software: The programs and libraries that provide tools to store, query, analyze, and visualize geographic data.
- Desktop GIS: QGIS, ArcGIS Pro, GRASS GIS.
- Spatial DBMS & Middleware: PostgreSQL/PostGIS, SpatiaLite, GeoServer.
- Geospatial Libraries & APIs: GDAL/OGR, PROJ, Shapely, GeoPandas, Leaflet, OpenLayers.
- Data: Often the most expensive and time-consuming component (typically accounting for – of project cost). It consists of two linked parts:
- Spatial (Geometric) Data: Coordinates, shapes, and topological relationships describing where features are.
- Attribute (Non-Spatial) Data: Descriptive tables describing what those features are (name, population, land-use code, elevation).
- People: The human expertise that designs, operates, and consumes GIS outputs—ranging from GIS analysts, cartographers, and database administrators to software developers and decision-makers.
- Methods (Procedures): The well-defined organizational workflows, data standards, quality-assurance rules, projection specifications, and analytical models that ensure repeatable, reliable results across an institution.
Core Functions (Subsystems) of a GIS
Every GIS implements four functional subsystems that form a continuous workflow from raw observation to decision support:
| Subsystem / Function | Key Operations | Typical Outputs |
|---|---|---|
| Data Input (Capture & Preprocessing) | Digitizing, scanning, GNSS logging, remote sensing import, georeferencing, format conversion, error cleaning | Clean, georeferenced spatial and attribute datasets |
| Data Storage & Management | Organizing layers in files or spatial DBMS, spatial indexing (R-tree, quadtree), topology maintenance, metadata management | Structured geodatabase supporting fast retrieval and multi-user edits |
| Data Manipulation & Analysis | Projection transformation, buffering, overlay, network routing, raster map algebra, spatial interpolation, spatial queries | Derived thematic layers, suitability maps, summary metrics |
| Data Output & Presentation | Thematic cartography, 3D scenes, web map tiles, summary reports, tables, interactive dashboards | Paper maps, GeoTIFFs, web services (WMS/WFS), decision briefs |
1.2 GIS, GI Science and geospatial technologies
While GIS refers to the software systems and operational tools used to work with spatial data, Geographic Information Science (GIScience) is the academic discipline that investigates the fundamental theoretical and computational principles behind those systems—such as spatial cognition, uncertainty, map projections, spatial statistics, and spatial data structures. Geospatial technology is the broader umbrella term covering all modern tools used to acquire, measure, analyze, and visualize Earth-referenced information, including GIS, remote sensing, GNSS, and mobile location sensors.
| Aspect | GIS (System) | GI Science (Science) |
|---|---|---|
| Primary Focus | Practical software, tools, workflows, and applications | Fundamental principles, theories, and algorithms of spatial information |
| Core Question | How do we use tools to map and solve this spatial problem? | What are the mathematical and cognitive foundations of spatial representation? |
| Outputs | Maps, spatial databases, suitability analyses, web map apps | Data models, error models, indexing algorithms, spatial ontologies |
Relationship with Allied Disciplines and Technologies
GIS sits at the intersection of several mature disciplines and acts as an integrator across them:
- Cartography: Provides the visual grammar, generalization rules, map projection theory, and symbolization principles that govern how GIS outputs communicate spatial patterns clearly. Whereas traditional cartography produces static maps, GIS stores dynamic models from which many different maps can be generated on demand.
- Remote Sensing (RS): Captures Earth observation imagery from satellites, aircraft, and UAVs across multiple spectral bands. Remote sensing acts as a primary data-acquisition engine for GIS (supplying land-cover maps, elevation models, and change-detection layers), while GIS provides ground-truth vector overlays and analytical context.
- Global Navigation Satellite Systems (GNSS): Constellations such as GPS, GLONASS, Galileo, and BeiDou provide accurate 3D coordinates and precise time anywhere on Earth. GNSS supplies real-time positioning for field data collection, surveying, and location-based services inside a GIS.
- Computer-Aided Design (CAD): Focuses on high-precision engineering drafting and 3D geometric modeling of built structures (buildings, bridges, circuits) usually in local Cartesian coordinates. GIS differs by using real-world Earth-referenced coordinate systems, maintaining topological connectivity across large geographic regions, and linking rich relational attribute tables to every geometry.
- Database Management Systems (DBMS): Conventional DBMS (e.g., PostgreSQL, Oracle) provide ACID transactions, concurrency control, indexing, and SQL querying for tabular data. A GIS extends a DBMS into a Spatial DBMS by adding geometric data types (
Point,LineString,Polygon), spatial indexes (R-tree), and geometric predicates (ST_Intersects,ST_Distance).
| Criterion | GIS | CAD | Conventional DBMS |
|---|---|---|---|
| Coordinate Reference | Real-world geographic or projected CRS | Arbitrary local drawing units | No inherent spatial coordinate system |
| Primary Purpose | Spatial analysis, modeling, thematic mapping | Precision architectural/engineering drafting | Transactional storage and retrieval of tabular data |
| Attribute Linkage | Tight one-to-one or one-to-many link per feature | Minimal text labels or layer tags | Rich relational tables without geometry |
| Topology & Scale | Supports multi-scale topology over entire regions | Focuses on exact local dimensions of an object | Not applicable |
1.3 Historical development and evolution
GIS evolved alongside computer hardware, database theory, and internet connectivity through five overlapping architectural generations:
- Pioneering Era and Mainframe/Desktop GIS (1960s–1980s):
- In the early 1960s, Roger Tomlinson developed the Canada Geographic Information System (CGIS) to inventory land capability, coining the term "GIS." Around the same time, the Harvard Laboratory for Computer Graphics created SYMAP and ODYSSEY, pioneering vector overlay and topological data structures.
- During the 1980s, affordable workstations and personal computers enabled standalone Desktop GIS (such as early ARC/INFO and GRASS). Processing, storage, and display all resided on a single machine using local files (e.g., Coverages, later Shapefiles).
- Enterprise GIS (1990s–2000s):
- Organizations needed multiple departments to edit and query the same authoritative datasets simultaneously without file corruption.
- Spatial extensions to relational databases (Oracle Spatial, ArcSDE, PostGIS) centralized vector and raster layers on dedicated database servers. Desktop clients connected over a Local Area Network (LAN), introducing multi-user versioning, role-based security, and standardized SQL access.
- Web GIS (Late 1990s–2000s):
- Moving GIS onto HTTP allowed any user with a web browser to view interactive maps without installing specialized desktop software.
- Open Geospatial Consortium (OGC) standards—such as WMS (Web Map Service) and WFS (Web Feature Service)—together with tiled slippy maps (introduced widely in the mid-2000s) turned maps into embeddable web services.
- Mobile GIS (2000s–Present):
- Smartphones and ruggedized tablets integrated GNSS receivers, cameras, inertial sensors, and cellular networks into portable terminals.
- Mobile GIS enabled real-time field inspections, offline-first map caching, crowdsourced mapping (OpenStreetMap), and consumer Location-Based Services (LBS) such as turn-by-turn navigation and ride-hailing.
- Cloud GIS (2010s–Present):
- Geospatial workloads migrated to elastic cloud infrastructure, combining object storage (S3-compatible buckets), Cloud Optimized GeoTIFFs (COG), distributed spatial engines (Apache Sedona), and serverless APIs.
- Platforms like Google Earth Engine allow petabyte-scale planetary analysis directly in the browser without downloading terabytes of satellite scenes locally.
| Era | Compute & Storage Location | Primary User Interface | Key Advantage | Limitation |
|---|---|---|---|---|
| Desktop GIS | Single local PC and local files | Rich desktop GUI (QGIS, ArcGIS) | Deep analytical toolset; works fully offline | Difficult multi-user collaboration and sharing |
| Enterprise GIS | Central spatial DBMS server on LAN | Desktop clients + internal portals | Multi-user concurrency, ACID integrity, centralized security | High licensing/IT overhead; bound to internal network |
| Web GIS | Application & map servers over HTTP | Web browser (Leaflet, OpenLayers) | Zero client installation; broad public dissemination | Historically limited complex editing/analysis in browser |
| Mobile GIS | Handheld device synced with cloud/server | Native mobile app with GNSS | Real-time field capture and location-aware services | Battery, screen size, and intermittent connectivity constraints |
| Cloud GIS | Elastic distributed cloud clusters & object store | Browser IDE, REST/OGC APIs, notebooks | Petabyte scalability, on-demand compute, instant global access | Requires internet connectivity; recurring cloud usage costs |
1.4 Spatial and attribute data; spatial vs. conventional queries
Every geographic entity represented in a GIS consists of two complementary components stored together: spatial data and attribute data.
Spatial Data vs. Attribute Data
- Spatial Data (Geometric / Geographic Data): Data that describes the location, shape, size, and spatial relationships of geographic features with respect to a coordinate reference system. Examples: for a hospital point, a polyline representing the Araniko Highway, or a polygon bounding Kathmandu Metropolitan City.
- Attribute Data (Non-Spatial / Asptial Data): Descriptive tabular information attached to each spatial feature via a unique Feature ID (
FID) or primary key. Examples: hospital name, bed capacity, road surface type, highway code, or ward population.
Spatial Queries vs. Conventional Database Queries
A conventional database query filters or aggregates records purely on attribute values using relational algebra (SELECT, WHERE, JOIN, GROUP BY). A spatial query evaluates geometric properties (area, length, distance) or topological relationships (contains, intersects, touches, within) on feature geometries, often combined with attribute predicates.
| Aspect | Conventional Database Query | Spatial Query |
|---|---|---|
| Filter Basis | Non-spatial column values (numbers, strings, dates) | Coordinates, geometry shapes, distances, and topological relations |
| Typical Operators | =, <, >, BETWEEN, LIKE, equi-join | ST_Contains, ST_Intersects, ST_DWithin, ST_Area, spatial join |
| Indexing Mechanism | B-tree, B+ tree, hash index (1D sorted keys) | R-tree, Quadtree, Grid index (2D/3D bounding boxes) |
| Example Question | "List all hospitals in Bagmati Province with beds." | "Find all hospitals within of the Ring Road that lie inside Ward 10." |
1
Suppose a municipal GIS stores school point features in a projected coordinate system (meters) as schools(id, name, students, geom) and ward polygons as wards(ward_no, geom).
- Distinguish between the spatial query and the attribute query needed to answer: "Find all schools with at least students located within of a fire station at ."
- Check numerically whether school at with students and school at with students satisfy both conditions.
Separating the attribute and spatial predicates:
- Attribute predicate (non-spatial):
students >= 500. This compares a numeric column in the attribute table without touching geometry. - Spatial predicate: Euclidean distance from to school point must satisfy (). In PostGIS SQL, this uses
ST_DWithin(geom, ST_MakePoint(1000, 2000), 5000).
Evaluating the distance for :
Since and , school satisfies both predicates.
Evaluating the distance for :
Although meets the attribute condition (), , so fails the spatial condition.
Only school satisfies both the attribute predicate (students >= 500) and the spatial distance predicate (). School fails the spatial predicate ().
1.5 Applications of GIS
Because almost every human and environmental phenomenon has a geographic footprint, GIS serves as a core decision-support platform across engineering, governance, and industry:
| Domain | Core Spatial Layers Used | Typical GIS Analyses & Decisions |
|---|---|---|
| Urban Planning | Cadastral parcels, zoning polygons, land use, building footprints, slope/DEM | Land-suitability analysis, zoning compliance, urban sprawl monitoring, transit-oriented development |
| Disaster Risk Management | Floodplains, fault lines, DEM, landslide inventories, population density, shelters | Hazard zonation, vulnerability mapping, evacuation route modeling, post-disaster damage assessment |
| Natural Resources | Forest cover, watersheds, soil types, mineral deposits, protected areas | Deforestation tracking, watershed delineation, timber volume estimation, wildlife corridor design |
| Transportation | Road networks (with speed/turn attributes), bus stops, traffic sensors | Shortest-path routing, origin–destination matrices, public transit accessibility, black-spot crash analysis |
| Utilities (Water, Power, Telecom) | Pipe/cable networks, transformers, valves, customer meters | Network trace (upstream/downstream isolation), outage impact mapping, asset maintenance scheduling |
| Public Health | Disease case locations, clinics, water sources, socioeconomic census tracts | Disease cluster/hot-spot detection, hospital catchment analysis, vector-borne outbreak modeling |
| Agriculture (Precision Farming) | Multispectral crop indices (NDVI), soil moisture grids, parcel boundaries | Variable-rate fertilizer/irrigation application, yield prediction, drought stress monitoring |
| Location-Based Services (LBS) | Real-time GNSS trajectories, Points of Interest (POIs), live traffic feeds | Geofencing alerts, nearest-facility search, ride-hailing dispatch, last-mile delivery optimization |
2
Exam-style question: Explain how a GIS integrates its five components and four functional subsystems in an urban flood risk management project.
Link the five components to the project:
- Hardware: GNSS rovers to survey flood marks, UAVs for high-resolution terrain capture, and a GIS workstation/server.
- Software: Desktop GIS (QGIS) for hydrological modeling and PostGIS/GeoServer for storing and publishing flood maps.
- Data: Spatial layers (DEM, river centerlines, building footprints, land-use polygons) linked to attribute tables (channel depth, building occupancy, floor count).
- People: Hydrologists, GIS analysts, municipal engineers, and emergency planners.
- Methods: Standard coordinate system (e.g., MUTM for Nepal), DEM hydro-conditioning procedures, and return-period flood depth classification rules.
Trace the four functional subsystems:
- Data Input: Collect UAV elevation points and historical flood-extent polygons; georeference and clean topology.
- Storage & Management: Store all layers in a spatial database with R-tree indexes and ISO metadata.
- Manipulation & Analysis: Overlay modeled -year flood polygons with residential building footprints (
ST_Intersects) and compute affected population attributes and shortest evacuation routes. - Output & Presentation: Publish color-coded flood hazard maps and an interactive web dashboard for emergency responders.
An urban flood risk GIS combines hardware (GNSS, UAVs, servers), software (QGIS, PostGIS, GeoServer), spatial/attribute data (DEM, rivers, buildings), skilled people, and standardized workflows across four subsystems—input, database management, spatial overlay/network analysis, and hazard map output—to identify exposed buildings and plan evacuation routes.
added
Define a Geographic Information System (GIS). List and briefly explain its five components and four functional subsystems.
Definition: A Geographic Information System (GIS) is an integrated computer-based system of hardware, software, spatial and attribute data, people, and methods designed to capture, store, manage, analyze, and visualize geographically referenced information.
Five components:
- Hardware: Computers, servers, scanners, GNSS receivers, and plotters used to acquire, process, and output data.
- Software: Desktop tools (QGIS, ArcGIS), spatial DBMS (PostGIS), and libraries (GDAL) providing spatial processing functions.
- Data: Spatial geometry (where features are) and non-spatial attributes (what features are).
- People: Analysts, database administrators, developers, and decision-makers.
- Methods: Institutional standards, quality rules, and analytical workflows.
Four functional subsystems:
- Data Input: Capturing, digitizing, scanning, and preprocessing spatial and attribute data.
- Data Storage & Management: Organizing geodatabases, spatial indexes, and metadata.
- Data Manipulation & Analysis: Performing spatial queries, overlays, buffering, and modeling.
- Data Output & Presentation: Producing maps, charts, reports, and web services.
A GIS is a computer-based system for capturing, storing, analyzing, and displaying geographically referenced data. Its five components are hardware, software, data, people, and methods, and its four subsystems are data input, storage/management, manipulation/analysis, and output/presentation.
Differentiate between GIS and GI Science, and explain how GIS relates to Remote Sensing, GNSS, CAD, and DBMS.
GIS vs. GI Science: GIS refers to the operational software tools and infrastructure used to solve spatial problems (e.g., creating a flood map in QGIS), whereas GI Science (GIScience) is the foundational science behind those tools—studying spatial data models, coordinate mathematics, uncertainty, and spatial algorithms.
Relationships with allied technologies:
- Remote Sensing: Supplies multispectral satellite/UAV imagery as primary raster input to GIS; GIS provides vector ground-truth context and multi-layer analysis.
- GNSS: Supplies real-time 3D coordinates for field surveys, asset mapping, and navigation inside GIS.
- CAD: Specializes in local-coordinate engineering drafting of structures; GIS differs by using Earth-referenced coordinate systems, maintaining regional topology, and linking rich relational attribute tables.
- DBMS: Provides ACID storage, indexing, and SQL querying; GIS extends DBMS with geometric data types, spatial indexes, and spatial predicates.
GIS is the practical software/toolset, while GI Science studies the underlying theories and algorithms of spatial information. GIS integrates Remote Sensing (imagery capture), GNSS (precise Earth positioning), and DBMS (structured storage and querying), while extending CAD with Earth-referenced coordinate systems, topology, and rich attribute databases.
Trace the historical evolution of GIS from desktop systems to enterprise, web, mobile, and cloud GIS, highlighting one key capability added at each stage.
Chronological progression:
- Desktop GIS (1960s–1980s): Began with pioneering systems like CGIS and SYMAP and matured on PCs; enabled standalone local file editing, overlay, and cartographic output on a single workstation.
- Enterprise GIS (1990s–2000s): Introduced spatial extensions to relational databases (PostGIS, Oracle Spatial, ArcSDE) so multiple users on a LAN could perform concurrent, transaction-safe editing on a centralized repository.
- Web GIS (Late 1990s–2000s): Brought interactive mapping to standard web browsers using HTTP and OGC services (WMS, WFS) plus tiled web maps, removing the need for desktop GIS installation by end users.
- Mobile GIS (2000s–Present): Integrated GNSS, sensors, and cellular connectivity on handheld devices for real-time field collection and consumer location-based services (LBS).
- Cloud GIS (2010s–Present): Leveraged elastic cloud compute, object storage, and platforms like Google Earth Engine for planetary-scale processing without local storage bottlenecks.
GIS progressed from Desktop GIS (single-user local analysis) Enterprise GIS (multi-user centralized spatial DBMS) Web GIS (browser-based maps via HTTP/OGC standards) Mobile GIS (GNSS-enabled field capture and LBS) Cloud GIS (elastic distributed processing over petabyte-scale geospatial archives).
A land-registry GIS stores rectangular agricultural parcels in projected meters. Parcel spans , and has attribute owner = "A. Gurung", crop = "Wheat".
- Identify which facts about are spatial data and which are attribute data.
- Compute the area of in hectares () and explain why finding "all wheat parcels larger than " is a hybrid spatial–attribute query.
Classifying spatial vs. attribute data:
- Spatial data: The bounding coordinates , defining the rectangular polygon geometry and its position in the projected CRS.
- Attribute data: The descriptive columns
owner = "A. Gurung"andcrop = "Wheat".
Computing the area of :
Why the query is hybrid: Filtering by crop = 'Wheat' is a conventional attribute predicate, whereas computing polygon area (ST_Area(geom) > 1000000) evaluates the spatial geometry. Combining both (WHERE crop = 'Wheat' AND ST_Area(geom) > 1000000) forms a hybrid spatial–attribute query, which satisfies ().
The coordinate bounds form the spatial data, while owner and crop are attribute data. Parcel has area and satisfies the hybrid query because crop = 'Wheat' (attribute filter) and ST_Area(geom) > 100 ha (spatial filter) both hold.
Select any three GIS application domains (e.g., disaster risk management, transportation, precision agriculture) and describe one specific spatial analysis task performed in each.
Disaster Risk Management: Overlaying a -year flood depth polygon layer with residential building footprints and population census blocks (ST_Intersects) to quantify exposed households and plan emergency shelter capacities.
Transportation: Building a topologically connected road network graph with segment lengths, speed limits, and turn restrictions to compute shortest-time ambulance routes (Dijkstra/) and -minute hospital service areas.
Precision Agriculture: Deriving Normalized Difference Vegetation Index () raster surfaces from multispectral satellite or UAV imagery and applying zonal statistics per farm management zone to schedule variable-rate fertilizer and irrigation.
- Disaster Risk Management: Flood hazard polygon overlay with building footprints to quantify exposed population.
- Transportation: Network shortest-path and drive-time service-area analysis for emergency routing.
- Precision Agriculture: Multispectral raster mapping and zonal statistics for variable-rate input management.