Lesson 2: Spatial Databases (PostgreSQL & PostGIS)

Traditional relational databases store text and numeric information, but they cannot efficiently manage geographic objects. PostGIS extends PostgreSQL by introducing spatial data types such as Point, LineString, and Polygon, together with powerful spatial indexes and functions.

It allows GIS professionals to perform buffer analysis, nearest-neighbour searches, overlays, routing, and spatial intersections directly using SQL, making it one of the most widely used open-source spatial database systems.

Advanced Spatial SQL Query Example

The following SQL example creates a spatial table and retrieves weather stations located within a 50 km buffer around a hazard zone.

-- Step 1 : Create Spatial Table

CREATE TABLE weather_stations (

    id SERIAL PRIMARY KEY,

    name VARCHAR(100),

    geom GEOMETRY(Point,4326)

);

----------------------------------------------------

-- Step 2 : Spatial Query

SELECT

    ws.id,

    ws.name,

    ST_Distance(

        ST_Transform(ws.geom,32644),

        ST_Transform(h.geom,32644)

    ) / 1000 AS distance_km

FROM weather_stations ws,

     hazard_zones h

WHERE

      h.name='Bay of Bengal Cyclone Path'

AND

ST_DWithin(

ST_Transform(ws.geom,32644),

ST_Transform(h.geom,32644),

50000

);

Why PostGIS is Important

• Supports millions of geographic features efficiently.
• Spatial indexing using GiST and R-Tree.
• Fast proximity and buffer searches.
• Network and routing analysis.
• Compatible with QGIS, ArcGIS and GeoServer.
• Perfect for enterprise GIS applications.

Lesson 3: Spatial Interpolation

Spatial interpolation is a geostatistical technique used to estimate unknown values at locations where measurements are unavailable. It assumes that nearby observations are more similar than distant ones, allowing continuous surfaces to be created from discrete sample points.

Interpolation is widely used in environmental monitoring, rainfall mapping, groundwater assessment, soil analysis, pollution studies, temperature modelling and agricultural planning.

Common Interpolation Methods

• Inverse Distance Weighting (IDW): Nearby observations have greater influence than distant observations.
• Kriging: A statistical interpolation method that models spatial autocorrelation and generally provides the highest accuracy.
• Spline: Produces a smooth continuous surface suitable for elevation and climatic variables.
• Natural Neighbor: Uses surrounding Thiessen polygons to estimate values while preserving local variation.
• Trend Surface: Fits mathematical polynomial equations for large regional trends.

Inverse Distance Weighting (IDW)

IDW assumes that sample points closer to an unknown location have greater influence than those farther away. It is simple, fast and commonly used for rainfall, groundwater level and air quality mapping.

Where:

• Z(x) = Estimated value
• Zi = Sample value
• Di = Distance from sample point
• p = Power parameter (usually 2)

Comparison of Interpolation Methods

Typical Applications

• Rainfall distribution mapping
• Groundwater level estimation
• Temperature and climate modelling
• Air pollution analysis
• Agricultural productivity mapping
• Soil nutrient prediction
• Environmental risk assessment

Lesson 4: Python & R Scripting for GIS

Python and R are two of the most widely used programming languages in modern Geographic Information Systems (GIS). They enable professionals to automate repetitive workflows, analyse large spatial datasets, perform statistical modelling, and develop advanced geospatial applications.

Python has become the industry standard for GIS automation through libraries such as GeoPandas, Rasterio, GDAL, Shapely, and ArcPy. R is widely used for spatial statistics, environmental modelling, and scientific research.

Why Learn Python & R?

• Automate repetitive GIS workflows.
• Process thousands of satellite images.
• Perform advanced spatial statistics.
• Create reproducible research projects.
• Build AI and Machine Learning workflows.
• Integrate GIS with databases and web services.
• Generate maps automatically.

Python Example (GeoPandas)

import geopandas as gpd

# Read Shapefile
gdf = gpd.read_file("districts.shp")

# Calculate Area
gdf["Area_sqkm"] = gdf.area / 1000000

# Display first rows
print(gdf.head())

# Export
gdf.to_file("district_area.shp")

R Example (sf Package)

library(sf)

districts <- st_read("districts.shp")

districts$Area_sqkm <- st_area(districts)

plot(districts["Area_sqkm"])

st_write(
districts,
"district_area.shp"
)

Popular Python GIS Libraries

Applications

• GIS Workflow Automation
• Remote Sensing Analysis
• Land Use & Land Cover Mapping
• Hydrological Modelling
• GeoAI & Machine Learning
• WebGIS Development
• Environmental Research

Lesson 5: Google Earth Engine (GEE)

Google Earth Engine (GEE) is a cloud-based geospatial analysis platform developed by Google. It provides access to petabytes of satellite imagery and geospatial datasets while allowing large-scale processing directly in the cloud without downloading data to your computer.

Using JavaScript or Python APIs, researchers can analyse decades of satellite observations, monitor environmental changes, detect land cover changes, calculate vegetation indices, and build large-scale geospatial applications.

Why Use Google Earth Engine?

• Cloud-based satellite image processing.
• No need for high-end computer hardware.
• Access to Landsat, Sentinel, MODIS and many global datasets.
• Fast processing of multi-year imagery.
• Ideal for research, climate studies and agriculture.
• Supports both JavaScript and Python APIs.
• Easy export to Google Drive and GIS software.

Example: NDVI Calculation using Sentinel-2

// Load Sentinel-2 Image Collection

var image = ee.ImageCollection('COPERNICUS/S2_SR')

.filterDate('2024-01-01','2024-12-31')

.filterBounds(geometry)

.median();

// Calculate NDVI

var ndvi = image.normalizedDifference(

['B8','B4']

);

Map.centerObject(geometry,10);

Map.addLayer(

ndvi,

{

min:-1,

max:1,

palette:['blue','white','green']

},

'NDVI'

);

Major GEE Datasets

Typical Applications

• Land Use & Land Cover Classification
• NDVI & Vegetation Monitoring
• Flood Mapping
• Drought Assessment
• Forest Change Detection
• Urban Growth Analysis
• Climate Change Research
• Agricultural Monitoring