Quiz20

Geography notes

Notes · Geography

5. Remote Sensing, Aerial Photographs, GIS & GNSS

Short, exam-focused revision notes. Read a section, then test yourself with a quiz.

8 min read9 sections

5. Remote Sensing, Aerial Photographs, GIS & GNSS

🎯
Key points: definition and process of remote sensing; EMR, atmospheric windows, scattering; spectral signatures; platforms, orbits, sensors and resolutions; satellites (world and Indian); aerial photographs – types, geometry, scale, overlap, stereoscopy, interpretation elements; satellite imagery, FCC, digital image processing and classification; GIS – definition, components, data models, process, analysis, software and applications; GPS and NavIC.

1. Remote Sensing – Concept ⭐⭐⭐

  • Remote sensing = the science and art of obtaining information about objects or areas on the earth's surface without physical contact, by recording reflected or emitted electromagnetic energy (Lillesand and Kiefer). Term coined by Evelyn Pruitt (1960s).
  • Sensors may be on the ground, towers, cranes, vehicles, drones, aircraft, balloons or satellites (not only aircraft and satellites).
  • Two basic processes: data acquisition (energy source, atmosphere, platform, sensor, ground station) and data analysis (processing, interpretation, GIS integration). GIS is part of analysis, not acquisition.
  • In India, satellite data are received and distributed by the National Remote Sensing Centre (NRSC), Hyderabad (ground station at Shadnagar); Bhuvan is ISRO's geoportal; training at IIRS Dehradun. Policy: Remote Sensing Data Policy (2011), Indian Space Policy 2023.

2. Electromagnetic Radiation ⭐⭐⭐

The electromagnetic spectrum: wavelength increases from gamma rays to radio waves.
The electromagnetic spectrum: wavelength increases from gamma rays to radio waves.
Region
Wavelength
Use
Ultraviolet
< 0.4 µm
Mostly absorbed by ozone; UV-B causes skin cancer
Visible
0.4–0.7 µm – blue 0.4–0.5, green 0.5–0.6, red 0.6–0.7
Blue: water penetration, coastal mapping; green: vegetation discrimination (green peak); red: chlorophyll absorption, soil vs vegetation
Near / reflected infrared
0.7–3.0 µm (NIR 0.7–1.3)
Vegetation vigour and biomass (high NIR reflectance), water body boundaries (water absorbs NIR)
Thermal infrared
3–14 µm (window 8–14 µm)
Emitted heat – surface temperature, fires, volcanoes, urban heat islands; sensed by radiometers/scanners, not photographic film
Microwave
1 mm – 1 m
Radar – penetrates clouds, works day and night, partly penetrates vegetation and dry soil
  • Order of increasing wavelength: gamma → X-ray → UV → violet/blue → green → red → NIR → thermal IR → microwave → radio.
  • Atmospheric windows = wavelength ranges that the atmosphere transmits well (visible, parts of NIR, 3–5 µm, 8–14 µm, microwave).
  • Scattering: Rayleigh (by gas molecules much smaller than the wavelength; affects short wavelengths – blue sky, red sunsets); Mie (dust, smoke, water vapour comparable to wavelength – haze); non-selective (large water droplets – white clouds, fog). Clouds are the biggest problem for optical sensors.
  • Spectral signature: characteristic reflectance curve of a feature – healthy vegetation absorbs blue and red, reflects green and strongly reflects NIR; water absorbs NIR (appears dark); dry soil reflectance rises steadily with wavelength. In thermal IR, water has the highest emissivity/lowest reflectance.
  • Additive primary colours of light: red, green, blue (RGB).

3. Platforms, Orbits, Sensors and Resolution ⭐⭐⭐

Orbit
Height
Features
Examples
Geostationary (geosynchronous)
about 35,786 km above the equator
Period = 24 h; views the same area continuously; coarse resolution
Weather and communication: INSAT, Kalpana-1, METEOSAT (Europe), GOES (USA), GOMS (Russia), GSAT
Sun-synchronous (near-polar)
about 600–900 km
Passes over each place at the same local solar time; global coverage; fine resolution
Landsat (705 km), IRS (IRS-1A/1B 904 km), SPOT (832 km), IKONOS (681 km), Sentinel
Sensor type
Meaning
Examples
Passive
Record naturally available energy – reflected sunlight or emitted heat
Cameras, multispectral scanners (MSS, TM, LISS), thermal radiometers
Active
Send their own energy and record the return
Radar (SAR – RISAT, Sentinel-1, NISAR), LiDAR (laser), radar altimeters, sonar
Scanning systems
Whiskbroom (across-track, rotating mirror – Landsat MSS/TM) vs pushbroom (along-track linear array of detectors – SPOT, IRS LISS)
ㅤ
Resolution
Meaning
Spatial
Smallest ground area/object distinguished – pixel size (e.g., 23.5 m, 5.8 m, 1 m)
Spectral
Number and width of wavelength bands (e.g., a red band); panchromatic < multispectral < hyperspectral
Radiometric
Sensitivity to differences in energy – number of grey levels (bits: 8-bit = 256 levels)
Temporal
Revisit frequency (e.g., 16 days Landsat, 24/5 days IRS)

Satellites – world and India ⭐⭐⭐

Satellite
Agency / year
Notes
TIROS-1 (1960), NOAA series
USA
Earliest weather satellites; NOAA AVHRR
Landsat-1 (ERTS-1)
NASA/USGS, 1972
First earth resources satellite; MSS bands 4–7 (green, red, two NIR); later TM (30 m), ETM+, OLI; Landsat-9 (2021)
SPOT-1
France (CNES), 1986
Pushbroom HRV, stereo
IRS-1A
India, 1988
LISS-I and LISS-II; IRS-1C (1995)
ERS-1, ENVISAT (2002), Sentinel (Copernicus)
Europe (ESA)
Radar and optical
IKONOS
USA (Space Imaging → GeoEye → DigitalGlobe/Maxar), 1999
First commercial high-resolution: 1 m PAN, 4 m MS
QuickBird, WorldView, GeoEye, OrbView
USA commercial
Sub-metre
Terra/Aqua MODIS
NASA
36 bands; daily global coverage
Indian EO milestones
Year
Aryabhata (first Indian satellite)
1975
Bhaskara-I (first Indian experimental earth observation satellite)
1979
IRS-1A
1988
IRS-1C (with PAN 5.8 m)
1995
Kalpana-1 (weather, geostationary)
2002
EDUSAT
2004
Cartosat-1 (stereo PAN 2.5 m, DEM)
2005
Oceansat-2
2009
Resourcesat-2/2A (LISS-III, LISS-IV, AWiFS)
2011 / 2016
Cartosat-2 series (< 1 m), Cartosat-3 (0.25 m)
2007–2019
RISAT (radar), EOS-04
2009–2022
NISAR (NASA–ISRO dual-frequency SAR)
Launched July 2025
  • Indian sensor resolutions: LISS-III 23.5 m, LISS-IV 5.8 m, AWiFS about 56 m, LISS-II 36.25 m, Cartosat-2 PAN about 0.65–1 m. Landsat RBV about 80 m, TM 30 m.
  • Indian satellite system components: INSAT/GSAT (communication and meteorology), Earth Observation (IRS, Cartosat, Resourcesat, Oceansat, RISAT, EOS), navigation (IRNSS/NavIC).
  • DEMs: SRTM (30 m / 90 m, 2000), ASTER GDEM (30 m), ALOS PALSAR (12.5 m), CartoDEM (from Cartosat-1, about 10–30 m).

4. Aerial Photographs ⭐⭐⭐

Type
Camera axis
Features
Vertical
Within 3° of vertical
Most used for mapping; scale nearly uniform on flat ground; stereo pairs
Low oblique
Tilted, horizon not visible
Trapezoidal ground coverage
High oblique
Tilted, horizon visible
Panoramic view
Terrestrial (horizontal)
Camera on the ground, horizontal axis
Close-range photogrammetry
Trimetrogon
One vertical and two high obliques simultaneously
Wide coverage (used in World War II)
Convergent
Two low obliques looking toward each other
ㅤ
Sonne's continuous strip
Film moves continuously past a fixed slit
Strip photography of routes
Composite
Several cameras combined
ㅤ
  • Film/emulsion types: panchromatic (B&W visible), B&W infrared, natural colour, colour infrared (false colour) – vegetation appears red; best for crop disease, plant stress, water pollution, soil moisture. Photographs are originally analogue (film); images are digital.
  • Best time for photography: around midday (sun moderately high, about 10 am–2 pm) – enough light and moderate shadows.

Geometry and measurements

Concept
Formula / fact
Scale of vertical photo
S = f / (H − h) (f = focal length, H = flying height above datum, h = ground elevation) = photo distance / ground distance. Scale varies with relief and tilt
Scale by comparison
Photo distance 10 mm, map distance 50 mm at 1:50,000 → ground 2.5 km → photo scale 1:250,000. Road 2 inches on photo, 60 ft on ground (720 inches) → 1:360
Relief displacement
d = h r / H – tall objects lean outward from the principal point
Height from parallax
h = (H − h₀) × dp / (P + dp)
Overlap
Forward (end) overlap about 60%; side (lateral) overlap about 20–30% – needed for stereoscopic 3-D viewing
Points
Principal point (centre, found from fiducial marks), nadir (point vertically below the camera), isocentre
Products
Photo mosaic (controlled / uncontrolled), orthophoto (corrected for relief displacement and tilt), photo index
Instruments
Stereoscope (lens, mirror), parallax bar, stereoplotter; photogrammetry = science of measurements from photographs

Elements of image interpretation ⭐⭐⭐

Element
Meaning
Example
Tone / colour
Brightness or hue
Water dark, sand bright; land cover and land use
Size
Relative dimensions
Distinguish houses from factories; zones of land use
Shape
Outline / form
Meanders, alluvial fans, stadium, railway curves
Texture
Frequency of tonal change (smooth or rough)
Grass smooth, forest rough
Pattern
Spatial arrangement
Orchards (regular rows), drainage patterns, planned towns
Shadow
Silhouette
Height and profile of tall objects – Qutub Minar, towers
Site / association
Location and relation to other features
Thermal plant near coal and water; ferry ghat by river
Height, location
ㅤ
ㅤ
  • Hierarchy (Estes): primary – tone/colour; secondary – size, shape, texture; tertiary – pattern, height, shadow; higher – site, association (some keys place site, situation and association at the secondary order).
  • Errors: dark tree shadows can be misclassified as water (shadows look darker and bluer).

5. Satellite Imagery and Digital Image Processing ⭐⭐

  • A satellite image is a raster of pixels; each pixel stores a digital number (DN) for each band.
  • False colour composite (FCC): standard FCC assigns NIR → red, red → green, green → blue, so healthy vegetation appears red, water dark blue/black, urban areas cyan/grey. Landsat MSS standard FCC uses bands 4, 5 and 7 (band 6 usually omitted); MSS bands 5 (red) and 7 (NIR) together show vegetation limits clearly.
  • Steps of digital image processing:
      1. Pre-processing (image restoration) – radiometric correction (sensor and atmospheric errors, noise removal), geometric correction (georeferencing, registration using ground control points).
      1. Enhancement – contrast stretching (linear: min–max, saturation/percentage stretch; non-linear: histogram equalisation), density slicing, spatial filtering (low-pass smoothing, high-pass/edge enhancement), multi-image manipulation.
      1. Transformation – band ratios, vegetation indices: NDVI = (NIR − Red)/(NIR + Red) (−1 to +1; dense vegetation high), principal components.
      1. Classification – supervised (analyst gives training sites: maximum likelihood, minimum distance, parallelepiped) and unsupervised (algorithm groups pixels into natural clusters: ISODATA, K-means, labelled later). Newer: object-based, machine learning.
      1. Accuracy assessment – error (confusion) matrix, overall accuracy, Kappa.
      1. Change detection – comparing multi-date images (land-use change, deforestation, urban growth).
  • Collection of data from photos and imagery: visual interpretation → ground truth verification → mapping (land use, geology, geomorphology, soils, forests, crops, water, urban), measurement of area, length, height, DEM generation.

6. Geographic Information System (GIS) ⭐⭐⭐

  • GIS = a computer-based system to capture, store, retrieve, analyse and display spatially referenced (geographic) data. First GIS: Canada Geographic Information System (1960s) by Roger Tomlinson ("father of GIS"); Harvard Laboratory for Computer Graphics; ESRI (1969).
  • Components (five): hardware (computer, digitiser, scanner, GPS, plotter), software, data (spatial + attribute), people (users, analysts), methods / procedures. (Landforms are not a component.)
  • Functional sub-systems (Marble): data input; data storage and retrieval; data manipulation and analysis (the heart of GIS); data output/reporting.

Data models ⭐⭐⭐

ㅤ
Raster
Vector
Representation
Grid of equal-sized cells (pixels) – tessellation; usually starts from the upper-left corner
Points (0-D), lines (1-D), polygons (2-D) by x,y coordinates
Advantages
Simple data structure; easy, efficient overlay; compatible with satellite images and DEMs; good for continuous data
Compact (less storage); high cartographic output quality; explicit topology; efficient for network analysis; accurate boundaries
Disadvantages
Large files; lower graphic quality; poor network analysis; resolution limits
Complex structure; overlays more difficult; less compatible with imagery; costly data capture
  • Data capture: digitising (vector) and scanning (raster) are data acquisition/input; also GPS, remote sensing, field survey, attribute tables, census data. Errors corrected by edge matching, topology building.
  • Georeferencing (assigning coordinates), projection definition, attribute data linked through IDs in tables (relational database).
  • Surfaces: DEM (raster grid of heights – all cells same size), TIN (triangulated irregular network – vector).

GIS analysis functions

Function
Meaning
Query
Select by attribute or location
Overlay
Combine layers (union, intersect) – site suitability
Buffering
Zone of a specified width around features (e.g., 500 m from a river)
Network analysis
Shortest route, service areas, stream flow in a basin
Aggregation / reclassification
Combine detailed categories into broader ones
Interpolation
Estimate values between points (IDW, kriging)
Terrain analysis
Slope, aspect, hillshade, watershed delineation
Intervisibility / viewshed
What can be seen from a point
Proximity, measurement, spatial statistics
Distances, areas, clustering

Software

  • GIS: ArcGIS / ArcView (ESRI), QGIS (open source), GRASS, MapInfo, GeoMedia, SuperMap; image processing: ERDAS Imagine, ENVI, IDRISI/TerrSet, eCognition; cloud: Google Earth Engine; web: Google Earth/Maps, Bhuvan. AutoCAD is CAD (drafting) software, not GIS. Excel's function wizard performs statistical operations.

Applications

Land-use/land-cover mapping, urban planning (master plans, Smart Cities), agriculture (crop acreage and yield – FASAL, crop insurance), forestry (forest cover – ISFR), water resources (watersheds, groundwater prospects), disaster management (flood, cyclone, landslide mapping), environment monitoring, mineral exploration, transport and routing, utilities, health (disease mapping), elections and census mapping, cadastral and property mapping (SVAMITVA drone surveys), PM Gati Shakti master plan, military, archaeology.

7. GNSS – Satellite Navigation

System
Country
Notes
GPS
USA
First satellite 1978; 24+ satellites in 6 orbital planes (about 20,200 km)
GLONASS
Russia
From 1982
BeiDou
China
From 2000; global (BDS-3) 2020
Galileo
European Union
From 2011 (civilian)
QZSS
Japan
Regional
NavIC (IRNSS)
India
Regional; 7 satellites (3 geostationary + 4 geosynchronous); covers India and about 1,500 km around; NVS series replacing old satellites
  • Chronological order: GPS → GLONASS → BeiDou → Galileo. GAGAN = India's satellite-based augmentation system (aviation). A GPS receiver needs signals from at least 4 satellites for a 3-D fix (trilateration).

⚡ Quick Revision

  • RS process: source → atmosphere → interaction → atmosphere → sensor → processing → users.
  • Visible 0.4–0.7 µm; thermal window 8–14 µm; microwave sees through clouds. Rayleigh → blue sky, red sunset.
  • Geostationary 35,786 km (INSAT, METEOSAT, GOMS); sun-synchronous (Landsat 705, SPOT 832, IRS 904, IKONOS 681 km).
  • Passive = natural energy; active = radar, LiDAR.
  • LISS-III 23.5 m; LISS-IV 5.8 m; AWiFS 56 m.
  • Landsat 1972 → SPOT 1986 → IRS-1A 1988 → IKONOS 1999.
  • Photo scale = f/(H − h); relief displacement = hr/H; 60% forward, 20–30% side overlap.
  • Shadow → tall objects; shape → meanders/fans; pattern → orchards.
  • Standard FCC: vegetation red. NDVI = (NIR − R)/(NIR + R). ISODATA/K-means = unsupervised.
  • GIS components: hardware, software, data, people, methods. Raster = simple, overlay, imagery; vector = compact, quality, network.
  • GPS 1978, GLONASS 1982, BeiDou 2000, Galileo 2011; NavIC = 7 satellites.

📝 Practice MCQs

  1. Which sensor records naturally available energy? (a) Active (b) Passive (c) SAR (d) LiDAR
Answer
(b) Passive
  1. On a standard false colour composite, healthy vegetation appears: (a) green (b) blue (c) red (d) black
Answer
(c) Red
  1. Which element of interpretation best identifies a tall tower on a vertical photo? (a) Tone (b) Texture (c) Shadow (d) Pattern
Answer
(c) Shadow
  1. Which is an advantage of the raster model? (a) Compact structure (b) Easy overlay (c) Network analysis (d) High output quality
Answer
(b) Easy overlay
  1. Spatial resolution of LISS-IV is: (a) 23.5 m (b) 5.8 m (c) 56 m (d) 1 m
Answer
(b) 5.8 m
  1. The scale of a vertical aerial photo is given by: (a) H/f (b) f/(H − h) (c) hr/H (d) (H − h)/f
Answer
(b) f/(H − h)
  1. Which is not a geostationary satellite? (a) INSAT (b) METEOSAT (c) Landsat (d) GOMS
Answer
(c) Landsat