Remote sensing sensors, platforms and Indian satellites

In short

A remote sensing system is a sensor carried on a platform: the sensor measures electromagnetic radiation reflected or emitted by the Earth, and the platform (ground, aircraft, drone or satellite) decides what area it sees, how often and at what detail. The topic covers sensor physics and classification, the four resolutions, scanning mechanisms, orbit types, and the Indian Earth-observation fleet run by ISRO with data distributed by NRSC (Hyderabad) through Bhoonidhi and visualised on Bhuvan. It is examined in GATE Geomatics (GE) Section on remote sensing fundamentals, in IIT/Anna University/JNTU B.Tech papers, and in the Survey of India, NRSC, NESAC, state remote sensing centre and ISRO scientist/engineer recruitment tests. Employers (NRSC, SAC, state RSACs, Survey of India, RMSI, Esri India, Genesys, ADANI/L&T geospatial teams, NHAI and smart-city consultants) expect a candidate to pick the right sensor and satellite for a given mapping scale.

Explained

What a remote sensing system measures

• Remote sensing: acquiring information about an object without physical contact, by recording electromagnetic (EM) radiation that the object reflects, scatters or emits.
• Energy path: source (Sun or the sensor itself) → atmosphere → target interaction (reflection, absorption, transmission) → atmosphere again → sensor → signal → data product.
• EM spectrum regions used: visible (0.4–0.7 µm), near infrared NIR (0.7–1.3 µm), shortwave infrared SWIR (1.3–3 µm), mid IR, thermal infrared TIR (8–14 µm), microwave (1 mm–1 m, bands P, L, S, C, X, Ku).
• Atmospheric windows: wavelength intervals of low absorption where sensing is practical — roughly 0.4–1.3, 1.5–1.8, 2.0–2.5, 3.0–3.6, 8–14 µm and the microwave region (largely all-weather).
• Spectral signature: the curve of reflectance against wavelength. Healthy vegetation is low in red (chlorophyll absorption) and high in NIR; water absorbs NIR strongly and looks dark; dry soil rises steadily with wavelength.
• Radiometric quantities: radiance (W m⁻² sr⁻¹ µm⁻¹) is what the sensor sees; reflectance is a dimensionless ratio 0–1; digital number (DN) is the quantised count stored.

Classification of sensors

• By energy source: *passive* sensors use natural radiation (Sun or Earth's emission) — cameras, multispectral scanners, thermal scanners, microwave radiometers; *active* sensors supply their own energy — RADAR, SAR, LiDAR, scatterometer, altimeter.
• By imaging geometry: *framing* systems (a camera or CCD array capturing a whole frame at once) and *scanning* systems (building the image line by line).
• By spectral detail: *panchromatic* (one broad band, best spatial resolution), *multispectral* (3–15 discrete bands), *superspectral* (15–50), *hyperspectral* (100+ narrow contiguous bands, e.g. ISRO's AVIRIS-NG campaigns and HySIS), *thermal* (emitted TIR).
• By output: imaging sensors (produce a picture) versus non-imaging sensors (radiometer, spectrometer, altimeter, sounder — a profile or a point value).
• Detector types: photon/quantum detectors (Si CCD, InGaAs, HgCdTe for TIR, usually cooled) and thermal detectors (bolometers).

Scanning mechanisms, IFOV and swath

• Whiskbroom (across-track) scanner: a rotating or oscillating mirror sweeps one detector across the track while the platform motion advances the line — e.g. Landsat MSS/TM. Few detectors, so calibration is uniform, but dwell time per pixel is short and moving parts wear out.
• Pushbroom (along-track) scanner: a linear CCD array covers the whole swath; the platform's forward motion builds the image. Longer dwell time → better signal-to-noise and finer resolution; no moving parts; but thousands of detectors need careful relative calibration (striping if one drifts). Used in LISS-III/IV, Cartosat, Sentinel-2.
• Step-and-stare / area array: a 2-D CCD or CMOS matrix images a frame — drone and aerial digital cameras, and Cartosat-3's advanced optics.
• IFOV (instantaneous field of view): the solid angle, β, a single detector sees; the ground patch it covers is the GRE/GSD. Narrow IFOV → fine resolution but less energy per pixel.
• FOV and swath: total FOV is the full angle imaged across-track; swath is its ground width. Fine resolution and wide swath conflict, so missions pair a wide-swath coarse sensor (AWiFS, 740 km) with a narrow-swath fine sensor (LISS-IV, 70 km).
• Along-track vs across-track stereo: Cartosat-1 carried two fore (+26°) and aft (−5°) cameras for same-pass along-track stereo for DEM generation; across-track stereo uses two passes and suffers radiometric mismatch.

The four resolutions

• Spatial resolution: smallest ground detail resolvable, normally quoted as the ground sample distance in metres — e.g. Cartosat-3 panchromatic about 0.25 m, LISS-IV 5.8 m, LISS-III 23.5 m, AWiFS 56 m, Oceansat OCM 360 m, INSAT-3D imager 1 km (visible). Mapping rule of thumb: usable scale ≈ GSD (m) × 2,000 to 5,000 as the map denominator.
• Spectral resolution: number and narrowness of bands. Narrow bands discriminate materials (hyperspectral mineral mapping) but collect less energy.
• Radiometric resolution: bit depth of quantisation — 7-bit (128 levels) in older IRS LISS, 10-bit (1,024) in LISS-III/IV, 12-bit (4,096) in Cartosat-3 and most new sensors. More bits = finer brightness discrimination, larger files.
• Temporal resolution (revisit): interval between two acquisitions of the same area — Resourcesat AWiFS about 5 days, LISS-III 24 days, LISS-IV 5 days with ±26° steering, Cartosat few days with agile pointing, INSAT-3D/3DR every 15–30 minutes in rapid-scan.
• Trade-off: sensor data rate is finite, so improving one resolution usually costs another; this trade-off is the most commonly asked conceptual question on the topic.

Platforms and orbits

• Ground-based: spectroradiometer (ASD/SVC) and tower-mounted sensors for signature libraries and vicarious calibration.
• Airborne: aerial cameras, airborne hyperspectral (AVIRIS-NG flown over India), airborne LiDAR; flexible, high resolution, weather-limited, costly per km².
• UAV/drone: centimetre detail over small areas; in India flights follow the Drone Rules, 2021 (as amended) with the Digital Sky platform, mandatory UIN registration, Remote Pilot Certificate for most categories, and the green/yellow/red airspace zone map.
• Spaceborne — low Earth orbit (LEO), 500–900 km: wide coverage, repetitive, systematic; most imaging missions.
• Sun-synchronous (polar) orbit: inclination slightly above 90° (about 98° at 800 km) so nodal precession matches the Earth's revolution; the satellite crosses the equator at nearly the same local solar time each pass (IRS descending node around 10:30 a.m.), giving comparable illumination between dates. Period about 100 minutes, roughly 14 orbits/day.
• Geostationary orbit (GEO), 35,786 km, equatorial, 0° inclination: period equals one sidereal day, so the satellite stays over a fixed longitude — continuous hemispheric view for meteorology (INSAT-3D/3DR/3DS) but coarse resolution and poor high-latitude geometry.
• Other orbital terms: repeat cycle (days to return to the same ground track), ground track, nadir and off-nadir viewing, ascending/descending node, perigee/apogee, highly elliptical and medium Earth orbits (navigation, not imaging).
• Orbit mechanics in brief: orbital period T = 2π√(a³/µ) with µ = 3.986 × 10¹⁴ m³ s⁻² — this one relation answers most numerical questions on altitude, period and ground-track speed.

Indian Earth-observation satellites and data policy

• History: Bhaskara-1/2 (1979, 1981) experimental; IRS-1A (1988) and 1B with LISS-I (72.5 m) and LISS-II (36.25 m) — India's first operational civilian series; IRS-1C/1D (1995, 1997) added a 5.8 m panchromatic camera and WiFS, making India a global supplier of high-resolution data at the time.
• Resourcesat-1 (2003), Resourcesat-2 (2011), Resourcesat-2A (2016): the land-and-water workhorses — LISS-IV (5.8 m, 70 km swath in mono and about 24 km in 3-band multispectral, steerable), LISS-III (23.5 m, 4 bands including SWIR, 141 km), AWiFS (56 m, 740 km, 5-day revisit). The backbone of crop inventory and land-use mapping.
• Cartosat series (cartography): Cartosat-1 (2005, 2.5 m stereo pair for DEMs), Cartosat-2 family (about 0.8 m pan), Cartosat-3 (2019, about 0.25 m pan, 1 m multispectral — India's sharpest civilian optical sensor).
• Ocean and atmosphere: Oceansat-2 and Oceansat-3/EOS-06 (2022) with Ocean Colour Monitor (13 bands, about 360 m), scatterometer for sea-surface winds and SSTM; SARAL (with CNES) altimetry; Megha-Tropiques; INSAT-3D/3DR (6-channel imager and 19-channel sounder) and INSAT-3DS (2024) for weather and cyclone tracking.
• Microwave: RISAT-1 (C-band SAR, 2012), RISAT-2B/2BR1 (X-band, with coarse-to-1 m modes), EOS-04/RISAT-1A (2022, C-band) — all-weather, day-and-night flood and crop mapping. NISAR (NASA-ISRO SAR, launched 2025) is a dual-frequency L- and S-band mission with free and open data for deformation and biomass; mission products were still ramping up through 2026.
• Other current missions: EOS-01 (radar imaging), EOS-04, EOS-06, EOS-08 (2024, with an EO infrared payload) — EOS-02 was lost in the SSLV-D1 launch failure of August 2022, HySIS (hyperspectral, about 55 bands in VNIR), and private Indian constellations (Pixxel's hyperspectral Firefly, SkyServe/Dhruva) now operating under IN-SPACe authorisation.
• Institutions and policy: ISRO builds and operates; NRSC, Hyderabad archives and sells data (Bhoonidhi portal; Bhuvan for visualisation); SAC, Ahmedabad develops payloads; IN-SPACe authorises private space activity and NSIL is the commercial arm, both under the Indian Space Policy, 2023. The National Geospatial Policy, 2022 and the February 2021 geospatial guidelines liberalised acquisition and dissemination of geospatial data by Indian entities, removing prior licensing for most ground and map data, with the finest freely shareable spatial accuracy set at 1 m horizontal and 3 m vertical; data finer than that may be created and held only by Indian entities, and stored in India. Survey of India remains the national mapping agency and runs the CORS network and the Indian vertical/horizontal datum.
• Limitations of spaceborne optical sensing: cloud cover (critical in the Indian monsoon), fixed revisit, no control over illumination, atmospheric effects, and the resolution–swath trade-off — the main reasons SAR, drones and aerial surveys remain necessary.

Key points

• Remote sensing - information from EM radiation without contact; a sensor on a platform.
• Passive vs active - passive uses solar or Earth emission; active (SAR, LiDAR) supplies its own illumination and works at night.
• Atmospheric window - wavelength band of low atmospheric absorption; sensors are designed to sit inside one.
• Spectral signature - reflectance-vs-wavelength curve; vegetation's NIR jump is the basis of NDVI.
• Whiskbroom - one detector, rotating mirror, across-track; uniform calibration, short dwell time.
• Pushbroom - linear CCD array, no moving parts, longer dwell, better SNR; risk of striping.
• IFOV - angular view of one detector; multiplied by altitude gives the ground resolution cell.
• Four resolutions - spatial (m), spectral (bands), radiometric (bits), temporal (days); improving one usually costs another.
• Sun-synchronous orbit - about 98° inclination at 800 km; same local solar time each pass, comparable illumination.
• Geostationary orbit - 35,786 km, equatorial; continuous view, coarse resolution; INSAT-3D/3DR/3DS.
• Resourcesat trio - LISS-IV 5.8 m, LISS-III 23.5 m, AWiFS 56 m; the Indian land-mapping workhorse.
• Cartosat - cartography series; Cartosat-1 stereo for DEMs, Cartosat-3 about 0.25 m panchromatic.
• RISAT/EOS-04/NISAR - Indian microwave missions; all-weather flood, crop and deformation mapping.
• Oceansat/EOS-06 OCM - 13-band ocean colour at about 360 m for chlorophyll and potential fishing zones.
• Swath vs resolution - wide swath means coarse pixels; missions carry both kinds of sensor.
• Radiometric resolution - 8-bit gives 256 levels, 12-bit gives 4,096; more bits distinguish subtle brightness.
• NRSC/Bhoonidhi and Bhuvan - the national archive, ordering portal and geoportal for Indian data.
• Policy - Indian Space Policy 2023, National Geospatial Policy 2022, IN-SPACe authorisation, Drone Rules 2021 for UAV platforms.

Formulas and examples

Example 1 — IFOV and ground resolution. A pushbroom sensor at H = 800 km has detector size d = 10 µm and focal length f = 2.0 m.

• β = d/f = 10 × 10⁻⁶ / 2.0 = 5 × 10⁻⁶ rad = 5 µrad.
• GSD = β × H = 5 × 10⁻⁶ × 800,000 = 4.0 m.
• If the array has 12,000 elements, swath = 12,000 × 4.0 = 48,000 m = 48 km.
• Flown instead on an aircraft at 3 km, GSD = 5 × 10⁻⁶ × 3,000 = 0.015 m = 1.5 cm, swath = 12,000 × 0.015 = 180 m.

Example 2 — orbital period and ground-track speed of an IRS-class satellite at h = 817 km.

• a = 6,378 + 817 = 7,195 km = 7.195 × 10⁶ m; a³ = 3.725 × 10²⁰ m³.
• T = 2π√(3.725 × 10²⁰ / 3.986 × 10¹⁴) = 2π√(9.345 × 10⁵) = 2π × 966.7 = 6,075 s ≈ 101.2 minutes.
• Orbits per day = 86,400/6,075 ≈ 14.2.
• v = √(3.986 × 10¹⁴ / 7.195 × 10⁶) = √(5.540 × 10⁷) = 7,443 m/s ≈ 7.44 km/s; ground speed ≈ 7.443 × 6,378/7,195 = 6.60 km/s.
• Equator shift between successive orbits ≈ 40,075 km / 14.2 ≈ 2,820 km — which is why a wide-swath sensor is needed for frequent coverage.

Example 3 — data volume of one LISS-III scene. Swath 141 km, GSD 23.5 m, scene 141 km along-track, 4 bands, 10-bit stored in 2 bytes.

• Pixels across = 141,000/23.5 = 6,000; along = 6,000 → 36 × 10⁶ pixels per band.
• Four bands → 1.44 × 10⁸ pixels; at 2 bytes each = 2.88 × 10⁸ bytes ≈ 275 MiB (about 288 MB decimal).
• The same area from LISS-IV at 5.8 m: 141,000/5.8 ≈ 24,310 pixels per side → 5.91 × 10⁸ pixels per band, i.e. about 16 times the LISS-III data for the same ground area.

Example 4 — choosing a sensor for a mapping scale. A municipality needs a 1:5,000 base map; plotting accuracy is 0.25 mm on paper.

• Ground equivalent = 0.25 mm × 5,000 = 1,250 mm = 1.25 m, so GSD must be about 1 m or finer → Cartosat-3 (0.25 m pan / 1 m MX) or drone imagery, not LISS-IV (5.8 m).
• For a 1:50,000 district land-use map: 0.25 mm × 50,000 = 12.5 m → LISS-III at 23.5 m is marginal for linework but standard for land-use classes; LISS-IV is comfortable.

Example 5 — revisit with off-nadir steering. LISS-IV has a 70 km swath on an orbit whose adjacent ground tracks are 117 km apart, with ±26° steering at 817 km altitude.

• Lateral reach = H tan 26° = 817 × 0.4877 ≈ 398 km either side, covering several adjacent tracks.
• A 24-day nadir repeat therefore becomes an effective revisit of about 5 days, at the cost of oblique geometry (larger pixels at the swath edge and relief displacement).

Where it is used

• Agriculture and food security: AWiFS and LISS-III drive the FASAL/CROP crop-acreage and yield programmes and the Mahalanobis National Crop Forecast Centre's forecasts; state agriculture departments and PMFBY crop-insurance assessment use them.
• Urban, infrastructure and utilities: Cartosat-3 and drone imagery for AMRUT/smart-city base maps, NHAI highway alignment and corridor monitoring, property-tax surveys and the SVAMITVA rural property mapping programme.
• Water, disaster and forests: RISAT/EOS-04 for monsoon flood extent, Oceansat for potential fishing zone advisories (INCOIS), Forest Survey of India's biennial State of Forest Report using LISS-III, NDMA and state SDMA response mapping.
• Mining, geology and groundwater: hyperspectral and SWIR data for mineral targeting (GSI), National Aquifer Mapping with CGWB, lineament mapping for borewell siting.
• Job roles: remote sensing analyst, GIS/geospatial engineer, photogrammetrist, image analyst, payload/data-products engineer at NRSC and SAC, scientist/engineer-SC in ISRO centres, consultants at RMSI, Esri India, Genesys, Tata Consulting Engineers, and state remote sensing application centres (KSREC, APSAC, MPSRSAC, GSDMA).
• Public products: Bhuvan, Bhoonidhi data ordering, NRSC's land-use/land-cover 1:50,000 database, and open NISAR and Sentinel data increasingly used alongside Indian archives.

Interview questions

• Q: What is the difference between a passive and an active sensor? A: A passive sensor records naturally available radiation (reflected sunlight or Earth's thermal emission); an active sensor transmits its own energy and measures the return, so it works at night and, for microwaves, through cloud.
• Q: Why is a sun-synchronous orbit preferred for Earth-observation satellites? A: Its precession keeps the equator-crossing local solar time nearly constant, so images of the same area on different dates have comparable sun angle and shadow, which makes change detection valid.
• Q: Compare whiskbroom and pushbroom scanners. A: Whiskbroom sweeps one detector across-track with a mirror — uniform radiometry but short dwell time and moving parts; pushbroom uses a linear array with longer dwell time, better SNR and finer resolution, but needs careful inter-detector calibration to avoid striping.
• Q: A satellite has 10 µrad IFOV at 700 km. What is its ground resolution? A: GSD = 10 × 10⁻⁶ × 700,000 = 7 m.
• Q: Why can't a sensor have fine spatial resolution, many narrow bands, high bit depth and daily revisit all at once? A: Each improvement reduces energy per pixel or multiplies the data rate, and the downlink, storage and swath are finite — so missions trade off, or fly complementary sensors like LISS-IV and AWiFS.
• Q: Which Indian satellite would you use for a 1:5,000 urban base map, and why? A: Cartosat-3 at about 0.25 m panchromatic, because a 1:5,000 map needs about 1 m or finer ground detail; LISS-IV at 5.8 m cannot carry that linework.
• Q: How was Cartosat-1 able to produce DEMs? A: It carried two cameras pointing fore and aft so that the same ground was imaged twice in one pass, giving along-track stereo with consistent illumination for photogrammetric height extraction.
• Q: What does radiometric resolution mean, and what does 12-bit give you? A: It is the number of brightness levels the sensor quantises into; 12 bits give 2¹² = 4,096 levels, letting you separate subtle differences such as shallow water depths or dark forest canopy.
• Q: What is the advantage of a geostationary platform, and its main cost? A: It views one hemisphere continuously, so INSAT-3D/3DR can image a developing cyclone every 15–30 minutes; the cost is kilometre-scale resolution and distorted geometry at high latitudes.
• Q: Where do you obtain Indian satellite data and what governs its use? A: From NRSC's Bhoonidhi portal (and Bhuvan for viewing), under the Indian Space Policy 2023 and National Geospatial Policy 2022; private operators work under IN-SPACe authorisation.
• Q: What does the AWiFS 5-day revisit buy an agriculture department? A: Enough repeat coverage during the kharif and rabi seasons to catch sowing progress and drought stress despite monsoon cloud, which a 24-day LISS-III cycle would miss.
• Q: Why does swath shrink as resolution improves on the same satellite? A: For a fixed detector count and data rate, a smaller ground cell per detector means the array covers less ground width — Cartosat-3's 16 km swath against AWiFS's 740 km is this relation in practice.

Common mistakes

• Treating resolution as a single number: a student quotes "5.8 m resolution" without saying whether the spectral, radiometric and temporal resolutions suit the task.
• Mixing up IFOV (an angle, in radians or microradians) with the ground resolution cell (a length, in metres) — always multiply by the altitude.
• Calling a sun-synchronous orbit "polar at 90°": the inclination is slightly retrograde (about 98°) precisely to obtain the precession.
• Saying a geostationary satellite is "stationary in space" — it orbits at 3.07 km/s; it is stationary only relative to the rotating Earth.
• Assuming higher spatial resolution is always better: AWiFS's coarse 56 m pixels give the 5-day revisit and 740 km swath that crop monitoring needs, and generate a fraction of the data.
• Quoting withdrawn or merged mission names — the newer missions are named EOS-01/02/04/06/08, and Oceansat-3 is EOS-06; dates and specifications of very recent missions should be verified against NRSC's current data catalogue.
• Forgetting that microwave is the only largely all-weather option, then proposing optical imagery for monsoon flood mapping.

Related topics

GIS and remote sensing · Digital image processing for remote sensing · Microwave and LiDAR remote sensing · Photogrammetry and drone surveys · Geodesy and map projections · Global navigation satellite systems (GNSS/NavIC).