🛰️ Remote Sensing — Principles and Sensors

This lesson builds core elective concepts in BSc Agriculture with practical applications and exam-oriented clarity.

---

Remote Sensing — Principles and Sensors

Definition of Remote Sensing

Remote sensing is the science and art of obtaining information about an object, area, or phenomenon through the analysis of data acquired by a sensor that is not in physical contact with the object. The information is typically derived from electromagnetic radiation emitted or reflected by the target.

In the context of agriculture, remote sensing enables the monitoring of crop health, soil conditions, water bodies, and land use across vast areas — repeatedly, objectively, and at low marginal cost per hectare.

---

The Electromagnetic Spectrum

Electromagnetic (EM) radiation travels as waves and is characterized by wavelength (λ, in micrometres — μm) and frequency (Hz). All EM radiation travels at the speed of light (3 × 10⁸ m/s).

Region	Wavelength Range	Remote Sensing Relevance
Gamma rays	< 0.001 nm	Not used in RS
X-rays	0.001–10 nm	Not used in RS
Ultraviolet (UV)	0.01–0.4 μm	Limited RS use (ozone sensing)
Visible	0.4–0.7 μm	Blue (0.4–0.5), Green (0.5–0.6), Red (0.6–0.7) — most RS cameras
Near Infrared (NIR)	0.7–1.4 μm	Strong vegetation response; NDVI computation
SWIR (Short-Wave IR)	1.4–3 μm	Leaf water content; crop stress; mineral mapping
Mid Infrared (MIR)	3–8 μm	Fire detection
Thermal IR (TIR)	8–14 μm	Surface temperature; canopy temperature; irrigation stress
Microwave	1 mm – 1 m	SAR radar; all-weather penetration; soil moisture
Radio waves	> 1 m	Not typically used in agricultural RS

---

Interaction with the Atmosphere

As EM radiation travels from the Sun to Earth's surface and back to the satellite sensor, it interacts with the atmosphere through:

Scattering

• Rayleigh scattering: Caused by gas molecules (N₂, O₂); inversely proportional to λ⁴; predominantly affects shorter wavelengths (blue) → sky appears blue; blue band satellite images are hazier
• Mie scattering: Caused by aerosols (dust, smoke, pollution); affects visible and NIR wavelengths; significant over agricultural areas with dust/haze
• Non-selective scattering: Water droplets in clouds; affects all wavelengths equally — clouds appear white; clouds block optical satellite sensors

Absorption

Certain atmospheric gases absorb specific wavelengths:

• Ozone (O₃): Absorbs UV (< 0.3 μm) — protects life but limits UV remote sensing
• Water vapour (H₂O): Major absorber at several IR wavelengths
• Carbon dioxide (CO₂): Absorbs at 4.3 μm
• Oxygen (O₂): Absorbs at 0.76 μm (oxygen absorption band — used to detect this band in plant physiology)

Atmospheric Windows

Wavelength ranges where the atmosphere is relatively transparent — these are the usable bands for remote sensing:

Atmospheric Window	Wavelength	Sensors
Visible	0.4–0.7 μm	Optical cameras, multispectral sensors
NIR	0.77–0.90 μm	Multispectral sensors (NDVI)
SWIR-1	1.55–1.75 μm	Landsat Band 5, Sentinel-2 Band 11
SWIR-2	2.08–2.35 μm	Landsat Band 7, Sentinel-2 Band 12
MIR	3.5–5.0 μm	Fire detection (MODIS)
TIR	8.0–14 μm	Thermal sensors (Landsat, ASTER, MODIS)
Microwave	0.1–30 cm	SAR radar (all-weather)

---

Interaction with Vegetation

Understanding how plants interact with EM radiation is fundamental to vegetation remote sensing:

• Blue and red wavelengths: Strongly absorbed by chlorophyll (a and b) for photosynthesis → healthy vegetation appears dark in blue and red bands
• Green wavelength (~0.55 μm): Partially reflected → healthy vegetation appears green to the eye
• NIR (0.7–1.3 μm): Strongly reflected (70–80%) by leaf mesophyll cell structure → sharp "red edge" rise from red to NIR is the most diagnostic feature of vegetation
• SWIR (1.4–2.5 μm): Absorbed by leaf water content → lower reflectance with higher water content; useful for drought stress detection

The contrast between low red reflectance and high NIR reflectance forms the basis of the NDVI — the most widely used vegetation index in agriculture.

---

Types of Sensors

Passive Sensors

Passive sensors detect radiation that originates externally — either reflected solar radiation (visible, NIR, SWIR) or emitted thermal radiation (TIR):

• Require sunlight for optical bands → limited at night and unable to see through clouds
• Examples: Multispectral cameras (Landsat OLI, Sentinel-2 MSI), thermal scanners (Landsat TIRS), panchromatic cameras

Active Sensors

Active sensors generate their own energy and detect the return signal:

• SAR (Synthetic Aperture Radar): Emits microwave pulses; measures backscatter; penetrates clouds and rain; works day and night; sensitive to surface roughness, moisture, and crop structure
• LiDAR (Light Detection And Ranging): Emits laser pulses (NIR or green); measures return time → 3D point cloud; measures canopy height, tree volume, terrain topography

Property	Passive Optical	Passive Thermal	Active SAR	Active LiDAR
Energy source	Sun	Earth/target	Sensor	Sensor
Cloud penetration	No	Partial	Yes	No
Night operation	No	Yes	Yes	Yes
Output	Reflectance image	Temperature image	Backscatter image	3D point cloud
Agriculture use	Crop mapping, NDVI	Irrigation stress	Soil moisture, flood	Biomass, canopy

---

Sensor Characteristics

Spatial Resolution

The size of the smallest feature that can be detected; equivalent to pixel size on the ground:

• Very high resolution (VHR): < 1 m — WorldView-3 (0.3 m), Pleiades (0.5 m), Cartosat-3 (0.25 m) — individual plant rows visible
• High resolution: 2–10 m — Sentinel-2 (10 m), SPOT-7 (1.5 m), PlanetScope (3 m)
• Medium resolution: 10–100 m — Landsat (30 m), IRS LISS-III (23.5 m)
• Coarse resolution: > 100 m — MODIS (250 m–1 km), NOAA AVHRR (1.1 km) — regional/global crop monitoring

Spectral Resolution

The number and width of spectral bands recorded:

• Panchromatic: Single broad band (visible); black-and-white image; high spatial resolution
• Multispectral: 4–15 discrete bands; Landsat (11 bands), Sentinel-2 (13 bands), RESOURCESAT (4 bands)
• Hyperspectral: 100–500+ contiguous narrow bands; PRISMA (Italy), EO-1 Hyperion, airborne AVIRIS; can detect specific biochemical properties; disease and nutrient stress

Temporal Resolution (Revisit Time)

How frequently the satellite revisits the same area:

• MODIS: Daily (Terra + Aqua)
• Sentinel-2 (A+B pair): 5 days at equator
• Landsat 8/9 (pair): 8 days; single satellite: 16 days
• IRS RESOURCESAT-2 AWiFS: 5 days
• PlanetScope constellation: Daily

Radiometric Resolution

The sensitivity of the sensor to detect differences in energy levels; measured in bits:

• 8-bit: 256 discrete values (older Landsat TM)
• 12-bit: 4,096 values (Landsat 8 OLI, Sentinel-2) — better detection of subtle reflectance differences
• 16-bit: 65,536 values (high-end hyperspectral sensors)

---

Key Satellites for Agriculture

Satellite	Country/Agency	Sensor	Spatial Res.	Key Bands	Revisit	Data Access
Landsat 8/9	USA (NASA/USGS)	OLI + TIRS	30 m (optical), 100 m (thermal)	11 bands incl. SWIR, TIR	8 days (pair)	Free (USGS EarthExplorer)
Sentinel-2A/2B	EU (ESA)	MSI	10 m (VIS, NIR), 20 m (RE, SWIR), 60 m (coastal)	13 bands incl. Red Edge	5 days	Free (Copernicus Open Access Hub)
MODIS (Terra/Aqua)	USA (NASA)	MODIS	250 m–1 km	36 bands; daily global	Daily	Free (NASA Earthdata)
IRS RESOURCESAT-2/2A	India (ISRO)	LISS-III + AWiFS	23.5 m (LISS-III), 56 m (AWiFS)	4 bands (Blue, Green, Red, NIR)	24 days (LISS-III), 5 days (AWiFS)	NRSC Bhuvan portal
Cartosat-3	India (ISRO)	PAN + MS	0.25 m (PAN), 1 m (MS)	PAN + 4 MS bands	4 days	Restricted / NRSC
RISAT-1A	India (ISRO)	C-band SAR	1–50 m	SAR (microwave)	~12 days	NRSC
NOAA AVHRR	USA (NOAA)	AVHRR	1.1 km	5 bands incl. TIR	Daily	Free
PlanetScope	USA (Planet Labs)	RGB + NIR	3–5 m	4–8 bands	Daily	Commercial (free for research)
Sentinel-1	EU (ESA)	C-band SAR	10 m	SAR (backscatter)	6–12 days	Free (Copernicus)

---

IRS Satellite Series (India)

The Indian Remote Sensing (IRS) satellite series, operated by ISRO and NRSC, has been the backbone of India's national agricultural monitoring since 1988:

• IRS-1A (1988): LISS-I sensor, 72.5 m resolution — first systematic crop mapping
• IRS-1B (1991): Repeat of 1A with improved instruments
• IRS-1C/1D (1995, 1997): LISS-III (23.5 m), WiFS (188 m), PAN (5.8 m) — major leap; LISS-III became standard for crop mapping
• RESOURCESAT-1 (2003): LISS-III + LISS-IV (5.8 m) + AWiFS (56 m)
• RESOURCESAT-2 (2011): Improved LISS-III and AWiFS; LISS-IV (5.8 m)
• RESOURCESAT-2A (2016): Backup/replacement; currently operational

NRSC's Bhuvan geoportal (bhuvan.nrsc.gov.in) provides free access to IRS and other satellite data for Indian users.

---

Overview

Remote sensing captures EM radiation reflected or emitted from Earth's surface. The atmospheric windows in visible, NIR, SWIR, TIR, and microwave regions define usable bands. Vegetation's strong NIR reflectance and low red reflectance (due to chlorophyll) are the basis for vegetation indices. Passive sensors need sunlight and are blocked by clouds; active SAR sensors operate in all weather. Four key resolution types — spatial, spectral, temporal, and radiometric — define a sensor's capabilities for a given application. For Indian agriculture, IRS RESOURCESAT-2 (LISS-III/AWiFS), Sentinel-2 (free, 10 m, 5-day), and MODIS (daily, 250 m) are the primary data sources.

---

Summary Cheat Sheet

Topic	Key takeaway
Main focus	Electromagnetic spectrum, atmospheric interactions, sensor types, spatial/spectral/temporal resolution, and key agricultural satellites.
Section context	Revise this lesson with the rest of Remote Sensing for stronger conceptual continuity.