UNIT 2: Remote Sensing & GIS - Exam-Focused Short Notes

MODULE 1: FUNDAMENTALS OF REMOTE SENSING

Electromagnetic Spectrum (EMS) & Radiation Principles

• Energy Source: Sun is the primary source for passive remote sensing. Active sensors (e.g., RADAR) provide their own energy.

• Atmospheric Interactions: Key processes affecting radiation reaching the sensor:

  • Absorption: Specific molecules (H₂O, CO₂, O₃) absorb specific wavelengths.

  • Scattering: Redirects radiation. Types:

    • Rayleigh Scattering: By small particles (< wavelength). Causes blue sky.

    • Mie Scattering: By particles ~ wavelength size (aerosols).

    • Non-selective Scattering: By large particles (> wavelength). Affects all wavelengths.

  • Transmission: Portion of radiation that passes through atmosphere unaffected.

• Spectral Reflectance/Radiance:

  • Spectral Reflectance (ρ): Ratio of reflected energy to incident energy for a specific wavelength.

$$ρ(λ) = \frac{\text{Reflected Energy}}{\text{Incident Energy}}$$

*   **Radiance (L):** Amount of electromagnetic energy flowing per unit area, per unit solid angle, per unit time. Measured by sensors.

*   **Importance:** Different Earth surface features have unique reflectance patterns across EMS, forming the basis for identification.

[!TIP] EXAM CRITICAL: Be prepared to sketch and explain characteristic spectral reflectance curves.

Spectral Reflectance Curves (HIGH PRIORITY)

• General Shape: Plot of Reflectance (%) vs. Wavelength (µm).

• Vegetation:

  • Low reflectance (5-10%) in visible (blue, green, red) due to chlorophyll absorption (peaks in blue & red).

  • High reflectance (~40-50%) in Near-Infrared (NIR) due to cell structure. Sharp rise at "red edge" (~0.7 µm).

  • Water absorption bands in Short-Wave Infrared (SWIR) at 1.4 µm and 1.9 µm.

  • Phenological Stages: Green vegetation shows classic curve. Senescent/dry vegetation has higher reflectance in visible and lower in NIR.

• Soil:

  • Generally increasing reflectance with wavelength (no sharp absorption features).

  • Curve shape influenced by moisture content, texture, organic matter. Dry soil has higher overall reflectance than wet soil.

• Water:

  • Very low reflectance in visible and NIR (< 10%).

  • Absorbs strongly in NIR and SWIR (appears dark).

  • Turbid/sediment-laden water has higher reflectance in visible (especially red) than clear water.

  • Suspended materials and bottom reflectance affect curve in shallow clear water.

Ideal Remote Sensing System & Components

• Definition: A system that accurately and efficiently records, processes, and interprets energy interactions between electromagnetic radiation and the Earth's surface.

• Block Diagram Components:

  1. Energy Source (Sun or active sensor).

  2. Atmosphere (Interaction path).

  3. Interaction with Target (Reflection, emission, absorption).

  4. Sensor (Detects and records energy).

  5. Platform (Carries sensor - satellite, aircraft, drone).

  6. Data Processing & Interpretation (Raw data to information).

[!TIP] COMMON PITFALL: Remember the sequence: Source → Atmosphere → Target → Sensor → Platform → Processing. Platform carries sensor; sensor detects energy.

Characteristics of RS Data

• Synoptivity: Ability to capture a large, contiguous area in a single view (e.g., entire state/country). Significance: Enables regional studies, macro-level planning.

• Repetivity: Ability to revisit the same area at regular intervals. Significance: Monitoring dynamic phenomena (crop growth, floods, deforestation).

• Advantages: Large area coverage, access to inaccessible regions, multi-spectral capability, temporal monitoring, objective data.

• Limitations: Weather dependency (clouds), atmospheric interference, interpretation requires ground truth, high initial cost, limited spatial resolution for some applications.

MODULE 2: PLATFORMS, SENSORS & RESOLUTIONS

Satellite Orbits & Types (HIGH PRIORITY)

Indian Satellite Program & Missions

• Chandrayaan-3 (Salient Features & Objectives):

  • Objective: Demonstrate safe and soft landing on lunar surface, rover operations.

  • Components: Lander (Vikram), Rover (Pragyan), Propulsion Module.

  • Salient Features: First mission to land near lunar south pole (~69°S). Indigenous technology. Rover conducted in-situ chemical analysis. Successfully demonstrated hopping maneuver.

• IRS Series (Indian Remote Sensing):

  • Series: IRS-1A/1B (LISS), IRS-P6 (Resourcesat), IRS-2 (Cartosat - high-res), etc.

  • Applications: Agriculture (crop inventory, drought), Forestry (deforestation, biomass), Water Resources (surface water, watershed), Urban Planning, Disaster Management.

• Earth Resource vs. Weather Satellites:

  • Earth Resource (e.g., IRS, Landsat): Focus on land surface. High spatial resolution (meters), moderate spectral resolution (multispectral), lower temporal resolution (days to weeks). Used for mapping, resource inventory.

  • Weather/Meteorological (e.g., INSAT, GOES): Focus on atmosphere & ocean. Low spatial resolution (km), high temporal resolution (minutes), specific spectral bands for clouds, water vapor, temperature. Used for forecasting, storm tracking.

Sensor Resolutions (HIGH PRIORITY)

• Spatial Resolution: Minimum separable distance between two objects on ground. Measured in meters (GSD - Ground Sample Distance). Influenced by: Sensor optics, detector size, platform altitude.

• Spectral Resolution: Ability to detect specific wavelength intervals. Number and width of bands (e.g., multispectral: 3-15 bands; hyperspectral: hundreds of narrow bands). Influenced by: Filter/detector design.

• Radiometric Resolution: Ability to detect differences in brightness. Number of digital levels (e.g., 8-bit = 256 levels). Influenced by: Sensor's signal-to-noise ratio, quantization.

• Temporal Resolution: Revisit time over same area. Influenced by: Orbit, sensor swath width, sensor pointing capability, latitude.

[!TIP] MEMORY AID: Spatial = Size (meters). Spectral = Spectrum (bands). Radiometric = Range of brightness (digital numbers). Temporal = Time (revisit).

MODULE 3: IMAGE ANALYSIS & CLASSIFICATION

Image Interpretation Techniques

• Visual Image Analysis:

  • Elements: Tone (brightness/color), Texture (roughness/smoothness), Shape (geometric form), Size (relative/absolute), Pattern (spatial arrangement), Association (relationship with other features), Shadow (shape/height).

  • Tools: Stereoscope: Used for viewing stereo-pair images to perceive 3D elevation and depth.

• Digital Image Analysis:

  • Concept: Computer-based processing of digital numbers (DN) in image pixels.

  • Workflow: Pre-processing (corrections) → Enhancement → Classification → Post-processing → Accuracy assessment → Interpretation.

[!TIP] CONTRAST: Visual uses human eye/brain; Digital uses algorithms/computers.

Image Classification

• Supervised Classification:

  • Process: User defines training sites (samples of known land cover). Algorithm learns spectral signature of each class. Classifies entire image based on these signatures.

  • Algorithms: Maximum Likelihood (most common, probabilistic), Minimum Distance, Parallelepiped.

• Unsupervised Classification:

  • Process: Algorithm automatically groups pixels into clusters (spectrally similar) without prior knowledge. User then interprets/assigns meaning to clusters.

  • Algorithms: ISODATA (Iterative Self-Organizing Data Analysis), K-Means.

• Key Difference: Supervised uses prior knowledge (training data). Unsupervised discovers natural groupings.

Image Pre-processing & Enhancement

• Geometric Corrections:

  • Systematic Errors: Predictable, caused by sensor/platform geometry (e.g., scan skew, Earth rotation). Corrected using sensor model.

  • Non-systematic Errors: Unpredictable, caused by platform instability, terrain displacement. Corrected using ground control points (GCPs) and polynomial warping.

• Radiometric Corrections: Atmospheric correction (converts radiance to reflectance), noise removal (destriping).

• Image Filtering (Convolution Masks):

  • Purpose: Enhance or suppress specific image features based on spatial frequency.

  • Types:

    • Low-Pass (Smoothing): Averages neighborhood. Reduces noise, blurs edges. Kernel: [[1,1,1],[1,1,1],[1,1,1]].

    • High-Pass (Edge Detection): Highlights rapid changes (edges). Kernel: [[0,-1,0],[-1,4,-1],[0,-1,0]] (Laplacian).

    • Directional (Edge Enhancement): Highlights edges in specific direction (e.g., N-S, E-W).

MODULE 4: GEOGRAPHIC INFORMATION SYSTEM (GIS) FUNDAMENTALS

Definition, Objectives & Components (HIGH PRIORITY)

• Definition: A computer-based system for capturing, storing, managing, analyzing, and displaying spatially referenced data. \boxed{\text{GIS is a system for handling geographic (spatial) data.}}

• Objectives: Efficient data management, spatial analysis for decision support, integration of diverse data sources, visualization, modeling.

• Key Components (5-6 P's):

  1. Hardware: Computer, storage, input/output devices.

  2. Software: GIS package (ArcGIS, QGIS), database management.

  3. Data: Spatial data (maps, imagery) + Attribute data (tables). Most critical & expensive component.

  4. People: Users, specialists, managers.

  5. Procedures: Methods, workflows, standards.

  6. Network: For data sharing and distributed processing (in modern GIS).

[!TIP] EXAM TIP: Always list all 5/6 components with a brief explanation. Emphasize "Data" as the core.

Spatial Data Models & Structures (VERY HIGH PRIORITY)

[!TIP] 14-MARK QUESTION STRATEGY: For "Compare Vector & Raster", use a detailed table as above, then add a concluding paragraph on application suitability (Vector for cadastre, roads; Raster for elevation, satellite images).

Coordinate Systems & Map Projections (HIGH PRIORITY)

• Geographic Coordinate System (GCS):

  • Uses latitude/longitude (angular units, degrees).

  • Based on spheroid/ellipsoid model of Earth (e.g., WGS84).

  • Datum: Defines the position of the spheroid relative to Earth's center (e.g., WGS84, Everest).

  • Units: Degrees. Distortion increases away from standard parallel (none in GCS, but distances/areas not uniform).

• Projected Coordinate System (PCS):

  • Projects GCS onto a 2D plane using a map projection.

  • Uses linear units (meters, feet).

  • Examples: UTM (Universal Transverse Mercator), State Plane, Albers Equal Area.

• Map Projection:

  • Definition: Mathematical method for representing curved Earth surface on a flat map.

  • Purpose: To create flat maps for measurement, navigation, display.

  • Importance: All flat maps have distortion. Choice of projection depends on preserving area, shape, distance, or direction (cannot preserve all).

• UTM Projection (Dec 2024 Q8ii):

  • Type: Conformal (preserves shape locally). Transverse Mercator family.

  • Global system dividing Earth into 60 zones (6° wide).

  • Central Meridian for each zone is true scale (no distortion).

  • Scale factor at central meridian = 0.9996 (slightly reduces scale to minimize overall distortion).

  • Units: Meters. False Easting/Northing added to avoid negative coordinates.

  • Standard parallels: None (cylindrical). Distortion increases away from central meridian.

Data Input, Management & Conversion

• Creating Shapefiles (Vector Data Input):

  1. Define attribute table (fields/columns).

  2. Digitizing: Convert paper maps to digital vector.

     • Manual Digitizing: Using digitizing tablet.

     • Heads-up Digitizing: Tracing features directly on-screen from scanned map/imagery.

  3. Assign attributes to each feature (point, line, polygon) during/after digitizing.

  4. Save as shapefile (.shp, .shx, .dbf, .prj files).

• Data Conversion in GIS:

  • Raster to Vector: Vectorization (tracing raster cells to create lines/polygons). Used for converting scanned maps or classified images.

  • Vector to Raster: Rasterization (converting features to grid cells based on cell value). Used for creating elevation grids, suitability maps.

  • Scanning: Creates raster image from paper map.

  • Georeferencing: Assigning real-world coordinates to raster image (using GCPs).

• Integration of Spatial and Attribute Data:

  • Spatial Data: Geometry (points, lines, polygons) stored in .shp file.

  • Attribute Data: Tabular data (e.g., name, area, population) stored in .dbf file.

  • Linkage: Unique Feature ID (FID) common to both files. GIS software joins them automatically when shapefile is loaded.

  • Result: Clicking a polygon on map displays its attribute record from table.

[!TIP] EXAM ANSWER: For "Procedure to input map and create shapefile," list: 1. Georeference source map (if scanned). 2. Choose appropriate theme (point/line/polygon). 3. Digitize features using heads-up or tablet. 4. Enter/edit attributes in table. 5. Save as shapefile set.

MODULE 5: RS-GIS INTEGRATION & SPATIAL ANALYSIS

Integration of Remote Sensing and GIS

• Conceptual Framework: RS provides the spatial data (thematic maps, base maps, change information). GIS provides the platform to store, manage, analyze, integrate RS data with other thematic layers (soil, dem, census).

• Problems of Using RS Data in GIS:

  1. Resolution Mismatch: RS image resolution (e.g., 30m) may not match vector data (e.g., cadastral maps at 1:1000). Requires resampling.

  2. Registration Errors: Geometric inaccuracies between RS image and base maps. Requires precise georeferencing.

  3. Thematic Accuracy: Classification errors in RS-derived maps propagate into GIS analysis. Requires accuracy assessment (confusion matrix, kappa).

  4. Data Format Compatibility: RS data often in raster formats (GeoTIFF); GIS uses both vector and raster. Requires format conversion.

  5. Radiometric/Atmospheric Variations: Images from different dates/sensors may have different brightness/contrast, affecting change detection.

Spatial Analysis Operations in GIS

• Overlay Analysis (Vector):

  • Concept: Integrating multiple thematic layers to find relationships.

  • Types:

    • Union: Outputs all features from both inputs. (A ∪ B).

    • Intersection: Outputs only features common to both inputs. (A ∩ B).

    • Identity: Outputs features of input A with attributes of overlapping B.

  • Applications: Land use suitability (soil ∩ slope ∩ land cover), watershed delineation (streams ∩ contour).

• Buffer Analysis:

  • Concept: Creating zones of specified distance around a feature.

  • Generation: Specify buffer distance and attributes. Can be single or multiple buffers. Can be dissolved (merged) or non-dissolved (individual rings).

  • Applications: Riparian buffer zones, noise pollution zones around roads, service area analysis for facilities.

• Other Operations:

  • Query: Selecting features based on attribute (SQL) or location (e.g., "within").

  • Measurement: Calculating length, perimeter, area.

  • Neighborhood Analysis (Raster): Focal/statistical operations (e.g., mean of 3x3 window).

MODULE 6: APPLICATIONS OF RS & GIS

Applications in Water Resources Engineering (HIGH PRIORITY)

1. Watershed Delineation & Morphometry: Using DEM (from RS) to automatically delineate basins, streams, calculate drainage density, slope.

2. Surface Water Mapping & Monitoring: Mapping lakes, reservoirs, rivers using multispectral imagery (water absorbs NIR/SWIR). Monitoring seasonal changes.

3. Groundwater Potential Zoning: Overlay analysis of factors: lithology, lineaments (from RS), slope, drainage density, land use.

4. Flood Risk Assessment & Mapping: Flood extent mapping (using satellite images during/after flood), floodplain delineation (using DEM + hydraulic models), risk zoning (combining flood depth, land use, population).

5. Reservoir Sedimentation: Estimating silt load using multi-temporal satellite data (change in reservoir area over time).

6. Irrigation Management: Crop water requirement estimation (using LULC & crop coefficients), command area monitoring, canal network mapping.

Land Use/Land Cover (LULC) Change Detection Procedure

1. Data Acquisition: Multi-temporal satellite images (same season to avoid phenology effects).

2. Pre-processing: Atmospheric correction, geometric correction, co-registration (images must overlay perfectly).

3. Classification: Supervised/unsupervised classification for each date to generate LULC maps.

4. Post-classification Comparison: Compare classified maps. Post-classification comparison is preferred over "direct differencing" to minimize errors from atmospheric/sensor differences.

5. Change Detection: Generate change matrix (contingency table) showing transitions from Class X (date1) to Class Y (date2).

6. Accuracy Assessment: Use ground truth points to calculate overall accuracy, producer's/user's accuracy, and Kappa coefficient for each classified map.

7. Interpretation & Mapping: Analyze change patterns, quantify rates, map change hotspots.

Other Key Application Areas

• Traffic Management & Transportation: Network analysis (shortest path), traffic flow modeling, corridor planning, accident hotspot analysis, public transport routing.

• Urban Planning: Urban sprawl mapping (change detection), infrastructure planning, land suitability analysis, disaster vulnerability (earthquake, landslide).

• Disaster Management: Flood mapping, cyclone damage assessment, drought monitoring, landslide susceptibility mapping.

• Agriculture: Crop inventory, yield estimation, drought assessment, soil mapping.

• Forestry: Forest cover mapping, deforestation monitoring, biomass estimation, wildfire risk mapping.

• Mineral Exploration: Lineament/fracture mapping (structural control), lithological mapping, alteration zone identification.

[!TIP] EXAM PATTERN: For application questions (7 marks), structure answer: 1. Brief intro to application area. 2. Specific RS & GIS tools/techniques used. 3. 2-3 concrete examples/outputs. 4. Concluding benefit.

Final Preparation Checklist:

• [ ] Draw and label spectral reflectance curves for soil, vegetation, water.

• [ ] Sketch ideal RS system block diagram.

• [ ] Create comparison tables for: Orbits, Vector vs. Raster, Supervised vs. Unsupervised.

• [ ] Memorize definitions: GIS, Synoptivity, Repetitivity, Map Projection, Overlay, Buffer.

• [ ] Know UTM projection characteristics (Conformal, Transverse Mercator, 60 zones, scale factor 0.9996).

• [ ] List 6 GIS components and 4 types of resolution.

• [ ] Outline LULC change detection steps and water resources applications.

• [ ] Practice 7-mark answers with clear headings and bullet points.