UNIT 2: REMOTE SENSING & GIS FOR URBAN & TOWN PLANNING

(Exam-Focused Short Notes)

1.0 FUNDAMENTALS OF REMOTE SENSING

1.1 Spectral Reflectance Characteristics

• Definition: The fraction of incident electromagnetic energy reflected by a surface, varying with wavelength.

• Typical Reflectance Curves:

  • Vegetation: Low reflectance in blue/red (absorption by chlorophyll), high in NIR (cell structure), very low in SWIR (water absorption).

  • Soil: Generally increasing curve from visible to SWIR; influenced by moisture, organic matter, texture.

  • Water: Low reflectance in visible/NIR (absorption), high in SWIR for turbid water.

  [!TIP] Exam Focus: Be ready to sketch and label these curves. Key absorption features: chlorophyll (0.45µm, 0.67µm), water (1.4µm, 1.9µm, 2.7µm).

• Factors Influencing Spectral Signatures:

  • Intrinsic: Chemical composition, moisture content, surface roughness.

  • Extrinsic: Sun angle, atmospheric conditions, sensor characteristics, viewing geometry.

1.2 Ideal Remote Sensing System

• Components: Energy Source → Atmosphere → Target → Sensor → Processing → User.

• Functional Requirements: Adequate spatial, spectral, temporal, and radiometric resolution; geometric accuracy; calibrated sensors.

• Block Diagram:

  
  [Sun/Active Source] → [Atmosphere] → [Earth Surface] → [Sensor] → [Ground Station] → [Processing] → [Information Product]
  
  

  [!TIP] Common Pitfall: Students often forget the "Atmosphere" as an interactive component, not just a barrier.

1.3 Electromagnetic Spectrum & Energy Interactions

• Wavelength Regions Used:

  | Region | Wavelength (µm) | Primary Use | |--------|----------------|-------------| | Visible | 0.4 - 0.7 | True-color imaging, basic land cover | | Near-IR (NIR) | 0.7 - 1.3 | Vegetation vigor, water boundary | | Shortwave-IR (SWIR) | 1.3 - 3.0 | Soil moisture, geology, burnt areas | | Thermal-IR | 3.0 - 14.0 | Surface temperature, urban heat islands | | Microwave | 0.1 - 100 cm | All-weather, soil moisture, topography |

• Atmosphere-Terrain Interactions: Absorption (by gases), scattering (Rayleigh, Mie), transmission windows.

2.0 REMOTE SENSING PLATFORMS & SENSORS

2.1 Satellite Orbits & Types

2.2 Sensor Resolutions

1. Spatial: Minimum separable distance on ground (e.g., 30m for Landsat). Defines detail level.

2. Spectral: Number and width of wavelength bands (e.g., multispectral vs. hyperspectral).

3. Temporal: Revisit period (time between observations of same area). Depends on orbit & sensor FOV.

4. Radiometric: Number of brightness levels (e.g., 8-bit = 256 levels). Defines sensitivity to energy differences.

2.3 Types of Earth Observation Satellites

• Earth Resource: Landsat (US), Sentinel-2 (EU), Resourcesat (India). Multispectral, moderate resolution.

• Weather: NOAA (US), INSAT (India). Geostationary, frequent temporal resolution.

• Indian Satellite Program (IRS Series):

  • Resourcesat-2/2A: Advanced LISS-III (23.5m), AWiFS (56m).

  • Cartosat: High-resolution panchromatic (≤1m) for mapping.

  • RISAT: Radar imaging (SAR) for all-weather observation.

• Case Study: Chandrayaan-3 (Lunar Mission):

  • Objectives: Safe & soft landing on Moon's south pole, rover operations, in-situ science.

  • Salient Features: Lander (Vikram), Rover (Pragyan), propulsion module; instruments for lunar surface analysis.

2.4 Platforms Comparison

3.0 IMAGE PROCESSING & CLASSIFICATION

3.1 Image Classification Approaches

3.2 Digital vs. Visual Image Analysis

3.3 Visual Interpretation Techniques

• Elements of Interpretation:

  • Tone/Color: Relative brightness/color.

  • Texture: Roughness/smoothness (e.g., residential = medium texture).

  • Pattern: Spatial arrangement (e.g., grid = agricultural, organic = informal settlements).

  • Shape: Geometric form (e.g., rectangular = fields/plots, irregular = natural).

  • Size: Relative dimension.

  • Shadow: Indicates height/object type.

  • Association: Relationship with other features.

• Stereoscope: Used with stereo pairs (overlapping images) for 3D perception → essential for topography, building height estimation.

3.4 Image Enhancement & Filtering

• Purpose: Improve visual interpretability or prepare for classification.

• Common Filters:

  • Smoothing (Low-pass): Reduce noise (e.g., mean filter).

  • Edge Enhancement (High-pass): Sharpen boundaries (e.g., Laplacian, Sobel).

  • Contrast Stretching: Expand histogram to use full dynamic range.

4.0 ERRORS & QUALITY IN REMOTE SENSING DATA

4.1 Systematic Errors

• Sensor-related: Calibration drift, detector non-uniformity.

• Atmospheric: Scattering, absorption (corrected using models like MODTRAN, dark object subtraction).

• Geometric: Platform instability, Earth rotation, terrain relief → corrected via GCPs (Ground Control Points) and orthorectification.

4.2 Non-Systematic (Random) Errors

• Sources: Random sensor noise, mixed pixels (partial coverage of multiple materials).

• Mitigation: Filtering (smoothing), using higher resolution data, fuzzy classification.

4.3 Data Quality Parameters

• Accuracy: Closeness to true value (assessed via ground truth/confusion matrix).

• Precision: Repeatability of measurement.

• Reliability: Consistency and trustworthiness for decision-making.

5.0 GEOGRAPHIC INFORMATION SYSTEM (GIS) FUNDAMENTALS

5.1 Definition, Objectives & Components

• Definition: A computer-based system for capturing, storing, analyzing, managing, and presenting spatial (geographic) data.

• Objectives: Support decision-making, spatial query & analysis, integrate diverse data sources, visualize scenarios.

• Core Components (5-P Model):

  1. Hardware: Computer, GPS, digitizer, plotter.

  2. Software: ArcGIS, QGIS, GRASS.

  3. Data: Spatial (maps, imagery) + Attribute (tables).

  4. People: Skilled users, managers, technicians.

  5. Procedures: Workflows, data standards, analysis models.

5.2 Data Sources & Collection Objectives

• Primary Data: Collected via GPS, ground survey, digitization.

• Secondary Data: Existing maps, satellite imagery, census data, government databases.

• Objectives in Urban Planning:

  • Inventory of infrastructure (roads, utilities).

  • Demographic and socioeconomic data integration.

  • Land ownership/parcel mapping.

  • Environmental baseline (slope, flood zones).

5.3 Integration of Spatial and Attribute Data

• How Linked: Unique Feature ID (e.g., Parcel_ID) connects spatial feature (polygon) to its attribute record in a table.

• Relational Database Management (RDBMS): Tables linked via keys. Allows complex queries (e.g., "Find all residential parcels >500 sqm with value <₹50 lakh").

6.0 SPATIAL DATA MODELS & STRUCTURES

6.1 Vector Data Model

• Features:

  • Point: Zero dimension (e.g., well, lamp post).

  • Line: 1D (e.g., roads, rivers).

  • Polygon: 2D (e.g., buildings, lakes).

• Topology: Rules defining spatial relationships (connectivity, adjacency, containment). Ensures data integrity (e.g., no gaps between parcels).

• Advantages: Compact storage, high accuracy, excellent for discrete features, maintains topology.

• Limitations: Complex for continuous surfaces (e.g., elevation), overlay analysis can be computationally intensive.

6.2 Raster Data Model

• Structure: Grid of cells (pixels) with values.

• Resolution: Cell size (e.g., 30m × 30m). Determines detail and file size.

• Advantages: Simple structure, easy for continuous data (e.g., DEM, temperature), fast overlay (map algebra).

• Limitations: Large file size for high resolution, less precise boundaries, topology not inherent.

6.3 Vector vs. Raster – Detailed Comparison

6.4 Data Conversion

• Raster to Vector: Vectorization (tracing edges). Tools: edge detection, thinning, polygonization. Challenge: Choosing appropriate threshold.

• Vector to Raster: Rasterization. Assign cell value based on feature (e.g., majority rule for polygons). Challenge: Choosing appropriate cell size to avoid information loss.

7.0 COORDINATE SYSTEMS & MAP PROJECTIONS

7.1 Coordinate Systems

• Geographic (Lat/Long): Angular measurements on sphere/ellipsoid. Units: degrees.

• Projected (e.g., UTM, State Plane): Cartesian (x,y) on a plane. Units: meters/feet.

  [!TIP] Key Point: GIS analysis requires projected coordinates for accurate distance/area calculations.

7.2 Map Projections

• Why Necessary? Earth is 3D, maps are 2D → distortion inevitable.

• Distortion Trade-offs:

  • Conformal: Preserves shape (angles) → e.g., Mercator. Distorts area.

  • Equal-Area: Preserves area → e.g., Albers. Distorts shape.

  • Equidistant: Preserves distance from one/two points.

• Universal Transverse Mercator (UTM):

  • System: 60 zones (6° wide), each with central meridian.

  • Features: Conformal, uses metric system, minimal distortion within zone.

  • Use in GIS: Standard for large-scale urban mapping within a zone. False easting/northing to avoid negatives.

7.4 Datum and Georeferencing

• Datum: Mathematical model of Earth's shape + origin point.

  • Horizontal: WGS84 (global), NAD83 (North America).

  • Vertical: Mean Sea Level (MSL) based.

• Georeferencing Raster Process:

  1. Identify Ground Control Points (GCPs) with known coordinates.

  2. Assign coordinates to GCPs in image space.

  3. Choose transformation (affine, polynomial).

  4. Resample image (nearest neighbor, bilinear, cubic convolution) to create new georeferenced raster.

  \boxed{\text{Georeferencing} = \text{Assigning real-world coordinates to an image}}

8.0 GIS DATA INPUT & MANAGEMENT

8.1 Digitization & Creating Shapefiles

1. Manual Digitizing: Tracing features from a hardcopy map on a digitizing tablet → creates vector data.

2. Scanning & Automated Vectorization:

   • Scan map → raster image.

   • Edge detection → Thinning → Vectorization (tracing).

3. Attribute Data Attachment: After creating geometry (shapefile), join table via Join operation using common key field.

8.2 Data Quality & Errors

• Topological Errors: Dangling nodes, undershoots, overlaps, sliver polygons.

• Attribute Errors: Missing values, incorrect data types, inconsistent coding (e.g., "Residential" vs "Res").

• Prevention: Data validation rules, topology checks, standardized attribute domains.

8.3 Metadata

• Definition: "Data about the data." Essential for data discovery and usability.

• Content (FGDC/ISO standards):

  • Identification (title, date, extent)

  • Data quality (lineage, accuracy)

  • Spatial reference (projection, datum)

  • Attribute information (field names, definitions, units)

  • Distribution info

9.0 SPATIAL ANALYSIS TECHNIQUES

9.1 Overlay Analysis

• Types:

  • Point-in-Polygon: Assigns polygon attributes to points (e.g., which parcel contains a well?).

  • Line-in-Polygon: Assigns polygon attributes to lines (e.g., road segment's land use zone).

  • Polygon Overlay (Union/Intersect): Combines two polygon layers → creates new polygons with combined attributes.

• Boolean Overlay: Uses logical operators (AND, OR, XOR) on binary suitability maps.

• Weighted Overlay: Assigns weights to criteria, scales values, multiplies, sums → suitability map.

9.2 Buffer Analysis

• Process: Create zone around a feature at specified distance.

• Urban Planning Applications:

  • Service Area: 500m buffer around fire station → coverage area.

  • Impact Zone: 100m buffer from highway → noise pollution study.

  • Environmental: Buffer around wetlands → development restriction zone.

9.3 Other Spatial Operations

• Spatial Query: Select features based on location (e.g., "all buildings within flood zone") AND/OR attribute (e.g., "all parcels > 1 acre").

• Proximity Analysis: Find nearest facility (e.g., hospital to residential cluster).

• Network Analysis: Route optimization, service area (using network datasets with connectivity). Brief intro: requires topologically correct road network with travel costs.

10.0 INTEGRATION OF REMOTE SENSING & GIS

10.1 Data Integration Workflow

1. Pre-processing of RS Data: Radiometric & atmospheric correction, geometric correction (orthorectification).

2. Importing: Load georeferenced imagery into GIS as raster layer.

3. Mosaicking: Stitch multiple adjacent scenes into single seamless image.

4. Classification: Perform supervised/unsupervised classification on imagery → creates thematic raster layer (e.g., LULC).

5. Vectorization (if needed): Convert classified raster boundaries to vector polygons for attribute editing/analysis.

6. Integration: Overlay classified LULC vector with other GIS layers (zoning, infrastructure).

10.2 Problems in Using RS Data in GIS

11.0 APPLICATIONS IN URBAN & REGIONAL PLANNING

11.1 Land Use/Land Cover (LULC) Change Assessment

• Workflow:

  1. Acquire multi-temporal satellite images (e.g., 2000, 2010, 2020).

  2. Pre-process (atmospheric, geometric correction).

  3. Classify each image into categories (e.g., Built-up, Agriculture, Water, Vegetation, Barren).

  4. Post-classification comparison: Detect changes by overlaying classified maps → change detection matrix.

  5. Model urban sprawl (e.g., using CA-Markov, SLEUTH).

• Key Metrics: Rate of built-up area expansion, loss of agricultural land, fragmentation indices.

11.2 Water Resources Management

• Watershed Delineation: Using Digital Elevation Model (DEM) → flow direction → flow accumulation → stream network → watershed boundaries.

• Surface Water Mapping: Use NIR band (water absorbs NIR → dark in false-color composite). Change detection for reservoir/lake shrinkage.

• Flood Risk Zonation: Overlay factors (slope from DEM, land use, drainage network, rainfall) using weighted overlay → flood susceptibility map.

11.3 Traffic & Transportation Planning

• Traffic Flow Mapping: Use high-resolution temporal RS data (e.g., from Google Traffic, which uses GPS probes) to map congestion patterns.

• Network Analysis in GIS:

  • Route Optimization: Shortest/fastest path for emergency services.

  • Service Area: Define catchment area for public transit stops.

  • Location-Allocation: Optimal site for new facility (e.g., bus depot).

11.4 Other Urban Applications

• Urban Heat Island (UHI): Use Thermal IR bands to map surface temperature → identify hotspots. Correlate with LULC (high built-up = high temperature).

• Green Space Mapping: Classify vegetation (NDVI from NIR/Red bands) → assess per capita green space, urban forestry planning.

• Disaster Management:

  • Earthquake: Map liquefaction susceptibility (soil type, water table, slope).

  • Landslide: Slope + geology + land use + rainfall → vulnerability zonation.

12.0 ADVANCED TOPICS & CURRENT TRENDS

12.1 Synoptivity and Repetitivity

• Synoptivity: Ability to capture a wide area in a single image → essential for regional studies.

• Repetitivity: Regular revisit cycle → critical for temporal monitoring (e.g., crop growth, urban expansion).

• Importance: Enables change detection, trend analysis, and timely intervention in planning.

12.2 High-Resolution Satellite Data

• Sensors: WorldView (0.3m pan), GeoEye, Pleiades, Cartosat-3 (0.25m).

• Applications in Urban Mapping:

  • Detailed building footprint extraction.

  • Road network mapping at street level.

  • Individual tree identification in urban forestry.

  • Informal settlement delineation.

12.3 LiDAR and Photogrammetry

• LiDAR: Active sensor (laser pulses) → highly accurate 3D point cloud → Digital Surface Model (DSM), Digital Terrain Model (DTM).

• Photogrammetry: From overlapping aerial photos → 3D models, DSMs.

• Integration with GIS: Provide elevation data for:

  • 3D city modeling (for visualization, solar potential, view shed analysis).

  • Flood modeling (accurate terrain).

  • Volume calculations (cut/fill for construction).

12.4 Web GIS and Cloud Platforms

• Platforms: ArcGIS Online, Google Earth Engine, QGIS Cloud.

• Benefits for Collaborative Planning:

  • Share maps/data via web browsers/ apps.

  • Real-time data integration (sensors, crowdsourcing).

  • Public participation portals (e.g., comment on proposed plans).

  • Scalable processing (cloud computing for big RS data like Landsat time series).

\boxed{\text{UNIT 2 CORE THEME: Integration of RS (data acquisition) & GIS (analysis) for evidence-based urban planning.}}