Remote Sensing, GIS, Photogrammetry and GPS in Geology
CSIR NET Earth Sciences Part C geology: geological remote sensing, band ratios, photogeology, photogrammetry, GIS overlay, DEMs and GPS positioning.
- 22 min
On this page · 14 sections
Prerequisites
- The physics of remote sensing (spectrum, windows, Wien’s law, spectral signatures, orbits, resolutions, radar, InSAR, classification accuracy) is covered in the Part B page Elements of Remote Sensing. This page assumes it and goes on to geological use.
- Folds, faults and joints from Stress, Strain, Folds, Joints and Faults: photogeology is mostly the recognition of these structures from above.
- Minerals and rock types from Minerals and Rocks, since spectral features come from specific ions and bonds.
The Hook: A Bleached Hill in a Satellite Scene
Geological Remote Sensing
The aim is to map what lies at the surface (rock, soil, alteration, structure) and infer what lies below. Three kinds of information are used:
| Information | Physical basis | Typical product |
|---|---|---|
| Spectral | electronic and vibrational absorptions in minerals | band ratios, hyperspectral mineral maps |
| Spatial and textural | erosion picks out resistant and weak rocks and structures | lineament maps, drainage maps, photogeological maps |
| Topographic | stereo images, InSAR, LiDAR | DEMs, slope, structural attitude from topography |
Why minerals have spectral features
| Process | Region | Diagnostic of |
|---|---|---|
| Charge transfer and crystal-field transitions of transition metals | visible and near infrared (0.4–1.3 μm) | Fe\(^{3+}\) in haematite, goethite, jarosite (absorption near 0.9 μm and a steep fall below about 0.55 μm); Fe\(^{2+}\) broad band near 1 μm in pyroxene and olivine |
| Vibrational overtones and combinations of OH, H\(_2\)O and CO\(_3\) | shortwave infrared (1.3–2.5 μm) | Al–OH near 2.2 μm (kaolinite, muscovite or sericite, montmorillonite), Mg–OH near 2.3 μm (chlorite, talc, serpentine, epidote), carbonate near 2.33–2.35 μm (calcite, dolomite), H\(_2\)O near 1.4 and 1.9 μm |
| Fundamental Si–O stretching (reststrahlen bands) | thermal infrared (8–12 μm) | silicate composition: the emissivity minimum moves to longer wavelength from quartz-rich (felsic) to mafic and ultramafic rocks |
Quartz and feldspar are nearly featureless in the visible and SWIR; they are mapped in the thermal infrared. This is why multispectral TIR sensors matter for lithological mapping.
Band ratios
Dividing one band by another cancels topographic shading (both bands are darkened equally on a shadowed slope) and enhances absorption features.
| Ratio (Landsat TM/ETM+ numbering) | Landsat 8–9 OLI equivalent | Highlights |
|---|---|---|
| 5/7 (1.65/2.2 μm) | 6/7 | hydroxyl-bearing clays, micas, alteration (absorb at 2.2 μm, bright in band 5) |
| 3/1 (red/blue) | 4/2 | ferric iron oxides, gossans |
| 4/3 (NIR/red) | 5/4 | vegetation (the basis of NDVI) |
| 5/4 | 6/5 | ferrous iron |
A colour composite of 5/7, 3/1 and 4/3 in red, green and blue is a classic alteration-mapping product: clay-rich areas appear red, iron oxides green, vegetation blue. Landsat TM bands, for reference: 1 blue, 2 green, 3 red, 4 NIR, 5 SWIR near 1.65 μm, 6 thermal, 7 SWIR near 2.2 μm.
Multispectral sensors have only one or two SWIR bands, so they detect “clay” but cannot tell kaolinite from alunite. Sensors with several SWIR bands (ASTER) or hyperspectral imaging spectrometers (hundreds of contiguous bands) can identify individual minerals by matching absorption shapes against a spectral library. ASTER has 14 bands: three VNIR bands at 15 m (with a backward-looking band for stereo), six SWIR at 30 m and five TIR at 90 m.
Other enhancements used in geology: principal component analysis (the “Crosta technique” selects the component that loads oppositely on the two bands bracketing an absorption feature), decorrelation stretch of thermal bands, and spectral angle mapping for hyperspectral data.
Thermal remote sensing in geology
A rock’s day–night temperature range depends on its thermal inertia
$$\boxed{P = \sqrt{K\rho c}}$$(thermal conductivity \( K \), density \( \rho \), specific heat \( c \); unit J m\(^{-2}\) K\(^{-1}\) s\(^{-1/2}\)). High-inertia materials (dense rock, water, wet soil) warm and cool slowly and show a small diurnal range; low-inertia materials (dry sand, pumice, loose soil) show a large range. Comparing day and pre-dawn images separates compact outcrop from loose cover, and moist (often fault-controlled) ground from dry ground. Thermal imagery also maps geothermal areas, coal-mine fires, lava flows and submarine groundwater discharge along coasts (cool fresh water against warm sea).
Radar and LiDAR for structure
Side-looking radar at low depression angle exaggerates topographic texture, so faults, fold hinges and joint sets stand out even under forest cover. L-band radar penetrates dry sand and has revealed buried drainage channels in deserts. LiDAR (airborne laser scanning) strips away vegetation to produce bare-earth DEMs that reveal fault scarps and old landslide scars. InSAR deformation mapping is covered in the Part B page.
Photogeology: Interpreting the Landscape
Photogeology applies the elements of interpretation (tone, texture, pattern, shape, size, shadow, site, association) to geological questions. The key clues:
| Clue | What it suggests |
|---|---|
| Resistant ridges and hogbacks | sandstone, quartzite, dykes, BIF |
| Smooth, low, fine-textured ground | shale, clay, weathered schist |
| Rounded, massive hills with sheet joints and few drainage lines | granite, gneiss |
| Stepped, flat-topped hills with layered cliffs | flat-lying lava flows (Deccan Traps), horizontal sedimentary rocks |
| Sinkholes, disappearing streams | karst limestone |
| Straight, continuous lineaments offsetting units | faults, shear zones; straight dark lines may be dykes |
| Repeated or curved ridges closing in a “V” or arc | plunging folds |
| Vegetation alignments and tonal lines | water-bearing fractures |
Drainage patterns
| Pattern | Look | Geological control |
|---|---|---|
| Dendritic | tree-like, branches at acute angles | homogeneous rock or flat-lying strata; no structural control |
| Trellis | long parallel main streams with short right-angled tributaries | folded, alternating resistant and weak beds |
| Rectangular | right-angle bends in both main stream and tributaries | joints or faults at right angles |
| Radial | streams flow outward from a centre | volcanic cone, dome |
| Annular | ring-shaped segments around a centre | dissected structural dome or basin with alternating beds |
| Centripetal | streams flow inward | crater, structural or tectonic basin |
| Parallel | many parallel streams | uniform steep slope, parallel landforms |
| Deranged | irregular, with lakes and swamps | recently glaciated terrain |
Drainage density (total stream length per unit area) is high on impermeable, easily eroded rocks (shale) and low on permeable rocks (sandstone, karst, coarse alluvium), so it doubles as a rough indicator of infiltration and groundwater potential.
Lineaments
A lineament is a mappable, simple or composite linear feature of the surface whose parts are aligned in a straight or slightly curving relationship, and which probably reflects a subsurface phenomenon: faults, fracture zones, dykes, shear zones, or bedding traces. Lineament density and intersections are routinely used as evidence layers for groundwater in hard-rock terrain (fractures are the aquifer) and for mineralisation (fluids rise along structures). Not every lineament is geological: roads, canals, field boundaries and power lines must be screened out.
Photogrammetry
Photogrammetry is measurement from photographs. Its geological uses are scale, heights, slopes, dips and DEMs.
Scale and coverage
For a vertical photograph with camera focal length \( f \), flying height \( H \) above datum and terrain at height \( h \):
$$\boxed{S = \frac{f}{H - h}}$$Scale varies across a photograph wherever the terrain height varies; only a flat terrain gives a single scale. A 152 mm lens flown 2,000 m above the terrain gives \( 0.152/2000 \), that is about 1:13,200.
The ground covered by a square photograph of side \( d \) at scale \( 1{:}S_N \) is \( (d\,S_N)^2 \). A standard 23 cm frame at 1:20,000 covers \( 4.6 \) km \(\times\) \( 4.6 \) km \( = 21.16 \) km\(^2\).
Overlap, air base and flight lines
- Forward overlap about 60% gives stereo coverage: every ground point appears on at least two photographs.
- Side overlap about 20–30% between adjacent flight lines avoids gaps.
- Air base \( B \) (distance between exposure stations) \( = (1 - \text{forward overlap}) \times \) ground length of a photo.
- Flight-line spacing \( = (1 - \text{side overlap}) \times \) ground width of a photo.
For the 4.6 km frame: air base \( = 0.4\times 4.6 = 1.84 \) km; line spacing at 30% side overlap \( = 0.7\times 4.6 = 3.22 \) km.
The base–height ratio \( B/H \) controls stereo depth perception: a larger \( B/H \) increases vertical exaggeration in the stereo model, which helps the eye pick out low relief but steepens apparent dips. Dips read under a stereoscope must therefore be corrected before use.
Relief displacement
The top of an object of height \( h \) appears displaced radially outward from the principal point (nadir, for a truly vertical photo):
$$\boxed{d = \frac{r\,h}{H}} \qquad \Rightarrow \qquad h = \frac{d\,H}{r}$$where \( r \) is the radial distance of the displaced top and \( H \) the flying height above the base of the object. Relief displacement is zero at the nadir and grows towards the edges. An orthophoto is a photograph rectified with a DEM so that relief displacement is removed and scale is uniform; it can be used directly as a map.
Stereoscopic parallax
The absolute parallax of a point is the shift of its image between two overlapping photographs, measured parallel to the flight line. Higher points have larger parallax. For an object with absolute parallax \( P \) at its base and a parallax difference \( \Delta p \) between top and base:
$$\boxed{h = \frac{H\,\Delta p}{P + \Delta p}}$$With \( H = 3{,}000 \) m, \( P = 90 \) mm and \( \Delta p = 1.5 \) mm: \( h = 3000\times 1.5/91.5 \approx 49 \) m. Parallax bars measure \( \Delta p \); repeated over a grid, parallax gives contours, and modern digital photogrammetry automates this by image matching to produce DEMs. Satellite stereo (Cartosat, ASTER band 3B, SPOT) uses the same geometry.
GIS: Turning Layers into Decisions
A geographic information system captures, stores, analyses and displays georeferenced data. Geological maps, drill-hole data, geochemical samples, geophysical grids, satellite images and DEMs all become layers in one coordinate system.
Data models
| Raster | Vector |
|---|---|
| grid of cells, each with a value | points, lines, polygons with attribute tables |
| images, DEMs, gridded geophysics, interpolated geochemistry | wells, faults, contacts, roads, administrative units |
| simple overlay and map algebra; file size grows with resolution | precise boundaries, topology (connectivity, adjacency), compact |
| a 60 km \(\times\) 60 km area at 30 m has \( 2000\times 2000 = 4\times 10^6 \) cells per band | size depends on number of features |
Topology records spatial relationships (which polygons share a boundary, which lines connect), allowing checks such as “every contact line closes a polygon” and network analysis.
Coordinates and projections
All layers must share a datum (a reference ellipsoid fixed to the Earth) and a projection. GPS works on WGS 84. Older Survey of India toposheets used the Everest spheroid; mixing datums without transformation shifts features by up to hundreds of metres. The Universal Transverse Mercator (UTM) system divides the Earth into 60 zones, each 6° of longitude wide, with a scale factor of 0.9996 on the central meridian and a false easting of 500,000 m; India spans UTM zones 42 to 47. Survey of India 1:50,000 toposheets each cover 15′ \(\times\) 15′ and 1:250,000 sheets cover 1° \(\times\) 1°.
Analysis operations
| Operation | Example in geology |
|---|---|
| Buffer | zone within 500 m of mapped faults |
| Overlay (union, intersect, map algebra) | lithology \(\cap\) alteration \(\cap\) magnetic high |
| Reclassification | slope classes 0–15°, 15–30°, > 30° |
| Interpolation | inverse distance weighting or kriging of soil geochemistry, water-table depth |
| Terrain analysis of a DEM | slope, aspect, hillshade, curvature, flow direction and accumulation, watershed boundaries |
| Distance and network | haul routes, distance to drainage |
Slope from a DEM: a rise of 40 m over 100 m horizontal is a gradient of 40% and an angle of \( \tan^{-1}(0.4) = 21.8° \). Do not confuse per cent slope with degrees: 100% slope is 45°.
DEM sources include SRTM (C-band InSAR from the Space Shuttle in 2000), ASTER GDEM (optical stereo), CartoDEM (Indian Cartosat stereo) and LiDAR.
Multi-criteria overlay: prospectivity and susceptibility
The commonest Part C GIS question is a weighted index overlay: each evidence layer is reclassified into scores, each layer gets a weight, and the cell score is the weighted sum
$$\boxed{S = \sum_i w_i\,r_i \quad (\textstyle\sum w_i = 1)}$$graph LR
A[Geological map] --> E[Reclassify each layer into scores]
B[Alteration from band ratios] --> E
C[Magnetic and gravity grids] --> E
D[Fault buffers from lineaments] --> E
E --> F[Assign layer weights]
F --> G[Weighted sum per cell]
G --> H[Prospectivity map]
H --> I[Validate against known deposits]The same method gives groundwater potential maps (geomorphology, lithology, lineament density, slope, drainage density, land use) and landslide susceptibility maps (slope, lithology, distance to faults and roads, rainfall, land cover). Data-driven alternatives compute weights from known occurrences: weights of evidence (Bayesian), logistic regression, fuzzy logic and machine learning. Validation against occurrences not used in training is essential.
GPS and GNSS
How a receiver finds its position
Each satellite broadcasts its orbit (ephemeris) and a time signal. The receiver measures the signal’s travel time \( \Delta t \) and multiplies by \( c \) to get a pseudorange (pseudo, because the cheap receiver clock is not synchronised with the satellites’ atomic clocks). Four unknowns (\( x, y, z \) and the receiver clock offset) need at least four satellites.
Scale of the problem: a travel time of 0.0700 s corresponds to about 20,985 km; a clock error of just 1 μs is a range error of about 300 m. That is why the clock offset must be solved for, not ignored.
The system
| Segment | Content |
|---|---|
| Space | nominally 24 GPS satellites in six orbital planes inclined at 55°, at about 20,200 km altitude, period about 11 h 58 min (half a sidereal day) |
| Control | ground monitoring stations and a master control station that update orbits and clocks |
| User | receivers: navigation, mapping, survey-grade geodetic |
Kepler’s third law checks the period: \( a = 6{,}371 + 20{,}200 = 26{,}571 \) km gives \( T = 2\pi\sqrt{a^3/GM} \approx 43{,}100 \) s \( \approx 11.97 \) h.
GPS broadcasts on L1 (1575.42 MHz), L2 (1227.60 MHz) and L5 (1176.45 MHz). Other GNSS: GLONASS (Russia), Galileo (Europe), BeiDou (China). India’s regional system NavIC (IRNSS) was designed with seven satellites, three geostationary and four in inclined geosynchronous orbits, covering India and the surrounding region. GAGAN is India’s satellite-based augmentation system for GPS, used in aviation.
Error sources and accuracy
| Source | Note | Remedy |
|---|---|---|
| Ionospheric delay | usually the largest; depends on electron content and is frequency dependent (dispersive) | dual-frequency receivers cancel it; models |
| Tropospheric delay | water vapour and dry air; not dispersive | models, estimation |
| Satellite clock and ephemeris errors | small residuals after control-segment updates | precise products, differential methods |
| Multipath | signal reflected from buildings, cliffs, water | antenna siting, choke-ring antennas |
| Receiver noise | — | carrier-phase measurement |
| Geometry (DOP) | poor satellite spread multiplies all errors | wait for better geometry |
Positional error \( \approx \) DOP \(\times\) user range error. With PDOP 2.5 and a 4 m range error, expect about 10 m. Selective Availability, a deliberate degradation of civilian accuracy, was switched off in 2000.
| Technique | Typical accuracy | Use in geology |
|---|---|---|
| Single-receiver code positioning | metres | field mapping, sample location |
| Differential GPS (DGPS) | sub-metre to metre | GIS data capture, geophysical station location |
| RTK (real-time kinematic, carrier phase with a nearby base) | centimetres | surveying, mine planning, landslide monitoring |
| Static geodetic (long sessions, precise orbits) | millimetres | plate motions, interseismic strain, volcano and landslide deformation |
Continuous GPS stations measure plate velocities directly: a station that moves 0.25 m in 5 years has a velocity of 50 mm yr\(^{-1}\). Networks across the Himalaya and peninsular India constrain how much convergence is stored as strain for future earthquakes.
Heights: ellipsoid versus geoid
GPS gives ellipsoidal height \( h \) above WGS 84. Maps and engineering use orthometric height \( H \) above the geoid (mean sea level). They differ by the geoid undulation \( N \):
$$\boxed{h = H + N}$$If GPS reads \( h = 85.0 \) m where \( N = -92.0 \) m, then \( H = 85 - (-92) = 177 \) m. The sign matters: where the geoid lies below the ellipsoid (\( N < 0 \)), orthometric height exceeds ellipsoidal height.
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Band ratio: cancels shading, enhances absorption | Band difference (as in NDVI numerator): keeps illumination effects unless normalised | |
| Multispectral: detects “clay” | Hyperspectral: identifies kaolinite vs alunite vs dickite | |
| Relief displacement: radial, from height, zero at nadir | Tilt displacement: from camera tilt, about the isocentre | |
| Absolute parallax: shift of a point between photos | Parallax difference: top minus base, gives height | |
| Orthophoto: relief displacement removed | Raw aerial photo: varying scale | |
| Raster: cells, map algebra | Vector: points, lines, polygons, topology | |
| Ellipsoidal height (GPS) | Orthometric height (map, mean sea level) | |
| Datum (ellipsoid + origin) | Projection (how the curved surface is flattened) | |
| DGPS: code corrections, sub-metre | RTK: carrier phase, centimetres |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- Which absorption feature identifies kaolinite and sericite in SWIR data?
- Which drainage pattern suggests a dissected structural dome with alternating hard and soft beds?
- What is the minimum number of GPS satellites needed for a 3D fix with an uncorrected receiver clock?
Answers
- The Al–OH absorption near 2.2 μm (Landsat TM band 7 region), which makes the 5/7 ratio bright.
- Annular. (Radial is for a cone or dome without the ring-shaped strike valleys.)
- Four: three coordinates plus the clock offset.
Level 2: Part B Style
Practice Q1. A vertical aerial photograph is taken with a 152 mm lens from 2,500 m above mean sea level over a plateau at 500 m. The photo scale on the plateau is closest to:
(a) 1:16,400 (b) 1:13,200 (c) 1:19,700 (d) 1:3,300
Answer: (b) 1:13,200
\( S = f/(H - h) = 0.152/2000 = 1/13{,}158 \). (a) uses \( H = 2{,}500 \) m (ignores terrain height); (c) uses \( H + h = 3{,}000 \) m.
Practice Q2. On a vertical photograph flown 1,900 m above the base of a chimney, the top of the chimney appears 95 mm from the principal point and is displaced 2.5 mm relative to its base. The chimney height is:
(a) 25 m (b) 50 m (c) 72 m (d) 95 m
Answer: (b) 50 m
\( h = dH/r = 2.5\times 1900/95 = 50 \) m.
Practice Q3. In a Landsat TM band-ratio composite (5/7, 3/1, 4/3 as R, G, B), an area of argillic alteration with abundant limonite and little vegetation would appear:
(a) blue (b) red–green (yellowish) (c) dark (d) pure blue–green (cyan)
Answer: (b)
Clays give high 5/7 (red) and limonite gives high 3/1 (green); low vegetation keeps 4/3 (blue) low. Red plus green appears yellowish. Cyan would need green and blue: iron oxide plus vegetation.
Practice Q4. A handheld GPS reads an ellipsoidal height of 85.0 m at a site where the geoid undulation is −92.0 m. The orthometric (mean sea level) height is:
(a) −7 m (b) 7 m (c) 177 m (d) 92 m
Answer: (c) 177 m
\( H = h - N = 85 - (-92) = 177 \) m. (a) adds the undulation with the wrong sign.
Level 3: Part C Style
Practice Q5. Aerial photography at 1:20,000 is planned with a 23 cm \(\times\) 23 cm format, 60% forward overlap and 30% side overlap. Which statements are correct?
- Each photograph covers 21.16 km\(^2\) of ground.
- The air base is 1.84 km.
- Flight lines are 3.22 km apart.
- Increasing forward overlap to 80% would increase the air base.
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
- 1 true: side on the ground \( = 0.23\times 20{,}000 = 4{,}600 \) m; area \( = 4.6^2 = 21.16 \) km\(^2\).
- 2 true: \( B = (1 - 0.6)\times 4.6 = 1.84 \) km.
- 3 true: spacing \( = (1 - 0.3)\times 4.6 = 3.22 \) km.
- 4 false: at 80% overlap \( B = 0.2\times 4.6 = 0.92 \) km; more overlap means a shorter air base and more photographs.
Practice Q6. In a stereo pair flown 3,000 m above the base of a cliff, the absolute parallax at the base is 90.0 mm and the parallax difference between top and base is 1.5 mm. A student also notes that the same cliff shows a relief displacement of 1.6 mm at 96 mm from the principal point on one photograph. Which statement is correct?
(a) Parallax gives about 49 m and relief displacement about 50 m; the two agree within measurement error (b) Parallax gives about 50 m but relief displacement gives about 5 m (c) Parallax gives about 4.9 m; the student has misplaced a decimal (d) The two methods cannot be compared because relief displacement needs the focal length
Answer: (a)
Parallax: \( h = H\Delta p/(P + \Delta p) = 3000\times 1.5/91.5 = 49.2 \) m.
Relief displacement: \( h = dH/r = 1.6\times 3000/96 = 50.0 \) m.
The two independent estimates agree within about 2%. (d) is wrong: neither formula needs \( f \) once \( H \) is known, because the ratios \( d/r \) and \( \Delta p/P \) are dimensionless.
Practice Q7. A GIS prospectivity model uses four layers with weights: lithology 0.4, alteration 0.3, magnetic anomaly 0.2, distance to fault 0.1. Scores (1–5) for three cells are:
| Cell | Lithology | Alteration | Magnetic | Fault |
|---|---|---|---|---|
| A | 4 | 3 | 5 | 2 |
| B | 3 | 5 | 2 | 4 |
| C | 5 | 3 | 2 | 1 |
Which statements are correct?
- Cell A has the highest score, 3.7.
- Cell B scores 3.5 and cell C 3.4.
- If the weights of lithology and alteration were swapped, cell B would rank first.
- Weighted overlay needs all layers in the same coordinate reference system and the same scoring scale.
(a) 1 and 2 only (b) 1, 2 and 4 only (c) 1, 2, 3 and 4 (d) 3 and 4 only
Answer: (c) 1, 2, 3 and 4
Original weights:
- A: \( 0.4(4) + 0.3(3) + 0.2(5) + 0.1(2) = 1.6 + 0.9 + 1.0 + 0.2 = 3.7 \)
- B: \( 1.2 + 1.5 + 0.4 + 0.4 = 3.5 \)
- C: \( 2.0 + 0.9 + 0.4 + 0.1 = 3.4 \)
Swapped (lithology 0.3, alteration 0.4):
- A: \( 1.2 + 1.2 + 1.0 + 0.2 = 3.6 \)
- B: \( 0.9 + 2.0 + 0.4 + 0.4 = 3.7 \)
- C: \( 1.5 + 1.2 + 0.4 + 0.1 = 3.2 \)
B now ranks first, so 3 is true. Statement 4 is a requirement of any overlay. The sensitivity of the ranking to the weights is why expert-assigned weights must be validated against known deposits.
Quick Revision Table
| Idea | Fact or formula |
|---|---|
| VNIR features | Fe\(^{3+}\) near 0.9 μm and below 0.55 μm; Fe\(^{2+}\) near 1 μm |
| SWIR features | Al–OH 2.2 μm; Mg–OH and CO\(_3\) 2.3 μm; H\(_2\)O 1.4, 1.9 μm |
| TIR | Si–O reststrahlen 8–12 μm; minimum shifts longer from felsic to mafic |
| TM ratios | 5/7 clays; 3/1 iron oxides; 4/3 vegetation (OLI: 6/7, 4/2, 5/4) |
| ASTER | 3 VNIR (15 m), 6 SWIR (30 m), 5 TIR (90 m); band 3B stereo |
| Thermal inertia | \( P = \sqrt{K\rho c} \); high P, small day–night range |
| Drainage | dendritic none; trellis folds; rectangular joints; radial cones; annular domes; centripetal basins |
| Scale | \( f/(H - h) \) |
| Coverage | \( (d\,S_N)^2 \); 23 cm at 1:20,000 = 21.16 km\(^2\) |
| Air base, spacing | \( (1 - p)L \), \( (1 - q)W \) |
| Relief displacement | \( d = rh/H \), radial, zero at nadir |
| Parallax height | \( h = H\Delta p/(P + \Delta p) \) |
| GIS | raster vs vector; buffer, overlay, interpolation, DEM terrain analysis |
| UTM | 60 zones of 6°; k = 0.9996; false easting 500,000 m; India zones 42–47 |
| Weighted overlay | \( S = \sum w_i r_i \) |
| GPS | ≥ 4 satellites; about 20,200 km; about 12 h; L1, L2, L5 |
| Clock error | 1 μs ≈ 300 m |
| Error | ionosphere largest (dual frequency removes); multipath; DOP × range error |
| Accuracy | code m; DGPS sub-m; RTK cm; geodetic mm |
| Heights | \( h = H + N \) |
Key Takeaways
What’s Next
- Engineering Geology: Dams, Tunnels, Landslides and Seismic Design: photogeology and GIS feed site investigation and landslide zonation.
- Mineral Exploration: Methods, Sampling and Evaluation: where alteration maps become drill targets.
- Natural Hazards: the hazards that InSAR, GPS and susceptibility mapping monitor.
- Back to the Applied Geology overview.