1 Concepts and foundations of Remote SensingChapter 1 Concepts and foundations of Remote Sensing Introduction to Remote Sensing Instructor: Dr. Cheng-Chien Liu Department of Earth Science National Cheng-Kung University

2 1.1 Introduction General definition of Remote Sensing:The Science and art of obtaining information about an object, area, or phenomenon through the analysis of data acquired by a device that is not in contact with the object, area, or phenomenon under investigation. e.g. reading process word  eyes  brain  meaning data  sensor  processing  information

3 1.1 Introduction (cont.) Collected data can be of many forms:variations in force distribution  e.g. gravity meter acoustic wave distribution  e.g. sonar electromagnetic energy distribution  e.g. eyes our focus: electromagnetic energy distribution

4 1.1 Introduction (cont.) Fig. 1.1 Generalized processes and elements involved in electromagnetic remote sensing of earth resources. data acquisition: a-f (§1.2 - §1.5) data analysis: g-i (§1.6 - §1.10)

5 1.2 Energy sources and radiation principlesFig. 1.3 electromagnetic spectrum  memorize Wave theory: c = nl c : speed of light (3x108 m/s) n : frequency (cycle per second, Hz) l : wavelength (m) unit: micrometer mm = 10-6 m

6 1.2 Energy sources and radiation principles (cont.)Fig. 1.3 (cont.) Spectrum : UV (ultraviolet) Vis (visible) narrow range, strongest, most sensitive to human eyes blue: 0.4~0.5mm green: 0.5~0.6mm red: 0.6~0.7mm IR (infrared) near-IR: 0.7~1.3 mm mid-IP: 1.3~3.0 mm thermal-IR: 3.0 mm~1mm  heat sensation microwave: 1mm~1m

7 1.2 Energy sources and radiation principles (cont.)Fig. 1.3 (cont.) Particle theory: Q = hn Q: quantum energy (Joule) h: Planck's constant (6.626x10-34 J sec) n: frequency Q = hn = hc/l  1/l implication in remote sensing:lQ  viewing areaenough area

8 1.2 Energy sources and radiation principles (cont.)Stefan-Boltzmann law: M = sT4 M: total radiant exitance from the surface of a material (watts m-2) s: Stefan-Boltzmann constant (5.6697x10-8 W m-2K-4) T: absolute temperature (K) of the emitting material blackbody: a hypothetical, ideal radiator totally absorbs and reemits all incident energy

9 1.2 Energy sources and radiation principles (cont.)Fig 1.4: Spectral distribution of energy radiated from blackbodies of various temperatures Area  total radiant exitance M T M (graphical illustration of S-B law) Wien's displacement law: lm=A/T  1/T lm : dominant wavelength, wavelength of maximum spectral radiant (mm) A: 2898 (K) T: absolute temperature (K) of the emitting material e.g. heating iron: dull red  orange  yellow  white

10 1.2 Energy sources and radiation principles (cont.)Fig 1.4 (cont.) Sun: T6000K  lm0.5mm (visible light) incandescent lamp: T  3000K  lm  1mm "outdoor" file used indoors  "yellowish“ need high blue energy flash  compensate Earth: T  300K  lm 9.7mm  thermal energy  radiometer l<3mm: reflected energy predominates l>3mm: emitted energy prevails Passive Active

11 1.3 Energy interaction in the atmospherePath length space photography: 2 atmospheric thickness airborne thermal sensor: very thin path length sensor-by sensor

12 1.3 Energy interaction in the atmosphere (cont.)Scattering molecular scale: d << l  Rayleigh scatter Rayleigh scatter effect  1/l4 "blue sky" and "golden sunset" Rayleigh  "haze" imagery  filter (Chapter 2) wavelength scale: d  l  Mie scatter influence longer wavelength dominated in slightly overcast sky large scale: d >> l e.g. water drop nonselective scatter  f(l) that's why fog and clod appear white why dark clouds black?

13 1.3 Energy interaction in the atmosphere (cont.)absorption absorbers in the atmosphere: water vapor, carbon dioxide, ozone Fig 1.5: Spectral characteristics of (a) energy sources (b) atmospheric effect (c) sensing systems atmospheric windows

14 1.3 Energy interaction in the atmosphere (cont.)important considerations sensor: spectral sensitivity and availability windows: in the spectral range  sense source: magnitude, spectral composition

15 1.4 Energy interactions with earth surface featuresFig 1.6: basic interactions between incident electromagnetic energy and an earth surface feature EI(l) = ER(l) + EA(l) + ET(l) incident = reflected + absorbed + transmitted ER = ER(feature, l)  distinguish features  R.S. in visible portion: ER(l)  color most R.S.  reflected energy predominated  ER important!

16 1.4 Energy interactions with earth surface features (cont.)Fig. 1.7: Specular versus diffuse reflectance specular  diffuse (Lambertian) surface roughness  incident wavelength: lI if lI << surface height variations  diffuse for R.S.  measure diffuse reflectance spectral reflectance

17 1.4 Energy interactions with earth surface features (cont.)Fig 1.8: Spectral reflectance curve (SRC) object type  ribbon (envelope) rather than a single line characteristics of SRC  choose wavelength characteristics of SRC  choose sensor near-IR photograph does a good job (Fig 1.9) Many R.S. data analysis  mapping  spectrally separable  understand the spectral characteristics

18 1.4 Energy interactions with earth surface features (cont.)Fig 1.10: Typical SRC for vegetation, soil and water average curves vegetation: pigment  chlorophyll  two valleys (0.45mm: blue; o.67mm: red)  green if yellow leaves  r(red)   green + red from 0.7 mm to 1.3 mm  minimum absorption (< 5%)  strong reflectance = f(internal structure of leaves)  discriminate species and detect vegetation stress l > 1.3 mm  three water absorption bands (1.4, 1.9 and 2.7 mm) water content  r(l)  r(l) = f(water content, leaf thickness)

19 1.4 Energy interactions with earth surface features (cont.)Fig 1.10 (cont.) soil moisture content  r(lwab)  soil texture: coarse  drain  moisture  surface roughness  r  iron oxide, organic matter  r  These are complex and interrelated variables

20 1.4 Energy interactions with earth surface features (cont.)Fig 1.10 (cont.) water near-IR: water r(lnear-IR)  visible: very complex and interrelated surface bottom material in the water clear water ® blue chlorophyll ® green CDOM ® yellow pH, [O2], salinity, ...  (indirect) R.S.

21 1.4 Energy interactions with earth surface features (cont.)Spectral Response Pattern spectrally separable  recognize feature spectral signatures  absolute, unique reflectance, emittance, radiation measurements, ... response patterns  quantitative, distinctive variability exists! identify feature types spectrally  variability causes problems identify the condition of various objects of the same type  we have to rely on these variabilities

22 1.4 Energy interactions with earth surface features (cont.)Spectral Response Pattern (cont.) minimize unwanted spectral variability maximize variability when required! spatial effect: e.g. different species of plant temporal effect: e.g. growth of plant  change detection

23 1.4 Energy interactions with earth surface features (cont.)Atmospheric influences on spectral response patterns sensor-by-sensor mathematical expression: r: reflectance E: incident irradiance T: atmospheric transmission Lp: path radiance E = Edir + Edif E = E(t)

24 1.5 Data acquisition and interpretationdetection photograph  chemical reaction simple and inexpensive high spatial resolution and geometric integrity detect and record electronic  energy variation broader spectral range of sensitivity improved calibration potential electronically transmit data record on other media (e.g. magnetic tape) photograph  image

25 1.5 Data acquisition and interpretation (cont.)data interpretation pictorial (image) analysis human mind  visual interpretation  judgment disadvantages: extensive training limitation of human eyes ® not fully evaluate spectral characteristics digital data analysis: digital image  2-D array of pixels digital number (DN) A-D signal conversion Fig 1.13: input voltage (V), sampling interval (DT), output integer DN range:8-bit: 0~255, 10-bit: 0~1023 easier for automatic processing, but limited in spectral pattern interpretation

26 1.6 Reference data R.S. needs some form of reference data Purposes:Analysis and interpretation calibration verification

27 1.6 Reference data (cont.) Collecting reference datashould be according to principles of statistical sampling design expensive and time consuming time-critical time-stable

28 1.6 Reference data (cont.) Collecting reference data (cont.)ground-based measurement principle of spectroscipy spectroradiometer  spectral reflectance curves (continuous) laboratory spectroscopy in-situ field measurement  preferred! four modes of operation: hand held, telescoping boom, helicopter, aircraft multiband radiometer (discrete) three-step process: calibration  known, stable reflectance measurement  reflected radiation computation  reflectance factor Lambertian surface bidirectional reflectance factor

29 1.7 An ideal remote sensing systemA uniform energy source A non-interfering atmosphere A series of unique energy/matter interaction at the earth's surface A super sensor A real-time data-handling system Multiple data users This kind of system doesn't exist!!!

30 1.8 Characteristics of real remote sensing systemenergy source active R.S.  controlled source passive R.S.  solar energy Both are not uniform and are fn(t, X) need calibration: mission by mission deal with "relative energy" atmosphere effects = fn(l, t, X) importance of these effects = fn(l, sensor, application) elimination/compensation  calibration

31 1.8 Characteristics of real remote sensing system (cont.)The energy/matter interaction at the earth's surface reflected/emitted energy  spectral response pattern  not unique!  full of ambiguity  difficult to differentiate our understanding  elementary level for some materials  non-exist for others

32 1.8 Characteristics of real remote sensing system (cont.)Sensor no super sensor limitation of spectral sensitivity limitation of spatial resolution Fig 1.17: (a) crop (b) crop + soil (c) two fields digital image  pure pixel + mixed pixel trade-offs photographic system: spatial resolution  spectral sensitivity  non-photographic system: spatial resolution  spectral sensitivity  platform, power, storage, ...

33 1.8 Characteristics of real remote sensing system (cont.)Data-handling system sensor capability > data-handling capability data processing  an effort entailing considerable thought, instrumentation, time, experience, reference data computer + human

34 1.8 Characteristics of real remote sensing system (cont.)Multiple data users data  information understand (a) acquisition (b) interpretation (c) use satisfy the needs of all data users impossible! R.S.  New and unconventional  not many users but as time  potential  limitation  users

35 1.9 Successful application of remote sensingPremise: integration many inventorying and monitoring problems are not amenable to solution by means of R.S.

36 1.9 Successful application of remote sensing (cont.)Five conceptions of successful designs of R.S. Clear definition of problem Evaluation of the potential for addressing the problem with R.S. Identify the data acquisition procedures Determine the data interpretation procedures and the reference data Identify the criteria for judging the quality of information

37 1.9 Successful application of remote sensing (cont.)Improvement of the success for many applications of R.S.  multiple-view for data collection  more information multistage (Fig 1.18) multispectral (multi sensors) multitemporal

38 1.9 Successful application of remote sensing (cont.)Example: detection, identification and analysis of forest disease and insect problems (multistage) space images  overall view of vegetation categories refined stage of images  aerial extent and position  delineate stressed sub-areas field-checked and documentation extrapolate to other area detailed ground observation  evaluate the question of what the problem is. R.S.  where? how much? how severe? ...

39 1.9 Successful application of remote sensing (cont.)Likewise, multispectral imagery  more information The multispectral approach forms the heart of numerous R.S. applications involving discrimination of earth resource types and conditions

40 1.9 Successful application of remote sensing (cont.)Multitemporal sensing  monitor land use change Summary R.S.  eyes of GIS (see §1.10) R.S.  transcend the cultural boundaries R.S.  transcend the disciplinary boundaries (nobody owns the field of "R.S.") R.S.  important in natural resources management

41 1.10 Land and geographic information systems (LIS, GIS)Definition GIS: A system of hardware, software, data, people, organizations, and institutional arrangements for collecting, storing, analyzing, and disseminating information about areas of earth LIS: A GIS having, as its main focus, data concerning land records

42 1.10 Land and geographic information systems (cont.)Definition (cont.) Other definitions: GIS: large area, regional, national or global LIS: small area, local, detailed data

43 1.10 Land and geographic information systems (cont.)GIS GIS  computer-based systems GIS  information of features GIS  geographical location data type: locational data attribute data

44 1.10 Land and geographic information systems (cont.)GIS (cont.) One benefit of GIS: spatially interrelate multiple types of information stemming from a range of sources Fig 1.19: example of studying soil erosion in a watershed various sources of maps land data files (slope, erodibility, runoff) derived data analysis output  high soil erosion potential

45 1.10 Land and geographic information systems (cont.)GIS analysis  overlay analysis aggregation buffering network analysis intervisibility perspective views

46 1.10 Land and geographic information systems (cont.)GIS  2 primary approaches raster (grid cell) pros: simplicity of data structure­ computational efficiency­ efficiency for presenting­ high spatial variability blurred boundaries cons: data volume­ limitation of spatial resolution  grid size topological relationship among spatial features  difficult vector (polygon) pros and cons: refer to raster

47 1.10 Land and geographic information systems (cont.)Digital R.S. imagery  raster format  easier for raster-based GIS  output raster format Plate 1: (a) land cover classification by TM data (b) soil erodibility data (c) slope information (d) soil erosion potential map red row crops growing on erodible soils on steep slopes the highest potential

48 1.10 Land and geographic information systems (cont.)Two wrong conclusions: must be raster format  wrong! GIS  conversion between raster and vector GIS  integration of raster and vector data must be digital format  wrong! visual interpretation of R.S. imagery  locate features  GIS GIS information  classification R.S. imagery two-way interaction between R.S. imagery and GIS R.S. & GIS  boundary becomes blurred!

49 1.11 Organization simple  complex short l  long lphotographic system  Chapter 2, 3, 4 non-photographic system  Chapter 5, 6, 7, 8

50 Elements of photographic systemsChapter 2 Elements of photographic systems Introduction to Remote Sensing Instructor: Dr. Cheng-Chien Liu Department of Earth Science National Cheng-Kung University Last updated: 13 March 2003

51 2.1 Introduction Advantages of aerial photographyImproved vantage point Capability to stop action Permanent recording Broadened spectral sensitivity Increased spatial resolution and geometric fidelity

52 2.2 Early history of aerial photography1840  use of photography for topographic surveying 1858  aerial photograph (balloon) 1860  Fig. 2.1: the earliest existing aerial photograph 1882  use kite to obtain aerial photograph

53 2.2 Early history of aerial photography (cont.)1890 the first kite aerial photograph 1906 Fig 2.2: the world-wide known aerial photograph obtained from kite 1890  the giant camera 1.4 x 2.4m 1903 Airplane 1909 Fig 2.3: The first aerial motion picture World War I & II Military purposes

54 2.3 Basic negative-to-positive photographic sequenceFig 2.4: generalized cross section of B&W photographic materials Silver halide grains Gelatin emulsion  photochemical reaction  latent image Base (support) Backing

55 2.3 Basic negative-to-positive photographic sequence (cont.)Fig 2.5: negative-to-positive sequence Negative film exposure: reverse geometry & tone Paper print enlargement: reverse geometry & tone Contact printing: only reverse tone Most aerial photographic paper prints Diapositives, transparencies

56 2.4 Processing black and white filmsFive steps Developing: developer solution Selective, alkaline reducing agents Molecular ionic state  pure atomic(black) state Stop bath: acidic solution Fixing fixer solution Remove unexposed silver halide grains  Harden the emulsion and render it chemical stable washing  free of any chemical residues drying  remove water Air drying or heat drying

57 2.5 Film exposure The simple cameraFig 2.6: comparison between pinhole and simple lens cameras Diaphragm  lens diameter Shutter  duration of exposure

58 2.5 Film exposure (cont.) Focus Equation of focusing Depth of field:Focal length : f object distance : o image distance : i Depth of field: f is fixed, charge o  change i, there exists a limited range of i  depth of field for aerial photography  o    i  f

59 2.5 Film exposure (cont.) Exposure equation of exposure:Film exposure: E (J mm-2) Scene brightness: S (J mm-2 s-1) Diameter of lens opening: d (mm) Exposure time: t (sec) Lens focal length: f (mm)

60 2.5 Film exposure (cont.) Aperture setting (f-stop) : F = f/dF  d   E  For a fixed value of E: Ft1/2 t  stop action, prevent blurring  the case of aerial photography d  F  useful under low light condition d  F  depth of field  Lens speed  F=f/dmax Example 2.2

61 2.5 Film exposure (cont.) Geometric factors influencing film exposureExtraneous effect those factors influence exposure measurements, but have nothing to do with true changes in ground cover type or condition Geometric atmospheric

62 2.5 Film exposure (cont.) Geometric factors influencing film exposure (cont.) Extraneous effect (cont.) Falloff  a distance from the image center a ground scene of spatially uniform reflectance does not produce spatially uniform exposure in the focal plane. Fig 2.7: factors causing exposure falloff

63 2.5 Film exposure (cont.) Geometric factors influencing film exposure (cont.) Vignetting effect internal shadowing resulting from the lens mounts and other aperture surfaces within the camera. It varies from camera to camera and varies with aperture setting for any given camera. Anti-vignetting filter (see §2.11)

64 2.5 Film exposure (cont.) Geometric factors influencing film exposure (cont.) Correction model  radiometric calibration (for given F) Photograph a scene of uniform rightness measure exposure at various q location identify the relationship Eq = E0cosnq Modern camera : n = 1.5 ~ 4

65 2.5 Film exposure (cont.) Geometric factors influencing film exposure (cont.) Object location Fig 2.8: Sun-object-image angular relationship Solar elevation, azimuth angle, viewing angle Fig 2.9: Geometric effects that cause variations in focal plane irradiance differential shading differential scattering specular reflection  extreme exposure few information should be avoid!

66 2.6 Film density and characteristic curvesRadiometric characteristics how a specific film, exposed and processed under specific conditions, responds to scene energy of varying intensity Important for photographic image analysis Tonal values  ground phenomenon (darkness) (crop yield) Photograph  visual records  many energy detectors (silver halide grains)

67 2.6 Film density and characteristic curves (cont.)Film exposure Instantly open  energy  reflectance  exposure Theoretically:  reflectance & fn(l) Unit: meter-candle-second (MCS) or ergs/cm2 MCS is an absolute unit, based on standard observer that is defined photometric. We will deal with “relative exposures” Transmittance:

68 2.6 Film density and characteristic curves (cont.)Film exposure (cont.) Opacity: O 1/T Density: D  log(O)  log (1/T) Transmission densitometer Reflectance densitometer Fig 2.11, Table 2.1 B&W film  AgBr Color film  3 dye layers  filter  max absorption D-logE curve Each film has a unique D-logE curve Also called H&D curve

69 2.6 Film density and characteristic curves (cont.)Fig 2.13 Components of a characteristic curve Gross fog: Dmin = Dbase + Dfog Toe Straight-line portion g  DD/DlogE g  contrast  explain! g  development t & T Shoulder Dmax The range of densities = Dmax - Dmin

70 2.6 Film density and characteristic curves (cont.)Film speed The sensitivity of the film to light Speed  exposure time  Speed  size of AgBr  resolution  Aerial film speed (AFS) AFS  1.5 / E0 E0 = E(D = Dmin) Effective aerial film speed Kodak aerial exposure computer

71 2.6 Film density and characteristic curves (cont.)Fig. 2.14: Exposure latitude The range of log E that will yield an acceptable image on a given film The range of variation from the optimum camera exposure setting that can be tolerated without excessively degrading the image quality Radiometric resolution The smallest difference in exposure that can be detected in a given film analysis Film contrast  exposure latitude  radiometric resolution 

72 2.6 Film density and characteristic curves (cont.)Densitometer (microdensitometer) Light source Aperture assembly Filter assembly Receiver Electronics Readout / recorder

73 2.6 Film density and characteristic curves (cont.)Types of Densitometer Spot Scanning Flatbed Rotating drum Output of densitometer Analog-to-digital (AD) Digital image D: 0~3 DN: 0~255 DIP Fig. 2.17: CCD scanner

74 2.7 Spectral sensitivity of black and white filmsB&W photographs Panchromatic film (Fig 2.18) Infrared-sensitive film (Fig 2.18) Boundary : 0.3~0.9 mm 0.9 mm : the photochemical instability of emulsion material 0.3 mm : Atmosphere absorption & scattering Grass lenses absorption  quartz lenses

75 2.7 Spectral sensitivity of black and white filmsApplication of UV photography in zoological research and management. (Fig 2.19) Harp seals on the snow and ice surface Adult harp seals  dark on both images Infant harp seals  only be dark on UV image Reliable monitoring of the change in population in harp seals.

76 2.7 Spectral sensitivity of black and white films (cont.)Limited applications of UV photography Mainly due to atmospheric scattering Monitoring oil spills

77 2.8 Color film Advantage of color film  more discriminableColor-mixing processes Psychophysical mechanisms  not fully understand We perceive all colors by synthesizing relative amounts of just three

78 2.8 Color film (cont.) Additive primaries : Blue, Green, RedBlue + Green  Cyan Blue + Red  Magenta Green + Red  Yellow Complementary color : choose one primary color and mix the others. Color TV  principle of additive color (human eyes)

79 2.8 Color film (cont.) Color photography  principle of subtractive color Cyan dye  absorb red Magenta dye  absorb green Yellow dye  absorb blue The subtractive color-mixing process: plate 2b

80 2.8 Color film (cont.) Structure and spectral sensitivity of color film Fig 2.20 Blue blocking filter Generalized cross section (Fig 2.20a) Spectral sensitivities of the three dye layers (Fig 2.20b) Color formation with color film (Fig 2.21)

81 2.9 Processing color filmsColor negative films Negative-to-positive sequence Similar to B&W negative film Color reversal films Directly produce positive image Color slides Color diapositives, color positive transparencies

82 2.9 Processing color films (cont.)Fig 2.22 : color reversal process Expose film First developer Re-expose to white light Color developer Bleach & fixer View image

83 2.10 Color infrared film Color infrared filmColor of dye developed in any given emulsion layer  (not necessary correspond to)  color of light to which the layer is sensitive Color infrared film 3 emulsion layers 0.7~0.9 mm False color

84 2.10 Color infrared film (cont.)Fig 2.23: Structure and sensitivity of color infrared film Blue blocking filter (yellow filter) Image color  ground reflectance (nearly equal sensitivity of all layers of the film to blue) Improve haze penetration  reduce Rayleigh scatter  filter out blue light

85 2.10 Color infrared film (cont.)Camouflage detection (CD) film WWII Healthy green vegetation  red (Plate3) Object painted green  blue (Plate3) Only when T is extremely high  IR film can record. Otherwise, IR film is responding to reflected IR energy that is not directly related to T Fig 2.25 Plate 4

86 2.11 Filters Filters Kodak Wratten filter numberTransparent (glass or gelatin) materials Absorption or reflection, eliminate or reduce the energy reading a film in selected portions of the spectrum Place in front of lens Kodak Wratten filter number

87 2.11 Filters (cont.) Absorption filter Bandpass filterOften used in film-filter combination E.g. use a UV-transmitting (Wratten 18A) filter to discriminate harp seals pups. (Fig 2.19) E.g. use a short wavelength blocking filter (high pass) to distinguish between natural grass and artificial turf (Fig 2.27) Bandpass filter Fig 2.28: typical transmittance curve for bandpass filter. Low pass absorption filters are not available!

88 2.11 Filters (cont.) Interference filters : reflect rather than absorbYellow filter  panchromatic film  reduce atmospheric haze B&W film Yellow filter  forestry Red or IR-only filter  delineate water bodies

89 2.11 Filters (cont.) Antivignetting filters:Strongly absorbing in central area and progressively transparent in circumferential area Usually built into other filters Color-compensation filter  aging Using filters  increase exposure Filter factors

90 2.12 Aerial Cameras Four basic types:

91 2.12.1 Single-Lens frame camerasMost common camera Photogrammetric mapping purpose High geometric image quality Film format size 230mm Film capacity 240mm x 120m Intervalometer Focal length : 90~210mm, most widely used: 152mm Long focal length: 300mm  high altitude Frame camera lense (measured along image diagonal) Normal angle (<75o) Wide angle (75o~100o) Super wide angle(>100o)

92 2.12.1 Single-Lens frame cameras (cont.)Principal components (Fig 2.31) Lens cone assembly Lens  bring light rays to focal plane Filter Shutter Diaphragm Body Magazine Supply reel Take up reel Film flattening mechanism Film-advancing mechanism

93 2.12.1 Single-Lens frame cameras (cont.)Principal components (cont.) Image motion compensation Moving the film across the focal plane at a rate just equal to the rate of image movement. Fig 2.32: the modular nature of modern aerial mapping camera system Fig 2.33: a vertical photograph (mapping camera) Fiducial marks Principal point

94 2.12.1 Single-Lens frame cameras (cont.)Large Format Camera (LFC) (NASA) Orbit altitude Space shuttle, free-flying spacecraft, aircraft Advanced image motion compensation mechanism 305-mm-focal-length lens 230x460-mm image format Space-hardened High resolution (3) low distortion (<15 mm) Fig 2.36 Fig 2.37 Fig 2.38

95 2.12.1 Single-Lens frame cameras (cont.)Metric Camera (ESA) Reconnaissance cameras Faithfully record details but not geometric fidelity Color-corrected lens  high quality color photographs

96 2.12.2 Multi-lens Frame CamerasMulti-band photographs photographs taken simultaneously from the same geometric vantage point but with different film-filter combinations. Fig 2.39: multi-lens frame cameras Fig 2.40: example B,G,R, IR Enhance contrast, but to optimize this contrast  choose the film-filter combination

97 2.12.2 Multi-lens Frame Cameras (cont.)Color additive viewers Fig 2.41, 2.42 Four projectors aimed at a single viewing screen Four B&W multi-band images in a positive transparency format Optically superimpose  color composite images Normally, use 3 projectors True or false color “Exotic” color display  enhance discrimination Plate 5: example of color composite

98 2.12.2 Multi-lens Frame Cameras (cont.)Camera filter colors Viewer filter colors Positive transparency-viewer filter combinations DIP Plate 12: six examples of Lansat TM data Multi-band photography use arrays of several single-lens frame cameras

99 Strip Cameras Fig 2.44 Moving film past a fixed slit in the focal plane Shutter  continuously open Inherant image motion compensation Width of slit  determine exposure Designed and good for low altitude and high speed military reconnaissance Permits obtainment of very detailed photography

100 Strip Cameras (cont.) Bad for high altitude and moderate speed  distortion Frame cameras improve in lens & image motion compensation  Strip Cameras have a very limited application.

101 2.12.4 Panoramic Cameras Fig 2.46:Similar to strip cameras, but rotate the lens or a prism to cover ground areas (Fig 2.45) Fig 2.46: panoramic distortion and scan positional distortion

102 2.12.4 Panoramic Cameras (cont.)Optical bar camera NASA. High altitude. Reconnaissance purpose 610-mm-focal-length lens Total FOV : 1200 (600) Film capacity : 2000m Altitude : 19800m Ground coverage : 34.3km x 2 Used extensively for high altitude aerial reconnaissance and Apollo missions

103 2.12.4 Panoramic Cameras (cont.)Pro  broad and detailed view of the ground Con  lack the geometric fidelity variations of atmospheric effect

104 2.12.4 Panoramic Cameras (cont.)Applications: USFS (Plate 9) Forest pest damage detection Plate 9 Timber salvage operations EPA Enviro-Pod : one vertical camera + one forward-looking camera Industrial pollutants hazardous waste sites emergency episodes

105 2.13 Electronic imaging Comparison between photographic and electronic imaging (Table 2.2) Charge-coupled devices (CCDs)  wider range Digital signal  storage, process, transmit. Kodak Professional DCS 200 digital camera Nikon camera body + Kodak camera back 1524 x (9x9 mm) 1/8000 sec 1.5 Million pixels For 35 mm film 2.5~3 Million pixels Fig 2.49, 2.50

106 2.13 Electronic imaging (cont.)Airborne Data Acquisition and Registration System 5000 (ADAR System 5000) A multi-spectral digital camera system (4 CCD Sensors) 0.012~0.3 mm in band width 739 x 478 1/60 ~ 1/2000 sec Ground resolution: 0.5~4m per pixel GPS monitoring Plate 6

107 2.13 Electronic imaging (cont.)Pro: Rapid turnaround time Images are immediately available & computer-ready Higher exposure latitude

108 2.14 Video recording Video recording  standard analog television signals are recorded on magnetic tape or disles Can use various cameras Follow NTSC RS-170 standard Tape format: super-VHS, Hi-8, HDTV..etc. Pros: Real-time viewing & immediately available Inexpensive media Audio track Recorded GPS information

109 2.14 Video recording (cont.)Cons: Poor spatial resolution Expensive equipment Cumbersome to index or handle tapes View: VCR AD converter  frame grabber

110 2.14 Video recording (cont.)Applications  timeliness is required in crop inventorying or disease detection Generalized agricultural, rangeland and natural resource management Analysis of hazardous waste sites Detection of soil conditions Wild rice mapping Trout stream monitoring Right-of-way monitoring Water quality studies Crop condition assessment

111 2.14 Video recording (cont.)Applications (cont.) Detection of forest insect and disease problems Irrigation mapping Detection of frost damage in citrus groves Fig 2.53 : example of 4 CCD array cameras Plate 7: example of video versus photograph Still video cameras Widely used in photojournalism No significant advantage for aerial imaging

112 2.15 Basic geometric characteristics of aerial photographsOrientation Vertical photographs  rarely obtainable Tilted photographs Oblique photographs High  image of the horizon Low

113 2.15 Basic geometric characteristics of aerial photographs (cont.)Taking vertical aerial photographs Fig 2.55 Flight lines (flight strips) nadir line Endlap  at least 50% to ensure total stereoscopic coverage Stereoscopic coverage Stereopairs

114 2.15 Basic geometric characteristics of aerial photographs (cont.)Taking vertical aerial photographs (cont.) Stereomodel Stereoviewing Intervalometer Stereoscopic overlap area 55% ~ 65% overlap  at least 50% endlap (Fig 2.56) Air base Base-height ratio  air base / flying height Vertical exaggeration  (Fig 2.57)

115 2.15 Basic geometric characteristics of aerial photographs (cont.)Taking vertical aerial photographs (cont.) Sidelap  at least 30% Block of photographs GPS navigation system control Index mosaic (Fig 2.59)

116 2.15 Basic geometric characteristics of aerial photographs (cont.)Scale of aerial photographs Photograph scale : one unit of distance on a photograph represents a specific number of units of actual ground distance Unit equivalents, representative fractions, ratios Example 2.3

117 2.15 Basic geometric characteristics of aerial photographs (cont.)Scale of aerial photographs (cont.) (Fig 2.60) Photographs taken over terrain of varying elevation will exhibit a continuous range of scales associated with the variations in terrain elevation Example 2.4 Example 2.5 Average scale

118 2.15 Basic geometric characteristics of aerial photographs (cont.)Comparative geometry of map & vertical aerial photograph Map  orthographic projection  map position Vertical photograph  perspective projection  relative horizontal (planimetric)positions Fig 2.61 Relief displacement: tops of objects are always displaced from their bases, this distortion is  hobject  1/H’  radial distance from the principal point Aerial photographs  (not directly)  map (chap 4) Ground coverage fn(camera format size, focal length, H’) Fig 2.62

119 2.16 Photographic resolutionSpatial resolution: an expression of the optical quality of an image produced by a particular camera system Influenced by Resolving power of film Camera lens Uncompensated image motion Atmospheric condition Film processing condition Fig 2.63: resolving power test chart

120 2.16 Photographic resolution (cont.)Resolving power of the film (lines/mm) The reciprocal of the center-to-center distance (mm) of the lines that are just “distinguishable” in the test chart  Contrast 

121 2.16 Photographic resolution (cont.)Modulation transfer function A microdensitometer is used to scan across images of a series of “square wave” test patterns (Fig 2.64) Spatial frequency modulation transfer function Complete curve (Fig 2.65) A “trade-off” between “speed” & “resolution” Dynamical spatial resolution of the total system Detection  recognition  identification Ground resolution distance

122 2.17 Conclusion Aerial photography Trend  digital recordingBackbone of remote sensing Pros Cons & limitations Trend  digital recording

123 Introduction to airphoto interpretationChapter 3 Introduction to airphoto interpretation Introduction to Remote Sensing Instructor: Dr. Cheng-Chien Liu Department of Earth Science National Cheng-Kung University Last updated: 16 April 2003

124 3.1 Introduction Airphoto interpretation: Historyraw data  human brain processing  information  communicate History Balloon photographs (1858) WWI  military reconnaissance tool WWII  CD film After WWII  wide spread

125 3.2 fundamentals of airphoto interpretationPhoto interpreter Training Experience Keen power of observation coupled with imagination & patience Thorough understanding of the phenomenon Knowledge of the geographic region Supporting materials Maps Field observations

126 3.2.1 elements of airphoto interpretationAirphoto interpretation departs from daily image The portrayal of features from an overhead, often unfamiliar perspective l outside visible range Unfamiliar scales and resolutions Basic characteristics: Shape Size Pattern

127 3.2.1 elements of airphoto interpretation (Cont.)Basic characteristics (cont.) Tone Texture Shadows  topographic variations  geologic landform Site: aid in the identification of vegetation types Association : e.g. a ferris wheel  amusement park

128 3.2.2 Photo interpretation strategiesDirect recognition e.g. identification of highway interchange Inference of site conditions: e.g. infer the buried gas pipeline  light-toned linear streals e.g. infer the type of crop  crop calendar Detective  put all evidence  solve a mystery The interpreter uses the process of convergence of evidence to successively increase the accuracy and detail of the interpretation

129 3.2.3 Airphoto interpretation keysHelp the interpretation in an organized and consistent manner. Two basic parts A collection of annotated or captioned stereograms A graphic or word descrioption Two general types Selective key Elimination key  more positive answer but may result in erroneous answers. Dichotomous key (Fig 3.1)

130 3.2.3 Airphoto interpretation keys (cont.)More easily constructed and more reliably utilized for cultural feature identification than for vegetation or landform identification Crop , tree identification  region-by-region, season-by-season.

131 3.2.4 Film-filter combinationsAffect the amount of information that can be interpreted from the image

132 3.2.5 Temporal aspects of photo interpretationVegetative growth, soil moisture  vary during the year Observe several time  better result

133 3.2.6 Photo scale Table 3.1: typical scales and areas of coverageSmall~1:50,000~medium~1:12,000~large Small  reconnaissance mapping, large area resource assessment, general resource management planning Medium  identification, classification, mapping of tree species, agricultural crop type, vegetation community and soil type Large  intensive monitoring of the damage caused by plant disease, insects or tree blowdown, emergency response to hazardous waste sills and for the intensive site analysis of hazardous waste sites.

134 3.2.6 Photo scale (cont.) NHAP I: NHAP II: NAPP:1980~1985, 1:58,000, 1:80,000, 13200m, leaf-off NHAP II: 1985~1987 , leaf-on. NAPP: 1:40,000, leaf-on, -off

135 3.2.7 Approaching the interpretation processPhotographic materials, interpretation equipment, goals of the interpretation  no single right way to do. Examples of requirement Identify and count Identify anomalous conditions Delineate discrete areal units Classification system or criteria  separate categories Minimum mapping unit (MMU) (Fig 3.2) From higher contrast one to lower one From the general to the specific delineate photographic regions. (tone, texture….)

136 3.2.8 photo preparation and viewingImportant factors: Relevant collateral sources of information (maps, field reports, ….) Good lighting & access to equipment Systematically labeled and indexed Boundary delineations Fiducial marles, road intersections  registration

137 3.2.8 photo preparation and viewing (cont.)Effective areas: Definition: central, bounded by lines bisecting the area of overlap with every adjacent photograph Advantages Cover entire photo without duplicate effort. The least relief displacement Construction  transfer points Disadvantages of lelineating effective areas on each photos  need twice efforts Sometimes, only for every other photo

138 3.3 Basic photo interpretation equipment3 purposes Viewing Making measurements Transfer interpreted information to base maps or digital databases

139 3.3 Basic photo interpretation equipment (cont.)Binocular vision  stereoscopic view  3-D view Stereopairs, stereograms Simple lens stereoscope (Fig 3.3) Test stereoscopic vision (Fig 3.4) (Table 3.2)  elevation One-weak eyesight Cannot get stereoscopic view But still can be a good interpreter  monocular view Viewing the stereogram without a stereoscope

140 3.3 Basic photo interpretation equipment (cont.)Lens stereoscopes Fig 3.3 Pros: portable, cheap Cons: cannot view the entire photo Lens spacing: 45~75mm Lens magnification: typically 2 power

141 3.3 Basic photo interpretation equipment (cont.)Mirror stereoscopes Fig 3.5 Pros: broader view, a pair of 240 mm photos, measurable. Cons: large and costly Magnification: 2~4 power

142 3.3 Basic photo interpretation equipment (cont.)Scanning mirror stereoscope Fig 3.6 Built-in provision Magnification: 1.5~4.5 power Zoom stereoscope Fig 3.7 Continuously variable magnication of 2.5~10(5~20) power. Image in each eyepiece  rotate 3600 Expensive, precision, high resolution

143 3.3 Basic photo interpretation equipment (cont.)Light table: Fig 3.8 For transparency Balance the spectral characteristics of the film and lamps for optimum viewing condition Color temperature 3500k  black body heat at 3500k Noon daylight~5500k Indoor tungsten bulb 3200k Distance measurement Low accuracy  cheaper e.g. a triangular engineer’s scale or metric scale

144 3.3 Basic photo interpretation equipment (cont.)Area measurement Extremely accurate measurement see §4.9, §4.10 Direct measurement Error sources  measuring device, relief, tilt  better to use vertical photos with low relief. Dot grid (Fig 3.9) Polar planimeter (Fig 3.10) Electronic coordinate digitizer (Fig 3.11)

145 3.3 Basic photo interpretation equipment (cont.)Interpretation information map Different size (map and photo) Zoom Transfer Scope (Fig 3.13) Color additive viewer (Fig 2.42)

146 3.4 Land use/ land cover mappingThe type of feature present on the surface of the earth Land use: Human activity or economic function associated with a specific piece of land. A knowledge of both land use and land cover can be important for land planning and land management

147 3.4 Land use/ land cover mapping (cont.)USGS land use and land cover classification system Land use and land cover should not be intermixed , but practically, land cover  land use but also need some additional information sources Use categories rather than specific information

148 3.4 Land use/ land cover mapping (cont.)USGS land use and land cover classification system (cont.) Designed criteria: 85% accuracy Same accuracy for the several categories Repeatable from one time of sensing to another Applicable over extensive areas Infer land use

149 3.4 Land use/ land cover mapping (cont.)USGS land use and land cover classification system (cont.) Designed criteria (cont.) Use for different time of a year Divisible categories Aggregation of categories Comparison with future data Recognize multiple uses of land

150 3.4 Land use/ land cover mapping (cont.)USGS land use and land cover classification system (cont.) Table 3.3: level I, II Also provide level III IV, but it is intended to let the local users to design level III, IV. (Fig 3.14) Reviewing and revising  more wetland classes Table 3.4: Representative image interpretation formats for various classification levels. General relationship, not restriction. Minimum size of land use/land cover units mapped at various classification levels. The smallest representative area on a map 2.5mm x 2.5mm.

151 3.4 Land use/ land cover mapping (cont.)Level I classes Urban or built-up land  take precedence Agricultural land  drained wet lands for agriculture Rangeland Forest land  tree-crown areal density>10% If has wet land characteristics  wet land category Water

152 3.4 Land use/ land cover mapping (cont.)Level I classes (cont.) Wetland Shallow water with submerged  vegetation water class Short-lived wetness or flooding  wetland Cultivated wetlands  agricultural land Uncultivated wetland  wetland Drained wetland for other purposes  other classes Barren land: vegetation or other cover <1/3 Tundra: treeless regions beyond the geographic limit of the boreal forest and above the altitudinal limit of trees in high mountain ranges. Perennial snow or ice areas

153 3.4 Land use/ land cover mapping (cont.)USGS land use/land cover classification system maps 1:250,000 For most categories: a minimum map unit 16 ha Some  1:100,000 Digital data: vector format raster format (grid cell size 4 ha)

154 3.5 Geologic and soil mappingComplex and variable earth surface Topographic relief and material composition Reflect the bedrock, unconsolidated materials, agents of charge. Rock type, fracture, effects of internal movement, erosion, Bear the imprint of the processes that produced them

155 3.5 Geologic and soil mapping (cont.)Geomorphological principles Airphoto interpretation & geological and soil mapping Identify and evaluate materials and structures. Geological mapping History of development: 1913, 1920, 1940….. Identify landforms, rock types, rock structure, portray geological units and structure and spatial relationship. explore mineral resource. Far below the surface and inaccessible region R.S.  potential area  drill holes

156 3.5 Geologic and soil mapping (cont.)Geological mapping (cont.) Multistage image interpretation: 1:250,000, 1:100,000  1:58,000~1:130,000  1:20,000 Lineaments: regional linear features  linear alignment of regional morphological features  streams, escarpments, mountain ranges, tonal features (fractures or fault zones). Scales: a few ~ hundreds of km Important in mineral resource studies  ore deposition Detection  angular relationship. (Fig 3.16)

157 3.5 Geologic and soil mapping (cont.)Geological mapping (cont.) Ronchi grid A diffraction grate: 78 lines/cm || grid  suppressed  grid  enhanced. Lithologic mapping The mapping of rock units Stereoscopic viewing  enhance See § 3.15

158 3.5 Geologic and soil mapping (cont.)Geological mapping (cont.) Geobotany The relationship between a plant’s nutrient requirements and 2 interrelated factors– the availability of nutrients in the soil and the physical properties of the soil, including the availability of soil moisture indirect indicator Distribution of vegetation  (indirect indicator)  composition of the underlying soil and rock materials Geobotanical approach to geologic mapping  Cooperative effort among geologists, soil scientists and field-oriented botanists Identification of vegetation anomalies related to mineralized areas.

159 3.5 Geologic and soil mapping (cont.)Geological mapping (cont.) Geobotany (cont.) Geobotanical anomalies: Anomalous distribution of species and/or plant communities Sturted growth and/or decreased ground cover Alteration of leaf pigment and/or physiographic process that produce leaf color changes. Anomalous charges in the phenologic cycle. e.g. early foliage change senescence in the fall. alteration of flowering periods, late leaf flush Taking photos several times during the year. Establishing “normal” condition  identify “anomalous” Band 1.6 um & 2.2 um are important for mineral exploration and lithologic mapping.

160 3.5 Geologic and soil mapping (cont.)Soil survey  resource information  land use planning Trained scientists + extensive field works + airphoto interpretation  identify soil & delineate soil boundaries. Airphoto interpretation  1930s Panchromatic aerial photos: 1:15,840~1:40,000

161 3.5 Geologic and soil mapping (cont.)Agricultural soil survey (Fig 3.17, Table 3.6) USDA, 1900s. 1957  publish 1980s  many counties  line maps or digital form Purposes: Estimating crops Evaluating rangeland suitability Determine woodland productivity Assessing wildlife habitat conditions Judging suitability for various recreational uses Judging suitability for various development uses

162 3.5 Geologic and soil mapping (cont.)The reflection of sunlight from bare soil surfaces Soil moisture content Soil texture Surface roughness Iron oxide Organic matter content

163 3.5 Geologic and soil mapping (cont.)Plate 8: different appearance of one field during one growing season Soil parent materials glacial meltwater deposits of stratified sand and gravel overlain by 45~150 cm of loess (wind-deposited silt) (a), (b), (c): corn plants 10 cm 2.5 cm rain fall uniform  patchiness dry  high infiltration & slight mound wet  low infiltration & receive runs off

164 3.5 Geologic and soil mapping (cont.)Plate 8: (cont.) (d), (e), (f): corn plants 2m little rain fall dry  leaves and stalks drying out and turn brown. wet  continuing to grow and still green Soil scientist  four classes. (Fig 3.18 & Table 3.7) Certain times of the year are better suited to aerial photography for soil mapping purposes than others.

165 3.6 Agricultural applicationsBig picture direct application Crop type classification (and area inventory) Spectral response and photo texture  identify crop type Require a knowledge of the developmental stages of each crop in the area to be inventoried  crop calendar Use photographs taken on several dates during the growing cycle for crop identification Color + infrared films are better than panchromatic film

166 3.6 Agricultural applications (cont.)Crop type classification (cont.) Stereoscopic coverage  plant height  discrimination Table 3.8: dichotomous airphoto interpretation key  use single-date panchromatic photography  only broad classes of crops are to be inventories. Fig 3.19: demonstrate the importance of date of photography, photo tone and texture, and stereoscopic coverage.

167 3.6 Agricultural applications (cont.)Crop condition assessment Large scale airphotos  documenting deleterious conditions  crop disease, insect damage, plant stress, disaster damage. Table 3.9: typical crop management information potentially obtainable from large scale color infrared aerial photographs.

168 3.6 Agricultural applications (cont.)Crop condition assessment (cont.) Detailed within field interpretations of soil and crop condition  fertilizer spreaders and irrigators  crop management activities  fn(geolocation) Detected plant diseases detected insect damage other detected damage: A difficult task  finer differences in spectral response

169 3.6 Agricultural applications (cont.)Crop yield estimation Simple & straightforward  area * yield/area  yield Complex  soil moisture, soil fertility, air and soil temperature, disease, insect stress,….. Crop yield prediction  climatic & meteorological conditions. Traditional approach: area * yield/area (airphoto interpretation) (field inspection) Direct approach  historical information  deviation

170 3.6 Agricultural applications (cont.)Other applications Determine areas  erosion control, weed control fertilizing, replanting, fencing or other remedial measures Taxation & real estate purposes Assessment of irrigation systems Farm livestock surveys

171 3.7 Forestry applications1/3 of the world’s land area Tree species identification More complex than crop identification. Complex mixture  more uniform Forest understory Step 1: elimination Step 2: establish groups Step 3: identify individual species Shape & size Fig 3.20: Silhouettes of forest trees Fig 3.21: Aerial views of tree crowns. Pattern & Shadows

172 3.7 Forestry applications (cont.)Tree species identification (cont.) Tone  relative tones Texture  tufted, smooth, billowy Fig 3.22, Fig 3.23 Black spruce coniferous, slender crowns, pointed tops. even height or change gradually carpetlike appearance Aspen deciduous, rounded crowns widely spaced, variable in size and density Balsam fir symmetrical coniferous, sharply pointed tops thicker than black spruce erratic changes in size  uneven and irregular pattern

173 3.7 Forestry applications (cont.)Tree species identification (cont.) Fig 3.22, Fig 3.23 (cont.) Fig 3.22: Black spruce  Aspen Fig 3.23: Balsam fir  Black spruce (mixture) Art Scale does matter Table 3.10, 3.11  Airphoto interpretation key Phenological correlations Changes in the appearance in the different seasons of the year e.g. separation of deciduous and evergreen trees e.g. difference in the time at which species leaf out

174 3.7 Forestry applications (cont.)Timber cruising Objective: determine the volume of timber that might be harvested from an individual tree of a stand of trees To be successful  skilled interpreter + aerial and ground data Photo measurements Tree height or stand height  measuring relief displacement or image parallax. Tree-crown diameter  distance measurements Density of stocking Stand area

175 3.7 Forestry applications (cont.)Timber cruising (cont.) Photo volume tables Multiple regression (extracted data, ground data) Volume of individual trees = fn(species, crown diameter, height) (Table 3.12) For large scale photos, scattered trees in open areas. Table 3.13: stand volume table

176 3.7 Forestry applications (cont.)Assessment of disease and insect infestations Panchromatic photos  often used Medium and large scale color and color infrared photos  most successful surveys Tree disease Insect damage Forest damage Plate 9: gypsy moth defoliation of hardwood trees.

177 3.7 Forestry applications (cont.)Additional applications: Success  high quality reference data

178 3.8 Rangeland applicationsDefinition: see § 2.4 Provide forage, support land use (agriculture, recreation housing) Range managers use air photos in much the same way as do foresters. Main concern: vegetation change over time Forage yield estimation.

179 3.9 Water resource applicationsWater resources: irrigation, power generation, manufacturing, recreation, … Airphoto  monitor  quality, quantity, distribution Interaction of sunlight and water Absorption=fn(l) near-infrared  a few tenths of a meter  dark tone 0.48~0.6um  best penetration (15~20m)  underwater haze red  few meters

180 3.9 Water resource applications (cont.)Analysis of underwater features White sand bottoms normal color film  blue-green color infrared (yellow filter)  blue Fig 3.24: color and color infrared photos for Hanauma Bay

181 3.9 Water resource applications (cont.)Water pollution detection Polluted water  impurities  limit its use Natural sources of pollution: e.g. minerals leached from soil and decaying vegetation Point source: e.g. industrial outfalls Non-point source: e.g. fertilizer and sediment runoff.

182 3.9 Water resource applications (cont.)Water pollution detection (cont.) Materials (excessive amounts)  water pollution Organic wastes Infections agents Plant nutrients Synthetic-organic chemicals Inorganic chemical and mineral substances Sediment Radioactive substances temperature

183 3.9 Water resource applications (cont.)Water pollution detection (cont.) Air photo  type and concentration?  discharge point & dispersion in some instances  it is possible to make valid observations about sediment concentrations using quantitative photographic radiometry coupled with the laboratory analysis of selective water samples. Fig 3.25: dispersal plume of silt-laden water flowing into a lake Air photos  enforcement of antipollution law Oil film on water (Fig 3.26) Oil slicks Oil sheens Oil rainbows

184 3.9 Water resource applications (cont.)Lake eutrophication assessment Trophic state Eutrophic state (nutrient rich) Oligotrophic (low nutrient, high Oxygen) Eutrophication: the general process by which lakes age Different people  different levels of eutrophication that can be accepted Air-photo  mapping aquatic macrophytes Table 3.14: interpretation key Algae concentration  good indicator of a lake’s trophic status

185 3.9 Water resource applications (cont.)Flood damage estimation Fig 3.27: multi-date sequence and aftereffects. A. normal appearance B. peak of a flood. C. 3 weeks after flooding D. 6 weeks after flooding Crop damage Streaked pattern of light-toned lines  direction of flood flow. Fig 3.28: For flood damage assessment B is clean than a Refer to Fig 6.12b for satellite image

186 3.9 Water resource applications (cont.)Other Applications Vegetation index Ground water location Groundwater discharge areas  well Groundwater recharge zone  protect Hydrologic watershed assessment Reservoir site selection Shoreline erosion studies Snow cover mapping Survey of recreational use of lakes and rivers

187 3.10 Urban and regional planning applicationsTimely, accurate and cost-effective sources of data (requirement) Population estimates. Air-photo  dwelling units of housing type Housing quality studies Statistical analysis of a set of environmental quality factors. e.g. house size, lot size,… Traffic and parking studies Traditional on-the-ground vehicle count Air-photo  shows distribution  area of congestion

188 3.10 Urban and regional planning applications (cont.)Various factors: RS photo  data  GIS Analysis Transportation route location Sanitary landfill site selection Power plant sitting Transmission line location Fig 3.29: urban change detection

189 3.11 Wetland mapping Significance of wetland Retain waterReduce flood level Trap suspended solids and attached nutrients  clear lake For wildlife (drinking and stopping) Species diversity and food web. Biological record recreation

190 3.11 Wetland mapping (cont.)Purpose (multi-or single-)  inventory 3 basic elements for identifying wetland Hydrophytic vegetation Hydric soils Wetland hydrology Fig 3.30 color infrared photo for wetland mapping Fig Wetland vegetation map Table 3.15 airphoto interpretation key.

191 3.12 wildlife ecology applicationsWildlife conservation Wildlife management

192 3.12 wildlife ecology applications (cont.)Wildlife habitat Combination of climate, substrate and vegetation Niche All mapping techniques are applicable “edges” delineation GIS Fig 3.32: Wildlife habitat types in Sheboygan Marsh 9 vegetation classes  5 wildlife habitat types Muskrat hut  in AV area, 100 white spots

193 3.12 wildlife ecology applications (cont.)Wildlife censusing Ground survey + aerial visual observation Problems of counting Quick decision? Aggregation? Low-flying aircraft  disturb Best method: vertical aerial photography Accurate counting Normal patterns of spatial distribution Permanent record Prolonged study  more information Only valid for those open areas during daytime

194 3.12 wildlife ecology applications (cont.)Wildlife censusing Scale=fn(animal size) Film-filter  high contrast (Fig 2.18) Digitize  computer-aid counting Fig 3.33: Prairie dog colony Fig 3.34: Beluga whales Calve Number & length Bachelor group

195 3.13 Archaeological applicationAirphotos  locate sites whose existence has been lost to history Both surface and subsurface features Surface features: Visible ruins: e.g. Stonehenge, castles, Indian dwelling. Mounds: e.g. bird-shaped and serpent-shaped Indian mounds. Rock structures e.g. Bighorn Medicine Wheel, mounds Fig 3.35: Nazca lines

196 3.13 Archaeological application (cont.)Subsurface features: e.g. buried ruins of buildings, ditches, canals, roads Tonal anomalies in soil moisture, crop growth, ephemeral difference in frost patterns Fig 3.36: ancient city of Spina A city of canals and waterways. Dark-toned linear feature  dense vegetation  wet location  former canals Light-toned rectangular areas  sparse vegetation over sand  brick foundation Light-toned linear feature  present-day drainage ditches.

197 3.13 Archaeological application (cont.)Subsurface features: Fertile loess soils over the white chalk bedrock (foundation) Fig 3.37: deep winter plowing  scraped the foundation  brought up chalk Fig 3.38: recent conversion from pasture to cropland  few fertilizer over the foundation  light toned.

198 3.14 Environmental assessmentHuman activities  adverse environmental effects. U.S. NEPA (1969) Environment impact statements Key items Environmental impact Cannot-avoided impact Alternatives Relationship(short-& long-term) Irreversible and irretrievable commitments

199 3.14 Environmental assessment (cont.)Principal biophysical effects of human activity on environment Change natural drainage conditions Change water turbidity & temperature Chemical pollutants Change in vegetation Change wildlife population and distribution US. RCRA

200 3.14 Environmental assessment (cont.)Airphotos EPA’s principal remote sensing tools e.g.: emergeray response photography to the spillage of hazard materials e.g.: site analyses of waste sites Historical photos Drainage conditions e.g.: locate potential sites for drilling and sampling of hazardous wastes e.g.: identification of failing septic systems

201 3.15 Principles of landform identification and evaluationSignificance of terrain characteristics Airphotos  Bedrock type Landform Soil texture Site drainage conditions Susceptibility to flooding Depth of unconsolidated materials over bedrock Slope of land surface.

202 3.15 principles of landform identification and evaluation (cont.)Soil characteristics Definition: e.g. 10m deposit (C)= 9m unaltered (B) + 1m weathered (A) Engineering: A+B+C Soil science (pedological): A+B Soil horizon A horizon: surface soil topsoil 0~60cm, 15~30cm Weathered horizon Most organic matter B horizon: subsoil 0~250cm, 45~60cm Fire-textured particles from A horizon C horizon: parent material

203 3.15 principles of landform identification and evaluation (cont.)Soil characteristics (cont.) Three origins of soil materials Residual soils Transported soils Organic soils Table 3.16: soil particle size designations. Silt+clay 50%=fine texture Sand+gravel> 50%= coarse texture For residual soil: texture  B+C For transported soil: texture  parent material USDA soil drainage classes (7 classes)  natural Artificial drainage

204 3.15 principles of landform identification and evaluation (cont.)Land use suitability evaluation Topographic characteristics For subdivision development: 2~6% Steep enough for good surface drainage Flat enough no site development problems 0~2%  may have some drainage problems 6~12%  more interesting but more costly >12%  problem in street and lot design, as well as domestic sewage disposal >20%  severe limitation For industrial park and commercial site  < 5 % Soil texture and drainage conditions Well-drained+coarse texture soils  few limitation

205 3.15 principles of landform identification and evaluation (cont.)Land use suitability evaluation (cont.) Groundwater tables Shallow  problems in septic tank At least 2m Depth of bedrock Prefer > 2m 1~2m may be feasible in some cases Soil-slope condition Slope stability analysis  not discussed Incipient landslide failure  detectable using airphoto

206 3.15 principles of landform identification and evaluation (cont.)Elements of airphoto interpretation for landform identification and evaluation Topography A distinct topographic change at the boundary between two different landforms. Drainage pattern and texture Indicators of landform, bedrock, soil, drainage conditions Fig 3.39: six basic drainage patterns  destruct ional from erosion Dendritic Rectangular Trellis Radial Cantripetal Deranged

207 3.15 principles of landform identification and evaluation (cont.)Elements of airphoto interpretation (cont.) Drainage pattern and texture (cont.) Fig 3.40 Coarse-texturedgood internal drainage Fine-texturedpoor internal drainage Erosion Gullies: The smallest drainage feature from airphotos A meta wide and a hundred meters long Formation rainfallnot percolate rivuletsrunoffenlarge Fig3.41 V-shape: sand & gravel U-shape: silty soils Long with gently rounded: silty clay, clay

208 3.15 principles of landform identification and evaluation (cont.)Elements of airphoto interpretation (cont.) Photo tone Brightness Use relative tone values Fig 3.54 Lighter tone  higher position Varying soil moisture content  different sunlight reflection  mottled tonal pattern Capillary action  tonal gradients  boundary sharpness Color or color infrared  same principle

209 3.15 principles of landform identification and evaluation (cont.)Elements of airphoto interpretation (cont.) Vegetation and land use Indicator, but in many cases, they are obscured The airphoto interpretation process Initially  consider all elements After some experience  subconsciously + instantaneously Humid climates  >50 cm rainfall per year Arid climates  <50 cm rainfall per year Airphoto interpretation  field operation Verify suggestion

210 3.16 Bedrock landforms Horizontally bedded sandstoneAirphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.43

211 3.16 Bedrock landforms (cont.)Horizontally bedded shale Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.44

212 3.16 Bedrock landforms (cont.)Horizontally bedded limestone Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.45

213 3.16 Bedrock landforms (cont.)Horizontally bedded granitic rocks Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.46

214 3.16 Bedrock landforms (cont.)Horizontally bedded lava flows Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.47

215 3.17 Aeolian landforms Sand dunes Formation Fig 3.48

216 3.17 Aeolian landforms (cont.)Sand dunes (cont.) Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.49

217 3.17 Aeolian landforms (cont.)Loess Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.51

218 3.18 Glacial landforms Formation:Repeated advances of glacial ice over 30% Glaciation forms: Valley glaciation Continental glaciation Ice move  abrade and pluck Glacial drift Materials deposite by glaciation

219 3.18 Glacial landforms (cont.)Till landforms Repeated advances of glacial ice over 30% Glaciation forms: Valley glaciation Continental glaciation Ice move  abrade and pluck Glacial drift Materials deposite by glaciation

220 3.18 Glacial landforms (cont.)End moraines Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.53

221 3.18 Glacial landforms (cont.)Basal ground moraine area Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.54

222 3.18 Glacial landforms (cont.)Drumlins Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.55

223 3.18 Glacial landforms (cont.)Eskers Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.56

224 3.18 Glacial landforms (cont.)Outwash sediments Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.57

225 3.18 Glacial landforms (cont.)Glacial lakebeds Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.58

226 3.18 Glacial landforms (cont.)Beach ridges Airphoto identification Topography: Drainage and erosion: Photo tone Vegetation and land use Fig 3.59

227 3.19 Fluvial landforms Formation:Flowing water  erosion + transportation + deposition fn(water velocity, particle size) Stream competence: the maximum-size particles a stream can transport at a given velocity Stream capacity: the maximum amount of materials the stream can transport and is related to stream volume

228 3.19 Fluvial landforms (cont.)Alluvial fans Airphoto identification Topography: fan shape Drainage and erosion: excellent internal drainage limited surface drainage with few gullies Numerous distributary channels Photo tone Generally light, but distributary channels may be darker Vegetation and land use Lack of vegetation May be heavier vegetation at base

229 3.19 Fluvial landforms (cont.)Alluvial fans (cont.) Fig 3.60 Slope: apex (10%)  valley (12%)  base (8%) Darker tone and vegetation near the base of fan

230 3.19 Fluvial landforms (cont.)Floodplain Airphoto identification Topography: Generally level with small downstream gradient Natural levees slightly higher position Slack water deposits in lowest position Drainage and erosion: Principal stream flow Wide floodplain  second stream High groundwater table Photo tone Point bar deposits, natural levee  light tone Slack water deposits  darker tone Oxbow  uniform gray tone Vegetation and land use Often agricultural use

231 3.19 Fluvial landforms (cont.)Alluvial fans (cont.) Fig 3.61 Present channel (PC) Abandoned channel (AC) Point bar deposits (PB) Oxbow lake (OX) Slack water deposit (SW)

232 3.19 Fluvial landforms (cont.)Delta Type Arcuate delta: e.g. the Nile Delta Birdfoot delta: e.g. the Mississippi Delta Fig 6.13

233 3.20 Organic soils Formation: CharacteristicsContinuous water saturation  limit the circulation of O2  decomposition of organic matter  accumulation  organic matter > mineralization  muck or peat Characteristics Typically begins in lakes or ponds Poor foundations for construction activities If over drain  irreversible hardening If too dry  fire hazard

234 3.20 Organic soils (cont.) Airphoto identification Found TopographyTopographic depressions (moraine, ground moraine, floodplain, oxbows) Depression between sand dunes, beach ridges Limestone sinkhole Kettle hole Topography Very flat, often sharp contrast with surrounding material

235 3.20 Organic soils (cont.) Airphoto identification (cont.)Drainage and erosion Poorly drain, few gullies Farm  artificially drain Photo tone, vegetation and land use Bare soil  dark tone Drained agricultural areas  distinctive pattern

236 3.20 Organic soils (cont.) Fig 3.62 Organic soilsGlacial till, drumlin Well drained, cultivated 2m ~ 5m Fibrous peat in a former glacial lakebed 80~95% organic matter, 5~20% mineral matter

237 Chapter 4 Photogrammetry Introduction to Remote SensingInstructor: Dr. Cheng-Chien Liu Department of Earth Science National Cheng-Kung University Last updated: 23 April 2004

238 4.1 Introduction Photogrammetry: Photogrammetry  Measurements MapsDigital elevation models Other derived products Photogrammetry  Where What areal extent

239 4.1 Introduction (cont.) SubjectsDetermining horizontal ground distances and angles from measurements made on a vertical photograph Determination of object height from relief displacement measurement Determination of object heights and terrain elevations by measurement of image parallax Use of ground control points

240 4.1 Introduction (cont.) Subjects (cont.)Generations of maps in stereoplotters Generation of orthophotographs and digital elevation models. Preparation of a flight plan to acquire aerial photography. Application of soft copy or digital photogrammetry.

241 4.2 Geometric elements of a vertical photographPhotogrammetry  Vertical photographs Unintentional tilts: <10 (<30) Fig4.1 Basic geometric elements of a vertical photo L: the camera lens exposure station f: the lens focal length X-axis: the forward direction of flight Y-axix: 900 counterclockwise from the positive x-axis O: the ground principal point ABCDE  abcde  a’b’c’d’e’ The x y photocoordinates

242 4.2 Geometric elements of a vertical photograph (cont.)Measurement of photocoordinates Triangular engineer’s scale  rudimentary problem Metric scale Glass scale  built-in magnifying eyepieces (Fig 4.2) Coordinate digitizer Comparator mono (Fig 4.3) stereo Precision: 1~5 mm

243 4.2 Geometric elements of a vertical photograph (cont.)Sources of error Lens distortion Atmospheric refraction Earth curvature Failure of the fiducial axes to intersect at the principal pt. Shrinkage or expansion Usually, correct this error Slight tilt  outweigh other sources Example 4.1: treat it as the problem of exchange rate

244 4.3 Determining horizontal ground lengths, directions, and angles from photo coordinatesFig 4.4(a). Displacement of terrain points Fig 4.4(b). Distortion of horizontal angles measured on photograph Relief displacement The datum plane: A΄B΄  a΄b΄ Terrain points AB  ab a΄b΄: the accurate scaled horizontal length and orientation of the ground line AB. Angle distortion: b΄c a΄  bca. b΄oa΄= boa (no distortion)

245 4.3 Determining horizontal ground lengths, directions, and angles from photo coordinates (cont.)Fig 4.5 determination of ground coordinates ∵△LOAA΄~△LOAa΄ ∴XA=(H-hA)xa/f Likewise: XB=(H-hB)xb/f ∵△LA΄A~△La΄a ∴YA=(H-hA)ya/f Likewise: YB=(H-hB)yb/f AB=[(XA-XB)2+(YA-YB)2]1/2 Example 4.2

246 4.3 Determining horizontal ground lengths, directions, and angles from photo coordinates (cont.)Fig 4.6 determination of line length and direction from ground coordinates Example 4.3

247 4.4 Relief displacement of vertical featuresFig 4.7: the radial nature of relief displacement Relief displacement  radial distance Fig 4.8 geometric components of relief displacement. ∵△AA΄A΄΄~△LOA΄΄ ∴D/h = R/H, d/r = D/R ∴h=dH/r Example 4.4 relief displacement  height

248 4.4 Relief displacement of vertical features (cont.)Premise: Truly vertical photo Accurate H Clearly visible objects Precise location of the principal point Accurate measurement technique Correcting the image positions of terrain points appearing in a photograph Example 4.5

249 4.5 Image Parallax ParallaxPrinciple: moving train  viewing window  relative movement  distance Fig 4.9: Parallax displacements on overlapping vertical photographs. Conjugate principal points  the flight axis (Fig 4.10) Parallax: pa= xa-xa΄

250 4.5 Image Parallax (cont.) Fig 4.11parallax relationships on overlapping vertical photos. Air base Parallax equation Example 4.6 Difference in elevation

251 4.6 Parallax measurement In example 4.6parallax  2 measurements required (cumbersome) Fig 4.12: single measurement  parallax Stereopair  photographs fasten down with flight aligned p=x-x΄=D-d  single measurement a and a΄ are identifiable Difficult to identify if the tone is uniform

252 4.6 Parallax measurement (cont.)Employment Fig 4.13: Floating mark principle Half marks Left one fixed and right one moves along the fight direction  fuse together  one mark floating Parallax bar: p=r+C where r= the parallax bar reading C=constant Determination of c: given p, measure r  C = p - r C = S Ci Usually use the two principal points Example 4.7

253 4.6 Parallax measurement (cont.)Parallax Wedge (Fig 4.16) Constitution: 2 converging lines on a transparent sleet Can be thought of as a series of parallax bar reading Fig 4.17 determination of the height of a tree using a parallax wedge Example 4.8 Measure absolute parallax

254 4.7 Ground control for aerial photographyHorizontal Vertical GPS  promising Accuracy is essential Cultural features, e.g. road intersection Ground survey  artificial target premarked

255 4.8 Use of ground control in determining the flying height and air base of aerial photographsFlying height determination Altimeter  approximate H. S= f /(H-h) Example 4.9 Ground control  H Given ground length AB elevations hA, hB focal length f. photocoordinates (xa, ya).(xb, yb) eg. (4,1)  (4,4)  H Iteration: H2=AB (H1-hAB) /AB1 +hAB where hAB: the average elevation of the two end points of AB Example 4.10

256 4.8 Use of ground control in determining the flying height and air base of aerial photographs (cont.) Air Base determination Ground control  B Given H & one vertical control point eq(4.10)  B Example 4.11 Given two control points  B Example 4.12

257 4.9 Stereoscopic plotting instrumentsPhotogrammetry  topographic maps Stereoplotters Concept: Stereopair photo: terrain  ray  lens  image plane Stereoplotter: photos  ray  terrain model  3D view Three components A projection system A viewing system A measuring and tracing system Fig 4.18: a direct optical projection plotter Image  tracing table  stereoview of terrain model Relative orientation  absolute orientation

258 4.9 Stereoscopic plotting instruments (cont.)Stereoplotters (cont.) Fig 4.19: three projectors  2 adjacement stereopairs to be oriented at once Anaglyphic viewing system. Color filter  red, cyan Only for panchromatic photo Polarized platen viewer (PPV) Polarizing filter Stereo image alternator (SIA) Rapidly alternate the projection of the two photos.

259 4.9 Stereoscopic plotting instruments (cont.)Tracing table platen Floating mark  raise and low Platen table height  terrain elevations Mapping features  pencil Compile contours

260 4.9 Stereoscopic plotting instruments (cont.)Viewing the photographs in stereo through a binocular system Mechanical or optical-mechanical projection plotters. Fig 4.20 Coordinatiograph Electronic image correlator Fig 4.21: analytical stereoplotter

261 4.10 Orthophotos Orthophotos Generation  analog orthophotosNo scale, tile relief distortions  Photomaps Best of both worlds Input to GIS Digital format Generation  analog orthophotos Differential rectification (Fig 4.22) Orthophotoscopes Orthophoto negative

262 4.10 Orthophotos (cont.) Fig 4.23an early version of a direct optical projection orthophotoscope Principle of operation

263 4.10 Orthophotos (cont.) Topographic orthophotomapFig 4.24: operating principle of direct optical projection Fig 4.25:contour line overlay orthophoto orthophotoscope Fig 4.26a: contour map Fig 4.26b: 3-D perspective view of the terrain Stereomates Fig 4.27: an orthophoto and a corresponding stereomate that may be viewed stereoscopically.

264 4.11 Flight planning Why need new photographs? Planning the flightOutdated Wrong season Inappropriate scale Unsuitable film type Planning the flight Weather  clear weather  beyond control Multi-task in a single clear day Time  10am~2pm  illumination max shadow min.

265 4.11 Flight planning (cont.)Planning the flight (cont.) Geometric aspects f Format size S Area size havg Overlap Side lap Ground speed Example 4.13 H΄ Location, direction, number of flight lines Time interval Number of exposures Total number of exposures

266 4.12 Soft copy photogrammetryDistinctions between traditional analog systems and digital systems Photographs  digital raster images Mathematical modeling (computer-based environment) Sources: digitized photos, digital cameras, electro-optical scanners, … Trend from now to future

267 Multispectral, thermal and hyperspectral scanningChapter 5 Multispectral, thermal and hyperspectral scanning Introduction to Remote Sensing Instructor: Dr. Cheng-Chien Liu Department of Earth Sciences National Cheng-Kung University Last updated: 7 May 2003

268 5.1 Introduction MSS (multispectral scanning) vs. MCS (Multiband camera system) Fig 2.42, 2.43 (MCS) Advantages Spectral bands: 0.3~1.4 mm with narrow band width Collection: same optical system for all bands. Calibration: electronically rather than photochemically. Storage and transmission

269 5.2 Across-track multispectral scanningAcross-track (whiskbroom) scanner: Fig 5.1: operation of a five-channels sensor  flight line Using a rotating or oscillating mirror (Fig 5.1b) Contiguous strips  2D image Dichroic grating  two forms of energy Thermal Nonthermal  prism  UV, Vis and near-IR Detectors  spectral sensitivity Signal  amplified  recorded  A-to-D  View-in-flight  storage  interpretation

270 5.2 Across-track multispectral scanning (cont.)An example of multispectral sensor IFOV (instantaneous field of view) the cone angle with which incident energy is focused on the detector (see b in Fig 5.1a) See “pure” and “mixed” pixels simultaneously. Resolution Spatial resolution: D=H΄b (Fig 5.2) Ground resolution element (resolution cell) Towards edge  resolution cell  (image distortion). For typical airborne MSS system: b = 0.5 ~ 5 (mrad)

271 5.2 Across-track multispectral scanning (cont.)Trade-off between spatial resolution and radiometric resolution. IFOV   signal-to-noise ratio  radiometric resolution   spatial resolution  Fig 5.3: eleven-band digital across-track MSS system. 7 of them  Landsat Thematic Mapper.

272 5.2 Across-track multispectral scanning (cont.)Fig 5.4: eleven-band MSS images (A) street  (B) grass: clear contrast in all bands Tonal reversal: from band 8~10 (near-IR) (C) water  (B) grass: difficult to differentiate in the Vis-band, but clean contrast in near-IR bands. (D) cloud shadow: channel 1:somewhat illuminated by Rayleigh scatter near-IR  low Rayleigh scatter  darkest thermal  no signal (moving clouds)

273 5.2 Across-track multispectral scanning (cont.)Fig 5.5: eight-band MSS image Pavement  grass: clear in Vis-band(1~3) tones reverse in near-IR(4, 5) and mid-IR (6, 7) River plume: clear in thermal band (8) Dock: clear in band (6~8)

274 5.3 Along-track multispectral scanningAlong-track (pushbroom) scanner Fig 5.6 A linear array of CCDs (charge-coupled device) No scanning mirror Size of detectors  ground resolution The smaller the better Each band  one array

275 5.3 Along-track multispectral scanning (cont.)Along-track (pushbroom) scanner (cont.) Pros: Longer dwell time (residence time) Stronger and greater range of signal Better spatial and radiometric resolution Fixed detector  geometric integrity  Size & weight & power  No moving part  reliability & life expectancy  Cons: Calibration Limited range of spectral sensitivity ( Underdeveloping

276 5.3 Along-track multispectral scanning (cont.)Fig 5.7 MEIS II the first airborne pushbroom scanner 1728-element linear arrays 8 spectral bands (0.39 mm~1.1 mm) IFOV 0.7 m rad TFOV 400 8-bit (256 DN)

277 5.3 Along-track multispectral scanning (cont.)Fig 5.8: MEIS II in stereomode. External mirror  forward-looking & aft-looking MOMS ( the 1st space borne pushbroom scanner)

278 5.4 Across-track thermal scanningThermal portion of the spectrum Atmospheric window  two ranges (3~5mm, 8~14 mm) Rapid response (<1-msec) Photon  electrical charge Dewar  cool detector 3 photon detectors in common use today Table 5.1: spectral sensitivity range

279 5.4 Across-track thermal scanning (cont.)Fig 5.9 Across-track thermal scanner schematic Recording procedure: Incoming energy Additional optics Focusing  detector Detector (encased by a dewar)  signal Amplify Display &record, amplified signal  modulate tube Scan Record Platen Film advanced rate = fn(V/H΄) Fig 5.10: a typical thermal scanner system

280 5.5 Thermal radiation principlesRadiant versus kinetic temperature Kinetic temperature: contact  internal manifestation  average translational energy of the molecules Radiant temperature: emitted energy  external manifestation  object’s energy state Blackbody radiation Fig 5.11: Spectral distribution of energy radiated from blackbodies of various temperatures Wien’s displacement law Stefan-Boltzmann law

281 5.5 Thermal radiation principles (cont.)Radiation from real materials No perfect blackbody Emissivity: e(T)  Mreal(T) / Mblack body(T) 0 < e < 1 e = fn(l, qviewing, T) Graybody e  fn(l) Selective radiator e = fn(l) Fig 5.12 Fig 5.13: (6~14mm) Water ~ blackbody (e = 0.98~0.99) Quartz ~ selective radiator

282 5.5 Thermal radiation principles (cont.)Radiation from real materials (cont.) Importance of M(l = 8~14 mm)  thermal sensing Atmospheric window M(T=300K) = 9.7 mm For broadband sensor  treated as graybody Table 5.2: typical values of emissivity over 8~14 mm Atmospheric effects Atmospheric windows: Fig 5.14 Thermal range : (3~5 mm, 8~14 mm) Atmospheric absorption & scattering  colder objects. Atmospheric emission  warmer object Biased sensor output Compensation  later this chapter

283 5.5 Thermal radiation principles (cont.)Interaction of thermal radiation with terrain elements. EI=EA+ER+ET  conservation of energy. 1 = a +r + t  normalize 1 = e +r + t  Kirchhoff radiation law (good absorbersgood emitters) 1= e +r  assumption of opaque to thermal radiation. For real body: M= esT4 Trad=e1/4Tkin Table 5.3: typical values Always underestimate T if neglect e effect Surface T (<50mm)  internal bulk T

284 5.6 Interpreting thermal scanner imagerySuccessful interpretations Rock type, structure Locating geologic faults Mapping soil type & moisture Locating irrigation canal leaks. Volcanoes. Evapo-transpiration from vegetation. Locating cold water springs. Locating hot water springs & geysers. Thermal plumes in lakes and rivers. Natural circulation patterns in water bodies. Forest fires. Locating subsurface fires in landfills or coal refuse piles

285 5.6 Interpreting thermal scanner imagery (cont.)Most application  qualitative Some application  quantitative Considerations of the time for acquiring thermal data Fig 5.15: generalized diurnal radiant T variations. Water  range of T is small Water  Tmax(water)  Tmax(rock)+1~2 Crossovers Thermal conductivity Thermal capacity Thermal inertia ( conductivity) Predawn imagery  preferred but difficult

286 5.6 Interpreting thermal scanner imagery (cont.)Fig 5.16: day time and night time imagery Water Land Trees Paved area Fig 5.17: day time thermal imagery Beach ridge(B) Lakebed soil(A) Trees( C ) Bare soil (D) Mowed grass (E)

287 5.6 Interpreting thermal scanner imagery (cont.)Fig 5.18: night time thermal imagery Cows Deers Metal roof Fig 5.19: high resolution daytime thermal image Helicopter shadows Fig 5.20: Daytime thermal imagery Hot water Circulations

288 5.6 Interpreting thermal scanner imagery (cont.)Fig 5.21 nighttime thermal image depicting building heat loss. Fig 5.22 Thermovision camera and display unit Fig 5.23 Thermal images showing streamline heat loss. NASA’s Thermal infrared Multispectral Scanner (TIMS). Plate 10(a). TIMS image Plate 10(b). A generalized lithologic map

289 5.7 Geometric characteristics of across-track scanner imageryAlong-track scanner  no mirror, fixed geometric relationship Across-track scanner  systematic and random geometric variations Spatial resolution and ground coverage Table 5.4 W= 2H΄tan

290 5.7 Geometric characteristics of across-track scanner imagery (cont.)Tangential scale distortion (TSD) Fig 5.24: Source of tangential scale distortion dq /dt=const doesn't mean dx/dt=const Fig 5.25: the effect of tangential distortion Constant longitudinal scale & varying lateral scale S-shaped sigmoid curvature Fig5.26: Comparison of aerial photo & across-track thermal scanner image Fig 5.27: tangential scale distortion compression near the top-bottom edge

291 5.7 Geometric characteristics of across-track scanner imagery (cont.)Tangential scale distortion (cont.) Fig 5.28 correction of TSD qp= ypqmax/ymax Yp= H΄tanqp Electronically or digitally correct TSD  rectilinearized images

292 5.7 Geometric characteristics of across-track scanner imagery (cont.)Resolution cell size (RCS) variations Fig 5.29 // fight direction  H΄b  Hq΄b =H΄secqb  flight direction  H΄b  Hq ΄secq=Hq΄b secq=H΄sec2qb Response time: time that a scanner takes to respond electronically to a change in ground reflected or emitted energy Variations of RCS can be 3~4 times along a scan line Limitation of RCS to the image analysis Object>pixel(RCS)  representative  analysis Compensation of irradiance fall-off due to RCS

293 5.7 Geometric characteristics of across-track scanner imagery (cont.)1-D relief displacement (RD) Fig 5.30: RD in a vertical aerial photo vs an across-track scanner image Dynamic continuous process  sensitive to the aircraft attitude deviations (Fig 5.34) Fig 5.31: across-truck thermal scanner image illustrating 1-D RD.

294 5.7 Geometric characteristics of across-track scanner imagery (cont.)1-D relief displacement (RD) Fig 5.32: 1-D RD and TSD Fig 5.33: Effect of nonsynchronized image recording rate and V/H΄ ratio Fig 5.34: distortions induced by aircraft attitude deviations Fig 5.35: effect of roll compensation

295 5.8 Radiometric calibration of thermal scannersThermal scanner image  lack of geometric integrity Take photo simultaneously Night time  use daytime photo (even old photos will do)

296 5.8 Radiometric calibration of thermal scanners (cont.)Two methods of radiometric calibration Internal blackbody source referencing Two controlled sources: cold & hot View the sources during every scan line Fig 5.36: configuration A typical mission: height 600m  0.30C but atmospheric condition can be 20C Air-to-ground correlation: Account for atmospheric effects Theoretical approach Empirical approach  the general approach Fig 5.37 a sample of calibration curve Fig 5.38 a thermal radiometer used for air-to-ground correlation measurements

297 5.9 Temperature mapping with thermal scanner dataMany applications  Map Two procedures Image-based Crude form: qualitative depiction low geometric and radiometric integrity Suffice for many application Tonal levels  densitometric analysis  rectilinerarized imagery. Numerically based N=A+BM=A+BeT4 N: DN A, B : to be determined e: emissivity T: kinetic temperature Two unknown (A, B)  two equations (data)

298 5.10 FLIR systems Forward-looking infrared (FLIR) system  oblique views of the terrain ahead of an aircraft Fig 5.39 Point forward  sweep across the scene of interest Military applications

299 5.11 Imaging spectrometry Hyperspectral  multispectralFig 5.40 Fig 5.41: Selected laboratory spectra of minerals Hyperspectral data  detailed identification of materials and quantification of their abundance Airborne Imaging Spectrometer (AIS) 128 channel from 1.2~2.4 mm with 9.3 nm band width 1.9 mrad, 4200m high, resolution 8x8m, swath 32 pixel

300 5.11 Imaging spectrometry (cont.)Airborne Imaging Spectrometer (cont.) Fig 5.42: AIS images Condominium complex a: school b: field Loss of detail in the atmospheric water absorption band (1.4mm) Watered courtyard lawn a (compared to unwatered field b) Water vapor absorption bands 0.94, 1.14, 1.4, 1.88 mm

301 5.11 Imaging spectrometry (cont.)Fig 5.43: selected laboratory spectra of green leaves Main components: chlorophyll and water Shape + peak+valley  species Fig 5.44: same as Fig 5.43 but single species Main components: lignin & holocellulose Difference  green or dry  holocellulose Assessing biomass Vegetation stress Carbon cycle Discriminating plant communities Phenological conditions

302 5.11 Imaging spectrometry (cont.)Airborne Visible-Infrared Imaging Spectrometer (AVIRIS) 224 channels, 0.4~2.45 mm with bandwidth 9.6nm Altitude: 20km, resolution 20m, swath: 10km Plate 11: hyperspectral image cube A common way to display hyperspectral data Color composite: 0.557(b), 0.815(g), 2.209(r) mm Fig 5.45 Three stages: green leaf initiation  development  change in plant pigment proportions  flowering  domancy  leaf loss  senescence Applications AVIRIS data  spectra matching  library  mineral type An export system-based analysis approach

303 5.11 Imaging spectrometry (cont.)Compact Airborne Spectrographic Imager (CASI) Commercial, programmable 288 channels from 0.4~0.9mm with bandwidth 1.8 nm IFOV= 1.2 mrad

304 5.11 Imaging spectrometry (cont.)Geophysical and Environmental Research Imaging Airborne Spectrometer (GERAIS) 63 channels, 0.4~2.48 mm Fig 5.46: GERAIS Image for identifying minerals Label A in channel 41  dark spot  alunite Alunite  dark from 41~44  absorption light from 49~52  reflection Red edge The slope of a reflectance spectrum from 0.68~0.76 mm Shift  change in the chemical and morphological status  heavy metals in the soil

305 5.11 Imaging spectrometry (cont.)DIAS-7915 ASDIS-204

306 5.12 Conclusion

307 Earth resource satellites operating in the optical spectrumChapter 6 Earth resource satellites operating in the optical spectrum Introduction to Remote Sensing Instructor: Dr. Cheng-Chien Liu Department of Earth Sciences National Cheng-Kung University Last updated: 28 May 2003

308 6.1 Introduction Remote sensing + space exploration (RS+SE)  interest and application over a wider range of disciplines Current application New technology  new or improved satellite/sensor  new application The most important outcome of RS+SE  observing earth  earth system

309 6.1 Introduction (cont.) This chapter  optical range  0.3 m m~14 m mLandsat Spot NOAA series

310 6.2 Early history of space imagingLudwig Bahrmann (1891): New or improved apparatus for obtaining Bird’s eye photographic views Alfred Maul (1907): gyrostabilization Alfred Maul (1912): 41kg, 200mm x 250 mm, 790m 1946~1950: V2 rockets 1960~ : TIROS-1, early weather satellite Not just look at but also look through

311 6.2 Early history of space imaging (cont.)1960s: Mercury, Gemini, Apollo Alan Shepard, 1961, 70 mm, 150 photos John Glenn, 1962, 35 mm, 48 photos. Later Mercury missions: 70 mm, 80 mm Gemini GT-4 mission: formal experiment directed at geology Tectonics, volcanology, geomophology. 1:2, photos Apollo 9: 4 camera array, electrically triggered. 140 sets of imagery

312 6.2 Early history of space imaging (cont.)Skylab 1973 Earth Resources Experiment Package (EREP) 6-camera multi-spectral array A long focal length “earth terrain” camera A 13-channel multispectral scanner A pointable spectroradiometer Two microwave systems. 35,000 images U.S.-USSR Apollo-Soyuz Test Project (ASTP)

313 6.3 Landsat satellite program overviewEarth Resources Technology Satellite (ERTS) 1967 ERTS-1, 1972~1978 Nimbus weather satellite  modified Experimental system  test feasibility Open skies principle Landsat-2, 1975 (ERTS-2)

314 6.3 Landsat satellite program overview (cont.)Table 6.1: Characteristics of Landsat 1~6 Return Beam Vidicon (RBV) camera systems Multispectral Scanner system (MSS) Thematic Mapper (TM) Enhanced Thematic Mapper (ETM) Table 6.2: Sensors used on Landsat 1~6 missions

315 6.4 Orbit characteristic of Landsat-1, -2, and –3Fig 6.1: Landsat –1, -2, and –3 observatory configuration 3m x 1.5m, 4m width of solar panels, 815 kg, 900 km Inclination = 90 To= 103 min/orbit Fig 6.2: Typical Landsat-1, -2 and –3 daily orbit pattern Successive orbits are about 2760km Swath: 185km Orbital procession  18 days for coverage repetition 20 times of global coverage per year

316 6.4 Orbit characteristic of Landsat-1, -2, and –3 (cont.)Sun-synchronous orbit 9:42 am  early morning skies are generally clearer than later in the day Pros: repeatable sun illumination conditions on the same day in every year Cons: variable sun illumination conditions with different locations and seasons  variations in atmospheric conditions

317 6.5 Sensors onboard Landsat-1, -2 and –33-Channel RBV 185km x 185 km Ground resolution: 80m Spectral bands: 1: mm~0.575 mm (green) 2:0.580 mm~0.680 mm (red) 3: mm~0.830 mm (NIR) Expose  photosensitive surface  scan  video signal Pros: Greater cartographic fidelity Reseau grid  geometric correction in the recording process

318 6.5 Sensors onboard Landsat-1, -2 and –3 (cont.)3-Channel RBV (cont.) Landsat-1: malfunction  only 1690 scenes Landsat-2  only for engineering evaluation  only occasionally RBV imagery was obtained. Landsat-3 Single broad band (0.505~0.75 u mm) 2.6 times of resolution improved: 30m  double f Two-camera side-by-side configuration with side-lap and end-lap. (Fig 6.4) Fig 6.5: Landsat-3 RBV image

319 6.5 Sensors onboard Landsat-1, -2 and –3 (cont.)4 Channel MSS 185km x 185km Ground resolution: 79m Spectral band: Band 4: 0.5 mm ~ 0.6 mm (green) Band 5: 0.6 mm ~ 0.7 mm (red) Band 6: 0.7 mm ~ 0.8 mm (NIR) Band 7: 0.8 mm ~ 0.9 mm (NIR) Band 8: 10.4~12.6 um  Landsat-3, failed Band 4~7  band 1~4 in Landsat-4, -5 Fig 6.6: Comparison of spectral bands

320 6.5 Sensors onboard Landsat-1, -2 and –3 (cont.)4 Channel MSS (cont.) Fig 6.7: Landsat MSS operating configuration Small TFOV  use an oscillating scan mirror A-to-D converter (6 bits) Pixel width: 56m x 79m  set by the pixel sampling rate (Fig 6.8) Each Landsat MSS scene  185km x 185km 2340 scan lines, 3240 pixels per line, 4 bands Enormous data Fig 6.9: Full-frame, band 5, Landsat MSS scene Parallelogram  earth’s rotation 15 steps Tick marks  Lat. Long. Annotation block Color composite: band 4 (b), band 5 (g), band 7(r) (Fig 6.6)

321 6.5 Sensors onboard Landsat-1, -2 and –3 (cont.)Data distribution Experiment  transitional  operational NASA NOAA NASA USGS EOSAT USGS Landsat-1,-2,-3 Landsat-4,-5,-6 Landsat-7 Department of Interior Department of Commerce Department of Defense Data receiving station Data reprocessing Data catalogue

322 6.6 Landsat MSS image interpretationApplications: agriculture, botany cartography, civil engineering, environmental monitoring, forestry, geography, geology, geophysics, land resources analysis, land use planning, oceanography, water resource analysis Comparison of Landsat & airborne image Table 6.4 Resolution Coverage Complementary not replacement 2-D, non-stereo mode

323 6.6 Landsat MSS image interpretation (cont.)Characteristics of MSS image Effective resolution  79m, (30m for Landsat-3) but linear feature with sharp contrast can be seen 1-D displacement relief (in E-W direction) Limited area can be viewed in stereo  study topographic High altitude + low TFOV  little RD  planimeter map E.g. World Bank, USGS. DMA, petroleum company

324 6.6 Landsat MSS image interpretation (cont.)Characteristics of MSS image (cont.) Band 5 (red)  better atmospheric penetration  detecting cultural features Band 4 (green)  deep, clear water penetration Band 6, 7  lineating water bodies (dark) The largest single use of Landsat MSS data  geologic studies  band 5.7

325 6.6 Landsat MSS image interpretation (cont.)Fig 6.10 : four Landsat MSS bands Extent of the urban area (B4, 5, light) Major road (B4, 5 light, not B6, B7 dark) Airport Asphalt-surfaced runways Four major lakes and connected river (B6, 7 dark) mid-July  algae  green  B4: similar to the surrounding agricultural land Agricultural field. (B5, 6, 7) Forest (B4, 5 dark)  winter images are preferred

326 6.6 Landsat MSS image interpretation (cont.)Fig 6.11: Landsat MSS band 5 December image 20 cm snow covered  all water bodies are frozen Snow covered upland and valley floors  light tone Steep, tree-covered valley sides  dark tone September image Identify forest area

327 6.6 Landsat MSS image interpretation (cont.)A hit-or-miss proposition Some events leave lingering trace Fig 6.12: Landsat MSS band 7 July image  200 m3/sec March image  1300 m3/sec  once every four years Fig 6.13: Mississippi River Delta Silt flow but vague boundary  band 5 Delineation of the boundary  band 7 Fig 6.14: short-lived phenomena Active forest fire in Alaska Volcanic eruption on Kunashir Island

328 6.6 Landsat MSS image interpretation (cont.)A hit-or-miss proposition (cont.) Fig 6.15: Extensive geologic features visible on MSS San Andreas fault, Six solid dots  earthquake > 6.0 Fig 6.16: Landsat MSS band 6 66-km-wide Manicouagan ring  212-million-year-old meteorite impact crater Fig 6.17: Landsat MSS images of Mt. St. Helens before and after its 1980 eruptions Fig 6.18: Landsat MSS image of Maritoba, Canada, showing tornado and hail scar Fig 6.19: Landsat MSS image of East kalimantan, Indonesia, showing tropical deforestation

329 6.7 Orbit characteristics of Landsat-4 and -5Fig 6.20: Sun-synchronous orbit of Landsat-4 and –5 Altitude: 900  705km Retrievable by the space shuttle Ground resolutions Inclination T=99min  14.5 orbit/day 9:45 am Fig 6.21: adjacent orbit space = 2752km 16-day repeat cycle 8-day phase between Landsat-4 and –5 (Fig 6.22)

330 6.8 Sensors onboard Landsat-4 and -5Fig 6.23: Landsat-4 and –5 observatory configuration MSS, TM 2000 kg, 1.5x2.3m solar panels x 4 on one side High gain antenna  Tracking and Data Relay Satellite system (TDRSS) Direct transmission  X-band and S-band MSS: 15 Mbps TM: 85 Mbps

331 6.8 Sensors onboard Landsat-4 and –5 (cont.)MSS Same as previous except for larger TFOV for keeping the same ground resolution (79m  82m) Renumber bands TM 7 bands (Table 6.4) DN: 6  8 bits Ground resolution: 30m (thermal band: 120m) Geometric correction  Space Oblique Mercator (SOM) cartographic projection

332 6.8 Sensors onboard Landsat-4 and –5 (cont.)TM (cont.) Bi-directional scan  the rate of oscillation of mirror dwelling time  geometric integrity signal-to-noise Detector: MSS: 6x4=24 TM: 16x6+4x1=100 Fig 6.24: Thematic Mapper optical path and projection of IFOV on earth surface Fig 6.25: Schematic of TM scan line correction process

333 6.9 Landsat TM Image interpretationPros: Spectral and radiometric resolution Ground resolution Fig 6.26: MSS vs TM Fig 6.27: All seven TM bands for a summertime image of an urban fringe area Lake, river, ponds: b1,2 > b3 > b4=b5=b7=0 Road urban streets: b4  min Agricultural crops: b4  max Golf courses

334 6.9 Landsat TM Image interpretation (cont.)Fig 6.27 (cont.) Glacial ice movement: upper right  lower left Drumlins, scoured bedrock hills Band 7  resample from 120m to 30m Plate 12 + Table 6.5: TM band color combinations (a): normal color  mapping of water sediment patterns (b): color infrared  mapping urban features and vegetation types (c)(d): false color

335 6.9 Landsat TM Image interpretation (cont.)Fig 6.28: Landsat TM band 6 (thermal infrared) image Correlation with field observations  6 gray levels  6T Plate 13: color-composite Landsat TM image Extremely hot  blackbody radiation  thermal infrared TM bands 3, 4 and 7

336 6.9 Landsat TM Image interpretation (cont.)Fig 6.29: Landsat TM band 5 (mid-infrared) image Timber clear-cutting Fig 6.30: Landsat TM band 3, 4 and 5 composite Extensive deforestation. Fig 6.31: Landsat TM band 4 image map 13 individual TM scenes + mosaic

337 6.10 Landsat-6 planned missionA failed mission Enhanced Thematic Mapper (ETM) TM+ panchromatic band (0.5~0.9 mm) with 15m resolution. Set 9-bit A-to-D converter to a high or low gain 8-bit setting from the ground. Low reflectance  water  high gain Bright region  deserts  low gain

338 6.11 Landsat ETM image simulationFig 6.32: Landsat ETM images

339 6.12 Landsat-7 Launch: 1999 Web site: http://landsat.gsfc.nasa.govLandsat 7 handbook Landsat 7 in orbit Depiction of Landsat 7

340 6.12 Landsat-7 (cont.) Landsat 7 Orbit Landsat data Orbital pathsSwath Swath pattern Landsat data

341 6.12 Landsat-7 (cont.) Payload Enhanced Thematic Mapper Plus (ETM+)Dual mode solar calibrator Data transmission TDRSS or stored on board. GPS  subsequent geometric processing of the data High Resolution Multi-spectral Stereo Imager (HRMSI) 5m panchromatic band 10m ETM bands 1~4 Pointable  revisit time (<3 days) Stereo imaging. 00~380 cross-track and 00~300 along-track

342 6.12 Landsat-7 (cont.) Application Monitoring Temperate ForestsMapping Volcanic Surface Deposits Three Dimensional Land Surface Simulations

343 6.13 SPOT Satellite ProgramBackground French+Sweden+Belgium 1978 Commercially oriented program SPOT-1 French Guiana, Ariane Rocket 1986 Linear array sensor+pushbroom scanning+pointable Full-scene stereoscopic imaging

344 6.13 SPOT Satellite Program (cont.)1990 SPOT-3 1993

345 6.14 Orbit characteristics of SPOT-1, -2 and -3Circular, near-polar, sun-synchronous orbit Altitude: 832km Inclination: 98.70 Descend across the equator at 10:30AM Repeat: 26 days Fig 6.33: SPOT revisit pattern at latitude 450 and 00 At equator: 7 viewing opportunities exist At 450: 11 viewing opportunities exist

346 6.15 Sensors onboard SPOT-1, -2 and -3Configuration (Fig 6.34) 223.5m, 1750 kg, solar panel: 15.6m Modular design High Resolution Visible (HRV) imaging system 2-mode 10m-resolution panchromatic mode (0.51~0.73mm) 20m-resolution color-infrared mode. (0.5~0.59mm, 0.61~0.68mm, 0.79~0.89mm)

347 6.15 Sensors onboard SPOT-1, -2 and –3 (cont.)HRV (cont.) Pushbroom scanning No moving part (mirror)  lifespan Dwell time  Geometric error  4-CCD subarray 6000-element subarray  panchromatic mode, 10m Three 3000-element subarrays  multi-spectral mode, 20m 8-bit, 25 Mbps Twin-HRV instruments IFOV (for each instrument)  4.130 Swath: 60km  2 - 3km = 117km (Fig 3.36) TFOV (for each instrument)  270=0.6045 (Fig 3.35)

348 6.15 Sensors onboard SPOT-1, -2 and –3 (cont.)HRV (cont.) Data streams Although 2-mode can be operated simultaneously, only one mode data can be transmitted  limitation of data stream Stereoscopic imaging Off-nadir viewing capability (Fig 6.37) Frequency  revisit schedule (Fig 6.33) Base-height ratio  latitude 0.75 at equator, 0.5 at 450 Control Ground control station  Toulouse, France  observation sequence Receiving station  Tordouse or Kiruna, Sweden Tape recorded onboard Transmitted within 2600km-radius around the station

349 6.16 SPOT HRV image interpretationFig 6.38: SPOT-1 panchromatic image 10m-resolution Cf: Landsat MSS 80m Cf: Landsat TM 30m (Fig 6.26) Cf: Landsat ETM 15m (Fig 6.32) Fig 6.39: SPOT-1 panchromatic image Plate14: merge of multispectral & panchromatic data Fig 6.40: SPOT-1 panchromatic image stereopair Plate 15: Perspective view of Alps SPOT stereopair + parallax calculation Plate 23 Fig 6.41: before and after the earthquake

350 6.17 SPOT –4 and –5 SPOT –4 Launched 1998Vegetation Monitoring Instrument (VMI) Swath: 2000km  daily global coverage Resolution: 1km Spectral band: b(0.43~0.47mm), g(0.5~0.59mm), r(0.61~0.68mm), N-IR(0.79~0.89mm), mid-IR(1.58~1.75mm)

351 6.17 SPOT –4 and –5 (cont.) SPOT – 5 Launched 2002Vegetation Monitoring Instrument (VMI) Swath: 2000km  daily global coverage Resolution: 1km Spectral band: b(0.43~0.47mm), g(0.5~0.59mm), r(0.61~0.68mm), N-IR(0.79~0.89mm), mid-IR(1.58~1.75mm)

352 6.18 Meteorological SatelliteMetsats Coarse spatial resolution  land-oriented system Very high temporal resolution of global coverage NOAA satellites  sun-synchronous GOES  geostationary  36,000km altitude DMSP

353 6.18 Meteorological Satellite (cont.)NOAA satellites Advanced Very High Resolution Radiometer (AVHRR) NOAA –6 ~ -12. (N-S) Even: 7:30AM crossing time Odd: 2:30 AM crossing time Table 6.6: characteristics of NOAA-6 ~ -12 Fig 6.42: Example coverage of the NOAA AVHRR Ground resolution: 1.1km at nadir AVHRR data LAC GAC Fig 6.43: Comparison of Spectral sensitivity

354 6.18 Meteorological Satellite (cont.)NOAA satellites (cont.) Fig 6.44: AVHRR images A: distortion  wide angle of view B: geometric correction Plate 16: NOAA AVHRR band 4 thermal image of the Great Lakes Fig 6.45: AVHRR images of the Mississippi Delta (a): present and past channels, future  Atchafalaya (b): Channel–1 (red), silky material  visible (c): Channel–2 (Near-IR), light tone  higher & drier (d): Channel–4 (thermal –IR) light tone  cooler Plumes of cooler river water

355 6.18 Meteorological Satellite (cont.)NOAA satellites (cont.) Plate 17: springtime NOAA-8 AVHRR color composite Applications of AVHRR in monitoring vegetation Use Ch-1 (0.58~0.68 mm) and Ch-2 (0.73~1.10 mm) A simple vegetation index VI=Ch2-Ch1 Normalized difference vegetation index NDVI = (Ch2-Ch1)/(Ch2+Ch1) Vegetated areas  large VI Clouds, water, snow  negative VI Rock, Bare soil  VI  0 For global vegetation  NDVI preferred  compensate the charging illumination conditions Plate 18: color-coded NDVI Select the highest NDVI during that period

356 6.18 Meteorological Satellite (cont.)NOAA satellites (cont.) Applications of AVHRR in monitoring vegetation (cont.) Applications: vegetation seasonal dynamics at global and continental scale, tropical forest clearance, leaf area index measurement, biomass estimation, percentage ground cover determination, photosynthetically active radiation estimation Other factors that might influence NDVI Incident solar radiation Radiometric response of the sensor Atmospheric effect and viewing angle  need further research

357 6.18 Meteorological Satellite (cont.)GOES (Geostationary Operational Environmental Satellites) NOAA + NASA 1974 36,000km USRS, ESA, NSDA Fig 6.46: GOES –2 visible band (0.55~0.7 mm) Frequency: 2/hour VI (daytime), IR (day and night)

358 6.18 Meteorological Satellite (Cont.)Defense Meteorological Satellite Program (DMSP) 1973 0.4~1.1 mm (VI+N-IR) Nighttime visible band  tune the amplifiers Fig 6.47: DMSP nighttime image Fig 6.48: Maps of population distribution

359 6.19 Ocean monitoring satellitesOcean  Land 2/3, but comparatively little is know Seasat (see §8.9) Nimbus –7 CZCS (Coastal Zone Color Scanner) 1978~1986 Proof of concept mission Table 6.7: CZCS bands  narrow bandwidth 825m resolution at nadir, 1566km swath Map phytoplankton concentrations and inorganic suspended matter N-IR  separate water from land

360 6.19 Ocean monitoring satellites (cont.)Japan Marine Observation Satellite (MOS)-1: 1987 MOS-1b: 1990 Table 6.8: Instruments included in MOS-1 and MOS-1b 4-Channel Multi-spectral Electronic Self-Scanning Radiometer (MESSR) 4-Channel Visible and Thermal Infrared Radiometer (VTIR) 2-Channel Microwave Scanning Radiometer (MSR) 909km altitude, revisit period:17days

361 6.19 Ocean monitoring satellites (cont.)Sea-viewing Wide-Field-of-View Sensor (SeaWiFS) 8-channel across-track scanner (0.402~0.885 mm) Ocean biogeochemistry NASA-orbital science corporation (OSC) 1998 – date Data LAC: 1.13km GAC: 4.52km 705km altitude, 2800km swath

362 6.20 Earth Observing SystemMission to Planet Earth (MTPE) Aims: providing the observations, understanding, and modeling capabilities needed assess the impacts of natural events and human-induced activities on the earth’s environment Data and information system: acquire, archive and distribute the data and information collected about the earth Further international understanding of the earth as a system

363 6.20 Earth Observing System (cont.)EOS (Table 6.9) ASTER CERES MISR MODIS MOPITT MODIS (Table 6.10) Table 6.10 Terra: 2000 Aqua: 2002

364 6.21 Fine-resolution satellite systemCORONA 1960 – 1972, declassified in 1995 KH-1 ~ KH-4B ~ KH-5 Camera + film Band and resolution Web site: Impacts

365 6.21 Fine-resolution satellite system (cont.)IKONOS 1999 by Space imaging Bands and resolution 1m-resolution 0.45 – 0.90 mm 4m-resolution 0.45 – 0.52 mm 0.52 – 0.60 mm 0.63 – 0.69 mm 0.76 – 0.90 mm Orbit: sun-synchronous Repeat coverage: 1.5 (1m) ~ 3 (4m) days

366 6.21 Fine-resolution satellite system (cont.)OrbView–3 and –4 OrbView-2: SeaWiFS Will be launched soon! Similar bands and resolution as IKONOS OrbView–4 200 spectral channels in the range 0.45 – 2.5 m m at 8m resolution

367 6.21 Fine-resolution satellite system (cont.)QuickBird 2001 by EarthWatch Inc. Bands and resolution 61cm-resolution 0.45 – 0.89 mm 2.44m-resolution 0.45 – 0.52 mm 0.52 – 0.60 mm 0.63 – 0.69 mm 0.76 – 0.89 mm

368 Chapter 8 Microwave sensingIntroduction to Remote Sensing Instructor: Dr. Cheng-Chien Liu Department of Earth Sciences National Cheng-Kung University Last updated: 4 June 2003

369 8.1 Introduction Microwave Features:1mm~1m  not micro at all Features: Penetration  any weather condition Irrelevant to visible light Active and passive, airborne and spaceborne

370 8.2 Radar development Radio detection and ranging (RADAR)pulse of microwave energy  objects  echoes  detect and ranging Nonimaging radar: e.g. Doppler radar  Doppler frequency shift  velocity Plan position indicator (PPI) circular display screen rotating antenna weather forecasting, air traffic control, navigation Poor spatial resolution  not appropriate for R.S.

371 8.2 Radar development(Cont.)Side-looking airborne radar (SLAR) Side-looking radar (SLR) Antenna  fixed below the aircraft  point to the side Fig 8.1: SLAR image Continuous strips  depicting large ground areas 1950s  military reconnaissance All-weather operating capability Active, day-or-night imaging system Declassification  time lag  non-military targets An active state of advancement

372 8.2 Radar development(Cont.)Applications of SLAR A complete survey of the Darien province of Panama Fig 8.2 1967 Persistent cloud cover Mapping Venezuela Fig 8.3 1971 Improve the accuracy of the country boundary Water resourcessource of several rivers Project Radam (Radar of the Amazon) 1971~1976 Geologic analysis, timber inventory, transportation route location, mineral exploration

373 8.2 Radar development (cont.)Applications of SLAR (cont.) Applications in ocean Determine wind, wave and ice conditions, internal waves Study ocean bottom contours Spaceborne radar remote sensing 1978  Seasat Shuttle Imaging Radar 1980s: Soviet Cosmos experiments 1991: Almaz-1, ERS-1, JERS-1 1991: Radarsat

374 8.3 SLAR system operation Fig 8.4: Operating principle of SLARSolid lines: radar pulse sent Dashed lines: return signals Signal from tree: later and smaller than signal from house The slant range SR=ct/2 SR: direct distance between transmitter and object

375 8.3 SLAR system operation (cont.)Fig 8.5: How to create a SLAR image fly speed Va synchronizer switch  transmitter  receiver transmitted pulse receive and process record

376 8.4 Spatial resolution of SLAR systemsGround resolution cell size  Pulse length  range direction Antenna bandwidth  azimuth direction Range resolution Fig 8.6: Dependence of range resolution on pulse length Pulse length PL Slant-range distance ABsr=ABcosq If ABsr >1/2PL separate signal  differentiable If ABsr <1/2PL overlapped signal  one large object

377 8.4 Spatial resolution of SLAR systems (cont.)Fig 8.7: slant-range resolution  ground range resolution Depression angle qd Look angle ql ABsr=Abcosqd AB=Rr=ABsr/cosqd =ct/2cosqd Example 8.1

378 8.4 Spatial resolution of SLAR systems (cont.)Azimuth resolution Fig 8.8: Dependence of Ra, b, GR Ra=GRb Example 8.2 Antenna beamwidth: b=l/AL ALbRa Physical length of antenna Brute force, real aperture, noncoherent radar e.g. l=5cm, b=10mrad  AL=5m if b =2mrad  AL=25m Simple for design and data processing Short l, short range, low altitude

379 8.4 Spatial resolution of SLAR systems (cont.)Azimuth resolution (cont.) ALbRa (cont.) Synthesizing and effective length of antenna Synthetic Aperture Radar (SAR) Complex Single physically short antenna+motion along the flight line  successive elements of a single, long synthetic antenna Near range  fewer elements Ra=constant  fn(Range) Another view of explaining how SAR operate (Fig 8.10) Ahead of the aircraft  upshifted f Behind the aircraft  downshifted f

380 8.5 Geometric characteristics of SLAR imageryDifferent from photo and scanner imagery Slant-range scale distortion Fig 8.11: Slant-range vs ground-range image format A=B=C but A1 GR=(SR2-H΄2)1/2 Range scale = fn(H΄) Azimuth scale = fn(Vair, VCRT) Inertial navigator and control system  strict control of flight parameters  reconcile and equalize these independent scales

381 8.5 Geometric characteristics of SLAR imagery (cont.)Relief displacement Fig 8.12: RD on SLAR versus photos Layover effect: a vertical feature lay over the closer features and appears to lean toward the nadir Fig 8.13: Effects of terrain relief on SLAR images Terrain slope steeper than lines perpendicular to the direction of the radar pulse  layover effect. D: no layover but foreshortening effect C: image of the front slope  foreshortened  0 B: layover effect, right side is facing away from the radar antenna  no return signal  dark A: layover effect, right side is also illuminated  weak return

382 8.5 Geometric characteristics of SLAR imagery (cont.)Parallax Fig 8.14: Flight orientation to produce parallax a: opposite side b: same side but different altitude  altitude parallax Fig 8.15: Stereo SLAR image, flying same flight line at different altitude

383 8.6 Transmission characteristics of Radar signalsTable 8.1: Radar band designation Letter codes  arbitrarily selected for military security Wavelength l  atmospheric attenuation/dispersion Precipitation echoes  D6/l4 D: drop diameter Applications: PPI  range of heavy rain e.g. l= 1cm  echo, l=3cm  no echo

384 8.6 Transmission characteristics of Radar signals (cont.)Polarization The signal can be filtered in such a way that its electrical wave vibrations are restricted to a single plane perpendicular to the direction of wave propagation Send: H,V Receive: H,V Like-polarized: HH, VV Cross-polarized: HV, VH Circular polarization Mode of polarization  details see §8.8

385 8.7 Earth surface feature characters influencing radar returnsGeometric characteristics: Fig 8.16: Effect of sensor/terrain geometry Local incident angle (Fig 8.17): qi Flat terrain: qi =ql Consider earth curvature: qi >ql Radar shadow  complete dark and sharp Factors that dominate radar image 0 < qi < 300: topographic slope 300 < qi < 700: surface roughness 700 < qi : radar shadows

386 8.7 Earth surface feature characters influencing radar returns (cont.)Geometric characteristics (cont.) Fig 8.18: Radar reflection from various surfaces Rayleigh criterion: SVHrms > l/8cosqi  Rough surfacediffuse reflectorsignificant return SVHrms  l/8cosqi  Smooth surfacespecular reflectorlow return Modified Rayleigh criterion: SVHrms> l/(4.4cosqi)  rough SVHrms< l/(25cosqi)  smooth Others: intermediate Table 8.2: various bands, various qi Corner reflector: double reflection  bright, sparkles (see Fig 8.22 and 8.23)

387 8.7 Earth surface feature characters influencing radar returns (cont.)Electrical characteristics Complex dielectric constant (DC)  reflectivity & conductivity Dry natural material: DC  3~8 Water: DC  80 Moisture  DC Plant: moisture  good reflectors Metal  DC (e.g. metal bridges in Fig 8.22a)

388 8.7 Earth surface feature characters influencing radar returns (cont.)Soil Response Soil moisture  DC  limit radar penetration Extremely dry soil condition  penetration of L-band (Fig 8.30) Vegetation response Size Vegetation canopy: leaves, stems, stalks, limbs,…. Underlying soil Wavelength Short (2~6cm)  sensing crop canopies and tree leaves Longer (10~30cm)  sensing tree trunks and limbs

389 8.7 Earth surface feature characters influencing radar returns (cont.)Vegetation response (cont.) Other factors: Moisture Like-polarized (HH or VV) penetrates more than cross-polarized Align in the azimuth direction qi

390 8.7 Earth surface feature characters influencing radar returns (cont.)Water and ice response: Smooth water surface  specular reflector  no returns Rough water surface  varying strengths of returns Wave moving toward or away from the radar system  easier to detect Sea ice  dielectric properties Ice age, surface roughness, internal geometry, temperature, snow cover…

391 8.8 Interpretation of SLAR imageryApplications of SLAR image Mapping major rock units and surficial materials Mapping geologic structure (folds, faults and joints) Mapping vegetation types Determining sea ice types Mapping surface drainage features l v.s. roughness Table 8.3 Fig 8.19

392 8.8 Interpretation of SLAR imagery (cont.)Intensity of return signal High  slopes facing aircraft, rough objects, high moisture, metal, urban building (corner reflection.) Low  smooth water, pavements, playas No  radar shadow Speckle: Grainy or salt-and-pepper pattern Random constructive and destructive interface  random bright and dark areas

393 8.8 Interpretation of SLAR imagery (cont.)Multiple-look processing Reduce speckle Average several independent images for the same area Amount of speckles  (number of looks)-1/2 The size of the resolution cell & number of looks

394 8.8 Interpretation of SLAR imagery (cont.)Fig 8.20: SLAR image Large synclinal mountain  upper left and center Flight on the top  lighter tone on the slopes facing up Return signals  vegetation surfaces Banding around mountain  alternation of bedrock types River and lake  dark tone HV shows less contrast than HH It’s not always possible to predict HH or HV is better

395 8.8 Interpretation of SLAR imagery (cont.)Fig 8.21 Basaltic lava flow C: The “Sunshine Basalt” flow  Sunshine Crater Darker tone on the HV image. D: The “Pisgah Basalt” flow  Pisgah Crater A: lighter tone on the HV image  greater density of vegetation B: boundary Playa (dry lakebed)  dark tone Gravel road

396 8.8 Interpretation of SLAR imagery (cont.)Fig 8.22a: small urban area Large building  corner reflection  high return Metallic bridge  high return River water  dark Rectangle field Bed sedimentary rock Fig 8.22b: horizontally bedded sedimentary rocks with loess cover Potential soil erosion  strip farming  contour lines

397 8.8 Interpretation of SLAR imagery (cont.)Fig 8.23: IRIS: high resolution mode 3m  6m resolution Wide swath mode: 18m  10m resolution Malldark Potomac Riverblack Airport Fig 8.24: Multiwavelength SLAR image Wooded area: diffuse reflectors in both X and L band Cultivated fields: diffuse reflectors in X band but acts as near-specular reflectors in the L band

398 8.9 Experimental Radar remote sensing from spaceSeasat-1 , (99 days) 800km, near-polar orbit SAR, L-band (23.5cm), HH polarization Swath: 100 km Resolution: 25m25m (range  azimuth) Original goals: Global sea surface wave field Polar sea ice conditions. Also revealed applications to Ocean: internal waves, current boundaries, eddies, fronts, bathymetric features, storms, rainfalls, windrows. Land: geology water resources, land cover mapping, agricultural assessment.

399 8.9 Experimental Radar remote sensing from space (cont.)Seasat-1 images Fig 8.25: Seasat SAR image of the English Channel near the straits of Dover. Tidal variations  7m, 1.5m/s Sand bars  hazardous Fig 8.26: Pack ice Banks Island (lower right-hand portion) Brighter ice areas  rough surface  older ice Darker areas  open water or recently frozen ice 3 days  15km Fletcher’s Ice Island: 7km  12 km, 157km in 2 month Fig 8.27: Appalachian mountains. L-band, 1:575,000 Sidelighting  auticlines and synclines

400 8.9 Experimental Radar remote sensing from space (cont.)Shuttle Imaging Radar SIR-A: 1981 260km SAR. L-band (23.5cm), HH polarization 9.4m antenna, 470~530 look angle Resolution: 40m 40m (range  azimuth) Swath: 50km Fig 8.28: Saudi Arabia & Iraq. (MSS vs SAR) Dry river channels  smooth, dry layer of wind- deposited silt  lots radar return  dark Outcropping carbonate rocks  rough angular surface  strong radar return. Fig 8.29: Eastern China White spots  villages Levee Fig 8.30: Sahara Desert Penetration of L-band wave in extremely dry material  underlying bedrock structure

401 8.9 Experimental Radar remote sensing from space (cont.)Shuttle Imaging Radar (cont.) SIR-B: 1984 Tiltable antenna (150 ~ 600) Assessing the effect of various incident angles Stereo images Resolution Azimuth: 25m Range: 14m at 600~ 46m at 150 Fig 8.31: Montreal Developed area  bright area River  dark area Bridge Long, striplike patterns of agricultural fields (lower right) Fig 8.32: Mt. Shasta. a: 600, b:qi=300 Young lava flow  unvegetated angular chunks of basalt Older lava flows  darker, more vegetated

402 8.9 Experimental Radar remote sensing from space (cont.)Shuttle Imaging Radar (cont.) Fig 8.33: perspective views of Mt. Shasta Generated from Fig 8.32 Successive views taken counterclockwise around the mountain Fig 8.34: Stereopair (450 and 540) Small stereo convergence angle (90) but excellent image Snow cover  dark Fig 8.35: northern Florida Flat, 45m mean elevation, sandy soil overlay weathering, limestone, sinkhole lakes Water bodies (W), clear-cut areas(C), powerline right-of-way (P), roads ( R), Pine forest (F), Cypress-tupelo swamps (S) Cypress-tupelo swamps: dark (580)  light (450)  lighter (280) Specular reflection from the standing water + tree trunks  complex corner reflector effect

403 8.9 Experimental Radar remote sensing from space (cont.)Shuttle Imaging Radar (cont.) SIR-C/X-SAR 1994 X-band(3cm), C-band(6cm), L-band (23cm) Antenna + shuttle  pointable Swath: 15-90km Resolution: 10~200m Plate Cosmos-1870 Experiment 1987

404 8.10 ALMAZ-1 ALMAZ-1 Fig 8.36: Almaz radar image 3-31-1991~10-17-1992USSR Commercial basis Altitude: 300km  360km S-band (10cm) L= 300 ~ 600 Resolution:10m ~ 30m 2 antennas Swath: 350km  2 Radiometric scanner (RMS) Fig 8.36: Almaz radar image

405 8.11 ERS Satellite Program Agency: ESA (European Space Agency)Orbit: 777km, Sun synchronous Design life: 3 years ERS-1: 1991 ERS-2: 1995

406 8.12 Sensors onboard ERS-1 C-band AMI (active microwave instrument)Ku-band radar altimeter An along-track scanning radiometer 3 modes of AMI IMAGE WAVE WIND

407 8.13 ERS-1 AMI image interpretationAMI vs SIR-A, -B and Almaz-1 Shorter : C-band Steeper qi: qi =230 VV Fig 8.37: ERS-1 radar image Canada/USA border Milk River Fig 8.38: Center pivot irrigation area  moisture  lighter Marsh  roughness, moisture  corner reflection  lighter

408 8.13 ERS-1 AMI image interpretation (cont.)Fig 8.39: mountainous region. Effect of layover  steep ql The Pacific Ocean Fig 8.40: area of regenerating forest clearcut. Clear cut  pasture  smooth grass surface  visible Reprocess  clearcut  light-colored River

409 8.13 ERS-1 AMI image interpretation (cont.)Fig 8.42: internal waves from the Atlantic Ocean to the Mediterranean Sea. 2km Different salinities  different layer Tide  current Fig 8.43: St. Lawrance River Wind  roughened surface  lighter tone Influence of an atmospheric front on water surface roughness  pattern (upper-left  lower-right) Small river mixed with St. Lawrence River  temperature difference  arc shape

410 8.13 ERS-1 AMI image interpretation (cont.)Fig 8.44: oil slick Oil films  dampening wave  darker

411 8.14 JERS-1 JERS-1 Fig 8.45: Mt. Fuji 1992568km, sun-synchronous orbits SAR, L-band (23cm), HH polarization Swath: 75km Resolution: 18m Expected lifetime: 2 years Fig 8.45: Mt. Fuji Snow-covered  dark Lake  dark Forest  lighter-toned

412 8.15 Radarsat: Radsat 1995 798km, sun-synchronous orbitSAR, C-band (5.6cm). HH polarization Swath and resolution: Table 8.4: Radarsat beam selection modes Fig 8.46: Radarsat imaging mode Stereo coverage Data storage and transmit Applications

413 8.16 Spaceborne radar system summaryTable 8.5

414 8.17 Radar remote sensing of VenusMagellan spacecraft 1989 Elliptical polar orbit: 2100km above the poles to 175km above the equator SAR, s-band Swath 16,000km long, 25km wide Resolution: 75m Fig 8.47: Mead impact crater (d=280km) Fig 8.48: radar stereopair of the crater Geopert-Meyer Same side at =150, 280 The edge of a ridge belt Planetary scientists Fig 8.49: Sapas Mons

415 8.18 Elements of passive microwave sensingPassive microwave sensing vs thermal sensing Similar principles  blackbody radiation theory (Fig 8.50) Use antenna Fig 8.51: Components of a passive microwave signal Emitted from the surface = fn(T, material) Emitted from the atmosphere Reflected from the surface Transmitted from the surface Passive microwave sensing fn(surface electrical, chemical and textural characteristics, bulk configuration and shape, viewing angle)

416 8.19 Passive microwave sensorsMicrowave radiometers Basic configuration (Fig 8.52) Switch  rapid, alternate sampling between the antenna signal and a calibration temperature reference signal Amplify weak signal Readout and recording Trade-off between antenna beamwidth and system sensitivity Apparent antenna temperature The system is calibrated in terms of the temperature that a blackbody located at the antenna must reach to radiate the same energy as collected from the ground

417 8.19 Passive microwave sensors (cont.)Scanners Scan transverse to the direction of flight Mechanically, electronically, multiple antenna array Fig 8.53: Passive microwave image Looks like thermal image, but bright  cold Agricultural fields Striping  irrigation Density  moisture

418 8.20 Applications of passive microwave sensingAdvantages Disadvantages Meteorology Oceanography Geology

419 8.21 LIDAR Fig 8.54: Principle of lidar bathymetryFig 8.55: Lidar returns measured over a forest canopy Laser-induced fluorescence (LIF) Single-channel laser source + multi-channel receivers Distinguish several plant groups