GEOS 107: The Planet Earth Session 7. Geographic Information Science: Remote Sensing, GIS and GPS/1. Dr. Mark J Chopping

Transcription

1 GEOS 107: The Planet Earth Session 7 Geographic Information Science: Remote Sensing, GIS and GPS/1 Dr. Mark J Chopping

2 Geographic Information Science We have 3 sessions on Geographic Information Science: REMOTE SENSING, GIS and GPS Today s s will cover just REMOTE SENSING This is a logical place to start, since much mapping is now done using remote sensing as the data source.

3 Geographic Information Science Movie on NPOESS: the National Polar Orbiting Operational Environmental Satellite System (NOAA/DoD/NASA) Compare and contrast with the NASA Earth Science movie you saw Make notes: what is the tone/mood of the movie? What is the message? What are the benefits of NPOESS? To whom? Will it be able to answer long or short-term questions/problems?

4 Geographic Information Science In case you didn t t know: GIS = Geographic Information Systems. A database storing maps of features above, on or below the Earth s s surface (hence the geographic), encoded in binary data (i.e. in a computer system). Implies not only storage but input, display, transformation, extraction, output, and analysis of these data. GPS = Global Positioning System. Doesn t actually position (move) anything but gives you your position on the planet to a very high precision (centimeters to meters). Uses 24 satellites.

5 GISci:Remote Sensing:Lecture Outline What You are Expected to Know Definition Sources of Radiation for Remote Sensing Remote Sensing Systems & Acquisition RS Systems/Resolution & Coverage Remote Sensing of Vegetation Common Applications Case Study: Scanning LIDAR The State-of-the-Art Further Reading & Surfing Questions to Ponder

6 GISci:Remote Sensing: What is Expected 1. You may be concerned at the level of knowledge expected of you with respect to Remote Sensing Science. 2. The difficulty is that most Earth Science textbooks treat Remote Sensing in a very flimsy way. This is no longer acceptable as in recent years Remote Sensing is contributing much more to the study of the Earth System (partly as a result of improvements in technology, e.g. satellite radar interferometry, imaging LIDAR and imaging spectroscopy). 3. The following page lists the topics within Remote Sensing that you are expected to become familiar with.

7 GISci:Remote Sensing:What is Expected 1. The Electromagnetic Spectrum: major regions imptt. for RS as less affected by atmosphere; and for vegetation remote sensing; 2. Classification of RS instruments (passive, active, imaging, non- imaging) and the platforms on which they are flown; 3. The Acquisition Methods of different sensor types (e.g. across- track scanning, whiskbroom, circular scanning, CCD, antennæ); 4. The effects of Acquisition Geometry (sun-target-sensor) for passive instruments which use solar wavelength (visible to near- infrared) radiation AND the definitions of related angle terms (nadir, zenith, azimuth, backscattering, forward-scattering); 5. Resolution and Coverage, including barriers to global coverage; 6. Vegetation Indices & Maximum Value Compositing; 7. Atmospheric Windows: the t role of the atmosphere in solar wavelength satellite remote sensing (briefly, what aerosols, water vapor and gases do to light as it passes thru the atmosphere).

8 GISci:Remote Sensing:Definition REMOTE SENSING Implies measuring something - anything! - without physical contact. Under a restricted definition for Earth Observation: sensed objects / areas are on or near the Earth s s surface observation is from above (airborne or spaceborne sensors) information is carried by electromagnetic radiation (radio waves, emitted infrared, visible and ultraviolet radiation) Essentially, remote sensing is a development of aerial photography (but what a development).

9 GISci:RS:Sources of Radiation Longer Shorter Electromagnetic radiation/1 Principally specified by wavelength (l)( ) or frequency (1/l) Ultraviolet - white clothes reflect; absorbs O 3 Visible - the light you use to see (violet-red) Near-infrared - u can t t see this but camcorders can Shortwave/MID-IR /MID-IR- veg.. moisture/stress; snow/clouds Thermal infrared - felt from sun, heaters, fires Far infrared - nothing special here (atm( atm.. opaque) Radio frequencies - microwave range is useful (not attenuated by water vapor, can see thru clouds). Most of the light reaching the Earth s s surface is concentrated in the VNIR: after interacting with atmospheric and surface materials, this radiation is absorbed, reflected or transmitted.. After transmission and absorption it can be emitted with a shift to longer wavelengths (e.g, thermal).

10 GISci:RS:Sources of Radiation Electromagnetic radiation/2 Most incoming light is in visible to near-infrared (VNIR) region. That s s why our eyes use it rather than other ls. VNIR radiation is reflected, absorbed or transmitted.. (if absorbed, interaction transforms to longer l) Near-infrared (NIR) light is invisible but non-negligible: leaves reflect and transmit to avoid overheating (some materials -- e.g. leaves -- are partially translucent in the NIR region even if opaque in the visible). Thermal IR radiation is emitted; emission depends on the Temperature and Emissivity of a material. Nothing is emitted in the VIS wavelengths at ambient temps. (300 K): things don t t glow until they get really hot! Surface of sun is 6000 K, so peak emission is ~0.5 µm red-orange light (color of sun).

11 GISci:RS:Passive Solar Radiation <<< SHORTER l LONGER >>> YOU SEE HERE! also showing atmospheric transmission, leaf spectral reflectance, absorption features and regions used by the three best-known satellite-based sensors. this feature is important for RS of vegetation!

12 GISci:RS:Passive Solar Radiation <<< LONGER l SHORTER >>> showing atmospheric windows where attenuation (db) is relatively low and which are useful for remote sensing from space.

13 GISci:RS:Remote Sensing Systems CLASSIFICATION OF REMOTE SENSING SYSTEMS A) PASSIVE using naturally-occurring radiation - sunlight or thermal emission B) ACTIVE using an artificial light source to illuminate the target 1) IMAGING SYSTEMS form a two-dimensional array corresponding to the brightness of the object or surface 2) NON-IMAGING SYSTEMS - don t. These systems sample points or profiles (see case study/slide 58). 1 2 A B

14 GISci:RS:Remote Sensing Systems Passive Sensor Systems Spaceborne / Airborne / Ground-based High / Low altitude Airborne aerial photography, scanners, CCD arrays: low-alt platforms: blimps, helicopters high-alt: aircraft, balloons Spaceborne scanners, CCD arrays, microwave antennas - all from satellites (LEO, GEO) - orbits from 400 km km; ~40,000 km towers, masts, cherry-pickers, blimps(?), hand-held radiometers, spectrometers Remote Sensing Active Sensor Systems Spaceborne / Airborne RADAR SLAR (side-looking synth.. aperture; imaging) SAR (synthetic aperture; imaging) Scatterometers (non-imaging) LIDAR Scanning LIDAR (imaging) Laser profilers (non-imaging) Sounding devices (SONAR) Ground-based Camera with flash {Camera ALL REQUIRE CALIBRATION AND VALIDATION

15 GISci:RS:Remote Sensing Systems Imaging Sensor Systems Spaceborne / Airborne / Ground-based High / Low altitude Airborne aerial photography, videography scanners (whiskbroom; Daedalus ATM) CCD arrays (pushbroom( pushbroom; ; SPOT HRV) coherent (SLAR, scanning LIDAR) Spaceborne scanners (whiskbroom; Landsat, AVHRR) CCD arrays (pushbroom( pushbroom; ; SPOT HRV) coherent (SAR) Ground-based Film/ digital camera, video from masts, cherry-pickers, blimps*, Remote Sensing * hybrid types Non-Imaging Sensor Systems Spaceborne / Airborne / Ground-based High / Low altitude Airborne radar altimeter laser profiler (Ritchie( Ritchie) MQUALS - MODIS QUick Airborne Looks Spaceborne radar altimeter radar scatterometer laser profiler (VCL*) Ground-based spectroradiometer,, IRT from masts, cherry-pickers, blimps*, hand-held

16 GISci:RS:Systems - Acquisition Cross-track (whiskbroom) & circular scanners (ATSR-2)

17 GISci:RS:Systems - Acquisition Along-track (pushbroom( pushbroom) ) & side-scanning techniques (SLAR)

18 GISci:RS:Systems - Acquisition Landsat TM spectra Multispectral sensors register response in a few channels

19 GISci:RS:Systems - Acquisition ASD FieldSpec spectra (ground) Imaging Spectroscopy: when each pixel of an image is recorded as a spectrum like this one a a LOT of data! Hyperspectral sensors register response in many (15 to 200+) channels

20 Earth & Environmental Studies Montclair State University Remote Sensing provides GISci:RS:Resolution & information on things a long way Coverage away - at different spatial & temporal scales AirPhoto (air) IKONOS lat w/pointing NASA MASTER (air) 250 m 2,500 m 25 m 10,000 m LANDSAT TM Every 16 days - w/o clouds NOAA AVHRR-LAC 2 x day/satellite x 2 satellites 100 km NOAA AVHRR-GAC 2 x day/satellite

21 Earth & Environmental Studies Montclair State University GISci:RS:Resolution & Coverage The BIG PICTURE from km (GEO) above the equator. Observes 42% of the surface.

22 Earth & Environmental Studies Montclair State University GISci:RS:Resolution & Coverage The SMALL PICTURE (QuickBird)

23 GISci:RS:Resolution & Coverage We are interested in the ENTIRE PLANET we need to use SENSOR/PLATFORM COMBINATIONS which provide a high geographic coverage. This means using a WIDE SWATH

24 GISci:RS:Systems:Acquisition:Geometry Acquisition Geometry is IMPORTANT!

25 GISci:RS:Systems:Acquisition Geometry

26 GISci:RS:Systems:Acquisition Geometry BRDF* Effects, Example 1: Spruce Forest Looking in the Backscattering direction: shadows are HIDDEN behind objects casting them; sensor sees a smaller proportion of shadows. Looking in the Forward- scattering direction: shadows are VISIBLE to the sensor to a greater or lesser degree * Bidirectional Reflectance Distribution Function which describes the angular distribution of reflected light over the upper hemisphere in relation to the angular distribution of incoming light. The Most Important effect is from SHADOWING by plants and microtopographic elements (e.g., rough soil elements; if you don t believe me see the next slide).

27 GISci:RS:Systems:Acquisition Geometry BRDF Effects, Example 7: from Space (AVHRR) (a) Before BRDF correction (b)after BRDF correction (a) images from 2 different orbits stitched together: in the left half the satellite sensor is looking in the backscattering direction (sensor looking away from the sun) and in the right half it is looking in the forward-scattering direction (towards the sun). (b) BRDF correction gives more reasonable map of reflectance. Courtesy Canadian Ctr. for Remote Sensing

28 GISci:RS:Resolution & Coverage For MODERATE RESOLUTION ( km) GLOBAL COVERAGE we need to use SENSOR/PLATFORM COMBINATIONS which provide a high geographic coverage but which also provide (WHAT?) high repeat rates (temporal sampling) so that we can fill in the gaps in our maps created by clouds

29 GISci:RS:Resolution & Coverage MODERATE RESOLUTION ( km) RS SYSTEMS HAVE PIXEL SIZES OF 250 m to 1.5 km (or >) AND HIGH REVISIT TIMES to obtain GLOBAL COVERAGE What does this mean for applications, e.g. global land cover mapping or within-biome mapping (e.g. forest type / semi-arid grassland species)?

30 GISci:RS:Systems - Acquisition MODIS and MISR on the NASA EOS Terra Satellite launched in December MODIS = MODerate resolution Imaging Spectrometer MISR = MultiAngular Imaging SpectroRadiometer

31 Forestry Agriculture GISci:RS:What about the practice? Crop stress & disease Yield forecasting Precision Agriculture w/gps Monitoring CRP (US) Monitoring Set-a-Side (EU) Stand stress & disease Inventory Planning Hydrology (runoff ) Geomorphology (erosion) R a n g e o f A p p l i c a t i o n s Natural Resourecs Natural Hazards Semi-natural environments (range, forest, reserves) Geological exploration (minerals/mining) Snow, rivers, lakes, water supply & quality (sediment loads, diatoms) Fire Weather (including tornadoes, floods) Volcanoes Earthquakes Landslides Airborne particulate matter (dust, PM10s) Tsetse fly and ticks (yes, really)

32 Remote Sensing : The State-of-the-Art REMOTE SENSING IN THE SOLAR WAVELENGTHS REMOTE SENSING AT THERMAL INFRA-RED WAVELENGTHS REMOTE SENSING AT MICROWAVE WAVELENGTHS REMOTE SENSING AND GEOGRAPHIC INFORMATION SYSTEMS Imaging spectroscopy ( hyperspectral( hyperspectral,, AVIRIS, Hyperion) LIDAR: light detection and ranging (VCL) Multiangular & BRDF sampling (MISR, CHRIS, POLDER) High spatial resolution imagery (IKONOS, QuickBird,, aerial photography) Land emissivity and LST from multispectral data (temperature-emissivity separation algorithm) Sea surface temperature (AATSR) Multi-polarimetric SAR, interferometry (ERS-1/2, ENVISAT); soil moisture (passive, e.g. SGP97/99) Integrated GIS - combining raster and vector models (e.g., CLEVER mapping project at Cambridge University)

33 Reading & Surfing THE Remote Sensing Tutorial: WHAT YOU CAN LEARN FROM SENSORS ON SPACECRAFT THAT LOOK INWARD AT THE EARTH AND OUTWARD AT THE PLANETS, THE GALAXIES AND, GOING BACK IN TIME, THE COSMOS.. (NASA) tutore.html Fundamentals of Remote Sensing Tutorial (Canadian Ctr. for Remote Sensing).