Study Guide: Remote Sensing / GIS Exam #2 Fall 2015 Semester. A Basic Introduction to Thermal Infrared Imaging

Transcription

1 Study Guide: Remote Sensing / GIS Exam #2 Fall 2015 Semester A Basic Introduction to Thermal Infrared Imaging Thermal infrared is the only common radiated (rather than reflected) remote sensing system Used primarily for monitoring weather systems & ocean surface temperatures A system which sees radiant (rather than kinetic) temperature of observed features Two kinds of heat energy Kinetic heat A measure of the random vibrations of atoms & molecules - Measured with thermometers Radiant heat The electromagnetic energy emitted by an object - Measured with radiometers Heat transfer mechanisms Conduction One atom or molecule to a neighboring atom or molecule Convection Vertical circulation caused by heat-related density differences Radiation Emission of EMR Infrared spectral bands Reflected 0.70 µm to 3.0 µm - Short & middle infrared See below Radiated 3.0 µm to 14.0 µm µm to 5.0 µm Minor CO 2 absorption Relatively high temperatures Volcanoes, wildfires, etc µm to 8.0 µm Major H 2 O [vapor] absorption Water vapor video loops µm to 14.0 µm Minor O 3 absorption Relatively low temperatures Clouds, animals, etc. Blackbodies Hypothetical objects that do two things: - Absorb 100% of all EMR wavelengths that reach its surface - Emit 100% of that energy at wavelengths determined entirely by its Kelvin temperature Wien s displacement law λ max = b / T kin Wavelength at which maximum energy is emitted Stefan Boltzmann law F b = σ. 4 T kin Radiant energy flux of a blackbody The basic concept & importance of emissivity in thermal infrared imaging Emissivity is a percent efficiency of heat radiation from real objects - F r = ε. σ. 4 T kin Emissivity ( ε ) ranges from 0.00 to 1.00 (0 to 100 percent) The example of the aluminum block - Perfectly uniform kinetic temperature but - Radically different radiant temperatures due only to the nature of its surface Additional thermal properties Thermal conductivity How rapidly heat energy travels through any substance Thermal capacity How much heat energy can be stored by any substance Thermal inertia How rapidly temperature changes in any substance The scientific basis for using the Landsat TM, ETM+ & OLI IR3 band Designed for a portion of sunlight reflected by surface features Useable to observe high-temperature features - Active lava flows, wildfires & Chernobyl-type nuclear reactor fires Study Guide: Remote Sensing / GIS Exam #2: Fall 2015 Semester, page 1

2 A Basic Introduction to Hyperspectral Imaging Three basic concepts Spectral bands Contiguous regions of the electromagnetic spectrum (e.g., blue, green, red etc.) - Spectral band width significantly affects energy flux on the sensors Multispectral Using < 16 contiguous regions of the EMS (e.g., TM/ETM+ bands ) - Spectral band widths are relatively large, resulting in relatively high energy flux & high SNR Hyperspectral Using > 16 contiguous regions of the EMS - Spectral band widths are relatively small, resulting in relatively low energy flux & low SNR Consumer digital cameras Everyday multispectral imaging Use of the Bayer filter A 2 x 2 array of pixels 1 red pixel, 2 green pixels, 1 blue pixel Multispectral ó hyperspectral classification Multispectral classification first defines n-dimensional colors & then looks at each pixel - Colors can be defined by the user or by the computer - There are relatively few defined n-dimensional colors (~ 8 to ~ 32 spectral classes) - Both hard (definitive class) & soft (percentage of 2 or more spectral classes) classifiers can be used Hyperspectral classification uses spectral libraries produced under carefully controlled lab lighting - > 100 spectral types common - Ability to identify & classify mineral spectral absorption lines A Basic Introduction to Radar Remote Sensing Radar is the only common active remote sensing system This gives it day and night imaging capability and cloud penetration This system sees most fundamentally - The electrical conductivity properties of observed features o Technically known as dielectric constant - The geometry (both micro- and macro-) properties of observed features o Micro- Smooth, intermediate & rough surfaces o Macro- One-surface, two-surface & three-surface specular reflectors The fundamental characteristics of and differences between Side-looking airborne radar (SAR) - Sometimes called brute force (using the actual physical length of the radar antenna) - Usually short wavelengths to maximize spatial resolution Synthetic aperture radar (SAR) - This uses the physical length of the radar antenna times the number of looks - This is equivalent to a radar laser (all emitted radiation is coherent ) - The issue of radar speckle (the salt and pepper effect The causes and effects of the strange spatial resolution characteristics of radar imaging Azimuth resolution decreases (expected) w/increasing distance from the radar system Range resolution increases (not expected) w/increasing distance from the radar system A Basic Introduction to LiDaR Remote Sensing LiDaR is becoming a common active remote sensing system, together with radar The acronym stands for Light Detection And Ranging Like radar, LiDaR has day/night capability but, unlike radar, no cloud penetration Like synthetic aperture radar (SAR), LiDaR employs coherent EMR (i.e., laser) Study Guide: Remote Sensing / GIS Exam #2: Fall 2015 Semester, page 2

3 - Like SAR, LiDaR can produce both DEM s & images Essential LiDaR technology Transmit a laser pulse & record the arrival time & intensity of its reflection Convert round-trip travel time to straight-line distance Convert straight-line distance to altitude LiDaR location parameters Phase-differenced kinematic GPS - Determines the precise position of the LiDaR instrument in three dimensions Inertial navigation system - Determines the precise orientation of the LiDaR instrument in three axes LiDaR system variables Laser wavelength & pulse frequency Scan frequency & angle Platform speed & altitude Three LiDaR modes First reflections - Tops of buildings & trees DEMs Intermediate reflections - Understory features in forested areas Structures, shrubs Last reflections - (Presumably) the ground surface DTMs LiDaR bathymetry Near IR reflects from water surface & blue-green (i.e., cyan) reflects from the bottom Differential Absorption LiDaR (DiAL) One wavelength is strongly absorbed & the other strongly reflected by the target gas (e.g., O 3 ) A Basic Introduction to Database Query Using Idrisi GIS databases are useless unless you can get new information from those databases The four-stage approach to solving geospatial problems - What I want/need - What I have - What I can get - What I do to get there Queries can be made of single or multiple input data layers Producing derivative map layers - Land cover/land use GIS layers - Digital elevation model (DEM) GIS layers The different numerical data types Boolean: 0 (unacceptable) or 1 (acceptable) - Often applied to ranges of values Real: Continuous values with decimal points - Almost always used for DEM s Byte: 2 8 values ranging from 0 to Almost always used for images The logical AND operator The mathematical operator is multiplication when applied to Boolean data layers - 0 times anything is 0, while 1 times anything is the original value Use of the RECLASS module in Idrisi RECLASS classifies or reclassifies the pixel values stored in raster images Study Guide: Remote Sensing / GIS Exam #2: Fall 2015 Semester, page 3

4 Use of the OVERLAY module in Idrisi OVERLAY produces a new image from the data of two input images Use of the AREA module in Idrisi AREA measures the areas associated with each integer category in an image Use of the GROUP module in Idrisi GROUP determines contiguous groupings of identically valued integer cells in an image Use of the IMAGE CALCULATOR in Idrisi This is usable only because Idrisi is raster-based - It can eliminate a lot of labor if used correctly The importance of making visual sense How making raster group files (.rgf) can help the process of making visual sense A Basic Introduction to Multi-Criteria Evaluation Using Idrisi Two basic categories of criteria Boolean criteria: Absolute, i.e., either totally unacceptable (0) or totally acceptable (1) - Only specific soil types, buffer zones Continuous criteria: Relative, i.e., ranging from 0.0 % to % - Actual distances (e.g., in meters or feet, kilometers or miles) from specified features New terminology Constraints - Something is prohibited or allowed with no other possibilities Factors - All possibilities are open but are moderated by their numerical value Handling factors & constraints The BUFFER module The RECLASS module Aggregation of factors & constraints All factors have been converted into constraints - The MCE module also allows for different constraints to have different weights A Basic Introduction to Map Algebra Using Idrisi The possibilities of using non-gis datasets to help solve GIS problems Tabular data, e.g., temperature and rainfall at various weather stations Theoretical relationships, e.g., mathematical equations based on the laws of Physics Empirical relationships, e.g., mathematical equations derived from field studies The Idrisi OVERLAY module Several relatively simple mathematical operations between two raster datasets - Arithmetic multiplication is the equivalent of the logical AND operator The Idrisi SCALAR module Does scalar arithmetic operations on a single raster image - Addition, subtraction, multiplication, division or exponentiation of pixel DN values The Idrisi TRANSFORM module Does non scalar arithmetic operations on a single raster image, e.g., log & trig functions The Idrisi IMAGE CALCULATOR module Combined functionality of OVERLAY, SCALAR & TRANSFORM modules - Usefulness of the Insert Image functionality The Idrisi EDIT module Study Guide: Remote Sensing / GIS Exam #2: Fall 2015 Semester, page 4

5 A very simple text editor similar to Windows Notepad - Idrisi recognizes only two columns of tabular data The Idrisi REGRESS module Produces the mathematical formula for a least squares regression line on tabular data - One way in which tabular data may be incorporated into derivative map layers The Idrisi CROSSTAB module A multiple overlay showing all combinations of the logical AND operation Awareness of physical/mathematical correlations Strong positive correlation between elevation & precipitation Strong negative correlation between elevation & temperature - Use of field data to determine local relationships A Basic Introduction to Distance and Context Operators Using Idrisi The basic concept of GIS distance (a.k.a. neighborhood or local) operators Buffer zone derivative layers and how they are calculated from binary raster layers Continuous distance derivative layers and how they are calculated from binary raster layers The basic concept of GIS context operators Slope and aspect as calculated from DEMs - Slope can be calculated in 4 (N, E, S and W) or 8 directions from each pixel in a DEM - Aspect gives the compass direction of the slope The Idrisi SURFACE module Calculates slope (in or %) and/or aspect (azimuth toward which a slope faces) - Uses a DTM as the input dataset The Idrisi GROUP module Identifies unique contiguous polygons in a raster image - Useful for calculating the area of classified raster images The Idrisi AREA module Calculates the area of contiguous polygons in a raster image - Ground resolution cell (GRC) size must be part of the metadata A Basic Introduction to Natural Disaster Monitoring and Modeling Important definitions Disasters: Extreme natural events that have happened Hazards: Extreme natural events that may happen Classification of disasters Purely natural, human-aggravated, human-induced & human-made Scale issues in disaster management Usability of RS & GIS technologies - Assessment of floods (e.g., radar), earthquakes (e.g., CIR), volcanoes (e.g., TIR) Examples of RS & GIS monitoring Chandeleur Islands, Mount St. Helens, iceberg calving A Basic Introduction to Modeling Earth s Biosphere The role of land cover maps Vegetation functional regions and seasonal dynamics - Transition from static to dynamic processes with improving technology Study Guide: Remote Sensing / GIS Exam #2: Fall 2015 Semester, page 5

6 Geology and soils that support plant and animal communities The importance of land/atmosphere interactions Absorption and emission of different wavelengths of EMR Friction between winds and trees Infiltration of water and evapotranspiration Static (equilibrium) systems and dynamic (non-equilibrium) systems Both spatial and temporal effects Multi-theme global databases Vegetation, hydrology, energy balance Possibility of monitoring and analyzing microenvironments Satellite sensing systems AVHRR, Landsat MSS, Landsat TM and ETM+, SPOT Cloud cover, sensor deterioration, satellite life span can generate problems A Basic Introduction to Wildlife Management Purpose of & plans for wildlife conservation Maximize plant & animal diversity in as many regions as possible Establish parks & reserves encompassing critical wildlife habitat Protect functional ecological units - Not entirely effective but expressive of values Mapping wildlife distribution Field, aerial & satellite observations - Almost always some degree of guesstimation - Habitats rather than animal counts due to temporal & resolution difficulties Wildlife resource requirements Food, water, shelter, nesting - Thus, vegetation maps often serve as animal species habitat maps - Multi-date NDVI images often used despite some inherent problems Wildlife resource suitability Uncertainties due to using vegetation as the only explanatory variable - Water [wetlands] useable for projects like Ducks Unlimited in Canada Determining cause & effect is very difficult & thus fraught with potential errors Errors of both omission & commission Accuracy of wildlife resource suitability maps Need to remember that models predict suitability rather than presence/absence Understanding factors influencing wildlife distribution Multiple scales involved: global range, home range, resource utilization Other considerations Static & dynamic models based on past data & extrapolating into future decades & centuries Spatial & temporal transferability of wildlife models Study Guide: Remote Sensing / GIS Exam #2: Fall 2015 Semester, page 6