GISCI 241 Chap.4 Electromagnetic Radiation and Multispectral Images
Electromagnetic Radiation and Multispectral Images
Electromagnetic radiation is the physical link between surface and sensor. This chapter relates wavelength and frequency through the speed of light, compares reflected, absorbed, transmitted and emitted energy, and explains how surfaces form wavelength-dependent spectral signatures. It then treats a multispectral image as a registered stack of raster bands, each storing measured intensity at every pixel.
Display colour is separated from physical response so false-colour images can be interpreted without mistaking screen channels for literal surface colour.
What this chapter covers
- 01
Wavelength, frequency and the speed of light
- 02
Nanometres, micrometres and unit conversion
- 03
Reflection, absorption, transmission and emission
- 04
Spectral signatures and environmental influences
- 05
Multispectral band stacks and pixel vectors
- 06
True-colour and false-colour composites
Worked example · free
Convert wavelength to frequency
- +1Convert 0.75 micrometres to 0.75 times 10 to the power negative 6 metres.
- +1Rearrange c equals wavelength times frequency to frequency equals c divided by wavelength.
- +1Substitute consistent units to obtain approximately 4.00 times 10 to the power 14 hertz.
- +1State that this identifies the band location, not a land cover without multiband and reference evidence.
Key terms
- Wavelength
- The distance between successive wave peaks, commonly expressed in nanometres or micrometres for optical sensing.
- Frequency
- The number of complete wave cycles per second, inversely related to wavelength.
- Reflectance
- The proportion or pattern of incoming electromagnetic energy reflected by a surface.
- Emission
- Electromagnetic energy released by a target, often related to its temperature and material properties.
- Band
- A defined interval of wavelengths sampled and stored as one raster layer.
- Multispectral image
- A co-registered stack of image bands measuring several wavelength intervals.
- Pixel vector
- The set of band values recorded for one pixel location across a multispectral image.
Electromagnetic Radiation and Multispectral Images FAQ
Why are wavelength and frequency inverse?
Electromagnetic energy travels at a fixed speed in the stated relationship, so more wave cycles must fit into each second when each cycle is shorter. The course expresses this as c equals wavelength times frequency. Keep units consistent and perform a direction check after every calculation.
Does a red false-colour pixel reflect red light?
Not necessarily. A false-colour composite can assign near-infrared or another band to the red display channel. The screen colour shows the rendering choice, while the physical interpretation concerns the input band's response. Read the band mapping before describing any surface process.
Is a spectral signature unique to a material?
A spectral response can help discriminate materials, but it changes with moisture, health, atmosphere, illumination, viewing geometry, season and mixed pixels. Treat a signature as evidence under defined conditions. Use several bands, representative references and independent validation before assigning a class or causal interpretation.
Why is one bright band not enough to identify land cover?
Different materials and conditions can produce similar response in one wavelength interval. Illumination, atmosphere, moisture, season and mixed pixels also shift the recorded value. Identification becomes stronger when a multiband pattern is compared with representative references under compatible conditions, but it still remains an evidence-based classification rather than a unique fingerprint.
Exam move
Memorise the relationship c equals wavelength times frequency and practise converting nanometres and micrometres to metres before substitution. Draw one horizontal spectrum with wavelength increasing and frequency decreasing. For every composite you encounter, write the input band numbers or wavelength regions beside the display channels.
Build simple response sketches for vegetation, water, soil and built surfaces, then annotate the conditions that can shift them. When a question presents one band value, resist assigning a class until you have a multiband pattern and context. Finish numerical answers with a direction check and finish interpretation answers with a statement separating measured response from surface meaning.
Create a unit ladder from metres to micrometres and nanometres, and solve conversions in both directions before using the wave equation. Estimate whether the resulting frequency should rise or fall before calculating; that direction check catches many exponent errors. Build a channel ledger for true-colour and false-colour composites.
Record which measured band feeds display red, green and blue, then explain a bright feature through those inputs rather than through its screen colour. For spectral signatures, compare curves across several bands and label reflection, absorption and transmission possibilities. Add moisture, illumination, atmosphere, season, view angle and mixed-pixel support as modifiers around each curve.
Practise one discrimination question where two materials overlap in visible bands but separate in infrared, and another where the same material changes condition. State why a multiband vector is stronger than one value but still requires reference evidence.
Finally, rehearse the full causal boundary: the sensor measures energy, preprocessing changes the values, a composite or index displays a pattern, and the analyst proposes a surface interpretation. Use that order in both numerical and written answers.