What is Remote Sensing?

A Guide
to Earth Observation
GISGeography Remote Sensing

Remote Sensing Satellite

A Guide to Remote Sensing

What is Remote Sensing?

It’s the multi-billion dollar question for an industry that’s reaching new heights.

Orthophotos, land cover classification maps, digital elevation models – these

are examples or remote sensing data sets.

We’ve looked at digital image classification. But what are some of the general
principles of remote sensing?

It’s time to connect the dots with this definitive guide to remote sensing.

What is Remote Sensing? Remote Sensing means you are attaining

information at a distance without physically being there.

Your Body Uses Remote Sensing

In fact, you’re using remote sensing by reading this right now…

Remote sensing means your acquiring information from a distance. Your body is
equipped with three remote sensors:
1. Your eyes see the reflected electromagnetic light.
2. Your ears hear sound, which are mechanical waves.
3. Your nose smells aromas, which is more of a chemical reaction.

With the sense of taste, your tongue physically touches food. With the sense
of touch, your hand physically touches an object.

So we can’t classify taste and touch as remote sensing.

Your eyes use remote sensing

to detect light and colors

The Electromagnetic Spectrum – All Possible

Wavelengths
The Electromagnetic Spectrum is the range of all possible wavelengths.

It’s composed of thousands of wavelengths of light. It spans from short

wavelengths (like X-rays) to long wavelengths (like radio waves).

Electromagne
tic Spectrum (Not Drawn to Scale)

Remote sensing enables you to see the invisible (and visible).

You read this post. Light bounces off the screen into your eyes as three
channels: red, green and blue.

The incident energy source could be the sun, lights, candles or a flashlight. Our
eyes see reflected energy in the visible spectrum (390-700 nm).

What if you were a goldfish or bumble bee?

 If you were a goldfish, you would see light a bit differently. A goldfish sees
infrared radiation (700 nm to 1mm) which is invisible to the human
eye.
 Bumble bees can see ultraviolet light (10 nm to 380 nm). Humans
don’t see ultraviolet radiation with our eyes and (UV-B harms us.)
We, humans made up these wavelength regions (spectral bands) for our own
purpose – to conveniently classify them.

Visible (red, green and blue), infrared and ultraviolet are descriptive regions in the
electromagnetic spectrum.

Each region is categorized based on its frequency (v) /wavelength ( ).

Therefore:

What are spectral bands? Spectral bands are simply groups of wavelengths.
Examples of spectral bands are: Ultraviolet, Visible, Near-infrared, Mid-infrared,
Thermal Infrared and Microwave.

Each Object Has Its Own Unique Spectral

Signature
Why even care about the electromagnetic spectrum?

The EM spectrum is important because each object reflects, transmits and

absorbs light differently, depending on its chemical composition.
Objects reflect light in bands of light we cannot see with our eyes – but sensors
can. Spectrometer record light that objects reflect into bands.

Plants are the color green because they reflect more green light. Healthy
vegetation reflects more near-infrared light and we use an index called NDVI to
help classify vegetation.

Near-infrared false color composite

Each object has it’s own unique chemical composition.

This is in tune with saying each object has it’s own spectral signature.
Differences in spectral signatures is how we tell objects apart.

Spectral signatures using different wavelengths in the EM spectrum gives us

the ability to learn more information about Earth’s features that we may have
not known.

What is a spectral signature? A spectral signature is the amount of energy

reflected in a particular wavelength. Differences in spectral signatures is how
we tell objects apart. Spectral signatures are driven by the objects chemical
composition.
Hyperspectral vs Multispectral
Multispectral and hyperspectral imagery gives the power to see as humans
(red, green and blue), goldfish (infrared), bumble bees (ultraviolet) and more.
This comes in the form of reflected EM radiation to the sensor.

The main difference between multispectral and hyperspectral is the number

of bands and how narrow the bands are.

Multispectral imagery generally refers to 3 to 10 bands that are represented

in pixels. Each band is acquired using a remote sensing radiometer.

Multispectral
Example: 5 wide bands (Image not drawn to scale)

Hyperspectral imagery consists of much narrower bands (10-20 nm). A

hyperspectral image could have hundreds of thousands of bands. This uses
an imaging spectrometer.

Hyperspectral
Example: Imagine hundreds of narrow bands (Image not drawn to scale)
Multispectral Example
An example of a multispectral sensor is Landsat-8. Landsat-8 produces 11 images with
the following bands:

Band 1: Coastal aerosol (0.43-0.45 um)

Band 2: Blue (0.45-0.51 um)
Band 3: Green (0.53-0.59 um)
Band 4: Red (0.64-0.67 um)
Band 5: Near infrared NIR (0.85-0.88 um)
Band 6: Short-wave Infrared SWIR 1 (1.57-1.65 um)
Band 7: Short-wave Infrared SWIR 2 (2.11-2.29 um)
Band 8: Panchromatic (0.50-0.68 um)
Band 9: Cirrus (1.36-1.38 um)
Band 10: Thermal Infrared TIRS 1 (10.60-11.19 um)
Band 11: Thermal Infrared TIRS 2 (11.50-12.51 um)

Hyperspectral Example
The TRW Lewis satellite was meant to be the first hyperspectral satellite
system in 1997. Unfortunately, NASA lost contact with it.

But later NASA did have a successful launch mission. The Hyperion imaging
spectrometer (part of the EO-1 satellite) is an example of a hyperspectral sensor.
The Hyperion produces 30-meter resolution images in 220 spectral bands (0.4-
2.5 um).

NASA’s Airborne Visible / Infrared Imaging Spectrometer (AVIRIS) is an example of a

hyperspectral airborne sensor. AVIRIS delivers 224 contiguous channels with
wavelengths from 0.4-2.5 um.

Multi means many bands (more than 3). It’s usually collected at a lower
spectral resolution. Hyper means hundreds of bands. It’s usually collected at
high spectral resolution.

What is Multispectral Data? Multispectral data generally has 3-10 wider

bands. An example is Landsat-8
What is Hyperspectral data? Hyperspectral has Hundreds of narrow bands.
An example is the Hyperion satellite.

Read More: Multispectral vs Hyperspectral Imagery Explained

Atmospheric Windows and Absorption Bands

Not all of the EM spectrum hits the Earth’s surface. Atmospheric absorption
prevents specific types of EM radiation.

Atmospheric Electromagnetic Window

Think of atmospheric windows like a curtain with holes:

The cloth from the curtain blocks specific wavelengths. In reality, the
atmosphere contains water vapor or carbon dioxide that absorbs x-rays,
gamma rays and other EM spectra. This is known as “absorption bands”.

But these are special types of holes that allow only specific types of sunlight to
freely pass. Radio waves can pass through. But x-rays cannot.
Holes in the curtain are the atmospheric window with specific bands of EM
spectrum can freely pass.

The cloth from the curtains block sunlight.

The point is:

Some types of light doesn’t reach Earth so engineers design sensors to

measure specific wavelengths.

What is an Atmospheric Windows? An atmospheric window is the portion of

the electromagnetic spectrum that can be transmitted through the
atmosphere.
What is an Absorption Band? An absorption band is the portion of the
electromagnetic spectrum that can be transmitted through the atmosphere.

Passive vs Active Remote Sensing

Passive sensors measure reflected sunlight that was emitted from the sun.
Active sensors have their own source of light or illumination and its
sensor measures reflected energy.

 Active sensors like the Radarsat missions create their own source of
illumination. It measures the energy that bounces back to the sensor.
 Passive sensors like the Landsat missions collects reflected energy that was
emitted by the sun emits. Passive sensors measure the reflected
energy at a specific frequency (v) (i.e. wavelength ).

Think of cameras as a passive AND active sensor:

Active Remote Sensing Camera Example

You hold your camera in your hand. Flash turned on. You take a picture.

What’s exactly happening here?

The camera sends light to the target. The light reflects off the target back to
the camera lens. This is the light that your camera measures.

You can think of active remote sensing like a handheld camera with the flash
turned on. But active remote sensing can be spaceborne satellites orbiting the
Earth or airborne on an aerial unit.
  Cameras are active sensors when the photographer uses flash. It
illuminates its target and measures the reflecting energy back to the
camera.
 Cameras are passive sensors when the photographer does not use the
flash. The camera is not providing the source of energy. It uses naturally
emitted light from the sun or lamp.

What is Passive Remote Sensing? Passive sensors measure reflected

energy emitted from the sun.
What is Active Remote Sensing? Active sensors illuminate their target and
measures its backscatter.

Read More: Passive vs Active Sensors in Remote Sensing

Satellite Orbits

Stare up 705 kilometers into the atmosphere at the right time and you can see
Landsat-7 or 8 satellites. This is a typical altitude of a satellite.

The height of the satellite above the Earth surface will determine the time it
takes for the orbit to take one complete orbit of the Earth. Orbital period
increases with satellite height.

But how do satellites orbit the Earth?

Geostationary orbits is a circular orbit Earth’s equator with a radius of

approximately 42,164 km intentionally matching the Earth’s rate of rotation.
(Example GOES satellites)

Sun Synchronous orbits is a geocentric orbit which combines altitude and

inclination keeping the angle of sunlight on the surface of the Earth as
consistent as possible. (Example Radarsat)

Polar Orbits passes above or nearly above both poles of Earth, but possibly
another body such as the Sun with an inclination of approximately 90 degrees
to the equator. (Example SPOT)

Resolution Properties
Spatial resolution is the detail in pixels of an image. Higher spatial resolution
means more detail and smaller pixel size. Lower spatial resolution means less
detail and larger pixel size.

Spatial Resolution
Comparison

Spectral resolution is comprised of spectral bands that are groups of

wavelengths. High spectral resolution are bands that are more narrow – such as
hyperspectral. Low spectral resolution are broader bands covering more of the
spectrum.

Temporal resolution refers to the measurement with respect to time of image

acquisition. A faster satellite revisit period means higher temporal resolution. A
slower revisit period refers to lower temporal resolution.

Remote Sensing Energy Mechanics

Some electromagnetic energy is absorbed in the atmosphere before hitting the
Earth’s surface. The EM radiation that actually makes it to Earth is called
incident energy (E ). But, now what happens?
i

There are three interactions that can happen with electromagnetic energy:
1) Electromagnetic waves bounce off the surface. This is called reflected
energy (E ). Think of flashing a light in a mirror.
r

2) Electromagnetic waves are absorbed and cease to exist. This is absorbed

energy (E ). Think of why black shirts absorbs more light than white shirts.
a

3) Electromagnetic waves goes through the object. This is called transmitted

energy (E ). Think of transparent objects.
t

Incident energy is a combination of reflected, absorbed and transmitted energy.

Incident Energy Formula:

E =E +E +E
i r a t

Incident energy interacts differently for different features on Earth. Earth’s

features have different proportions of energy being reflected, absorbed and
transmitted. These differences give our eyes and sensors the capability to
differentiate objects on Earth at different wavelengths. Sensors detect reflected
energy:

Reflected Energy Formula:

E =E –E –E
r i a t

Just like how our eyes see colors. We see things because of the way
electromagnetic energy bounce off objects into our eyes. Sensors are similar in
that they detect reflected energy at specific wavelengths (visible, infrared,
ultraviolet). The energy reflected that a sensor measures is equal to the energy
incident minus the energy absorbed and transmitted.

What is the proportion of reflected energy back to the sensor compared to the
incident energy from the sun?

This proportion is called spectral reflectance (p). Spectral reflectance is

different at specific wavelengths and is the main principle of remote sensing

Spectral Reflectance Formula:

p=E /E
r i

Remote Sensing Applications

There is no shortage of how remote sensing is being implemented in different
industries:

Satellite and other remote sensing information is fundamentally important if we

are going to solve some of the major challenges of our time.

From navigating through the Arctic to measuring soil moisture – there is a

remote sensing solution for almost every world issue. Remote sensing puts a
wealth of information into the hands of decision makers.

If we are going to take on some of the biggest challenges of Earth in the near
future, we need remote sensing to cover that much ground.