Remote Sensing

Definitions of Remote Sensing

The term “remote sensing” was coined in the 1950s by Ms. Evelyn Pruitt, a geographer in the U.S. Office of Naval Research. At this time, technology was rapidly evolving to record information about the Earth from aerial and satellite sensors. Today, with thousands of satellites in orbit, remote sensing has undoubtedly become a foundational research tool across many disciplines.

Fig. 1 - False color composite satellite image of the Nabro Volcano in Eritria. With remote sensing, temperature measurements can be displayed visually.

How Does Remote Sensing Work?

You might be wondering -what is actually being remotely sensed? That would be electromagnetic radiation or EMR.

Buckle up for a quick physics review about the electromagnetic radiation spectrum!

The sun emits shortwave EMR in the form of photons. Shorter wavelengths have higher frequencies and higher energy, while longer wavelengths have lower frequencies and lower energy. The Earth's atmosphere absorbs some of the sun's incoming energy and slowly releases it back out as long wave EMR.

We see this energy as visible light on Earth, and we feel this energy as heat in the air, depending on the wavelength of the photons.

Fig. 2 - Remote sensing detects and records electromagnetic radiation.

All incoming radiant energy is either absorbed, reflected, or transmitted through an object, depending on its wavelength and the object's material. Remote sensing works by recording values of EMR that reach a satellite or aerial sensor after being emitted or reflected from Earth.

For example, snow on a mountain top will reflect a very high percentage of incoming visible light (that's why it appears so bright and white). In contrast, a nearby stand of trees on the mountain will appear darker and thus does not reflect as much visible light. Remote sensing satellites can record these differences in reflectance values, and they can be interpreted to identify land cover types from afar.

Remote Sensing Techniques

Remote sensing technology is diverse and includes various techniques tailored to obtain and interpret the different spectral signatures of objects. However, all remote sensing techniques rely on three principal components.

The Three Components of Remote Sensing

1. Target: an object, surface, or area that is to be measured.

2. Electromagnetic Energy: this can be one or many parts of the electromagnetic spectrum.

3. Sensor: a device that detects and records electromagnetic information.

Types of Remote Sensing

The two types of remote sensing refer to differences in how sensors obtain EMR data.

Passive sensors are useful for capturing satellite imagery of the Earth's surface, especially within the visible and infrared portions of the spectrum. These sensors are optimized to detect the sun's shortwave energy as it is reflected out. However, shorter wavelengths are more susceptible to atmospheric scattering because they are small enough to collide with gas particles. That's where active remote sensing comes in.

By emitting energy directed at an object, active sensors can detect a clearer spectral signature with less scattering. Weather radars rely on active remote sensing because long wave lengths can pass through the atmosphere's gases and clouds with less scattering than shorter wavelengths. The returning reflected energy can be detected with less "noise" created by scattering.

Recording and Visualizing Remote Sensing Data

Energy arriving at a satellite's sensor is recorded within a grid. You can think of this grid as pixels in an image. A value is assigned to each pixel in the grid, and this value represents the intensity of the arriving energy.

Many satellites used to study Earth can detect a wide range of wavelengths. Wavelength types are grouped into bands, which allow ERM data to be stored in separate grid layers.

The visible spectrum is typically split into three separate bands: blue, green, and red. Other common useful bands include near and mid-infrared.

Humans see visible light as a combination of red, green, and blue light. When we assign the blue band to be represented visually with blue color, the green band with green color, and the red band with red color, we get a true color composite image with the appearance of natural colors.

Fig. 4 - True color composite image of the San Francisco Bay Area.

If we want to better distinguish areas of vegetation in the San Francisco Bay Area, we can instead utilize the near infrared band and represent its values with red color. This is called a false color composite image. Notice how areas with vegetation stand out more in the second image.

Fig. 5 - False color composite image of San Francisco Bay Area.

Remote Sensing and GIS

Remote sensing allows for the collection and recording of EMR data, while geographic information science (GIS) systems are used to store, visualize, and interpret the data.

GIS systems transform grids of reflectance values into visual representations with pixels. This raster data can then be manipulated in GIS software for countless types of analysis.

The grid of geographic data supplied by remote sensors can be analyzed in its raster format or transformed to other formats used in GIS to answer spatial questions.

Remote Sensing Examples

With spatial analysis in GIS, remote sensing has become a powerful tool for learning about the Earth. Below are several examples of how remote sensing is used in GIS to answer important questions.

Quantifying Deforestation with Landsat Satellites

NASA's Landsat program began with the launch of the first Landsat satellite in 1972. Landsat satellites have passive sensors that detect and store ERM within a variety of bands.

To understand how deforestation has progressed in the Amazon Rainforest, Landsat imagery throughout the years can be added into GIS software for spatial analysis. Notice the differences between the satellite image of Rondônia, Brazil on the left from 1975, and the image of the same area taken in 2012.

Fig. 6 - Landsat Images of Rondônia, Brazil from 1975 and 2012. The left shows dense forest cover in 1975, and the right shows settlement patterns and signs of deforestation.

In this example, GIS was used to classify land cover types and quantify their change in area over time. Researchers found that over 70,000 km² of Rondônia's forest has been removed since 1975.

Mayan Ruins Discovered with LiDAR

Light detection and ranging or LiDAR is a form of active remote sensing, and it can be used to create 3D models. A strong pulse of light is first emitted at a target, and the returning energy's location and intensity are recorded. This type of sensing occurs relatively close to Earth's surface, typically with sensors attached to planes or drones.

LiDAR can capture fine details and can even penetrate forest canopy to reveal what is hidden below dense vegetation, so researchers have been using this technology to search for ruins across the world.

Fig. 7 - 3D topographic model of the Aguada Félix ruins site discovered with LiDAR.

One such discovery involving LiDAR revealed a large Mayan ruin of a ceremonial site called Aguada Félix in Mexico². The slight variations in topography from the ruins were not detectable by far away orbiting satellites.