Remote Sensing in the Classroom — Getting Started Guide
A quick introduction into remote sensing for teachers who would like to use satellite images in class
A guest blog post by Pierre Markuse
Remote sensing is finding its way into more and more classrooms. Which is a good thing, as satellite images allow students to take a look with their own eyes and gain knowledge by actually seeing what otherwise would maybe just have been talked about in class. And remote sensing is by no means limited to science classes; On the contrary, many subjects can benefit from satellite images, just think about images of refugee camps for social studies or politics classes, or the possibility of custom scripting for computer science classes.
But while some classrooms are starting to make use of satellite data, in many other classrooms it is not yet happening. Teachers often don’t know about the free availability of satellite data. And if they do, some of them are still stopped by the false belief that you would have to be an expert to make use of it.
This blog post is meant to give you a quick introduction into what remote sensing is and how it is done. It will cover the basic principles but is not meant as a guide on how to get or process images, however, at the end of this blog post, you will find a list of helpful resources to get you started with your endeavor of getting images yourself using the Sentinel Hub EO Browser.
Getting students to use the EO Browser can be an engaging way of communicating knowledge, while also making them accustomed to working with modern tools.
Still, even when only used by you as a teacher or lecturer (for whatever reason, maybe time constraints or lack of suitable computer equipment), it is an awesome tool to get meaningful and relevant satellite images for your topic of interest which you can use and share with your students. The availability of the latest images also makes it possible to talk about recent or ongoing events, something that just isn’t possible with textbooks as an image source.
How we take a look at Earth
Planes, drones, and satellites
Satellites are usually the best tool when it comes to looking at Earth. However, on smaller scales, airplanes and drones can deliver remote sensing data as well and are widely used. They usually deliver higher resolution data, because they’re much closer to the ground. However, they mostly lack the continuity of satellite observations, as well as the growing archive of past observations, which are an important tool for science to recognize and analyze patterns and changes in our environment.
We will focus on satellites here. Nowadays there is a multitude of remote sensing satellites orbiting Earth. Some of them in near-polar orbits (often in sun-synchronous orbits, so the satellite will pass over the same area at the same local mean solar time, making images more consistent and easier to compare), giving us images from different parts of Earth as they orbit, others in geostationary orbits, showing us the same part of Earth all the time (usually weather satellites). But how do we now take a look at Earth? Here are some of the most common ways.
These are images taken using a camera sensor. Probably the most well-known and used type of satellite imagery. Usually, remote sensing satellites take multiple images in different “bands”, which are frequency windows. To get a true-color image (basically an image looking like a normal photograph) for example, the satellite is taking one image in the red part of the visible spectrum of light, one in the green part, and one in the blue part. These three monochromatic images are then combined into an RGB image. In the case of a true-color image, the acquisition in the red part of the spectrum is assigned to the red channel of the image, the one in the green part to the green channel, and the one in the blue part to the blue channel, leaving us with an RGB image that more or less looks like what we would see with our eyes. Often, but not always, these images are then processed a bit more to show more saturated colors and better contrasts so we can better pick up details.
Atmospheric effects on satellite images
A type of post-processing is also used to mitigate atmospheric effects (although usually in a scientific correct and automated way). Satellite images are affected by atmospheric scattering in different ways, one way being Rayleigh scattering of the light which is stronger at shorter (blue) wavelengths, leading to a bluish tint and haziness of satellite images to a varying degree.
Some satellite data is already atmospherically corrected, for example, Sentinel-2 L2A data. This is especially important if you want to measure things using indices, in that it eliminates (or at least mitigates) the effects of different atmospheric situations and therefore cancels out the effects of those on your measurements.
When it comes to just looking at Earth, for example, floodings, wildfires, volcanoes, or basically any type of landscape, atmospherically corrected data can help in making the images look more natural and clear. However, to just get telling images, you can easily work with data that is not atmospherically corrected and just correct it yourself to look natural.
Most satellites are not limited to frequency bands in the red, green, and blue part of the spectrum, but can also take near-infrared and shortwave-infrared images. All those images can be freely combined into a resulting image, focusing on different aspects of what you would like to look at on Earth. Shortwave-infrared data, for example, can be used to show hot spots, and can, therefore, be helpful in finding fires or viewing volcanoes. Different satellites have different abilities when it comes to those frequency bands they can take images in.
See above the thirteen different frequency bands of Sentinel-2A, the first of the two Sentinel-2 satellites of the European Copernicus programme. The American Landsat satellites have comparable, though not identical, bands, in many cases making it possible to use data from both sources. This is helpful because it shortens revisitation times and because Landsat has an archive of satellite images going back to the 70s (Landsat 1 was launched as Earth Resources Technology Satellite 1 on July 23rd, 1972).
As mentioned, those bands can be combined freely in any combination. Bands 2, 3, and 4 are important for true-color images, the many red bands are used in agriculture and forestry to get information about the status of crops and plants. Near-infrared bands also play an important role in vegetation analysis. Band 10 can be used to detect Cirrus clouds, and bands 11 and 12 are useful for cloud and smoke penetration, finding hot spots, moisture content, and in geology. All these usages are by no means exhaustive, you can find some more information further down when we will be talking about some of the presets in the EO Browser.
How close and how often do we take a look?
Aside from differences in their ability to image different frequency bands, satellites also differ in their spatial and temporal resolution. Some, usually commercial, satellites have spatial resolutions of 30 to 50 centimeters, while most non-commercial satellites are in the range of 10 to hundreds of meters. Temporal resolution can also vary a lot, while some weather satellites provide an image of the same area every 5 to 10 minutes, polar-orbiting satellites (or satellite systems consisting of multiple satellites), imaging the whole Earth, have temporal resolutions from one day (usually lower resolution) to multiple days (medium resolution).
The two active European Sentinel-2 satellites have revisitation times between three and five days, depending on your area of interest.
Synthetic Aperture Radar (SAR) images
These images are taken using synthetic aperture radars. Radar has the advantage that it can see through clouds, and therefore is not dependent on the weather situation in your area of interest like an optical satellite.
Basically, the satellite is sending a signal and is recording the reflections of this signal, creating an image from that reflection data. SAR images are usually not as easy to interpret as optical images, especially for beginners, but can be immensely helpful because they are not dependent on the weather.
Some of the preset visualizations in the EO Browser and custom scripts might be helpful here to create images that are easier to understand even for people with no experience. We would encourage you to also include SAR data in your class, especially when working with older students. Take a look at the following image visualizing floodings in Mozambique, caused by Cyclone Idai. By using a script it was possible to get the colors to look more familiar to people (green = Land, blue = water), so most people will have no problems to see and understand what is going on here.
See below an image of one of the European Sentinel-1 SAR satellites.
Data from other sensors
Some satellites use different types of sensors to measure specific things. Sentinel-5p, for example, can measure different atmospheric values including, but not limited to, carbon monoxide (CO), nitrogen dioxide (NO₂), and ozone (O₃).
This data might often provide valuable insides. For example, you can see the carbon monoxide emissions of wildfires or big cities and can see how these emissions are carried around the planet during and after the actual event.
Processing satellite data
Since you have decided to make use of the EO Browser in your class, processing images will probably be solely done within the EO Browser itself using either one of the many presets or a custom script.
There are other tools to process satellite images, but I believe no other tool to be as easy and yet as powerful as the Sentinel Hub EO browser. Whether you just want to quickly get an image, or whether you would like to analyze images using indices, perhaps even using your own custom script, the EO Browser can do it.
The EO Browser offers a lot of preset visualizations for different scenarios and a helpful Education mode to guide you to the right decisions and help in getting results. Using these visualizations it is easy even for total beginners to get useful results and good images.
Those presets usually include true-color images and other well-established and used band combinations that are useful to analyze what is happening in the image.
As you can see in the image, for Sentinel-2 you get 8 different visualizations in the normal mode. These will help you to quickly get meaningful images even if you are a total beginner when it comes to remote sensing. And those 8 different visualizations aren’t the only ones available.
Once you switch to the Education Mode, the EO Browser will provide you a selection of presets, suiting the theme (Agriculture, volcanoes, wildfires, geology, and many more) you have chosen. This way it guides you to choose band combinations that are useful in looking at what you have selected.
The Education Mode is activated by clicking on the graduation cap in the top right corner and selecting it from the menu that opens.
You can find more information about the Education mode and some other newly added features of the EO Browser here in this blog post.
Some of the presets you find are standard presets, used by many scientists to visualize a scene, other presets are variations of a custom script available in the custom script Repository and “hardcoded” into a preset for beginners.
Older K12 and university students might also make use of the EO Browser’s capability of custom scripting. Writing a script that uses the available data in some new way that is advantageous to show what you would like to show.
Aside from available custom scripts, encouraging your students to write their own custom scripts could be quite helpful to further understanding of how remote sensing data can be used.
Tutorial and “getting into” scripting
A good tutorial into custom scripting you can find here. I encourage you to download the more extensive PDF version and to also take a look at the many already available custom scripts in the Custom Script Repository to get a feel for how to write them.
Feel free to experiment and just give it a go!
Low hardware requirements on your site
The EO Browser, as well as its custom scripting functionality, have very low hardware requirements on the user site, so even older school computers should have no problems running it. Basically, the only thing you need is a computer able to run a recent web browser version. Some schools give out tablets to their students, and the EO Browser will even run on most tablets or even smartphones.
To get even more from the results the EO Browser can deliver, sometimes it might be useful to make use of other tools. To further enhance images regarding brightness, saturation, contrast, sharpness, and other visual aspects as well as adding annotations, an application such as Adobe Photoshop could be used. Since this is a commercial application it will not always be available in class, but there is an alternative in the form of GIMP, which is free and open-source. Not something you immediately have to use, but surely interesting for older K12 and university students.
Older K12 and university students could also be introduced to a geographic information system. Here QGIS is a good choice, as it is also free and one of the most used applications for this purpose, with lots of extensions and many helpful guides available on the Internet. Google Earth is another free option, and for many people is easier to get into.
Possible applications in the classroom
Some applications are immediately obvious, like science classes or geography classes. But as stated before, satellite images could be used in so many different classes and applications. Politics, journalism, and social studies classes can benefit from them, for example by imaging refugee camps or the effects of natural disasters. Computer science classes can benefit from the custom scripting aspect of the EO Browser. Environmental monitoring is another possible application, which could take place in different classes.
You can probably think of many more applications in which the latest satellite images could be helpful in class. And the EO Browser can be an interesting tool for students of a broad age range. Younger students could profit from images provided by the teacher, while older students could process images on their own and even write their own scripts.
Resources to get started
A few links to some useful information, focusing on processing images using the Sentinel Hub EO Browser.
- Some basic information about the Sentinel Hub EO Browser and how to use it (Click)
- An overview of new capabilities and recently added functions for the EO Browser by Monja Šebela (Click)
- The Sentinel Hub Education pages, providing some information about different remote sensing applications (Click)
- A really good custom script tutorial online (with the possibility of downloading an even more extensive PDF version) by Monja Šebela (Click)
- Our Custom Script Repository, featuring a multitude of custom scripts for different applications and as examples to get into scripting yourself (Click)
- “It’s a faaaake… — Or not?”, a blog post by Pierre Markuse talking about the possibilities and dangers of fake satellite images (Click)
- “Looking at Wildfires (and more…) — An Introduction”, a post about how to get and process (not only) wildfire images by Pierre Markuse (Click)
- “Visualizing Wildfires and Burn Scars with the Sentinel Hub EO Browser V2”, an introduction into the Wildfire script by Pierre Markuse (Click)
- “Water Quality Information for Everyone”, an introduction into the Ulyssys Water Quality Viewer script by András Zlinszky and Gergely Padányi-Gulyás (Click)
- “Environmental Monitoring of Conflicts using Sentinel-2 data”, showing some nice applications of the EO Browser by Wim Zwijnenburg (Click)
- “Mapping deforestation with Sentinel Hub”, an introduction, written by and using a script by Nicolas Karasiak (Click)
- My Flickr with now over 1,600 satellite images for you to use in class or just get ideas from where to look yourself (Click)
- The Sentinel Hub Flickr account, featuring many great satellite images (Click)
Even more help
But even aside from our accounts, the Remote Sensing community on Twitter is an awesome experience. If you ask politely, you will almost always find experts willing to help you with problems or give you an opinion.
Pierre Markuse is a freelance remote sensing expert and blogger with a special interest in optical remote sensing, his images being used by numerous media outlets.
How to make satellite images more approachable to the general public, communicating scientific backgrounds, and the usage of satellite images in journalism are additional fields of interest.