Distinguish 500 Wavelengths from Space! Contribution of Hyperspectral Satellites [Space Brothers Collab #26]
“Space development” describes a wide range of developments and their objectives. In this series, we will contemplate the present and future of space development with Astronaut Serika Ito, who has been appointed Chief Dream Officer of Warpspace.
In this 26th edition, we welcome David Gauthier, an advisory board member of Orbital Sidekick, an American startup that is building a constellation of hyperspectral satellites, to introduce new satellite data use cases.
”Hyperspectral Satellite” to Determine the Types of Plants and Minerals on the Earth’s Surface from Space
Serika: Hello David, it is a pleasure to meet you!
The term “hyperspectral” is one that I’ve heard a lot recently and has been brought up many times in this series of articles. Could you tell us again what kind of sensor is hyperspectral?
David: A hyperspectral sensor is a sensor that is capable of spectroscopy of light at hundreds of wavelengths. Because they can distinguish spectral wavelengths more finely than the sensors on most satellites, they can identify the types of plants, minerals, gases, and other substances that cover the Earth’s surface.
The recent increase in the number of hyperspectral satellites is due, in part, to the easy availability of micro-electronic components and the affordability of launching small satellites. Dan Katz, CEO, and co-founder of Orbital Sidekick, founded Orbital Sidekick in 2016 with the idea that data collected from space by small hyperspectral satellites could change the world for the better, not only for governments but also for climate change and energy companies.
We are developing and operating a constellation of Global Hyperspectral Observation Satellites, or GHOSts, that is equipped with a hyperspectral sensor that can observe 500 unique wavelengths at 8m ground resolution (one pixel in an image represents 8m x 8m of information on the Earth’s surface).
We are now operating a three-satellite constellation, launched in 2023. We will launch three more satellites soon and intend to launch more satellites in the future. We are currently able to observe twice a day in some areas and only once a week in others, but as the number of satellites increases, we will be able to observe at a higher frequency.
Pipeline Inspection in a Few Days
Serika: Hyperspectral images, which are spectral of as many as 500 bands, must have a large data volume compared to conventional satellite images. Is there anything you are trying to do to get the observed data to the ground quickly?
David: The Orbital Sidekick satellite is equipped with artificial intelligence (AI) and has an onboard processing function that scans the observed spectral data immediately for a few known chemical signatures and can notify the ground quickly with a short message that those materials are present (methane, for example).
This is much quicker than waiting for a group of collected image cubes (each slice of the spectrum generates its image, so one collected area produces a stack of 500 images called a “cube”) to download through a ground terminal and be transferred to the cloud before starting to exploit them with analysis software on the ground. This machine-learning model is proprietary to Orbital Sidekick, and I think it’s one of our strengths.
Serika: That’s great! What are some of the uses for Orbital Sidekick’s hyperspectral images?
David: Let me introduce three use cases. The first is the detection of greenhouse gases. We operate a platform for detecting gas leaks and monitoring emissions.
The product first detects greenhouse gas leaks from pipelines carrying oil and natural gas. Some pipelines are thousands of kilometers long, such as those that run through North America or across Eurasia. Therefore, pipeline inspections have required months of time-consuming monitoring by aircraft and drones. With hyperspectral satellites, on the other hand, the entire pipeline can be scanned in a matter of days, and continuous monitoring allows for early detection of greenhouse gas leaks. This has the dual benefit of saving the customer money while also helping to protect the planet.
Second, we need to find more materials that can help us produce clean energy. For example, hyperspectral imagery can be used to find deposits of lithium. Fuels that are not sustainable and pollute the environment, such as oil, coal, and natural gas, are what’s called “dirty energy”, or environmentally hazardous energy that can only be used once, but if we can support the mining of environmentally friendly resources such as lithium which is used to make batteries, we can contribute to the transition to “clean energy” that is renewable. If we can support the discovery of more resources like lithium, and at the same time monitor mining operations to help them reduce any pollution from the mining itself, we can contribute to the transition to clean energy and a healthier planet.
Examples of Hyperspectral Imaging Applications Expanding with Ideas
The third is the automation and use of data for agriculture. Hyperspectral imagery can detect soil moisture content and tell whether farmland is heading for drought or has too much water supply. Detecting the chemicals released by crops and their reflected spectra can also give us an idea of crop stress conditions.
Serika: It sounds like we are listening to the voice of the trees and crops from space.
David: That’s for sure. I have studied fungi and mushrooms, and it is said that using chemicals many plants and fungi communicate with each other and share nutrients. Orbital Sidekick may be able to hear the voices of the trees from space!
If we are going to automate agriculture, we need to monitor the farmland to make sure it is functioning properly. Of course, it is necessary to install sensors on the ground to monitor the situation at the same time, but observations from space are also effective. If we can combine ground-based sensors with hyperspectral satellites, we can expect a synergistic effect. For example, there is an initiative to install IoT sensors on the vines of a remote grape farm and send imaging instructions directly from the sensors to a satellite to monitor the health of the grapes.
Serika: This could be used for a variety of things, depending on your ideas! How do you hope to contribute to society through hyperspectral satellites?
David: Orbital Sidekick wants to lead the transition to clean energy and help maintain a healthy global environment for a sustainable future by building satellite constellations and analyzing satellite data. Our program provides free data to educators, researchers, universities, research institutions, and non-profit organizations that are working on environmental issues and building a sustainable society. If you are interested, please contact us.
— — — — — — — — —
The guest for the 26th edition of our series of conversations with Astronaut Serika was David Gauthier of Orbital Sidekick. Stay tuned!