Hello, and welcome to the Earthbound! All subscribers get a weekly post on three interesting earth observation papers. Our paid community receives more in-depth articles focusing on one research study or article, instead of three. If you want to join either of these communities, feel free to subscribe.
Upcoming posts:
April 19th, 2024: EO Research Update
April 22nd, 2024 (paid) : Understanding marine coastal hazards with EO
The dark side of EO….you might have guessed it, but we’re talking about the environmental impacts of the earth observation sector. I reviewed a paper on a similar topic two weeks ago in my weekly EO updates.
In this article, we look at a short piece published in Nature. If you want to read the full article, you can find it here. As this piece was short, our summary will be a quick one.
According to the author, the number of satellites focused on collecting EO data has ballooned by 600% over the past ten years.
In the past, the majority of instruments were government-funded. Now, commercial instruments are becoming more prevalent.
Space junk is becoming a bigger concern as the number of instruments grow.
This junk is not only polluting our space, but it may also interfere with sensors that are currently trying to collect new data
When the satellites are launched, they also release a lot of rocket emissions that impact our climate
The space junk AND operational instruments can interfere with astronomical observations (I didn’t know about this one).
Satellite-driven light pollution
The cloud computing era allows for more “big computing” research, but the environmental impact of cloud data centres is likely not being accounted for in environmental assessments of EO missions
What about all the collected data, that grows exponentially each decade?
Maintaining the data over decades (reprocessing, etc) will be left to new generations
As more data enters the mix, users will need to rely on Deep Learning to process it, which also uses a lot of energy to run
A scary example from this article: Copernicus's single data holding centre emits the CO2 equivalent of 100 economy seats from London to New York!
According to the author, it is well known that EO data centres harm the environment in which they’re located. As such, we need a lot more transparency about where they’re located and how much energy they use. Unsurprisingly, Google and Sentinel-Hub are not open about the locations of these centres.
Takeaways
With the skyrocketing (pun intended) growth of EO, the environmental impacts can no longer be ignored
Space legislation is outdated - it needs to consider data legacy needs and environmental impacts of the satellite and the associated data
Commercial companies need to start sharing information about their data holdings.
That’s it for this week! Thanks for reading.
Other stuff: If you’re interested in learning how to work with remote sensing and GIS data, here’s a course I created last year: