Optical cameras can’t see through clouds. In Part 2, we explore Synthetic Aperture Radar (SAR), the LEO technology that allows satellites like Sentinel-1 and ICEYE to image the Earth at night and through storms.
In Part 1 of this series, we introduced the “Speed Demons” of the satellite world: Low Earth Orbit (LEO) satellites. We learned that because they fly so close to Earth (under 2,000km), they are fast, low-latency, and perfect for modern tech stacks.
But there is a problem with traditional satellites that most people don’t think about.
Clouds.
At any given moment, about 67% of the Earth is covered by clouds. If you are relying on an optical satellite (essentially a giant DSLR camera in space) to monitor a flood in India or a port in London, you are often out of luck. If it’s cloudy, or if it’s night time, your data is just a black square or a fluffy white blur.
Developers hate incomplete data. So, how do we solve this? We stop looking with our eyes and start looking with Radar.
Welcome to Part 2. Today, we are diving into Synthetic Aperture Radar (SAR)—the LEO technology that can see through clouds, smoke, and total darkness. To understand it, we’ll look at the two heavyweights of the industry: the public workhorse (Sentinel-1) and the private sniper (ICEYE).
What is SAR? (The Camera that Beeps)
To understand SAR, you have to understand the difference between Passive and Active sensing.
- Passive Sensing (Optical): This is like your phone camera. It needs an external light source (the Sun). No sun? No picture.
- Active Sensing (SAR): This is like a camera with a flash, or a bat using echolocation. The satellite shoots its own energy down to Earth and measures how much bounces back.
Why is it called “Synthetic Aperture”?
Here is the physics trick. To get a high-resolution radar image from space, you would physically need an antenna that is several kilometers long. Obviously, we can’t launch a 5km-long antenna into space.
Instead, we use the speed of LEO. Because the satellite is moving so fast (17,000 mph), it transmits pulses rapidly as it flies. By processing the “history” of these pulses together, we can mathematically simulate (synthesize) a giant antenna. We trick the physics into thinking the antenna is huge, allowing us to see crisp details like cars or ships from hundreds of kilometers away.
Why SAR Lives in LEO
You will rarely find a SAR satellite in GEO (Geostationary Orbit). Remember, GEO is 35,000km away.
Radar follows the Inverse-Square Law (actually, for radar it’s the inverse-fourth power). If you double the distance, the signal gets 16 times weaker. To shoot a radar pulse from GEO and get a bounce back, you would need a nuclear power plant attached to your satellite.
LEO is the only place where SAR makes sense. The proximity allows these satellites to blast the Earth with microwave pulses and listen for the echo without draining their batteries in 5 seconds.
Example 1: The Public Workhorse – Sentinel-1
If you are a developer looking to get into SAR data without spending a dime, Sentinel-1 is your best friend.
Operated by the European Space Agency (ESA) as part of the Copernicus program, Sentinel-1 is actually a pair of satellites (1A and 1B/1C) that dance around the globe.
- The Tech: It uses C-Band Radar. Think of C-Band as the “medium frequency.” It is fantastic at general-purpose monitoring. It offers a great balance between coverage (seeing a wide area) and detail.
- The Superpower: Interferometry (InSAR). By comparing two images of the exact same spot taken days apart, Sentinel-1 can detect ground movement down to the millimeter.
- Use Case: This is the go-to tool for scientists. When a volcano inflates before an eruption, or when the ground sinks due to groundwater extraction (subsidence), Sentinel-1 sees it. It’s also the primary tool for flood mapping because radar bounces off smooth water differently than rough ground, making floods pop out clearly even through hurricane clouds.
For Developers: You can access this data for free via the [Copernicus Open Access Hub] or Google Earth Engine.
Example 2: The Private Sniper – ICEYE
If Sentinel-1 is a wide-angle floodlight, ICEYE is a laser pointer.
ICEYE is a Finnish “New Space” company that changed the game. Before them, people thought SAR satellites had to be massive, bus-sized behemoths like Sentinel-1. ICEYE shrunk them down to the size of a suitcase.
- The Tech: It uses X-Band Radar. This is a higher frequency with a shorter wavelength. It doesn’t penetrate vegetation as well as C-Band, but it provides incredible resolution.
- The Superpower: Revisit Rate. Because their satellites are small and “cheap,” ICEYE launches swarms of them. Instead of waiting 6 days for a satellite to fly over, an ICEYE constellation might fly over your target every few hours.
- Use Case: This is for the commercial and defense sector. Insurance companies use it to verify if a specific roof was damaged in a storm. Oil traders use it to measure the shadow inside floating oil tanks to estimate global oil supply.
For Developers: This is premium data. You usually access it via an API where you pay per square kilometer or per “tasking” request.
Conclusion
SAR transforms the Earth from a visual photograph into a dataset of physical properties (roughness, moisture, shape).
By understanding the difference between the open C-Band data of Sentinel-1 and the high-res X-Band swarms of ICEYE, you can choose the right tool for the job. Do you need to map a flood across an entire state for free? Use Sentinel. Do you need to count the number of cars in a parking lot at 2 AM? Use ICEYE.
But as we launch more of these active sensors and mega-constellations into LEO, the orbit is getting crowded. Really crowded.
Join us for the finale, Part 3: The Future, where we discuss the dark side of the New Space economy: Space Junk, the Kessler Syndrome, and what happens when the sky becomes too full to fly.
Sources
- ESA Sentinel Online: https://sentinels.copernicus.eu
- ICEYE Technology: https://www.iceye.com/technology
- NASA – What is SAR?: https://www.earthdata.nasa.gov
AlphaEarth Arctic Arctic Navigation Data Science Deep Learning DGX DGX Spark Earth Observation FP4 Precision GB10 Geospatial Data GIS Ice navigation system Latency Optimization Local LLM Machine Learning Maritime Logistics navigation Northern Sea Route NRT Processing NSR NVIDIA NVIDIA DGX Spark Passive Microwave Path Algorithm Pathfinding Algorithms Polar Navigation python RAG Remote Sensing Route RouteView SAR SAR Data SAR Imagery Satellite satellite imagery sea Ice Sea Ice Analysis Sea Ice Drift Sea Ice Mapping Sentinel-1 Synthetic Aperture Radar Unified Memory vLLM
Leave a Reply