Critical Uses of Space-based SAR Data
Satellite data impacts nearly every facet of our lives — from the logistics that deliver Amazon packages to your doorstep, to the fresh food on your plate made possible by agricultural tech. Beyond daily conveniences, satellite data is a critical tool for monitoring the globe and making world-changing decisions.
When people think of imagery from space, optical imagers often come to mind. Imagery data from these ‘passive’ satellites is powerful, plentiful, and proves useful for applications like precision agriculture and environmental monitoring. However, there are some limitations related to cloud cover, precision, and light levels.
By contrast, satellites equipped with Synthetic Aperture Radar (SAR), are ‘active’ imagers that send pulses of microwaves that ‘bounce back’ to the spacecraft, allowing the satellite to collect precise information about ground features, and even those below Earth’s surface. Because SAR satellites can operate at any time and can effectively “see through” cloud cover, these instruments provide more detailed data and allow for greater coverage with fewer sensors.
In this series of blogs, we take a closer look at some of the most fascinating, and critical, uses of space-based SAR satellite data.
1. Landmine Detection
Reaching wide-scale use during WWII, landmines have remained a devastating tool, killing or maiming more than 5,000 people annually, and injuring 15,000-20,000 in total every year. Of those casualties, 85% occur in Egypt, Angola, and Iran, where collectively more than 48 million landmines are currently buried.
Most landmines can't be detected by the human eye and optical satellites and drones alone are unable to detect them without additional penetrating technology. Thus, space-based SAR has become a go-to tool for the detection of landmines, thanks to its ground penetration capabilities and wide-area coverage.
An overview of SAR landmine detection:
Regions of Interest (ROIs) are extracted from large SAR data strips
Data of each ROI is parsed for mine-like targets; some mine-like targets are landmines, others are noise (“false positives”)
Landmine probability is determined for each ROI, with high false alarm rates being common
Data is further parsed to isolate landmines from noise in the extracted ROIs
In-orbit feature learning
Feature edges and textures and learning over time reveals which structures in the data are most useful for discrimination
2. Oil Spill Detection
Although the number of large oil spills has largely decreased since the 1970s, between 2010 to 2019 there was an average of 1.8 large oil spills from tankers each year. In addition to costing US$2-8 billion on average to clean each large spill, they are extremely destructive to ecosystems — especially environments that are already fragile.
The number of large spills has been decreasing thanks in part to technological advancements and cooperation between companies and governments, which has been aided by EO technology like SAR.
Large oil spills (>700 tonnes) are undoubtedly devastating and headline-grabbing, but most oil pollution doesn’t occur during large spills. Instead, it comes from equipment run-off and smaller spills (7-700 tonnes) and leaks (<7 tonnes).
“Thousands of oil spills occur in U.S. waters each year. Most of these spills are small, for example when oil spills while refuelling a ship. But these spills can still cause damage, especially if they happen in sensitive environments, like beaches, mangroves, and wetlands.” — NOAA
This means that high-quality imagery, like that from SAR satellites, is required to detect smaller issues that can contribute to overall pollution levels and transform into much larger problems. SAR is therefore considered a prime method of detecting oil spills and leaks, as well as examining, recording, and tracking spills over time.
Thanks to its wide-swath coverage and capability to image any time of day and night and through cloud cover, SAR is perfect for predicting and tracking subsequent travel of oil contaminants to nearby coastal areas. This ability can be augmented by onboard, in-space machine learning and AI, which can detect spills quickly to minimize their impact.
The fallout from oil spills and leaks include:
'Oiling’ physically harms plants and animals:
Oil coats wings, leaving birds unable to fly
Insulating properties of animal fur is stripped away, leading to hypothermia
Significant threat to keystone species like arctic cod
Oil toxicity causes severe health problems in humans:
Immune system effects
Heavy cleanup costs paid by government and corporations:
Government cleanup budgets and restoration resources strained
Example: The BP oil spill in 2010 cost approximately US$64 billion, including everything from criminal charges and penalties to cleanup efforts
An overview of how oil spill detection works:
Surface Changes Detected
SAR data indicates the “roughness” of earth or water surface
Oil film stifles small waves and SAR detects that the energy reflected back to the satellite is less than a normal wave, so anomalies appear “smooth and dark” in contrast to the “normal” areas
Algorithms applied to the data help to classify textures on the ocean’s surface, looking for noise that can include sea ice and vegetation
Once a detection has been made and confirmed, further data analysis is performed to classify the information
3. Water Resource Management
Image Credit: NASA Earth Observatory
Water management has never been more important. From Flint, Michigan, to rural Mozambique, 1 in 10 people on the planet — 785 million — lack access to clean water.
Freshwater will continue to be an increasingly precious resource as human populations grow, populations flock to arid urban centers, freshwater reserves deplete, large-scale agriculture demands more irrigation, and wetlands remain threatened.
Measuring groundwater with optical satellite imagery has historically proven to be difficult, time-consuming, and with questionable accuracy that can “vary among image interpreters.” (Zhang et al.)
“Accurately quantifying surface water extent in wetlands is critical to understanding their role in ecosystem processes. However, current regional- to global-scale surface water products lack the spatial or temporal resolution necessary to characterize heterogeneous or variable wetlands.” — Huang, et. al.
As a result, SAR imagery has become increasingly common for use in water management. It's now a strategic tool for governments, institutions, and corporations alike, which use the data for specific water management applications like:
Agricultural irrigation planning
Mapping surface water
Tracking transformation of wetlands and aquifers
Detecting changes in permafrost
What's next? Improved automated classification, tracking, and learning over time — including onboard processing, AI, and machine learning — which will result in more timely and useful data.
SAR water management highlights:
Faster mapping thanks to SAR’s sensitivity to water content helps to map surface water and measure soil moisture content
Accuracy over 95%, for improved monitoring over time
L-band SAR penetrates soil to determine the water saturation level for applications like irrigation management and crop health, as well as near real-time flood monitoring.
Digital elevation models can then be determined and applied to characterize watersheds and hydrological changes
4. Missile Detection & Counter-Terrorism
The average level of global peacefulness deteriorated in 9 of the last 13 years, according to the Global Peace Index (GPI), which found that 87 countries improved and 73 countries deteriorated in 2021. Although the recent change was relatively small, the report notes that many of the conflicts that had begun to subside in the past decade were replaced by new tensions and crises, as a result of COVID-19 and associated economic uncertainty.
Areas of political tension and economic crisis are notorious breeding grounds for terrorist groups. And these aggressive organizations often leave larger footprints of destruction than their small size might suggest. Small factions of interconnected assailants (hiding in foothills and caves for instance) are extremely difficult to detect, never mind track or subvert.
Over the past decade, 90 countries experienced increased terrorist activity, and only 50 have recorded less, according to GPI results.
Because of its unique detection capabilities, SAR satellite data has proven to be a critical tool in counter-terrorism. With SAR data, tracking vehicle movement — like those used to transport and hide missiles — can be achieved regardless of weather conditions, camouflage, and levels of daylight. This results in better decision-making thanks to improved detection and longitudinal tracking.
Example - missile launchers hidden in mountain caves:
High-resolution SAR L-band allows detection of objects beneath trees, foliage, and camouflaged surface material
Situation assessment & camouflage detection: Ground surface ‘depression’ and disturbed earth assessment detects where heavy objects (like missiles) are moved to and from; can also detect decoys.
When moved in and out of the cave, the ground will have been depressed under the vehicle, which is detected by the SAR satellite
Radar Sight: Beyond Eyes in the Sky
Space-based Earth Observation is one of the most powerful tools we have to improve the world. Where optical imagery data presents challenges related to coverage, SAR has filled this gap and then some. With the onset of onboard processing, in-space machine learning and AI, and more efficient data delivery, the benefits of SAR are expected to compound over the next several years.
Stay tuned for Part II of our series, where we take a look at the use of space-based SAR for climate science, earthquake detection, market analysis, and planetary study.
SpaceAlpha Insights ("Alpha") is a Vancouver-based space company that's developing next-gen Earth Observation satellites for heightened geospatial intelligence. Alpha's SAR-XL satellites will deliver unprecedented global insights for missions related to environmental monitoring, security, logistics, and more.