Introduction to Radar Interferometry (InSAR) – Prof. Ramon Hanssen (Part 1/4)
Kickstarting a four-part series, this first article lays the foundation of radar interferometry (InSAR) on what phase is, why fringes matter, and how we turn repeat satellite passes into millimeter-scale insights. Over the next days, we’ll dive into parts 2,3 &4. Stay tuned to get the next chapters.
Radar interferometry (InSAR) extracts phase information from SAR images to reveal millimeter-scale ground and asset motion. In this primer, Prof. Ramon Hanssen, TU Delft professor and SkyGeo Co-Founder; explains how amplitude vs. phase works, what interferogram “fringes” really mean, why baseline controls height sensitivity, and how repeat-pass InSAR enables continuous monitoring of the built environment.
Why InSAR matters: from “pretty pictures” to decisions
Most early SAR applications used amplitude (image brightness) for mapping. InSAR taps the phase of the radar signal: initially “random” per image to measure tiny changes between repeat passes. Subtract those phases correctly, and you uncover deformation with millimeter precision, across cities, corridors, and critical assets.
Bottom line: InSAR scales: from regional change detection to asset-level diagnostics, without field crews or daylight.
Amplitude vs phase (and why phase feels “abstract”)
- Amplitude = signal strength (how bright a pixel looks).
Useful for interpretation (roads, buildings, fields). - Phase = angle of the complex radar return.
Looks random in a single image, but becomes meaningful when differences are taken between two dates (repeat-pass).
Think of each pixel as a sum of many tiny reflections (building edges, roofs, poles). Alone, their phase is arbitrary; between dates, consistent reflectors reveal true movement.
Interferograms & fringes: reading the “contours of change”
When you subtract the phase of two SAR images, you get an interferogram.
- Color “fringes” (cyclic 0 – 360°) show changes linked to topography or motion.
- Count fringes like tree rings: more fringes = larger difference (height or displacement, depending on the setup).
Baseline & height ambiguity: sensitivity 101
- Baseline (B⊥) = the perpendicular distance between two acquisition geometries.
- Larger B⊥ – higher sensitivity to elevation (DEM mapping).
- Height ambiguity = meters of height per full phase cycle; smaller is better for topography.
For deformation monitoring, we typically use repeat-pass interferometry with controlled baselines and time series models to isolate motion rather than topography.
From randomness to reliability: why repeat-pass works
If nothing changes on the ground (and geometry is aligned), subtracting phases should yield zero, that predictability is the foundation of InSAR.
- Stable reflectors (buildings, bridges, paved surfaces): consistent time series.
- Unstable reflectors (vegetation, open water): filtered out.
- Result: dense, high-quality point clouds over man-made structures and infrastructure.
Beyond earthquakes & volcanoes: the built environment advantage
Classic InSAR hit headlines with earthquakes and volcano inflation/deflation. The biggest operational impact today is in the built environment:
- Bridges & viaducts: differential motion vs. approaches.
- Rail & road: settlement, seasonal heave, construction impact.
- Dams, levees, tailings: deformation trends and alerts.
- Pipelines & utilities: ground/structure mismatch that strains connections.
- Urban portfolios: prioritize inspections, estimate risk, and plan capex.
InSAR lets operators measure, estimate, monitor, detect, categorize, anticipate, and respond, turning pixels into asset decisions.
Practical nuggets (demystified)
- Complex numbers: each pixel is a vector with amplitude (length) and phase (angle).
- Phase differencing: cancel “randomness,” reveal change.
- Fringes: cyclic color bands = repeated phase cycles.
- Look direction: ascending vs. descending passes see motion from different sides; combine wisely to resolve horizontal/vertical trends.
- Time series: more acquisitions is tighter velocity/acceleration estimates and robust seasonal modeling (e.g., gas storage up/down cycles).
What you can do with InSAR
- Benchmark & trend: Is an asset stable, subsiding, or accelerating?
- Differentiate: Is the street moving while the building is pile-founded and stable?
- Prioritize: Rank bridges, embankments, and corridors by risk for inspection.
- Anticipate & plan: 3 mm/yr becomes 3 cm/decade: act before thresholds bite.
- Respond: Detect anomalies early and dispatch targeted fieldwork.
Watch the source lecture
Video: Introduction to Radar Interferometry 1/4 Prof. Ramon Hanssen
This first video includes:
00:00 Intro
11:01 Complex numbers & SAR
18:09 SAR SLC observations
23:24 Satellite radar interferometry
28:43 Applications: the European Ground Motion Service & the Dutch Surface Motion Map
32:54 What can we do with it?
41:58 Why should we continuously monitor?
44:50 InSAR intuitive approach: geometry
55:40 Reference phase (flat earth phase)
1:05:00 Interferometry: deriving the equations
1:24:00 Q&A
Key takeaways
- InSAR turns phase differences into millimeter-level motion.
- Baseline governs height sensitivity; repeat-pass time series reveal deformation.
- The built environment is where InSAR delivers outsized operational value.
- With dense, reliable points over man-made structures, you can prioritize, anticipate, and act.
Event context: where this training happened
This lecture was delivered within the 12th ESA Training Course on Earth Observation 2022, held June 27–July 1, 2022 at Riga Technical University (Latvia). Organized by ESA with Latvia’s Ministry of Education and Science, the Institute of Environmental Solutions, and RTU, the course mixed theory and hands-on labs using Copernicus Sentinel-1 (SAR) and Sentinel-2 (optical). Topics spanned land cover, forestry, agriculture (including climate impacts), InSAR for terrain motion (e.g., gas), ship detection, and integrated applications. Attendance was free, English-language, and targeted to researchers, students, and early-career professionals, with a cohort of around 35 participants. eo science for society
Conclusion
InSAR improves awareness: from regional change to asset-level decisions, by converting radar phase into actionable insights & motion intelligence. This is where it gets interesting: InSAR Data alone is not enough: You need the expertise and interpretation of globally certified experts in your field.
-“Yes. I see the red and green dots, but what do I do with them?”
How to understand with the accuracy that each Industry needs? If you manage bridges, rail, dams, pipelines, or urban assets, this method gives you the when, where, and how much needed to prioritize work, control risk, and defend budgets. Only when understood correctly by experts… Like SkyGeo.
See how this applies to your portfolio: book a short SkyGeo walkthrough.
Thanks to the EO Open Science YouTube channel (https://www.youtube.com/@eoopenscience4372) for uploading and maintaining the recording of Prof. Hanssen’s session.
