InSAR
- Introduced by
- Gabriel, Goldstein, and Zebker
- Year
- 1989
- Type
- Radar interferometry for millimeter-precision surface deformation
- Family
- Radar remote sensing and deformation monitoring
What does it do?
Interferometric Synthetic Aperture Radar (InSAR) is a radar remote sensing technique that measures millimeter-scale ground surface deformation by analyzing the phase difference between radar images acquired from slightly different orbital positions.
Pioneered by Gabriel, Goldstein, and Zebker in 1989, InSAR has become essential for earthquake rupture characterization, volcanic monitoring, landslide detection, and subsidence quantification.
The intuition behind the method
A radar satellite transmits microwave pulses and records the reflected signal. The phase of the returning signal depends on the distance traveled and surface roughness. By comparing radar images acquired from similar but not identical orbital positions (separated by a few hundred meters perpendicular to the look direction), interference patterns (fringes) emerge. Each fringe represents a 2π phase change, corresponding to a wavelength of surface displacement (~2.8 cm for C-band radar at 5.6 cm wavelength). This extremely precise measurement of topography and deformation requires careful removal of atmospheric and orbital errors but enables stunning resolution of earthquake ruptures and ground subsidence.
History
Pioneered by Gabriel et al. (1989) using ERS-1 data, InSAR remained primarily academic until the 1995 Kobe earthquake, where dramatic interferograms revealed earthquake rupture with unprecedented detail. The technique gained momentum through the 1990s and 2000s, with major applications to earthquake ruptures (1999 Izmit, 2004 Sumatra), volcanic deformation, and slow-slip events. The 2014 launch of Sentinel-1 (free data, global coverage, 12-day repeat) democratized InSAR, enabling operational services for earthquake and volcano monitoring. Recent advances include SBAS (small baseline subset) time-series analysis and machine learning for phase unwrapping.
When to use it
Use InSAR to measure co-seismic (earthquake) rupture patterns, post-seismic relaxation, volcanic deformation, slow-slip events, and ground subsidence from mining or aquifer depletion. InSAR works best on coherent terrain (rock, vegetation); urbanized areas also show good coherence. Avoid InSAR over water (low coherence) or high-relief mountains (decorrelation due to geometric differences). Combine with GPS and seismic data for multi-parameter constraints.
Limitations and considerations
- Coherence loss: rapid surface changes (vegetation growth, water level changes, heavy rain) destroy phase coherence, rendering images unusable
- Atmospheric sensitivity: water vapor variations in the troposphere introduce phase noise; difficult to remove without external atmospheric data
- Phase wrapping ambiguity: large deformations (> one wavelength, ~2.8 cm) cause phase wrapping; distinguishing 2πN + phase requires other constraints
- Layover and shadow: steep terrain causes radar backscatter ambiguities (layover) or signal dropout (shadow), limiting usable area
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Millimeter accuracy: resolves deformation of centimeters to millimeters over wide areas
- Cloud-independent: radar penetrates clouds, enabling observation in rainy or cloudy regions
- Wide area coverage: satellite images cover hundreds of kilometers in a single pass
- Rapid revisit: modern satellites (Sentinel-1) have 6–12 day repeat cycles, enabling weekly deformation monitoring
Applications
- Earthquake rupture characterization: mapping co-seismic slip distribution on fault planes, testing mechanical models, and constraining earthquake magnitudes
- Volcanic deformation monitoring: detecting magma intrusion beneath volcanoes and forecasting eruption probability through ground uplift rates
- Landslide and slope stability: identifying slow landslide movement and deformation preceding catastrophic failure, enabling early warning
- Subsidence monitoring: tracking urban subsidence from groundwater or aquifer depletion, mining, and hydrocarbon production for infrastructure and water resource planning
How it works
How it works
Two SAR images of the same area are acquired from slightly different satellite positions (master and slave images). Data are from radar satellites (ERS, Envisat, Sentinel-1, ALOS-2) with repeat intervals of days to weeks.
The slave image is co-registered to the master with sub-pixel precision. Orbital errors are estimated and corrected using precise satellite trajectory data.
The complex interferogram is computed by multiplying the master image by the complex conjugate of the slave image. This interferogram contains phase variations due to topography, deformation, atmosphere, and noise.
Topography is removed using a digital elevation model (DEM). Atmospheric effects (water vapor, ionosphere) are estimated and subtracted. Phase unwrapping converts wrapped phase (-π to +π) to continuous deformation.
The final deformation map reveals ground motion: positive phase indicates motion toward the satellite; negative indicates motion away. Deformation is mapped at 5–100 m resolution with millimeter accuracy for well-coherent targets.
Common pitfalls
- Misinterpreting atmospheric phase as ground deformation; careful multi-image stacking and external atmospheric data are essential
- Incorrect phase unwrapping leading to systematic errors of integer multiples of 2π; validation against GPS is critical
- Assuming zero deformation before the first SAR image; earthquakes or other events may have occurred before data acquisition
Frequently asked questions
Why do earthquake rupture patterns visible in InSAR sometimes disagree with seismic moment tensor solutions?
Seismic moment tensor solutions are averaged representations assuming simple fault geometry; InSAR reveals the actual detailed spatially-variable slip distribution. Asperities (strong patches) and barriers on the fault produce slip heterogeneity that moment tensors cannot resolve. Joint analysis combining both datasets is more complete.
What is phase coherence and why does it matter?
Coherence is the correlation between master and slave radar images, ranging from 0 (no correlation) to 1 (perfect correlation). High coherence (>0.5) is needed to extract reliable phase information. Coherence degrades due to decorrelation: temporal (surface changes between acquisitions), geometric (perpendicular baseline too large), or thermal (instrument noise). Low-coherence regions appear as noise in deformation maps.
Can InSAR detect slow-slip events (silent earthquakes)?
Yes, in favorable cases. Slow-slip events release moment gradually over days to weeks, producing continuous ground deformation. Well-positioned InSAR can detect cumulative slip of centimeters. Recent studies of Cascadia, Hikurangi, and other subduction zones have detected slow slip using InSAR stacking of multiple acquisitions.
What is the difference between ascending and descending satellite passes?
Ascending passes image from east to west (satellite moving north); descending passes image from west to east (satellite moving south). This difference in look direction means east-west and vertical deformation are separable by combining ascending and descending InSAR data: E-W motion is sensitive to geometric differences; vertical motion is more uniform between passes.
Related methods
Sources
Gabriel, A. K., Goldstein, R. M., & Zebker, H. A. (1989). Mapping small elevation changes over large areas: Differential radar interferometry. Journal of Geophysical Research, 94(B7), 9183-9191.
Open DOI referenceCopied.Massonnet, D., & Feigl, K. L. (1998). Radar interferometry and its application to changes in the Earth's surface. Reviews of Geophysics, 36(4), 441-500.
Open DOI referenceCopied.