Interferometric Synthetic Aperture Radar

A schematic diagram depicting a radar satellite orbiting above the Earth illuminating a swath of ground on the right-hand side of the direction of travel. The illumination forms a radar image of the ground surface within the swath area. At a later time the satellite returns and records a second radar image. Between the two image acquisitions an earthquake has occurred and the ground surface has moved. For a particular image pixel the phase component of the second image has a sub-wavelength shift compared to the first image. These phase shifts are mapped over the entire imaged area in what is called the interferogram.

Figure 1.
Two SAR images of the same area
are acquired at different times.
If the surface moves between the
two acquisitions a phase shift is
recorded. An interferogram maps
this phase shift spatially.

Interferometric Synthetic Aperture Radar (InSAR) is a geodetic technique that can identify movements of the Earth's surface. Observations of surface movement made using InSAR can be used to detect, measure, and monitor crustal changes associated with geophysical processes such as tectonic activity and volcanic eruptions. Ground subsidence caused by anthropogenic influences such as groundwater or hydrocarbon extraction can also be identified with InSAR. When combined with ground-based geodetic monitoring, such as Global Navigation Satellite Systems, InSAR can identify surface movements of millimetre to centimetre scale with high spatial resolution.

InSAR can be used for a wide range of surface deformation studies, for example:

  • Subsidence and uplift induced by anthropogenic activities such as groundwater or hydrocarbon extraction, or reinjection into reservoirs during carbon capture and storage
  • coseismic deformation caused during an earthquake
  • postseismic and interseismic deformation on crustal faults between earthquakes
  • inflation/deflation of subsurface magma chambers preceding volcanic eruptions
  • monitoring surface movements in urban environments.
The two interferograms are displaying different representations of the same data. The wrapped interferogram displays the data modulo 2¿ radians and therefore a series of rainbow-coloured fringes is shown. It is not easily evident from this interferogram where the greatest ground displacement has occurred. The unwrapped interferogram has a continuous colour scale mapped over a range of -40 to 40 radians. In this interferogram one can see that the greatest negative range change (i.e. uplift) occurs across a 50 km stretch of the coastline; closest to the offshore Manokwari trench to the north. Two positive range change lobes (i.e. subsidence) are located approximately 20 km to the east and west of the zone of negative range change. Regions further than 50 km from the coastline have only a small amount of relative range change.

Figure 2.
A wrapped (A) and unwrapped (B)
interferogram of an earthquake
doublet that occurred in West Papua,
Indonesia created using data from
the Japanese ALOS satellite. The
magnitude 7.6 and 7.4 earthquakes
occurred on 03 January 2009 within
3 hours of each other and were
caused by subduction on the
offshore Manokwari Trench, which
is located north of the coastline.
Unwrapped phase in radians can
be converted to 'range change' or
displacement in millimetres with
knowledge of the satellite radar

Three images of the same area are shown for the dates 28 June 2006, 19 May 2008 and 19 April 2010. In the first image the entire region shows close to zero displacement. In the two subsequent images, two anomalous zones approximately 1 km across have an increased displacement, with greater magnitude for the most recent image. The maximum positive displacement is between 100-120 mm.

Figure 3.
InSAR time-series product showing
cumulative surface displacement
over time for a small region in the
southern New South Wales coal
fields. The 1-dimensional
displacement observations are in
the line-of-sight of the satellite;
the slanted path between the
ground and satellite position.
The positive polarity of the
signal at two anomalous zones
indicates movement away from
the satellite (i.e. subsidence).

How InSAR Works

InSAR uses two or more Synthetic Aperture Radar (SAR) images of an area to identify surface movements through time. Remote sensing satellites that collect SAR imagery transmit pulses of microwave energy to the Earth's surface and record the amount of backscattered energy. The use of microwave energy provides an all-weather capability because of its low sensitivity to clouds and rain.

SAR images contain information on the Earth's surface in the form of the amplitude and phase components of the backscattered radar signal. The amplitude image records information on the terrain slope and surface roughness, while the phase image records information on the distance between the satellite and the Earth's surface.

Differential InSAR uses two SAR images of the same area acquired at different times. If the distance between the ground and satellite changes between the two acquisitions due to surface movement, a phase shift will occur (Figure 1).

When mapped spatially the phase shift is a 'wrapped' signal within a range of 2¿ radians that appears as a series of interference fringes in an interferogram (Figure 2A). When this interferogram is unwrapped, the number of fringes is summed to give a continuous field of relative phase change (Figure 2B). When first processed, the initial interferogram contains a number of signal components, such as residual signals due to the orbital geometry of the satellite and signals due to the different atmospheric conditions at the time of the two acquisitions. After processing a network of interferograms, the signal component originating from surface movement can be isolated.

InSAR Products

By combining a network of multiple interferograms over a region, velocity map and time-series products can be generated (Figure 3). A velocity map gives the surface movement for each image pixel averaged over the total observation period whereas the time-series product shows the history of surface positions for a pixel at each acquisition time. The former is useful for mapping geophysical processes that occur steadily through time, for example the build-up of strain at a locked crustal fault zone. The latter is useful for detecting geophysical processes that vary considerably through time and may cause fluctuations in the direction of surface movement, for example the inflation and deflation of a magma chamber beneath an active volcano.