Let’s take a closer look at how SAR displacement data is gathered and measured.
The movement of objects and the ground surface between satellite passes is measured with a calculation. This calculation is the phase difference between successive images, rather than direct displacement output. Displacements less than half a wavelength are easy to determine. However, displacements greater than half can lead to uncertainty.
Displacement output between two SAR images can be obtained with DInSAR. This process involves correcting topography using a digital elevation model. Further processing to remove residual topographical errors, orbital errors, atmospheric effects, and data noise produces a displacement map with centimetric precision.
Analysis of a time series of images (normally at least 20) increases displacement precision. This is because corrections for errors such as atmospheric effects, residuals, and topographical errors are improved.
There are several algorithms to do this. They are suited to different conditions.
These algorithms enable spatiotemporal analysis of the radar phase across multiple images. This produces displacement data with millimetric precision. Depending on the frequency band, accuracy is 2-4mm.
Displacements obtained from InSAR can come from a pair of images or a time series. These displacements are in the line of sight (LOS) direction. This means that the displacement is along a straight line from the satellite antenna to the object on the Earth's surface.
The ascending and descending orbits have different LOS directions on opposite sides of the vertical axis at known incidence angles. Displacement data is derived from ascending or descending images. Positive values denote movement towards the satellite and negative values, movement away from the satellite.
On its own, LOS displacement data provides an indication of patterns of movement. If movements are usually vertical, like heave or subsidence of a horizontal ground surface, it is easy to figure out vertical displacements.
The true pattern of movement may be difficult to deduce if the direction of displacement is unknown. For example, when the ground is sloping. One set of LOS displacements may not be enough to determine the pattern.
Once the data of orbit LOS displacement at known incidence angles is obtained, it can be resolved into two directions. These are true vertical and horizontal (east-to-west) direction. Alternatively, it can be resolved into other directions, such as the direction of east-to-west sloping ground.
Temporal interpolation is necessary. This is because the ascending and descending images were not taken simultaneously. Additionally, spatial interpolation is needed. This is because there may be variations in the distribution of measurement points between the ascending and descending images.
Apart from displacement measurement, SAR data can be used for other practical applications such as soil moisture. The amplitude of SAR signals reflected from the ground varies based on several factors.
These include surface roughness, slope angle, and the dielectric constant of the ground. Soil moisture affects the dielectric constant. When all other conditions are held constant, variations in the amplitude of signals from successive images indicate changes in soil moisture.
In-situ measurements can be used to determine the absolute values of moisture content. These measurements can then be used to calibrate the variation of soil moisture with SAR amplitude. This information can then be used to produce useful maps of soil moisture.
SAR data has a useful application in geotechnical engineering. Geohazards, such as landslides and swelling clays, are significantly impacted by changes in soil moisture.
Large data files containing displacement, soil moisture and other data associated with their geolocation are created and combined. This is done to develop geohazard susceptibility maps.
The best way to view these maps is by uploading them to a web-based geographical information system (GIS). They can be viewed overlaid on maps, optical satellite images, or 3D surface models. GIS also allows the plotting of temporal graphs and profiles setting warning trigger levels and downloading specific data sets of interest.
Want to learn about how retroactive SAR analysis can be applied to the conservation of heritage sites? Read this article next...
Comments