Can someone please explain to me how InSAR is capable of such high accuracy deformation studies (millimetre accuracy) but SAR DEMs are nowhere near as accurate? Is it down to the resolution? Is there a high accuracy LiDAR or GNSS derived DEM used as the base map? Thanks in advance.
Hi Conor,
To make a DEM, and to estimate heights, you need to 'measure' the time it takes for the radar wave to travel from and to the satellite. With all the atmosphere and some other uncertainties, this has a precision of a few meters to tens of meters.
In InSAR we look at the phase changes mostly, under the assumption that the total deformation between two observations is not greater than about half the wavelength (typically 30mm). So we don't need the absolute heights (as in a DEM) to determine the deformation, we just keep track of changes in the phase of the wave arriving back at the satellite. We do need to correct for a lot of things, such as atmosphere and, indeed the terrain heights, to be able to interpret the InSAR results at sub-millimeter precision.
Actually in the course of an InSAR analysis, it's also standard to derive an ad-hoc DEM. (useful for height estimation and geolocation of the reflection points) but again, the accuracy is in the order of 1-2 meter precision, typically.
Hope this helps,
Jos
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology
- Research Papers