Temporal Subset SBAS InSAR Approach for Tropical Peatland Surface Deformation Monitoring Using Sentinel-1 Data

Tropical peatland in Southeast Asia has undergone rapid degradation and shows large
subsidence due to oxidation and peat shrinkage. The measurement of those deformations is thus
valuable for evaluating the peat condition and assessing peat restoration. The time series interferometric
synthetic aperture radar (TInSAR), especially with the small baseline subsets (SBAS) method, is
capable of measuring long-term deformation. However, the dynamic surface scatterers often change
in tropical peatland, which degrades the coherent scatterer (CS) distribution density. This article
presents a simple and efficient TInSAR approach that enhances the CS density under such dynamic
surface scatter variation based on the SBAS method. In the presented approach, a long-time series
of single-look complex images is separated into subsets, and deformation estimation is performed
for each subset. The effectiveness of this simple solution was investigated by InSAR simulation
and validated using SAR observation data. We applied the subset SBAS approach to the three-year
Sentinel-1A C-band SAR dataset acquired over tropical peatland in Indonesia. The analyses showed
an improved number of CSs for the introduced subset approach. We further introduce the color
representation of CS temporal behavior per subset for visual interpretation of scatterer change.