Mapping of Mean Deformation Rates Based on APS-Corrected InSAR Data Using Unsupervised Clustering Algorithms

Different interferometric approaches have been developed over the past few decades to process SAR data and recover surface deformation, each approach has advantages limitations. Finding an accurate reliable interval for preparing mean deformation rate maps (MDRMs) remains challenging. The primary purpose of this paper is implement application consisting three unsupervised clustering algorithms (UCAs) determining best from SAR-derived data, which can be used interpret long-term processes, such as subsidence, identify displacement patterns. Considering Port Harcourt (in Niger Delta) study area, it was essential remove sources error in extracting signals spatially ionospheric tropospheric delays, before using UCAs obtain its characteristics real data. Moreover, another advanced integration method (AIM) atmospheric phase screen (APS) correction enhance obtained through different processing approaches, including SARs (two-pass interferometry, InSAR) multitemporal interferometry (n-pass DInSAR; permanent scatterer (PSI); small baseline subset (SBAS)). Two methods were chosen evaluate find technique with create MDRM: first one compare corrected by AIM vertical component GPS station, showed providing 58%, 42%, 28% matching GNSS station outputs InSAR, PSI, SBAS, respectively. Secondly, similarity measures Davies–Bouldin index scores implemented SBAS K-medians chosen. Based on a 500 m radius around expressed up 5.5% better patterns than map techniques.