New Open-Source Python libraries for Radar Interferometry Data Processing and Analysis

Modern satellite missions continuously generate extensive observation datasets for Interferometry Synthetic Aperture Radar (InSAR), which is a crucial technology for monitoring ground surface deformation. The efficient processing and analysis of these extensive InSAR datasets poses two computational challenges: 1) the growing volume of the datasets that needs to be incorporated into the data processing workflow, and 2) the integration of contextual information associated with the InSAR data to reveal the mechanisms driving deformation.   

To address these challenges, we present two open-source Python libraries: SARXarray [1] and STMTools [2]. They facilitate common InSAR data processing tasks and are developed as extensions of the open-source Python library, Xarray, which handles labelled multi-dimensional arrays and is well-suited to the space-time nature of InSAR data. SARXarray is designed to work with coregistered raster stacks, such as SLC or interferogram stacks, offering functionalities like multi-looking, coherence computation, and coherence scatterers selection. STMTools, on the other hand, focuses on large spatio-temporal datasets in the form of a Space-Time Matrix (STM) [3], for instance, coherent scatterers. It can query background contextual data, such as geospatial polygons, and add the attributes-of-interest to the corresponding STM. Furthermore, both SARXarray and STMTools support data chunking and lazy evaluation, enabling the scaling up of the data processing pipeline and parallel processing of larger-than-memory data across various computational infrastructures. 

[1] Ku, O., et al., sarxarray [Computer software]. github.com/MotionbyLearning/sarxarray 

[2] Ku, O., et al., stmtools [Computer software]. github.com/MotionbyLearning/stmtools 

[3] Bruna, M. F. D., van Leijen, F. J., & Hanssen, R. F. (2021). A Generic Storage Method for Coherent Scatterers and Their Contextual Attributes.