G004-0037
A Web-Based Interface for Geodetic Imaging with Large SAR Data Sets using NASA High End Computing
Wednesday, 9 December 2020
Poster
Susan E Owen1, Rowan Biessel2, Lei Pan1, Hook Hua3, David P Bekaert4, Zhangfan Xing1, Andrew Zhang5, Marjorie Lucas5 and Mohammed Karim1, (1)NASA Jet Propulsion Laboratory, Pasadena, CA, United States, (2)Alaska Satellite Facility, Fairbanks, AK, United States, (3)Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, United States, (4)JPL/NASA/Caltech, Pasadena, CA, United States, (5)Jet Propulsion Laboratory, Pasadena, United States
Abstract:
We are developing community synthetic aperture radar (SAR) analysis tools that will allow scientists to run interferometric SAR (InSAR) processing and analysis problems on NASA’s High End Computing (HEC) resources. The interface to the HEC computer is through the automated data system implemented by the JPL-Caltech Advanced Rapid Imaging & Analysis (ARIA) project. The ARIA science data system has achieved the required maturity to enable mass production of InSAR data products, and has currently processed over 20,000 scenes on NASA’s Pleiades computing platform. The community interface will provide a range of analysis options through a straightforward web-based interface and deploys the selected analysis to the HEC automatically. This capability effectively brings NASA supercomputers to the desktop of the Solid Earth scientist, reducing the burden associated with large scale InSAR analysis and with finding the necessary storage and computing resources. We will be extending ARIA’s capabilities to the research user community, giving them access to generate standardized SAR data products from the increasingly large Sentinel-1A/B data archive.
Recent efforts on this project have focused on realizing this web-based guided user interface. This website will not only provide a means for NASA-funded researchers to utilize the HEC and ARIA systems in tandem, but will also guide the user in creating a network of InSAR products that is connected both spatially and temporally for potential geodetic time-series analysis. We are developing the interface with the following components: a tool for generating a desired AOI that follows a Sentinel-1/B acquisition track schema; a wizard that allows users to select existing AOIs, define pairing strategies, and query a processing estimate; and a page that provides updated information about the user’s queried and submitted requests. Once a user queries a processing request, ARIA generates an estimated on-demand processing load by enumerating over the given AOI-Track and InSAR pairing strategies, staging a list of jobs to be run. At the user’s discretion this request can be either discarded or submitted for processing at the NASA's Pleiades Supercomputer facility. When finished processing, these products will be accessible via the existing ARIA Products facetted search tool.
