Funded by NSF Office of Cyberinfrastructure and EarthCube programs, the collaborative GeoSCIFramework (GSF) project aims to improve earthquake, tsunami, and volcano early warning applications by applying big data analytics and machine learning methods to large streams of real-time data from a mix of seismic, geodetic-related sensors, and differential interferometric synthetic aperture radar (DInSAR) satellite imagery. It will provide researchers with a suite of datasets and a means to detect, monitor, and analyze geophysical activity over a region of interest. The work presented here focuses on the longer-term evolution of volcanic processes by developing DInSAR time series over Hawaii from November 2015 to April 2021. Our automated processing routine uses the Small Baseline Subset (SBAS) method and is based off of Generic Mapping Tools (GMT5SAR) and the Generic InSAR Analysis Toolbox (GIAnT) software [Kelevitz et al., 2021; Corsa et al., 2021]. We recently containerized the ISCE2 Stack Processor and MintPY on the Summit supercomputer for more efficient processing. Our final DInSAR time series is then integrated with Global Navigation Satellite System (GNSS) data using the Ordinary Kriging interpolation method to reveal 3D motions of Earth’s surface and to refine the accuracy of our results [Samsonov et al., 2006; 2008]. We present improved, assimilated DInSAR + GNSS time series and the associated uncertainty analysis. Our current work consists of modeling interferometric products in order to generate a robust, synthetic training data set for the GSF machine learning algorithm.