Processing methodology for the ITS_LIVE Sentinel-1 ice velocity products

Abstract

The NASA MEaSUREs Inter-mission Time Series of Land Ice Velocity and Elevation (ITS_LIVE) project seeks to accelerate understanding of critical glaciers and ice sheet processes by providing researchers with global, low-latency, comprehensive and state of the art records of surface velocities and elevations as observed from space. Here we describe the image-pair ice velocity product and processing methodology for ESA Sentinel-1 radar data. We demonstrate improvements to the core processing algorithm for dense offset tracking, “autoRIFT”, that provide finer resolution (120 m instead of the previous 240 m used for version 1) and higher accuracy (20 % to 50 % improvement) data products with significantly enhanced computational efficiency (>2 orders of magnitude) when compared to earlier versions and the state of the art “dense ampcor” routine in the JPL ISCE software. In particular, the disparity filter is upgraded for handling finer grid resolution with overlapping search chip sizes, and the oversampling ratio in the subpixel cross-correlation estimation is adaptively determined for Sentinel-1 data by matching the precision of the measured displacement based on the search chip size used. A novel calibration is applied to the data to correct for Sentinel-1A/B subswath and full-swath dependent geolocation biases caused by systematic issues with the instruments. Sentinel-1 C-band images are affected by variations in the total electron content of the ionosphere that results in large velocity errors in the azimuth (along-track) direction. To reduce these effects, slant range (line of sight or LOS) velocities are used and accompanied by LOS parameters that support map coordinate (x/y) velocity inversion from ascending and descending slant range offset measurements, as derived from two image pairs. After the proposed correction of ionosphere errors, the uncertainties in velocities are reduced by 9 %–61 %. We further validate the ITS_LIVE Version 2 Sentinel-1 image-pair products, with 6-year time series composed of thousands of epochs, over three typical test sites covering the globe: the Jakobshavn Isbræ Glacier of Greenland, Pine Island Glacier of the Antarctic, and Malaspina Glacier of Alaska. By comparing with other similar products (PROMICE, FAU, and MEaSUREs Annual Antarctic Ice Velocity Map products), as well as other ITS_LIVE version 2 products from Landsat-8 and Sentinel-2 data, we find an overall variation between products around 100 m yr−1 over fast-flowing glacier outlets, where both mean velocity and variation are on the order of km yr−1, and increases up to 300–500 m yr−1 (3 %–6 %) for the fastest Jakobshavn Isbræ Glacier. The velocity magnitude uncertainty of the ITS_LIVE Sentinel-1 products is calculated to be uniformly distributed around 60 m yr−1 for the three test regions investigated. The described product and methods comprise the MEaSUREs ITS_LIVE Sentinel-1 Image-Pair Glacier and Ice Sheet Surface Velocities: version 2 (DOI: https://doi.org/10.5067/0506KQLS6512, Lei et al., 2022).