Embedded Files
Image: Sitemap by Me. Using Landsat, World Glacier Inventory, and JIRP GNSS datasets
Validating the accuracy of Sentinel-1b microwave backscatter SAR for tracking temperate alpine glacier velocities
Validating the accuracy of Sentinel-1b microwave backscatter SAR for tracking temperate alpine glacier velocities
(In Progress)
(1) Hehlen, M., (2) Randall, (3) A. Kelly, J. (4) Alexander, K., (5) Kaufmann, J.(1) University of Vermont, (2) Boston University, (3) California State University, L.A., (4) Dartmouth College, (5) Juneau Icefield Research Program
ABSTRACT
This project, advised by Dr. Joshua Kelly (CalState LA), is concerned with spatially and temporally accurate ground truthing of ice offset and speckle tracking used to calculate ice stream velocities worldwide using ESA's Sentinel-1b satellite. S1b contains a microwave backscatter synthetic aperture radar sensor system capable of active remote sensing of glacier through cloud cover and darkness (to differentiate it from Landsat-based offset tracking used in GoLive and Its_Live). However, due to high melt-water concentrations within temperate snow pack, as well as near surface ice lensing, albedo can change enough to make velocity tracking unreliable. This study aims to quantify the error in S1b ice velocity tracking by using sub cm GNSS measurements with observations during nadir flyover. Initial findings show statistical significant low correlation between GPS and S1b measurements during both rainstorms and bluebird flyover windows. Continued study in 2022 will look to the historic record of GNSS measurements during the Sentinel-1 era to correlate observations and quantify error during recorded conditions to improve world-wide accuracy in temperate glacier regions.
Figure 1: Polar Scatter Plot for speed and direction of all 2022 GNSS and nearest neighbors in S1.Figure 2: Map of Taku Glacier Survey. Red colored arrows show GNSS stake displcement over the flyover window. Colored squares and contained arrows are S1 derived velocity. Attached scatterplot for Taku subsetFigure 3: Map of Vaughn-Lewis Glacier Survey. Consistent symbology with Fig 2, but note the change in color gradient velocity magnitude (much slower). Scatterplots show subsets for big and little vaughn-lewis glaciers which bisect the Camp 18 nunatak.Figure 4: Comparison of GNSS measurements at VL with annual ITS_Live Landsat tracking. Note Landsat cannot track velocity during long night winters or during precip events with storms, weighting readings toward bluebird flyover intervals such as the one mapped.
Page updated
Google Sites
Report abuse