Greenland Ice Mapping Project: ice flow velocity variation at sub-monthly to decadal timescales

We describe several new ice velocity maps produced by the Greenland Ice Mapping Project (GIMP) using Landsat 8 and Copernicus Sentinel 1A/B data. We then focus on several sites where we analyse these data in conjunction with earlier data from this project, which extend back to the year 2000. At Jako...

Full description

Bibliographic Details
Published in:The Cryosphere
Main Authors: I. Joughin, B. E. Smith, I. Howat
Format: Article in Journal/Newspaper
Language:English
Published: Copernicus Publications 2018
Subjects:
Online Access:https://doi.org/10.5194/tc-12-2211-2018
https://doaj.org/article/654def6937c945b7b1be631be2d74410
Description
Summary:We describe several new ice velocity maps produced by the Greenland Ice Mapping Project (GIMP) using Landsat 8 and Copernicus Sentinel 1A/B data. We then focus on several sites where we analyse these data in conjunction with earlier data from this project, which extend back to the year 2000. At Jakobshavn Isbræ and Køge Bugt, we find good agreement when comparing results from different sensors. In a change from recent behaviour, Jakobshavn Isbræ began slowing substantially in 2017, with a midsummer peak that was even slower than some previous winter minima. Over the last decade, we identify two major slowdown events at Køge Bugt that coincide with short-term advances of the terminus. We also examined populations of glaciers in north-west and south-west Greenland to produce a record of speed-up since 2000. Collectively these glaciers continue to speed up, but there are regional differences in the timing of periods of peak speed-up. In addition, we computed trends in winter flow speed for much of the south-west margin of the ice sheet and find little in the way of statistically significant changes over the period covered by our data. Finally, although the consistency of the data is generally good over time and across sensors, our analysis indicates that substantial differences can arise in regions with high strain rates (e.g. shear margins) where sensor resolution can become a factor. For applications such as constraining model inversions, users should factor in the impact that the data's resolution has on their results.