01 Jul 2021

01 Jul 2021

Review status: a revised version of this preprint is currently under review for the journal ESSD.

A new Greenland digital elevation model derived from ICESat-2

Yubin Fan1,2,3, Chang-Qing Ke1,2,3, and Xiaoyi Shen1,2,3 Yubin Fan et al.
  • 1Jiangsu Provincial Key Laboratory of Geographic Information Science and Technology, Key Laboratory for Land Satellite Remote Sensing Applications of Ministry of Natural Resources, School of Geography and Ocean Science, Nanjing University, Nanjing, 210023 China
  • 2Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing, 210023 China
  • 3Collaborative Innovation Center of South China Sea Studies, Nanjing, 210023 China

Abstract. Greenland digital elevation models (DEMs) are indispensable to fieldwork, ice velocity calculations and mass change estimations. Previous DEMs provided Greenland elevation information for different periods, but long temporal coverage introduced additional time uncertainty to scientific research. To provide a high-resolution DEM with a definite time, approximately 5.8 × 108 ICESat-2 observations from November 2018 to November 2019 were used to generate a new DEM for both the ice sheet and glaciers in peripheral Greenland. A spatiotemporal model fit process was first performed at 500 m resolution. To improve ICESat-2 data utilization, DEMs with 1 km and 2 km resolution across all of Greenland and an additional 5 km resolution in southernmost Greenland were used to fill the DEM gaps. Kriging interpolation was used to fill the remaining 2 % of void grids that were insufficiently observed by ICESat-2 measurements. IceBridge mission data acquired by the Airborne Topographic Mapper (ATM) Lidar system were used to evaluate the accuracy of the newly generated ICESat-2 DEM. Overall, the ICESat-2 DEM had a median difference of −0.48 m for all of Greenland, which agreed well with the IceBridge data, and the performance in the calculated and interpolated grids was similar. Better accuracy could be observed in the northern basins due to the larger proportion of calculated grids with 500 m resolution. The ICESat-2 DEM showed significant improvements in accuracy compared with other altimeter-derived DEMs. Compared to the DEM generated by image pairs, the accuracy was also significantly higher than those of the 1 km ArcticDEM and TanDEM. Similar performance between the ICESat-2 DEM and 500 m ArcticDEM indicated the high accuracy and reliability of the ICESat-2 DEM. Moreover, the ICESat-2 DEM performed better on northern aspects than the 500 m ArcticDEM. Overall, the ICESat-2 DEM showed great accuracy stability under various topographic conditions, hence providing a time-accurate DEM with high accuracy that will be helpful to study elevation and mass balance changes in Greenland. The Greenland DEM and its uncertainty are available at (, Fan et al, 2021).

Yubin Fan et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on essd-2021-183', Anonymous Referee #1, 15 Jul 2021
  • RC2: 'Comment on essd-2021-183', Anonymous Referee #2, 11 Oct 2021

Yubin Fan et al.

Data sets

Greenland DEM and its uncertainty maps based on ICESat-2 Yubin Fan, Chang-Qing Ke, Xiaoyi Shen

Yubin Fan et al.


Total article views: 765 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
564 182 19 765 9 13
  • HTML: 564
  • PDF: 182
  • XML: 19
  • Total: 765
  • BibTeX: 9
  • EndNote: 13
Views and downloads (calculated since 01 Jul 2021)
Cumulative views and downloads (calculated since 01 Jul 2021)

Viewed (geographical distribution)

Total article views: 650 (including HTML, PDF, and XML) Thereof 650 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 23 Jan 2022
Short summary
A new digital elevation model of Greenland was provided based on the ICESat-2 observations acquired from November 2018 to November 2019. A model-fit method was applied within the grid cells within different spatial resolutions to estimate the surface elevations with a modal resolution of 500 m. We estimated the uncertainty with a median difference of −0.48 m for all of Greenland, and the advantage of accurate time can benefit studies of elevation change and mass balance in Greenland.