TL;DR
This paper compares the precision and accuracy of elevation data from the ICESat laser altimeter and the Envisat radar altimeter over two key Antarctic regions. It finds that ICESat generally offers higher precision, but the difference between the two satellites provides insights into Envisat's signal penetration depth into the snowpack.
What Problem Does This Paper Address?
Accurately monitoring the Antarctic ice sheet is crucial for understanding global sea-level rise. Satellite altimetry has been a vital tool for this, but different sensor types (laser vs. radar) have varying characteristics. This paper addresses the need to quantify and compare the precision and potential biases of two prominent altimetry missions, ICESat (laser) and Envisat (radar), over distinct Antarctic regions (ice-gaining Lambert-Amery System - LAS, and ice-losing Amundsen Sea Sector - ASS) to better understand their respective strengths and limitations for ice sheet monitoring.
Methods
The authors employed a crossover analysis technique to compare elevation data from ICESat and Envisat. This involved identifying points where satellite tracks intersect. For each satellite, the standard deviation of elevation residuals at these crossover points was calculated to determine its precision. The mean difference between ICESat and Envisat elevations at crossovers was used to quantify their deviation. Data preprocessing included filtering, projection, and track classification. Crucially, Envisat data underwent slope correction, and ICESat data received saturation correction. To handle the sparser ICESat data, a 20km x 20km grid was used for overlap analysis, while a denser 10km x 10km grid was applied for Envisat.
Datasets
- ICESat GLA12 dataset: This is the Level 3 dataset for the Geoscience Laser Altimeter System (GLAS) on the Ice Cloud and Land Elevation Satellite (ICESat), covering Antarctic and Greenland ice sheets. The study utilized data from campaigns 1 through 16, spanning from February 2003 to October 2008.
- Envisat Level 2 data: This dataset comes from the Envisat Radar Altimeter 2 (RA-2) instrument. The study period for Envisat data was from September 2003 to December 2008.
Key Insights and Ideas
- Laser vs. Radar Precision: ICESat, with its laser altimeter, consistently demonstrated higher precision than Envisat's radar altimeter across both study regions. This is expected, as laser altimeters typically offer finer resolution.
- Regional Precision Differences: Precision generally improved at higher latitudes for both satellites, likely due to more regular track coverage. The LAS region showed better precision than the ASS region for both instruments.
- Envisat's Penetration Depth: The difference in elevation measurements between ICESat and Envisat provides an estimate of Envisat's radar signal penetration depth into the snowpack. This difference was found to be around 0.43 m in the ASS and 0.53 m in the LAS.
- Terrain Complexity Matters: The ASS region, characterized by more complex terrain and significant ablation, exhibited a larger deviation between the two satellites. This suggests that the slope correction applied to Envisat data might not fully account for all terrain-induced errors in highly variable areas.
Key Results
- ICESat Precision:
- Amundsen Sea Sector (ASS): 8.9 cm to 17 cm
- Lambert-Amery System (LAS): 6.7 cm to 14.3 cm
- Envisat Precision:
- Amundsen Sea Sector (ASS): 0.81 m to 1.57 m
- Lambert-Amery System (LAS): 0.46 m to 0.81 m
- ICESat-Envisat Deviation:
- Amundsen Sea Sector (ASS): Mean difference of 0.43 m ± 7.14 m
- Lambert-Amery System (LAS): Mean difference of 0.53 m ± 1.23 m
- The LAS region showed better agreement (lower standard deviation of difference) between ICESat and Envisat compared to the ASS region.
Limitations and Future Work
- Limited Scope: The study focused on precision and deviation but did not analyze factors like slope, ice velocity, or elevation change trends, which could influence the results.
- Future Work: The authors suggest that correlation analysis incorporating these additional factors would be a valuable next step. They also propose that the good precision of ICESat makes it suitable for detailed ice sheet monitoring, while Envisat's data, despite lower precision, is still valuable for applications like DEM generation where meter-level accuracy is sufficient.
Summary updated