Invited - Keynote: Monitoring Antarctica's Ice Shelves and Active Subglacial Lakes with Satellite Altimetry
Scripps Institution of Oceanography
Ice Shelf Thickness Change from CryoSat-2
1CPOM, University of Leeds, United Kingdom; 2Mullard Space Science Laboratory, University College London, UK
Floating ice shelves that fringe the majority (74%) of Antarctica's coastline provide a direct link between the ice sheet and the surrounding oceans, and changes in their constitution have been shown to influence the flow of inland ice due to their buttressing effect. This process has become increasingly important over recent decades as Antarctic ice shelves have thinned, retreated, and collapsed – events that have been recorded largely by European satellites. At the Antarctic Peninsula, ice shelf retreat has been observed throughout the satellite era (18% over 50 years), and large sections of the Larsen-A, Larsen-B, and the Wilkins Ice Shelf collapsed, catastrophically in 1995, 2002, and 2008, respectively. In the Amundsen Sea, ice shelves at the terminus of the Pine Island and Thwaites glaciers have thinned at rates in excess of 5 meters per year for more than two decades. Both signals are indicative of long-term changes in the regional climate, and have impacted on the ice inland. CryoSat-2 has repeatedly surveyed 49% of the coastal margins of Antarctica six or more times within the first three years of operation. That is six and five times more that ENVISAT (8%) and ICESat (10%), and will continue to be improved on if CryoSat-2 continues to live beyond its original mission lifetime. Further to this the CryoSat-2 synthetic aperture radar interferometry (SARIn) mode has a smaller footprint size (~300m by 1km) than previous radar altimetry missions, increasing the spatial resolution of the measurements made by radar altimeters. We use CryoSat-2 to map ice thickness change on Antarctic ice shelves by exploiting the dense spatial sampling and repeat coverage provided by the SARIn mode data acquired by CryoSat-2 from 2010 to the present day. We find that ice shelf thinning rates exhibit large fluctuations over short time periods, and the improved spatial resolution of CryoSat-2 enables us to resolve the spatial pattern of thinning with ever greater detail in Antarctica.
Elevation Changes of the Greenland Ice Sheet from 2013 to Present - CryoSat-2 vs. SARAL/ALtiKA
1DTU Space, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark; 2Nansen Environmental and Remote Sensing Center, 5006 Bergen, Norway
For more than six years, CryoSat-2 has measured elevation changes of the Greenland Ice Sheet. The nature of the onboard Ku-band radar enables CryoSat-2 to measure climate dependent changes in the upper snow/firn cover of the interior parts of the ice sheets. Hence, the retrieved waveform is a convolution of both surface scattering and volume scattering. This hampers the direct interpretation of surface elevation change, and further ice sheet-wide mass balance. However, if the surface and volume signals can be de-convoluted, CryoSat-2 would provide the most comprehensive record of the state of the Greenland Ice Sheet in the present warming climate. Since Ku-band radar altimeters are sensitive to snow characteristics, CryoSat-2 could provide information about firn densification, a crucial parameter for the conversion from volume change to mass balance. Changes in the firn densification is usually modelled, and independent observations are needed to validate this product.
To capture the many different regimes of firn conditions on the Greenland Ice Sheet, we use data from the French/Indian satellite SARAL. The altimeter ALtiKa, onboard SARAL, operates in Ka-band frequency that reduces surface penetration comparing to Ku-band altimeters. Initial studies have shown trend difference in the derived 3-year elevation changes for the two satellites. Understanding the differences in detail is the key to utilizing the full potential of CryoSat-2 data to both provide detailed surface elevation changes and the climate dependent changes in the firn. We focus our investigation on the difference in 3-year trend and time series of the elevation change over selected areas of the Greenland ice sheet. The interpretation of the differences will be supported by firn modeling, which previously has been used to derive Greenland mass balance from ICESat data. This firn model is further developed to also provide conceptual knowledge of possible penetration depths of the CryoSat-2 Ku-band altimeter.
Finally, this 3-year record of coinciding measurements from the two satellites may shed light on the needs for future multi-sensor satellite missions, to assess the state of the Greenland Ice Sheet and eventually to improve the elevation change products of the ongoing ESA Greenland Ice Sheet CCI project.
Extending Antarctic Ice Shelf Height Change Time Series using Cryosat-2
1Scripps Institution of Oceanography, La Jolla, California; 2Earth & Space Research, Oregon, USA
Continuous time series of surface height observations derived from satellite radar altimetry over the Antarctic ice shelves show significant and complex patterns of interannual variability. Such changes, which have been shown to correlate with climatic drivers (such as the El Niño-Southern Oscillation), can potentially impact the dynamics of the grounded ice sheet behind the floating ice shelves. It is, therefore, vital to continue the currently available 18-year observational record (1994-2012) from the ERS 1/2 and Envisat missions through the CryoSat-2 (CS-2) period (2010-present). However, due to the differences in instrumentation and orbit configuration between CS-2 (carrying a Delay/Doppler altimeter) and Envisat (which carried a conventional altimeter), the same processing techniques are not optimal for deriving height changes from CS-2 measurements. Here, we evaluate the potential of several processing methods, and present time series for various Antarctic ice shelves. We use the ~2-year overlap period between Envisat and CS-2 records to validate the independently derived height changes. We also compare the measured elevations with lidar observations from Operation IceBridge (2009-present). We present a case study on the Larsen-C ice shelf, which shows significant variability over decadal timescales up to the present time, to demonstrate the importance of continuous and long-term observations.
Satellite Altimetry of Greenland and Antarctic Ice Sheets: 40 Years of Advances and Challenges
NASA Goddard SFC, United States of America
Following the early suggestion of using satellite radar altimetry for mapping ice-sheet elevations (Robin, 1969) and for measuring elevation changes to determine ice-sheet mass balance (Zwally, 1975), both of these goals have been achieved using satellite radar and laser altimeter measurements (e.g. Shepherd et al., 2012 and Zwally et al., 2015). Measurements of ice-sheet elevations to ±65 began with the ocean-radar altimeters first on GEOS-3 in 1975 (Brooks et al., 1987) and to ±72on SeaSat in 1978 (Zwally et al., 1983) and GeoSat in 1985 (Zwally et al., 1989). The effect of off-nadir returns within the radar beam-limited footprints on causing slope-induced elevation errors (Robin, 1969) was shown in the measurements (Brooks et al., 1987), followed by approximate corrections made to SeaSat data (Brenner et al., 1983). Range errors in the altimeter’s automatic range-tracking algorithm were first corrected for SeaSat data with a waveform-fitting procedure called retracking (Martin et al., 1983), which was later applied more universally to ocean and ice measurements. The radar altimeter on ERS-1 and ERS-2 was the first with improvements for ice measurements including more adaptive range tracking and an ice mode with wider range gates. Some of the first papers to describe elevation change measurements with estimates of ice sheet mass balance included Zwally, 1989 using SeaSat and GeoSat data and Wingham et al., 1998; Zwally et al., 2005; and Davis et al., 2005 using ERS-1 and ERS-2 data. Wingham et al., 1998 was the first to apply a backscatter-based correction for seasonal and interannual variations in the penetration depth of the radar signal into the ice-sheet firn. The depth-penetration correction was further developed (Zwally et al., 2005; Yi et al., 2011) and its success is demonstrated by the close agreement of dH/dt over Lake Vostok in East Antarctic with that from ICESat laser altimetry (2.02 vs 2.03 cm/yr) (Zwally et al., 2015). ICESat (Zwally et al., 2002) with a laser altimeter and CryoSat-2 (Wingham et al., 2006) with an advanced multi-mode radar altimeter were the first satellites specifically designed for ice measurements and flown in near-polar orbits. Remaining challenges include: 1) developing more successful penetration-depth corrections for EnviSat and CryoSat that account for the complications of the interactions of their linear-polarized radar signals (oriented cross-track to the satellites) and the directions of the surface slope (e.g. Remy et al., 2012) and 2) developing long-lifetime laser-altimeter missions.