Monitoring the hydrological activities of subglacial lakes is critical to understanding the subglacial hydrological system and evaluating the internal mass changes of the Antarctic ice sheet. Drainage or filling events of active lakes lead to elevation changes in the ice surface. These changes can be observed by satellite altimetry, but the monitoring must be conducted continuously since the water movements in active subglacial lakes may occur frequently. We used CryoSat-2 Baseline-D and ICESat-2 data from 2010 to 2020 to obtain the time series of the ice surface elevation changes for 17 active lakes. We also evaluated the uncertainty of the time series derived from the CryoSat-2 data by cross-validation. The mean and RMS of the biases between the CryoSat-2-based and ICESat-2-based time series are generally less than 0.3 m and 1.0 m, respectively. However, the mean and RMS are greater over the lakes with rough ice surfaces, such as Whillans6, KT1, Mac3, and Slessor23. The drainage and filling events continue exhibiting in the extended period and the hydrological activities of SLW, L12, Whillans6, L78, and Mac1 occurred periodically. Furthermore, we inferred the hydrological connections between the lakes combining simulated water pathways.
The identification and activity monitoring of Antarctic subglacial lakes contribute to the analysis of the mass balance and sea level change. Remote sensing techniques, especially satellite altimetry, are ideal ways for identifying active subglacial lakes and indirectly monitoring their hydrological activity. Based on ICESat-2 altimetry data, we propose two methods to generate the rates of ice surface elevation changes for identifying active subglacial lakes in Antarctica. We identify 15 new active subglacial lakes and provide denser and more precise outline detail for known subglacial lakes in the Mercer and Whillans Ice Streams and Slessor Glacier regions. Then, we use ICESat-2 and earlier CryoSat-2 data to get time series of ice surface elevation changes for subglacial lakes to further understand their hydrological processes. Results show that subglacial lakes have exhibited periodic activity and at least two fill–drain cycles in the Mercer and Whillans Ice Streams region. Owing to data limitations, the time series shows less periodic lake activity in the Slessor Glacier region. However, subglacial water movements are similarly active and have various activity patterns.
Any unmodelled or mismodelled subdaily signals left in the model may not only affect the instantaneous site positions and the associated estimates, but also propagate into spurious seasonal signals, contaminating the daily coordinate time series. To demonstrate how subdaily ‘error’ in the modelling of the solid Earth-tide affects the estimates of tropospheric zenith total delay (ZTD) and how it propagates into long-period signal in the daily GPS time series, we analyse GPS observations collected between 2009 and 2013 for 13 sites in the coastal regions of Antarctica using the GAMIT/GLOBK 10.6 software. We find that ZTD differenced time series, with amplitude at 2 mm level, have inverse correlation with the input K1 correction, and the corresponding admittances range from 6% to 14%; Propagated spurious annual signals are evident in the vertical component of coordinate differenced time series, with amplitudes at the mm level and admittances of around 2–11%.
1 January 2026 in IEEE Transactions on Geoscience and Remote Sensing