|Paper title||Investigating the Applications of Singular Spectrum Analysis for Satellite Altimetry Derived Surface Elevation Change Time Series over Ice Sheets|
|Form of presentation||Poster|
Ice sheets store vast amounts of frozen water, capable of raising sea levels by over 60 m if fully melted. Meltwater runoff can also affect a range of glaciological and climatic processes including ocean driven melting, fjord dynamics and large-scale ocean circulation. As global temperatures continue to increase, accurately estimating the mass balance of ice sheets is vital to understanding contemporary and future sea-level changes.
Satellite altimetry measurements can provide us with continental-scale observations of surface elevation change (SEC). Once combined with firn densification models, this data record allows us to make estimates of ice mass losses to the oceans. Time series of surface elevation change, produced using altimetric data, are commonly generated in a simplistic manner, such as through averaging measurements in time. Here, we will explore the potential of employing more advanced statistical methods of time series analysis to improve the generation and interpretation of time series. One of these techniques is singular spectrum analysis (SSA), a model-free spectral estimation method for decomposing time series into the sum of different signal components. This method allows us to separate the unstructured residual components from the long-term trend and dominant oscillatory modes, such as seasonal cycles.
In this presentation, we will present two case studies that investigate how SSA can be applied to surface elevation change time series derived from satellite altimetry, which have formed part of methodological development undertaken within ESA’s Polar+ SMB feasibility study. (1) SSA shall be employed to remove noise from decade long CryoSat-2 radar altimetry SEC time series for areas of the Greenland Ice Sheet to improve their quality. The smoothed time series shall be validated against in situ and airborne datasets. (2) We will apply SSA to the long-term altimetry record for Antarctica to identify dominant periodicities longer than 2 years. This will aid our interpretation and allow us to investigate links to ocean and atmospheric circulations.