These data underpin the Fraser et al., 2026 publication in Nature Communications: "Revealing the Antarctic marginal ice zone: a decade-long wave-in-ice climatology". They are presented as one .csv file per year, each containing around 8500-9000 rows. Each row corresponds to the data retrieved from a SARAL half-orbit (pole to pole). Each row presents the along-track summary fields used to estimate the wave-affected marginal ice zone width. The marginal ice zone width is estimated from the difference in latitude between the ice edge and the latitude of the "inner MIZ limit". The inner MIZ limit is estimated from AltiKa Ka-band radar altimeter waveforms: where the waveform of returned power becomes suffiently peaky and "steep", we interpret this as being the point where wave passage becomes undetectable.

Antarctic Landfast sea ice (fast ice) is important climatologically, biologically and for logistics for short time-scale anomalies. Until recently, there hasn’t been an accurate, high-resolution fast ice extent dataset which can support an analysis on drivers of fast ice and most studies only investigate fast ice on limited regions of Antarctica in a limited time scale. There is a need to extend the spatial and temporal studying coverage to provide detailed information on the Antarctic coast over a longer period. This is the first detailed analysis to identify and quantify correlation between the environmental anomaly and fast ice anomaly mainly in the east Antarctic coast. By examining regional/local fast ice extent in in east Antarctic coast in the context of the broader and/or remote-teleconnected atmospheric circulation/properties using spatial correlation techniques, a strong correlation between NINO3 region and Lützow-Holm Bay fast ice and similar and significant correlation of regional scale factors from Lützow-Holm Bay to Mawson Coast mainly are found. The results of this thesis suggest that the pack ice, atmospheric factors and oceanic factors are important for interpreting fast ice anomalies. To identify and quantify correlation between the pack ice, temperature at 2m, wind at 10m, snow fall anomaly, sea surface temperature anomaly, ocean heat content anomaly and fast ice anomaly, backward multiple linear regression is conducted to demonstrate some predictive fast ice driver information by quantifying the correlation between different drivers and fast ice anomaly. The multiple linear regression also suggests that oceanic influences including pack ice are generally more important than atmospheric influences. Future experiments could be conducted to interpret fast ice anomalies in the context of the ocean mainly.