Seasonal and interannual variability of the landfast ice mass balance between 2009 and 2018 in Prydz Bay, East Antarctica

Landfast ice (LFI) plays a crucial role for both the climate and the ecosystem of the Antarctic coastal regions. We investigate the snow and LFI mass balance in Prydz Bay using observations from 11 sea ice mass balance buoys (IMBs). The buoys were deployed offshore from the Chinese Zhongshan Station (ZS) and Australian Davis Station (DS), with the measurements covering the ice seasons of 2009-2010, 2013-2016, and 2018. The observed annual maximum ice thickness and snow depth were 1.59 +/- 0.17 and 0.11-0.76 m off ZS and 1.64 +/- 0.08 and 0.11-0.38 m off DS, respectively. Early in the ice growth season (May-September), the LFI basal growth rate near DS (0.6 +/- 0.2 cmd(-1)) was larger than that around ZS (0.5 & PLUSMN; 0.2 cmd(-1)). This is attributed to cooler air temperature (AT) and lower oceanic heat flux at that time in the DS region. Air temperature anomalies were more important in regulating the LFI growth rate at that time because of thinner sea ice having a weaker thermal inertia relative to thick ice in later seasons. Interannual and local spatial variabilities for the seasonality of LFI mass balance identified at ZS are larger than at DS due to local differences in topography and katabatic wind regime. Snow ice contributed up to27 % of the LFI total ice thickness at the offshore site close to ground icebergs off ZS because of the substantial snow accumulation. Offshore from ZS, the supercooled water was observed at the sites close to the Dalk Glacier from July to October, which reduced the oceanic heat flux and promoted the LFI growth. During late austral spring and summer, the increased oceanic heat flux led to a reduction of LFI growth at all investigated sites. The variability of LFI properties across the study domain prevailed at interannual timescales, over any trend during the recent decades. Based on the results derived from this study, we argue that an increased understanding of snow (on LFI) processes, local atmospheric and oceanic conditions, as well as coastal morphology and bathymetry, are required to improve the Antarctic LFI modeling.