Observing, detecting and downscaling the winter coastal polynya characteristics over the Ross Sea sector of Antarctica

A polynya is an area of open water or reduced concentration of sea ice surrounded by either concentrated sea ice or land ice. Polynyas, with their characteristic kilometers, hold significant importance both locally and globally. They are persistent and recurring areas of open water/thin ice, spanning tens to tens of thousands of square kilometers. They are often seen as sites of intense ocean–atmosphere heat exchange and as ice production factories. Hence, polynyas are important small-scale features to study scale interactions of climate. Given their significance, it is crucial to quantify the satellite-derived polynya information accurately as well as understanding their modulating factors on different scales. This is the motive of the study.

Polynyas in their early evolution phase are generally narrow and occur at scales likely too fine to be detected by widely used passive microwave (PMW) radiometric sensors. In the thesis, polynya areal information over the Western Ross Sea is derived at 40 m spatial resolution from high-resolution synthetic aperture radar (SAR) Sentinel-1. Two automated machine-learning algorithms, supervised (a rule-based approach) and unsupervised (a combination of texture analysis with k-means clustering), are utilized to accurately identify the polynya areas. The work highlights discrepancies from PMW sensors (NSIDC and AMSR2) and compares SAR-based algorithms (rule-based and texture-based) with manually delineated polynya areas obtained through Sentinel-1. Analysis using PMW sensors revealed that NSIDC overestimates larger polynyas and underestimates smaller polynyas compared to AMSR2. The Chapter presents a precise identification of polynya presence and area using Sentinel-1 SAR observations, especially in clear cases and cases when PMW data miscalculates the polynya’s presence. Of the SAR-based algorithms, the rule-based approach was more accurate than the texture-based approach at identifying clear polynyas when validated against manually delineated regions.

Limitations and advantages of the three satellite products on different spatio-temporal scales in retrieving of polynya area are highlighted. Although Sentinel-1 provided accurate information of all, the period of temporal coverage and frequency further constrained the utility of this dataset. Therefore, as a suitable intermediary between NSIDC and Sentinel-1, AMRE/2 dataset is used for further examination of large-scale circulation in modulating polynya activity.

From the AMSRE/2 SIC datasets, coastal polynya variability in Ross Sea (RS) over winter is identified in an empirical orthogonal function (EOF) analysis of high-resolution PMW estimates. Observed patterns of polynya variations are related to surface regional extreme-winds from the ERA5 reanalysis to understand local atmospheric effects and to 500hPa geopotential heights to capture relationships with the large-scale hemispheric flow. The extra-tropical atmospheric circulation is thus strongly linked to coastal SIC variability and to polynya activity via its impact on Extreme Winds. Each of the three polynyas in the region (Ross Sea Polynya - RSP, Terra Nova Bay Polynya - TNBP and McMurdo Sound Polynya - MCMP) respond differently to the position of large-scale circulation patterns. There is evidence of increases/decreases in polynya area associated with specific patterns, particularly the Amundsen Sea Low (ASL) and its influence on Ross ice shelf air stream (RAS) and barrier wind regimes. Here the shift in longitudinal positions of the ASL is observed to alter different local wind regimes of the RS. These findings provide a causal pathway between large-scale modes from extra tropics and their modulation of small-scale sea-ice processes within the various polynya of the RS.

The subsequent study aimed to further investigate the source of the shifts in the longitudinal position of ASL, whether it is influenced by tropical modes of variability. Extra tropical response to ENSO related convection anomalies provided better understanding on the high-latitude pressure gradient shifts. A strong ocean-atmospheric coupling from tropical diabatic heating is observed to influence the extra-tropical to high-latitude wind patterns, via Rossby wave trains (RSW, observed as alternating high-low patterns) eventually affecting the longitudinal positions of ASL. This shift in the position of ASL modulates the local wind regimes of the RS like RAS and barrier wind patterns.

Convection anomalies from different ocean basins, Maritime Continent (MC) and Indian ocean modulate atmospheric variability, with varying sizes of pressure gradients (low, ASL) in West Antarctica. For the ASL (deep and intense) to exhibit broader coverage from the Bellingshausen Sea to RS, convection anomalies ought to be triggered in MC eventually propagating poleward (via RSW through Australia and south-east of New Zealand). Such anomalies displayed consistent behavior in all three polynya zones of the RS, dominant over RSP. When anomalies are Indian ocean/Western North Pacific related, westward shift of ASL (strengthened over north of RS) with relatively smaller size than in previous case, via RSW propagating from South of Australia is observed. Such anomalies may be related to dipole wind anomalies (resulting from ZW3 and ENSO in different conditions) over West Antarctica, eventually resulting in opposite polynya behavior in the regions western RS/MCMP and eastern RS/TNBP. Lastly, for the western RS (WRS) only polynya activity triggered by RAS and barrier wind regimes locally, the ASL is observed further east from its climatological position, towards Peninsula.

Therefore, the findings from this thesis underscores the sensitivity of winter coastal polynyas (that occur at small-scale) in the RS to hemispheric-scale teleconnections. In future, with change in ENSO conditions, in terms of frequency and intensity can affect polynyas of RS. This thesis highlights active polynya activity derived as reductions of SIC associated with a cold-phase of ENSO. Conversely, with an intense or a warm-phase of ENSO, the polynya activity in the RS maybe weakened. This further affects the ocean-circulation and sea-ice production, as well as sea-ice variability in other parts of Antarctica. However, relative amounts of contribution need further research. The study thus emphasizes polynya relationship with shift in atmospheric convection and circulation in and outside tropics.