Multi-scale observations of the co-evolution of sea ice thermophysical properties and microwave brightness temperatures during the summer melt period in Hudson Bay

• Published in: Elementa: Science of the Anthropocene
• Main Authors: Madison L. Harasyn, Dustin Isleifson, Wayne Chan, David G. Barber
• Format: Article in Journal/Newspaper
• Language: English
• Published: BioOne 2020
• Subjects: sea ice; passive microwave remote sensing; unmanned aerial vehicle; brightness temperature; melt period; hudson bay; Environmental sciences; GE1-350; Hudson Bay; Hudson; Sea ice
• Online Access: https://doi.org/10.1525/elementa.412; https://doaj.org/article/bfbf9be56bf544b894c804509b640a49

Monitoring the trend of sea ice breakup and formation in Hudson Bay is vital for maritime operations, such as local hunting or shipping, particularly in response to the lengthening of the ice-free period in the Bay driven by climate change. Satellite passive microwave sea ice concentration products are commonly used for large-scale sea ice monitoring and predictive modelling; however, these product algorithms are known to underperform during the summer melt period due to the changes in sea ice thermophysical properties. This study investigates the evolution of 'in situ' and satellite-retrieved brightness temperature (TB) throughout the melt season using a combination of 'in situ' passive microwave measurements, thermophysical sampling, unmanned aerial vehicle (UAV) surveys, and satellite-retrieved TB. 'In situ' data revealed a strong positive correlation between the presence of liquid water in the snow matrix and 'in situ' TB in the 37 and 89 GHz frequencies. When considering TB ratios utilized by popular sea ice concentration algorithms (e.g., NASA Team 2), liquid water presence in the snow matrix was shown to increase the 'in situ' TB gradient ratio of 37/19V. 'In situ' gradient ratios of 89/19V and 89/19H were shown to correlate positively with UAV-derived melt pond coverage across the ice surface. Multi-scale comparison between 'in situ' TB measurements and satellite-retrieved TB (by Advanced Microwave Scanning Radiometer 2) showed a distinct pattern of passive microwave TB signature at different stages of melt, confirmed by data from 'in situ' thermophysical measurements. This pattern allowed for both 'in situ' and satellite-retrieved TB to be partitioned into three discrete stages of sea ice melt: late spring, early melt and advanced melt. The results of this study thus advance the goal of achieving more accurate modeled predictions of the sea ice cover during the critical navigation and breakup period in Hudson Bay.