Evaluation of the AMSR2 Ice Extent at the Arctic Sea Ice Edge Using an SAR-Based Ice Extent Product

Passive microwave (PM) and synthetic aperture radar (SAR) observations are essential tools for providing long time series of sea-ice cover information, including concentration (SIC) extent (SIE). Large uncertainties have been revealed in PM SIC/SIE products the marginal ice zone (MIZ) during melting season, where fusion with SAR data could be effective improving accuracy due to its high spatial resolution ability preserve detailed distributions. A comprehensive comparison is needed better fusion. This study evaluates one SIE products, advanced scanning radiometer 2 (AMSR2) product retrieved arctic radiation turbulence interaction (ARTIST) sea (ASI) algorithm, using a neural-network-based throughout year 2019. First, we present key results three assessment parameters, overall (OA), error-of-ice (EI), edge location distance (LD), then estimate optimal SIC segmentation threshold AMSR2 ASI SIE. Based on OA EI, annual average 12.24%, winter 9.25%, summer 16.43% obtained regarded as by excluding cases large uncertainties. Second, found perform identifying thin melt ponds, while NN has detection brash frazil ice. We introduce parameter fragmentation fraction (IFF) analyze primary impact factors behind different performances. It that ratio LD IFF distinguish aforementioned conditions, thus hints combining complementary advantages two