Spatiotemporal analysis of arctic sea ice area variations using multisource satellite remote sensing data

This study systematically investigates Arctic sea ice melting patterns and their climatic correlations with sea surface temperature (SST) and surface air temperature (SAT) by integrating multi-source remote sensing data fusion techniques with spatiotemporal data analysis, based on long-term (1980–2022) remote sensing imagery and reanalysis datasets. The main conclusions are as follows: (1) The Arctic sea ice area (SIA) continues to decrease at an average annual rate of 2.83×104 km2. After 2004, it entered a phase characterized by persistent negative anomalies; the retreat rate during the melting season (June to September) (−5.5×104 km2·a-1) is significantly higher than that during the freezing season, with September showing the most pronounced changes; (2) There are notable regional differences in SIA retreat rates; specifically, the central Arctic Ocean exhibits the slowest SIA retreat rate while the Kara Sea and Barents Sea show the fastest rates; (3) A strong negative correlation exists between SIA and SST (R = -0.944), as well as between SIA and SAT (R = -0.768). SST has an immediate impact on ice melt, whereas SAT changes show a 1–2-month lag effect, a pattern that can serve as critical input for improving climate model parameters. These results offer technical references for optimizing remote sensing-based sea ice monitoring algorithms, improving the accuracy of climate simulation models, and facilitating data-driven decision-making in Arctic climate research, with significant relevance to advancements in spatiotemporal data analytics and environmental modeling.