ABSTRACT Field measurements within ice covers during the melting period are scarce, often constrained by safety concerns leading to uncertainty in model parameterizations of ice melt. To address these limitations, a Floating Research Station (FRS) was constructed in a small subarctic lake near Yellowknife, Canada to monitor ice processes year‐round. Using the FRS, the objective of this study was to delineate key processes influencing ice melt and decay through evaluating heat budget components over three melt seasons: April 1–May 31 of 2023, 2024, and 2025. Our results show that the decay process was both thermally and mechanically driven, with mechanical decay from ice collapse occurring at internal ice porosities of 0.31–0.35 and accounting for 24%–48% of total ice loss. Thermal melt at the surface (0.6–2.1 cm d −1 ) was driven by surface heat absorption (13–160 W m −2 ) and losses via net longwave fluxes (−95.2 to −6.6 W m −2 ). Bottom melt (0.1–0.6 cm d −1 ) was caused by modest mean daily water‐to‐ice heat fluxes (< 15 W m −2 ). Melt season lengths varied between 35 and 49 days, and break‐up dates up to 16 days, and were dependent on air temperatures, albedo, and net insolation. Results from this study can be used to improve and constrain melt period parameterizations in ice models while providing needed validation and calibration data in northern, high‐latitude environments.
Rafat et al. (Sun,) studied this question.