In this study the authors developed an approach for estimating frozen ground surface temperature underlying snow in Arctic tundra environments. Their approach is based on the inversion of a relatively simple microwave emission model (MEM), tailored to arctic tundra winter conditions, and driven by SMOS L-band brightness temperature observations. Differing MEM types are applied; the first assuming homogeneous surface conditions and the second applying a fractional water cover (FW) correction for snow and ice covered water bodies to reduce associated bias on ground temperature retrievals. The retrievals were derived and evaluated against global reanalysis data (ERA-5) and in situ temperature measurements across 21 Arctic tundra reference sites in northern Alaska permafrost locations from 2011-2020. The results reveal the important influence of FW on SMOS ground temperature retrievals in the Arctic and demonstrate an effective method for temperature retrievals spanning a range of FW conditions characteristic of the Arctic tundra winter environment.

Overall, the paper is well written with clearly explained methods and interesting and well justified results, and conclusions. The figures and table summaries are clearly depicted and provide adequate support to the major points of the paper. The study also shows the potential for effective monitoring of Arctic winter ground temperatures beneath the tundra snowpack from SMOS and other low frequency satellite microwave radiometers using a relatively simple MEM with minimal ancillary data requirements. Significant broader impacts from this study include the potential for satellite based monitoring of winter soil temperatures in the Arctic, which are generally poorly characterized in global land models and from available sparse monitoring sites. Winter ground temperatures also have strong science value due to amplified Arctic warming trends and the strong association of ground temperatures on permafrost and soil carbon stability. I consider the paper suitable for publication in it’s present form pending consideration of the following minor revisions.

Section 1: Include a concise statement of the broader science objective or goal of the study at the end of the Introduction section.

Section 2.4: Given the enhanced influence of topography on microclimate heterogeneity at high latitudes, consider adding the terrain elevation heterogeneity surrounding site location grid as an additional factor that may help explain differences between the relatively coarse resolution satellite and reanalysis observations, and the in situ site measurements.

Section 4.2.2:  Do the sites with higher bias share similar features that may help account for the larger temperature error? E.g., sites located along coastlines near open ocean or in complex foothills topography may be expected to have larger apparent bias than relatively flat inland locations.  

p.2, Ln 47: “microwave microwave” should be “microwave”.

p.19, Ln 351: “different for” should be “different from”.