Overall Review

The article provides a comparison between 10 different wet-snow dielectric mixing models and how the choice can influence the retrievals of liquid water content at L-band frequency. The study was conducted for Greenland’s percolation zone, utilizing SMAP (rSIR) brightness temperature and two in-situ forced surface-energy mass balance models to assess robustness. The combination of models and satellite observations is unique and novel, and answers an important question about the choice of wet-snow dielectric models for the estimation of liquid water amount. The manuscript is generally well organized and richly referenced, but a careful read reveals several minor errors that should be addressed before publication.

Major Strengths & Novelty

1. First side-by-side comprehensive comparison of ten-dielectric formulations (Debye-like, power-law, empirical, etc.) explicitly focused on the retrieval of LWC.
2. Link to operational satellite – SMAP rSIR Tbs
3. Quantitative assessments against the surface energy models.

Specific Questions and Technical Errors

1. Abstract – “Sihvola power-law mixing model showed an overall better performance than the other models for the 2023 melt season” – consider including metrics.
2. Which SMAP product was used (rSIR) can be mentioned in the introduction, in the last paragraph.
3. Duplicate equation number (for eq. 8 – mentioned at L137 and L155), and then subsequent equation numbers should be changed.
4. Typo in L273 Ks<<Ks, instead of Ks<<Ka.
5. Methods – The Hallikainen model was derived at 3-37 GHz; authors can justify its usage/extrapolation to 1.4GHz
6. Typo – Table 1: Key Parameters – “Depolarizaion” should be “Depolarization”.
7. Sihvola, misspelled at L99, L123, L166 as Sihivola.
8. Eq 20 refers to both e-folding depth and attenuation coefficient.
9. Hallikainen et. al. 1984 (L344) is not mentioned in the bibliography; are the authors referring to Hallikainen et. al. 1986? If so, the date should be changed.
10. I suggest making the zoomed-in version on the right in Fig. 2
11. L427 “The Colbeck model provides the lowest estimates for the entire LWA range,” referencing Fig. 4. However, in Fig. 4f, the Colbeck model appears to provide a higher estimate than Hallikainen. (A zoomed-in inset for Figs. 4, and 5,6 would be helpful).
12. Table 3 GEMB column is missing.
13. Line 550, “All three methods…”, is it referring to Fig.9?
14. Can include a plot of observed and simulated tb, to check the loss.
15. Line 885 (+more) Miller, J.Z. has year 2020a, but 2020b is missing, I see that at Line 897 Miller, O., et. al, has the year 2020b. but no corresponding 2020a.

Please see attachment.