General comments

The work aims to estimate the frequency of landslides in the US using available landslide inventories. Landslides are those triggered by earthquakes and precipitation.

The manuscript overall is well written and the objectives of the work are clear. The approach is also promising. Nevertheless, the part describing the models and techniques used (section 2.4) is somewhat cryptic and not easy to read for those with non-advanced skills on statistical distributions, such as negative binomial, applied to overdispersed data. I strongly suggest expanding the part on the models used, giving a more accessible explanation related to the purpose of the work.

Figures (especially 1, 2 and 6) are too dense of information and not easily readable. As a consequence, figure captions are also too long. Please consider moving some figures to a Supplementary Material section.

In addition, as a general rule, figures must be cited in advance in the text. For multiple maps/graphs in the same figure, they must also be cited in the text in the order given.

Specific comments

See attached PDF.

The study addresses a significant gap in current national-scale landslide susceptibility research by prioritizing the frequency of landslides, which is essential for effective hazard and risk management planning.  The manuscript is well-structured and states its objectives clearly, emphasizing the importance of informative landslide hazard estimates for mitigation planning and risk reduction at a national level. The authors employ Bayesian negative binomial regression to model landslide frequencies at the county level using predictor variables that comprise susceptible area, earthquake potential, precipitation frequency, and ecological region. Authors highlight the extreme diversity of landslide susceptibility across the United States, with frequencies described as ranging from a few incidents in some areas to high concentrations in others. There are some minor issues to be modified in the manuscript before publication. The methodology should be presented more concisely and provide the reader with an explanation for further advancing the proposed methods. The proposed manuscript is a significant addition to landslide hazard assessment in that it provides more complete and spatially resolved frequency data than have been feasible with national-scale studies. It is a publication standard, providing valuable information that can be utilized to direct targeted risk reduction and mitigation efforts across the United States.