General comments

The manuscript provides an overview of new dataset that was deployed specifically as part of the SWOT Ca/Val, but has endured beyond this period. The datasets and instruments are well explained, as are the quality control parameters. One thing that is unclear is whether any data available in real time to the user? Or is all data only available after instrument recovery (i.e., post-processed from the memory card). 

Specific questions/issues

• Line 13: It would be good to define when this dataset will be ‘periodically’ available
• Line 92: Consider replacing ‘right shore’ with a cardinal direction – i.e., ‘eastern shoreline’
• Figure 1: It might be helpful in Figure 1 to include a rough outline of which section is seasonally covered in ice
• Line 150-165: It would be helpful to be a bit more consistent in your reporting of error. At lines 150-165 you being by reporting ‘expected’ errors less than Xcm (Xsigma); but then you mention +/- X cm at line 164. Are the errors from the latter sensors not able to be characterised in the same way?
• Line 250: This idea about truncation is a bit confusing – does this simply refer to areas that were not surveyed (i.e., did not have observations)? If so, then perhaps just clarify that other (comparable) sites outside of the study area may experience water level truncation as well?
• Line 260-264: define all depths for wave observations in text
• Line 275: is this the standard sampling plan for AZMP wave buoys? Or customised to this site? Regardless, it should be explained why this sampling plan is used (i.e., to target specific wave frequencies that dominate this region?)
• Line 305: it’s unclear how wave calculations are conducted over 30min when they appear to be collected in 10min bursts (with 5 min gaps). Please expand on this
• Line 311: is ‘provenance’ wave direction?
•  

Technical corrections

• Line 267: small typo – should be AZMP not AMZP
• Line 274: perhaps add the specific wave buoys you’re referring to here with these dimensions
• Line 276: small typo – ‘a245 ‘ ?

This paper provides a valuable and dense dataset of 18 water level measurements (barometrically-corrected), supplemented with 13 GNSS-Interferometric Reflectometry sensors, and 3 directional wave buoys for validation of SWOT’s nadir and wide-swath altimetry products. Location is the St. Lawrence Estuary and Saguenay Fjord in Quebec

This is a significant effort - deploying and maintaining the instruments for this dataset. Other data are mentioned such as - high-frequency radars deployed during SWOT Cal/Val period and the AirSWOT campaign conducted in August 2023 (Simard et al., 2023).

Why aren't the radar measurements or the from the AirSWOT campaign available?

The description of the processing is appreaciated as well as the multi-sensor intercomparisons.

However, for a paper providing a benchmark dataset for SWOT, it should include actual SWOT comparrisons with the measured data.

Particularly, the issue of geoid transformation is the most ambiguous, and the paper mentions "care must be taken" and points to the SWOT handbook. A user wanting to actually do the SWOT comparison has to reconstruct the chain themselves. For a benchmark dataset I find it as a real omission and all the workflow should be provided (codes included) to directly compare the measured data with SWOT.

This would make the dataset easy to apply and be used along with SWOT.

check also AMZP" vs AZMP inconsistency