Comment on acp-2021-592

The authors focus on the declines and peaks in NO2 pollution during the multiple lockdown phases in the New York metropolitan area and disentangle the contribution of anthropogenic emissions sources and role of meteorology. In general, I find this manuscript to be of interest for publication and appropriate for Atmospheric Chemistry and Physics. I have one main concern and several minor suggestions for improvement listed below that should be considered by the authors before publication.

The authors focus on the declines and peaks in NO2 pollution during the multiple lockdown phases in the New York metropolitan area and disentangle the contribution of anthropogenic emissions sources and role of meteorology. In general, I find this manuscript to be of interest for publication and appropriate for Atmospheric Chemistry and Physics. I have one main concern and several minor suggestions for improvement listed below that should be considered by the authors before publication.
The authors use total column NO2 in this study and a few references are made to tropospheric column NO2 throughout the manuscript. The authors can be specific about whether the "column NO2" refers to "total column NO2" or "tropospheric column NO2". I have pointed out a few instances below. Typically, studies use tropospheric column NO2 to relate to changes in NOx emissions. How do the values of total column NO2 compare to tropospheric column NO2 for New York metropolitan area? The authors should consider discussing why total column NO2 is used and the possible impacts on the results.

Minor comments
Line 68 Population and area of New York metropolitan area?
Line 81-86 Do both these studies use total column NO2? Or tropospheric column NO2?
Line 85 "0.4° radius". Also mention in kms to compare to 100km radius in previous sentence.
Line 101 "4% yr -1 decrease …" is in tropospheric column NO2. How would the trend values be for total column NO2?
Line 106 "…highest national NO2 levels." The authors can give value of annual mean NO2 and compare to the recently updated WHO guidelines.
Line 117 Here and everywhere else, the authors mention high-frequency observations from Pandora but do not provide any value.
Line 131 Section heading can be changed to "Materials and Methods" as the subheadings also focus on the various datasets.
Line 135 Last assessed date for the URL.
Line 169 Tropospheric columns of?
Line 173 Filtering criteria such as TCNO2>0 can lead to a positive bias in the mean TCNO2. Is it possible to quantify how much data is removed because of this filtering criteria and if the positive bias is large?
Line 187 The authors use total column NO2, but the retrieval steps are also given for tropospheric vertical column. Line 266 URL or DOI for MTA data?
Line 297 "Variability in TCNO2 also decreased…" except New Brunswick?
Line 324-326 This statement starts suddenly and the value for NO2 changes need to be given before explaining why they are lower compared to NOx changes.
Line 339-344 The authors can consider giving a brief description of how these changes in NOx emissions were obtained (either here or in section 2).
Lines 484-492 Interesting result! Table 1 The table caption can include details of the time. "Present" may be replaced by "05/2021" Additional footnotes can be used on the column of "Temporal range of data" to mention data unavailability. For example, data from PSI#135 not available for Jan-Mar 2020 is mentioned in Figure 4 and should also be mentioned in table 1. The value of stdev for PSI#56,#69 for Sept-Nov reads as "0.24x". Check for typo. Lastly, are both the PSIs at Queens, New Brunswick and New Haven in operation for the time period stated and is the data from both of them used?