This study focuses on improving SO2 top-down emission inventory over snow-covered surface. The snow-covered Vertical Column Densities (VCDs) are involved in SO2 emission estimation process for the first time. By creating new AMFs and then VCDs under snow-covered conditions, SO2 emissions of snow-impacted sources can be reevaluated, as more measurements in winter season become available. As a major anthropogenic source of SO2, emissions from power plants in high-latitude regions have shown significant improvement. I recommend publication after some minor corrections mentioned below.

At the end of section 2.3, I suggest to place the extension of SO2 observations by including snow-covered pixels in a broader context by also referring to the NO2 product of OMI and TROPOMI, which is already including snow-covered pixels based on similar principles (see the TROPOMI ATBD of NO2, and Van der A et al., 2020, https://doi.org/10.1038/s41612-020-0119-z)

Line 66: “…and volcanic SO2 is is used …”. An “is” too many.

Line 141: An 10% empirical correction is applied to the OMI VCDs. Is this a positive (+10%) or negative correction (-10%)?

Line 246: “…norther Russia…” should be “… northern Russia…”

Line 457: “In summary, is it worth…” should be “In summary, it is worth…”

The manuscript " Estimation of anthropogenic and volcanic SO2 emissions from satellite data in the presence of snow/ice on the ground" from Fioletov et al. focusses on improving existing SO2 emission inventories based on satellite data by explicitly including measurements over snow, which are usually excluded in these inventories. These inventories therefore likely underestimate emissions of sources at high latitudes, which are covered by snow for part of a year.

The authors have generated new site-specific AMFs for snow conditions and evaluated the impact on the SO2 inventory, showing a significant improvement.

I recommend publication after some minor corrections:

Line 64: " ...for assessment the efficiency... " - it should be " ...for assessment OF the efficiency..."

Line 66: "... volcanic SO2 is is used ..." - it should be "... volcanic SO2 is used..."

Section 2.3, Line 172ff: I suggest to move your discussion about the SO2 lifetime from Sect. 3.3 line 404ff here, since it gives a very good explanation why you use the same decay time for snow and snow-free conditions and it fits better in this section...

Line 193: What is the reason for using CRF<0.3 for cloud&snow-free conditions? 0.3 is really high and you will certainly use pixels which are partly covered by clouds, especially for the big pixel size of OMI... What happens if you use e.g. CRF<0.1?

Line 202ff: I a msising information here which Snow/Ice information for TROPOMI data you use. You describe it for OMI, but not for TROPOMI

Line 202ff: TROPOMI data also contains VIIRS cloud data, that you could use and that provides more accurate results on CF...