This study investigates the cloud processing of hydroperoxymethyl thioformate (HPMTF) in the marine boundary layer (MBL) using six-week measurements of cloud fraction and marine sulfur species, along with an observational constrained chemical box model. HPMTF is an intermediate oxidation product of dimethyl sulfide (DMS), the major sulfur species emitted from the ocean. By examining the loss of HPMTF to clouds, this study can support a more accurate quantification of carbonyl sulfide (OCS) and sulfur dioxide (SO2) formation and their contributions to new particle formation, and also improve parameterizations in global chemical transport models.  

The manuscript is well written and accounts for a variety of sources that could cause uncertainness. However, there are two major concerns.

1. The measurements were conducted below a height of 10 m above ground level. A major concern is how well this can represent conditions near cloud levels. The manuscript indicates that this study is more relevant for well-mixed MBL conditions. Given the short lifetime of HPMTF, it would be helpful to perform a careful comparison of its lifetime and that of DMS with the MBL turnover time to understand the vertical profiles of these key sulfur species throughout the day. Specifically, the observation that the [DMS]/[HPMTF] ratio was low in cloud-free conditions but significantly higher below the cloud deck should be supported by an illustration of the vertical mixing dynamics of both DMS and HPMTF. This would help determine whether the observed changes are due to cloud processing of HPMTF or are instead a result of vertical mixing and dilution effects.
2. Another major concern is the method of calculating the cloud loss of HPMTF by comparing modeled results with observations. First, diurnally averaged measurements of meteorological factors and chemical species are used to constrain box modeling, which could lead to a significant bias, especially when comparing with measured HPMTF under varying atmospheric conditions. Although the study includes a simulation with time-varying temperatures, other key meteorological and chemical factors also vary and can significantly impact HPMTF concentrations. Second, as shown in Figure S1, the modeled results exceed the calibrated measurements of HPMTF, which indicates the calculation of differences will generate negative values to represent cloud loss thus doesn’t make sense. The manuscript attributes this discrepancy to a potential 60% underestimation of measured HPMTF, which is not convincing and warrants further discussion. Also, it may be beneficial to incorporate a cloud term directly into the model box to facilitate a more accurate comparison and a clearer insight into the cloud processing of HPMTF.