This is a very nice paper presenting detailed observations of primary surface ocean sulfur species for the Peruvian (Humboldt) coastal upwelling region.  The paper is well written and the methods conform to expected standards in this field.  I have a few general comments below.  I find very few punctuation or other formatting issues.

General Comments / Suggestions:

line 17, 25-28 and elsewhere:  The conclusion that seawater DMS was ‘relatively low’ seems too vague and imprecise.  Better to just state the mean/range/variance of the observations and compare this to previous measurements.  I wouldn’t stress a comparison with Lana et al. 2011 too much (PMEL database is better).  As you have mentioned, there isn’t much data from the “Humboldt Current Coastal’ province, as defined in Lana et al., and their seasonal extrapolation for this province from limited data is just a best guess.  This is a region where we expect seawater DMS concentration to vary quite a bit, spatially and temporally (on seasonal, yearly and perhaps decadal timescales).  Combined with previous measurements, these data provide a better picture of seawater DMS in this region than the Lana et al. gridded product.  Since this study doesn't represent extremes in the ENSO cycle it would be interesting to know how well previous studies have sampled ENSO variability.  Can the authors discuss their study and previous measurements from the ENSO perspective a bit more?  Are we still far from a representative sample of DMS variability during ENSO extremes?  Should this be a focus of future studies?

line 19 and elsewhere:  the terminology 'flux density' is a bit odd and not typically used in the air-sea flux community.  It's potentially confused with usage like 'spectral density' for power spectra, etc.  Better to just say 'flux' or 'fluxes'

lines 129-131: Although it doesn't make much difference to the conclusions of your study, we should really discourage the use of transfer models like Nightingale 2000 for DMS flux estimates.  Numerous direct studies of DMS air sea flux have been conducted over the past decade or more and the relationship with wind speed (or friction velocity) is closer to linear, especially over the wind speed range of your study.  Transfer models based on highly insoluble tracers don't represent DMS transfer well (see Fig 3 in Bell et al. 2013).  Better to use a model that's actually validated with direct DMS flux observations.  The COARE gas transfer model has been used (e.g. Bell et. al. 2013, Blomquist et al. 2017) but for this paper a simple linear relationship is probably fine: e.g. Huebert et al. 2010, Fig 3 or Blomquist et al. 2017 Fig 5.

line 269: I'm confused by the reference to 'terrestrial DMS sources'.  What are these potential sources?  I'm not aware of any, especially in the arid coastal climate of Peru and Chile.  A more likely source of high atmospheric DMS variability would be hotspots close to shore, which might be implicated if trajectories in the marine boundary layer follow the coastline for some distance.

Minor notes:

line 15: don’t need a comma after ‘present’

References:

Bell et al., 2013, https://doi.org/10.5194/acp-13-11073-2013

Blomquist et al. 2017, https://doi.org/10.1002/2017JC013181

Huebert et al. 2010, https://doi.org/10.1029/2009GL041203

The manuscript “Dimethylated sulfur compounds in the Peruvian upwelling system” presents results from two different ship cruises in the Peruvian upwelling region focusing on observations of marine sulfur species. Both sea water and atmospheric DMS measurements are presented as well as DMSP and DMSO concentrations. I believe that the manuscript is very interesting and well written and I therefore recommend the paper for publication after the following minor comments have been addressed:

General comments:

Comparison of atmospheric DMS concentrations: In the manuscript, you note that there is a good correlation between the two employed techniques but that the results are not on a 1:1 line. I would recommend discussing a bit further for example mentioning that the PTR measurement is showing higher concentrations. Besides, while the PTR directly measures the DMS concentration in air there might be potential losses of DMS in the other technique. Was there a possibility to do a comparison in the lab of the two techniques? This could also help to find out if the instruments actually worked well. It would also help to know which E/N ratio was used for the PTR as that has an influence on the performance of the instrument and if it was calibrated before and after the measurements. Which m/z was used to track the DMS signal in the PTR?

Regarding the atmospheric DMS concentrations, I was wondering if you could compare your data to results from satellite data?

I was wondering about the roughness of the sea during the cruises. For concentrations of gas and particulate species in the marine atmosphere, the structure of the surface of the ocean is quite important. Figure 7 shows wind speed that as is mentioned in the manuscript is often correlated with DMS concentrations. What about wave breaking? Do you have measurements of wave height or where the waves started breaking? Was there wave breaking in the regions defined as “coastal stations”?

Specific comments:

Sometimes there is a strange thing with the font. For example on page 3, line 76: the symbols in “°S” are too close to each other, a similar issue occurs with the word “Niño” on page 3, line 91 and in later occurrences of the word.

Page 2, line 43: in the sentence starting with “Some studies..” there is only 1 reference. Either change to “The study by xx .. “ or add more references.

Page 9, line 269: you mention “terrestrial DMS sources” – did you mean terrestrial sulfate sources? Or what would terrestrial DMS sources be?

The manuscript investigated the variations of dimethylsulfide (DMS), dimethylsulfoniopropionate (DMSP), and dimethyl sulfoxide (DMSO) in the Peruvian upwelling system. This study will help us understand better the biogeochemical cycling of dimethylated sulfur compounds in the upwelling regions off Peru. The authors' study indicated that nutrient availability appeared to be the main driver of the observed variability in the surface DMS distributions in the coastal areas. However, there are some conclusions drawn that might need further discussion and revision. I would like to provide the detailed comments below.

Line 17-18 Why did not the authors take samples of DMSP and DMSO in October 2015?

Line 158-160 “The N:P ratio, defined as the ratio of the sum of nitrate (NO₃^-) and nitrite (NO₂^-) to dissolved phosphate (PO₄^3-) for both cruises, is a good indicator of nutritional status: high/low N:P ratios indicate nitrogen repletion/limitation.” Why didn't authors consider ammonium (NH₄⁺) which was also an important dissolved inorganic nitrogen? Would it make a significant effect on the conclusions if considering the concentration of NH₄⁺?

Line 167-168 “with the most abundant phytoplankton groups being diatoms (45 %), haptophytes (24 %), and chlorophytes (18 %) (2018)”. “(2018)” here needs a reference.

Line 168-169 “N:P ratios were generally between 8‒13 in the Peru upwelling region during SO243, indicating slightly limiting nitrogen conditions.” How to define the nitrogen repletion/limitation, what’s the N:P ratio range?

Line 202-203 “In contrast to our observations, Zindler et al. (2012) reported a general decreasing trend of DMSPt concentrations with decreasing N:P ratios (1‒12). This may be because the response to nitrogen limitation differs among specific algae groups.” What are the dominant algal groups in Zinder et al.'s (2012) study?

Line 205-207 “variability at the species or genus level might result in different responses under nitrogen limitation.” Here a reference is needed.

Line 227-228 “Generally, both N:P ratios and Ndef significantly correlated with coastal DMS values in the surface waters”. Did the DMS values and the concentrations of solo nutrients exhibit any relationship?

Line 255-256 “A comparison with DMS data from other EBUS and the Arabian Sea illustrates that DMS concentrations off Peru (up to 44 nmol L^-1) are higher”. Why did not the authors discuss the difference in DMS between Peru 1982 and this study? The DMS concentrations in this study were significantly lower.

Line 279 “which might be a result of the discrepancy for DMS flux densities between the two cruises”. The explanation was out of place. Don't put the cart before the horse. The discrepancy for DMS flux densities was influenced by the DMS concentrations in seawater and atmosphere and wind speeds. Therefore, the difference in the DMS concentration in the atmosphere could not be attributed to the discrepancy for DMS flux densities. Please explain it reasonably.

Line 289-290 “On average, the mean DMS flux densities in the first and third regions during SO243 are slightly higher/comparable to 5.5 µmol m^-2 d^-1 reported from Marandino et al. (2009)” Where did the first and third regions represent? Please define them.

Line 342-Line 343 and Line 358-360. Please check the references carefully.