|This is an interesting work on soil GHG fluxes in an underrepresented region; it does not only provide a good (and short) dataset but also some insights into soil N2O uptake. I have had a look to the previous version and the comments of the reviewers and I agree with them that the experimental set up and the dataset fits better into a “preliminary data” paper. The paper is nicely and clearly written, and it is honest with regard to its limitations; at the same time, serves in suggesting ways forward for increasing our understanding on the patterns of GHG fluxes in tropical montane forest soils.|
Probably one of the most relevant changes I suggest is with regard to the title. “Varying sources and sinks” seems to me that you look at different components within the forest, but this is not the case. I suggest to make explicit mention to the soil (e.g. –only a suggestion- “Ideas and perspectives: patterns of soil CO2, CH4 and N2O fluxes along ….”) and probably to montane forests.
Further, I understand the rationale of using GWP to compare between GHGs, but I strongly doubt the usefulness of including CO2 fluxes since, as you mentioned in the paper, it is only a part of the story and do not reflect inputs to the soil.
Finally, you highlight quite prominently the topographic position as a key driver controlling spatial variability (e.g. L32, L186); while this is true, other sources of variability may be equally important (e.g. degradation/forest management, hydrological status at the catchment level, soil types, exposition and associated microclimate) and should be taken into account when proposing “broader studies”.
For the rest, I only have a couple of minor comments, which are depicted below.
L41: Remove “(anaerobic)”
L44-46. I am missing a citation here, so I can´t check on my own. I guess this is the C sink of the forest, and not of the soil; since you previously referred to the soil, it is misleading, unless you are able to infer what proportion is stored in the vegetation and in the soil (or below- and aboveground).
L70: Probably merge the two paragraphs.
L75: Remove “more” before “severe”. Remove “even though forests can be managed to mitigate climate change as well”, it is not really relevant here. Eventually, you can state that it is important to understand the feedbacks to come up with appropriate forest management options to mitigate climate change.
L93: I suggest to change “budget” by “fluxes”.
L102: Zimmermann et al 2009(Eur J Soil Sci, doi: 10.1111/j.1365-2389.2009.01175.x) worked also on an altitudinal gradient in the Andes and pointed towards changing C allocation patterns. The paper might be also interesting for the summary table (I was not involved in this work).
L119: Remove “on the other hand”. Reformulate the second sentence. The lowest flux is the closest to 0, and this is not what you want to point out.
L139: The range is wrong, I think, it should go from negative to positive.
L143: This is also similar to what Gerschlauer (Biogeosciences) found in a gradient in the Kilimanjaro. I don´t remember if they found relationships between N2O fluxes and the isotopic signature in the soil, but may be worth looking at it (not involved either in the work).
L152: Probably “support” or similar, rather than “confirm”
L186: Remove “For instance”
L194: Why bi-weekly? To me, the sampling should cover seasonal fluctuations (dry vs. wet season), but also at a finer scale.
Fig 2: Note: “the dotted vertical line” instead of “the dotted x axis”
Fig 3: Even if the three GHG are represented using CO2-eq, consider an axe-break instead on the zoom-in view.
Table S1. This is a great compilation of information. However, I find difficult to compare the studies. I suggest to add more columns, in order to separate between measured and estimated annual, and also a field with comments/period of measurement). I also suggest not to use the publication as a basis, but the site (e.g. Garcia Montiel et al 2004 measured on different sites, so include a line for each, with the same reference). For me, manual and dynamic chambers are not mutually exclusive. Do you rather mean manual vs. automated? In any case, probably the frequency is more important than the method per se (e.g. manual chambers can have daily, or monthly frequency).