|Soil fauna: Key to new carbon models by Filser et al. |
I reviewed the initial version of this ms and believe this revision is much improved, particularly with your clear statements about the independent effects of fauna on SOM dynamics (as compared to predictions from landuse, climate and soil and vegetation properties), and your discussions about scale and the shape of model functions. There is also clearer justification for the inclusion of the KEYSOM COST action in the main text.
Overall, the need for faunal incorporation in SOM models is therefore more clearly elucidated. However, I still wonder whether review 1's comments about modellers appreciating the way the material is communicated still holds true i.e. whether the paper will get to its target audience. However, the authors clearly feel that some of these comments, and my earlier comments a) - d), are outside their scope; I still believe that widening the paper's scope (and therefore addressing these comments more directly) would maximise the paper's impact with the modelling community. However, perhaps KEYSOM will do this in time, and ultimately it is the authors' decision as to their desired scope.
I have a few concerns that I believe still require addressing for the ms to increase its potential reach.
Although the main text justifies the incorporation of the COST action, it is not so clearly demonstrated in the Abstract, and the way it is currently incorporated in the Abstract still reads like an 'advert' to me, while the last sentence re: field experiment, experimental data and so forth are too vague (in my opinion) for an Abstract.
I believe the abstract would be stronger, if KEYSOM is still to be included, with a conclusion such as:
"We argue that explicit consideration of the soil fauna is essential to make realistic modelling predictions on SOM dynamics and to detect expected non-linear responses to global change. We present a decision framework, to be further developed through the activities of KEYSOM, a European COST action, for when mechanistic SOM models should include soil fauna. The research activities of KEYSOM, such as field experiments and literature reviews, together with dialogue between empiricists and modellers, will inform how this is to be done."
You could also include KEYSOM in the keywords if you want it to be searchable.
The above comment is of course a suggestion and what I wanted to know as a reader.
Key insights: Although some sections within this part now clearly demonstrate the link between the key insight and SOM dynamics (and then the importance of soil fauna in modifying the key insight) e.g. 2.6 on soil depth, other parts go straight into how animals affect the key insight, without stating how the key insight affects SOM. I realise the latter is included in Schmidt, but I think it might be clearer for readers to show how each insight affects SOM (briefly as with soil depth, not exhaustively), and then introduce the substantial body of the review regarding soil faunal effects on the key insight.
Table 1: I found the examples in Table 1 very interesting. However, I did not follow how many of the earthworm examples in 2. Humic substances, were to do with this insight, as they appeared most related to aggregate stability. It is also not immediately clear how e.g. casting 40-50 t/ha/yr influences 'physical heterogeneity' and so on and so forth.
As another example, for 'Various or mixed groups' what does microbial grazing leading to leaching of DOC and NH4+ have to do with 'molecular structure'? Don't some of these substances already exist in the absence of animals, whereas your points, at least as I understood the main text, are to do with animals modifying the molecular structure of SOM and thus influencing its decomposition dynamics?
In other words, for all of the examples, make it really clear how they relate to the key insights and the explanations in the main text.
I am surprised there isn't at least some discussion as to whether other processes could be important in improving SOM models (e.g. dissolved and particulate organic matter leaching, soil erosion, better parameterisation of litter inputs), and not just the requirement to include animals. I think it would be good to briefly mention alternative explanations for current model discrepancies; I am not suggesting these are reviewed.
I like the insertion of Figure 3. However, please explain what 'Etc' means. I would consider taking out the shaded miniature on the figure as I think it unnecessarily confuses and modify the main text accordingly. However, retain an explanation about what happens when data are not available at different scales. I think I found this somewhat confusing as to my geographical mind, 'biome' is a given scale e.g. boreal forest, tropical forest etc. Also, the flow from 6 to 7 to 8 is not clear, given the additional white box with 'compare output with previous models/versions'. I would consider redrawing this to a linear flow, similar to A) and B).
MINOR COMMENTS / GRAMMATICAL CORRECTIONS
I was surprised by the standard of the English in parts of the ms; I would suggest giving it a careful proof read before any final submission and, if accepted, publication.
Line 57: Suggest "Fauna control..." [fauna are plural, as are data later in the ms (line 423)], rather than "The fauna controls"
Line 59: "We show lots of quantitative examples" sounds very colloquial; I would suggest deleting and just stating "We demonstrate a very strong impact of soil animals on carbon turnover"
Lines 88-89: key to understandING and predictING changes in global carbon cycling
Suggest "Some years ago" on line 96 be the start of a new paragraph.
Line 108: Suggest "A correlative large scale field study has shown that including soil animal activities could help clarify discrepancies in existing carbon models"
Line 116: Suggest split this sentence to:
"Here we use the 'key insights' proposed by ...on the relationship of soil fauna to SOM dynamics. Our review justifies the relevance of incorporating soil fauna into SOM models"
When you state, on line 118, that there have been repeated reviews it begs the question why therefore you have done this review. Are SOM dynamics different to 'geological and pedological processes' - if so, I would state it explicitly, or just remove this sentence.
Line 123 - the Dorn reference to ants seems far too detailed for an Introduction - I would place it within an appropriate part of the review of key insights.
Line 147/148 - I presume 'humification' means, in this context, 'decomposition dynamics of animal faeces'. If so, I would put the parentheses after animal faeces, not where it is presently.
It may also be worth clarifying what the relationship is (if there is one) between 'humification', 'humic substances' and 'humus' (both of which are introduced in the next section). Also, you define humic substances on line 371 in a more explicit way than in section 2.2 - I would insert the definition on line 371 into section 2.2.
Line 218: Suggest delete 'any'.
Line 294 - still unsure of what the relationship of tensile strength is to SOM dynamics - please elucidate.
Line 331-333: Reference(s) required for suggestion that humid grasslands are "among those ecosystems that are most severely affected by global environmental change"
Line 341: "more detailed information on their biology is required". State why this is required for the modelling. As pointed out by review 1 (for the initial submission), it is very difficult for the modelling community to include lots and lots of details as it makes model interpretation and validation problematic. Your calls for more detailed biological information therefore needs to be justified from a modelling perspective.
Line 397: I still find the CENTURY example oddly phrased. You discuss 'implementing earthworm activity' and then state what happens 'without earthworms', all in the same sentence. Please rephrase.
Line 444: Unclear what 'richness' refers to. Could mean species richness, although could also mean e.g. soil fertility. I assume you mean fertility but not sure...
Line 472: THE main aim of KEYSOM...of soil fauna activity into SOM models...
Line 493 - provide a bit more detail about 'the development of a simple SOM model' - I presume you mean a model that explicitly incorporates faunal processes in it, but you don't actually state this.