Wood-Based Compost Affects Soil Fertility and the Content of Available Forms of Nutrients in Vineyard and Field-Scale Agroecosystems

Round 1

Reviewer 1 Report

The paper presented for evaluation concerns a very current problem concerning the use of waste organic matter for soil fertilization. The research was carried out on two farms with different production systems - conventional and organic. In an organic farm, the soil was fertilized only with wood-based compost, while in a conventional farm, mineral fertilizers and a small dose of manure were applied. The control object was unfertilized and uncultivated soil.

The article requires some additions:

1. Provide more information on the control soil, whether it came from any of the farms and what its properties were before setting up the experiment
2. In order to fully assess the effect of fertilization on selected physical, chemical and biological properties of the soil, soil characteristics in both farms before starting the research should be provided. Although the farms are adjacent to each other, they have a different production system, so even before the experiment was set up, soil properties could have varied significantly.

Author Response

We acknowledge the comment and we had initially taken into consideration such option. However, as several examples of reports assessing either soil chemistry or microbial ecology aspects have demonstrated, there is a conceptual built-in bias in comparing two situations under both ‘across-time’  + ‘across-treatment’ variables. The reason is due to the fact that time itself brings about per se several changes even in the same soil, irrespective from its treated or untreated condition. This is due to: inner consumption of organic matter and carbon emissions, anthropogenic fall-out deposition of nitrogen, sulphur and several other mineral elements, airborne microbial cells and spores immigration with daily discharge ruled by meteorological events. These could also irregularly include dust particles from other continents as often seen by Saharian trans-Mediterranean wind outbreaks.  As consequence, when analysing the same site across time, if this has also received agricultural treatments at farm scale, it would not be feasible to uncouple the net effect due to the local practices, from that due to the parallel global shifts in chemistry and biology of the area. For this reason, to avoid such double-variable confounding effect, the choice is to rely on the comparison of treated vs. untreated plots sampled at the same single time elapsed from the onset of the treatment, which is what we did in this experimentation. Therefore the clean controls are to be considered the ones that we show in each fourth column of Table 2 and Table 3 (Vineyard: Conventional, Organic, Untreated, and Field-scale: Conventional, Organic, Untreated, respectively). As regards the control soil, being located adjacently to the treated ones, and sharing the same geology as bedrock substrate (carbonate plateau) and the same pedological classification, the same climate and indirect atmospheric inputs, it is regarded as isogenic to the treated plots and differing only for the management practices, which are the rationale and object of this report.

Reviewer 2 Report

Overall is a comprehensive paper, did a lot of measurement, although it seems only meaningful in terms of soil aspects since this paper does not mention anything about crop aspects. Just some minor suggestions.

1. I wonder what are the initial physical and chemical property differences between OV and CV, OFS and CFS, as well as untreated soil in year 2015 rather than 2018 since you added huge amount of compost (60 t/ha), adding another table with the initial value would make this paper more complete.
2. Ca, Fe, K are higher in compost treated soil, is this mainly due to the extreme high content of these elements in compost you used? If this is the case, there is really no point to discuss these higher or lower, instead, if you have the yield data of the crops in these field, it would be more meaningful to compare the yield differences in this case (if yield is affected by these soil properties).
3. In your PCA figure, Fe locates in the left part, which seems negatively correlated with organic treated field, while in table 2, 3, OV and OFS are both have much higher Fe, any suggestion for this?

Author Response

Reviewer 2

Overall is a comprehensive paper, did a lot of measurement, although it seems only meaningful in terms of soil aspects since this paper does not mention anything about crop aspects. Just some minor suggestions.

ANSWER : Being the sites not experimental areas but real operative farms we did not withdraw their harvests (destined to the market) for quantitative or qualitative analyses as the focus of this research were soil chemistry balances. 

I wonder what are the initial physical and chemical property differences between OV and CV, OFS and CFS, as well as untreated soil in year 2015 rather than 2018 since you added huge amount of compost (60 t/ha), adding another table with the initial value would make this paper more complete.

ANSWER : We acknowledge the comment and we had initially taken into consideration such option. However, as several examples of reports assessing either soil chemistry or microbial ecology aspects have demonstrated, there is a conceptual built-in bias in comparing two situations under both ‘across-time’  + ‘across-treatment’ variables. The reason is due to the fact that time itself brings about per se several changes even in the same soil, irrespective from its treated or untreated condition. This is due to: inner consumption of organic matter and carbon emissions, anthropogenic fall-out deposition of nitrogen, sulphur and several other mineral elements, airborne microbial cells and spores immigration with daily discharge ruled by meteorological events. These could also irregularly include dust particles from other continents as often seen by Saharian trans-Mediterranean wind outbreaks.  As consequence, when analysing the same site across time, if this has also received agricultural treatments at farm scale, it would not be feasible to uncouple the net effect due to the local practices, from that due to the parallel global shifts in chemistry and biology of the area. For this reason, to avoid such double-variable confounding effect, the choice is to rely on the comparison of treated vs. untreated plots sampled at the same single time elapsed from the onset of the treatment, which is what we did in this experimentation. Therefore the clean controls are to be considered the ones that we show in each fourth column of Table 2 and Table 3 (Vineyard: Conventional, Organic, Untreated, and Field-scale: Conventional, Organic, Untreated, respectively). As regards the control soil, being located adjacently to the treated ones, and sharing the same geology as bedrock substrate (carbonate plateau) and the same pedological classification, the same climate and indirect atmospheric inputs, it is regarded as isogenic to the treated plots and differing only for the management practices, which are the rationale and object of this report.

Ca, Fe, K are higher in compost treated soil, is this mainly due to the extreme high content of these elements in compost you used? If this is the case, there is really no point to discuss these higher or lower, instead, if you have the yield data of the crops in these field, it would be more meaningful to compare the yield differences in this case (if yield is affected by these soil properties).

ANSWER: Yes, the addition of compost has led to an increase in plant available macro and micronutrients, and this was in line with the results found in previous studies (lines 263-265). As mentioned above we did not collect yield data also because we focused on the effect on soil properties being this report targeting the Special Issue “Management of Soil Organic Carbon for Soil Health in Agroecosystems". Anyway, as crop productivity is concerned, the interesting aspect of this work is that the addition of compost has led to an increase in numerous plant available elements without affecting their total content, compared to that induced by mineral fertilization and control.

In your PCA figure, Fe locates in the left part, which seems negatively correlated with organic treated field, while in table 2, 3, OV and OFS are both have much higher Fe, any suggestion for this?

ANSWER: The observation is correct, the FeDTPA is located on the left side of the PCA figure. But it happens that his eigenvector has a negative value (see Table 4). As a consequence the UNT, CV and CFS feature lower values in FeDTPA than OV and OFS.

Round 2

Reviewer 1 Report

I agree with the explanations of the Authors of the paper posted in response to the reviews. However, explanations about the research methodology should be included in the Material and methods section.

Author Response

Dear Review,

thanks for your suggestions.

We have now included in the text the explanation given in the previous round of refereeing. 

lines 119-133: "In terms of controls, as concerns the possibility of soil analyses at time zero, we opted for a different choice. Indeed as several examples of reports assessing either soil chemistry or microbial ecology aspects have demonstrated, there is a conceptual built-in bias in comparing two situations under both ‘across-time’ +‘across-treatment’ variables. The reason is due to the fact that time itself brings about per se several changes even in the same soil, irrespective from its treated or untreated condition. This is due to: inner consumption of organic matter and carbon emissions, anthropogenic fall-out deposition of nitrogen, sulfur and several other mineral elements, airborne microbial cells and spores immigration with daily discharge ruled by meteorological events. These could also irregularly include dust particles from other continents as often seen by Saharian trans-Mediterranean wind outbreaks. As consequence, when analyzing the same site across time, if this has also received agricultural treatments at farm scale, it would not be feasible to uncouple the net effect due to the local practices, from that due to the parallel global shifts in chemistry and biology of the area. For this reason, to avoid such double-variable confounding effect, the choice was to rely on the comparison of treated vs. untreated plots sampled at the same single time elapsed from the onset of the treatment. As regards the control soil, being located adjacently to the treated ones, and sharing the same geology as bedrock substrate (carbonate plateau) and the same pedological classification, the same climate and indirect atmospheric inputs, it is regarded as isogenic to the treated plots and differing only for the management practices, which are the rationale and object of this report. " 

Author Response File: Author Response.docx