the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 May 2023
Soil organic carbon stocks did not change after 130 years of afforestation on a former Swiss Alpine pasture
Abstract. Soil organic matter (SOM) plays an important role in the global carbon cycle, especially in alpine ecosystems. However, ongoing forest expansion in high elevation systems potentially alters SOM storage through changes in organic matter (OM) inputs and microclimate. In this study we investigated the effects of an Picea abies L. afforestation chrono-sequence (40–130 years) of a former subalpine pasture in Switzerland on soil organic carbon (SOC) stocks and SOM dynamics. We found that SOC stocks remained relatively constant throughout the chrono-sequence, with comparable SOC stocks in the mineral soils after afforestation and previous pasture (SOC40-year-old forest = 11.6 ± 1.1 kg m−2, SOC130-year-old forest = 11.0 ± 0.3 kg m−2, and SOC pasture = 11.5 ± 0.5 kg m−2). However, including the additional carbon of the organic horizons in the forest, reaching up to 1.7 kg m−2 in the 55-year-old forest, resulted in a slight in-crease in overall SOC stocks following afforestation. We found that the soil C:N ratio in the mineral soil increased in the topsoil (0–5cm) with increasing forest stand age, from 11.9 ± 1.3 in the grassland to 14.3 ± 1.8 in the 130-year-old forest. In turn, we observed a decrease in soil C:N ratio with increasing depth in all forest stand ages. This suggests that litter-derived organic matter (C:N from 35.1 ± 1.9 to 42.4 ± 10.8) is likely incorporated and translocated from the organic horizon to the mineral topsoil (0–10 cm) of the profiles. As roots had very high C:N ratios (pasture 63.5 ± 2.8 and forests between 54.7 ± 3.9 and 61.2 ± 2.9), particulate root-derived organic matter seems to have a minor influence on forest soil C:N ratio and thereby on SOC stock accumulation in the mineral soil. These results suggest that, although the afforestation only moderately affected the SOC stock, there is an apparent alteration in the SOC dynamics through changes of the litter composition caused by the vegetation shift. We conclude that spruce afforestation on a former subalpine pasture does not necessarily change the total SOC stock and that consequently there is no SOC sequestration on a decadal to centennial scale.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1057 KB)
-
Supplement
(187 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1057 KB) - Metadata XML
-
Supplement
(187 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-645', Anonymous Referee #1, 19 Jun 2023
The manuscript shows the effects of a 130-year afforestation of alpine pasture on SOC stocks in the Swiss Alps. Authors found that afforestation did not necessarily change SOC stocks but clearly altered SOC dynamics. In general, the manuscript gives clear conclusions and has high scientific value. To improve the manuscript quality, the way of presenting data in figures, tables, and the Result section could be improved. In addition, the conclusion regarding words such as “increased”, “no influence”, and “moderately affected” can be connected with statistical significance (P values) in a more consistent and convincing way. Finally, several places showing inconsistent contents among figures & tables, results, and discussions should be addressed.
Abstract:
Line 13-14 “SOC 20-year-old forest” and later
Please be consistent with the usage of subscripts.
Line 22 “moderately affected the SOC stock”
I am not sure if “moderately affected” is a precise way to describe your results because SOC stocks remained constant in mineral soils but has a slight increase (P = 0.229) in organic horizons. When combining things together, I find a “slight increase in overall SOC stock” in line 15, a “moderately affected the SOC stock” in line 22, and “there is no SOC sequestration on ...” in line 25. I would suggest finding a better way to present them in order to avoid any potential confusion.
Introduction：
It might be better to divide the Introduction into several paragraphs. For example, different paragraphs might focus on topics such as the significance of studies on relevant topics, the effects of afforestation on SOM, current knowledge gaps in the relevant study area, and so on.
Line 43-44 “OM accumulates in living biomass”. It is a bit confusing to mention living biomass. Do you mean tree biomass? It could also be soil microbial biomass.
Line 75-76
To be prudent, it is suggested to specify “litter-derived carbon input” as something like aboveground or foliage litter because root litter is also a part of the litter.
Material and methods:
Line 115-123 and Line 143
How is SOC stock calculated? Is it corrected by the volume of stones? Based on the content of Line 115-123, it seems that soils collected by steel cylinders contained stones, which were removed later while passing through the 2mm sieve. I guess the bulk density should be corrected by stone volume. Is it possible to make it clearer?
Line 119
Were bulk densities measured from organic samples? You calculated C stocks of organic horizons in the Result section.
Line 144-149
Based on Table 1, it seems the depth 0 cm started from the top of mineral soils. It looks like the formula only calculates SOC stock of the mineral soils. Were SOC stocks of organic horizons calculated using the same formula?
Line 151-155
If I understand correctly, forest age and soil depth are both fixed effect predictors. Is this right? Where can I find the outcomes of mixed-effect models from Tables & figures and the Result section?
Results:
3.1 and 3.2
It is not common that the results start with two sections based on supplementary materials. Maybe some information in the supplementary materials can be incorporated into figures and tables.
Line 159 and later
It is a suggestion that it may be not necessary to present F values and P values at the same time whenever you make comparisons. Maybe only P value is enough; alternatively, even omit P values if you mention that the significant difference means P < 0.05. Instead, consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Line 159 and 167
Table S2 and Table S3?
Line 166
It seems that pinecones and branches do not match the names in Table S3. Branch = wood and twigs? Pinecones = spruce cones?
Line 175 and later
It is not necessary to mention P adj <0.95 all the time.
Line185-188
The P value seems not to be 0.801 but (nearly-) significant if you look at the values of combined SCO stocks of organic horizons. (i.e. 1.7±0.2, 1.3±0.2, and 0.8±0.1).
Discussion:
Line 244 “… a significant lower SOC stock at 15 to 25 cm…”
Is it from Table 1? It will be easier for readers if significant differences could be found directly from tables and figures.
Paragraph 1 until Line 250:
I suggested making it clear that you are talking about mineral soils (next paragraph for organic horizons).
Line 262 and after:
I think it might be helpful to make it clear when we can say something is non-significantly, significant, or unchanged. This does not mean that we need to draw a massive line between P < 0.05 and P > 0.05, but we should more or less be sure when we can say that A is higher than B. For example, it is not significant when P = 0.669 (line 245), but it is a (slightly) increased SOC sequestration when P = 0.229 (line 15, 195 and 262). I know that the latter might be a key conclusion for this manuscript, but it would be better to present it in a more convincing way.
Line 274 and after “…C:N ratios of Oi horizons increased…”
Table 1 should be Table 3?
Please be aware in lines 203-204 you said C:N ratios were not significantly different between forest ages, but you mentioned here C:N ratios increased with age. Values of 42.4±10.8 indicated a very large variation within the samples of 130-year-old soils. This is probably the reason why it is not statistically significant.
Line 287 and after
In general, organic matter input from the root is comprised of root litter and root exudates, from which the exudate is considered as the more important belowground input (Sokol et al. 2018 doi: 10.1111/nph.15361). Root exudates contain a lot of easily-decomposable (high-quality) compounds, which promote the microbial transformation of these compounds and the formation of microbial-derived compounds as stabilized SOC with low C:N ratios (Cotrufo et al. 2013, you cited). It is possible that the intensive (or effective) utilization of root exudates promotes the formation of stable microbial-derived organic matter, and therefore, the C:N ratios were still low. Therefore, my suggestion is that try to discuss it in a more convincing way with the consideration of the transformation of root exudates
Figure and Table:
In general, there is no information of significant difference in figures and table. This makes it difficult to link figures and tables to the Result and Discussion sections. Please consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Table 1
When combined organic and mineral soils, SOC was 11.6 + 0.8 =12.4 for 40-year-old soil and 11.0 + 1.3 = 12.3 for 130-year soil, respectively. In line 194, the 40-year-old soil and 130-year-old soil had 7% and 11% higher SOC stocks than pasture soil, respectively. Is this a mistake? I think both organic and mineral soils should be included when compare total SOC stocks.
Table 2
63.5 + 02.8, remove ‘0’.
Citation: https://doi.org/10.5194/egusphere-2023-645-RC1 -
AC1: 'Reply on RC1', Tatjana Carina Speckert, 19 Jul 2023
RC1: The manuscript shows the effects of a 130-year afforestation of alpine pasture on SOC stocks in the Swiss Alps. Authors found that afforestation did not necessarily change SOC stocks but clearly altered SOC dynamics. In general, the manuscript gives clear conclusions and has high scientific value. To improve the manuscript quality, the way of presenting data in figures, tables, and the Result section could be improved. In addition, the conclusion regarding words such as “increased”, “no influence”, and “moderately affected” can be connected with statistical significance (P values) in a more consistent and convincing way. Finally, several places showing inconsistent contents among figures & tables, results, and discussions should be addressed.
Author's response: Thank you for the helpful comments and feedback! To improve the presentation of the figures, we will add a map of the location within Switzerland, and we will increase the points in Figure 1. In Figure 2 and 3 we will enlarge the font size and correct the corresponding caption. In the result section, we will avoid terms like “moderately”, “slightly” etc. in case there is no statistical significance, and we will correct inconsistencies between statements and the corresponding p-values.
RC1: Abstract:
Line 13-14 “SOC 20-year-old forest” and later
Author's respose: Thanks for pointing out the inconsistency of the formatting and we will fix it throughout the revised text.
RC1: Line 22 “moderately affected the SOC stock”
I am not sure if “moderately affected” is a precise way to describe your results because SOC stocks remained constant in mineral soils but has a slight increase (P = 0.229) in organic horizons. When combining things together, I find a “slight increase in overall SOC stock” in line 15, a “moderately affected the SOC stock” in line 22, and “there is no SOC sequestration on ...” in line 25. I would suggest finding a better way to present them in order to avoid any potential confusion.
Author's response: We agree that using terminology such as “moderately” or “slightly” is not precise. We removed all the qualitative judgements and only state that there is no SOC stock change with different forest age.
RC1: Introduction：
It might be better to divide the Introduction into several paragraphs. For example, different paragraphs might focus on topics such as the significance of studies on relevant topics, the effects of afforestation on SOM, current knowledge gaps in the relevant study area, and so on.
Author's response: We agree with the suggestion and will divide the introduction into paragraphs organized by topic.
RC1: Line 43-44 “OM accumulates in living biomass”. It is a bit confusing to mention living biomass. Do you mean tree biomass? It could also be soil microbial biomass.
Author's response: Indeed, we meant tree biomass instead of living biomass, and will change the text accordingly.
RC1: Line 75-76
To be prudent, it is suggested to specify “litter-derived carbon input” as something like aboveground or foliage litter because root litter is also a part of the litter.
Author's response: We fully agree and will change the wording to “needle litter-derived carbon input”.
RC1: Material and methods:
Line 115-123 and Line 143
How is SOC stock calculated? Is it corrected by the volume of stones? Based on the content of Line 115-123, it seems that soils collected by steel cylinders contained stones, which were removed later while passing through the 2mm sieve. I guess the bulk density should be corrected by stone volume. Is it possible to make it clearer?
Author's response: Thank you for your questions. We calculated SOC stock as described in the M&M section. We correct SOC stock to stone content to the final calculation of bulk volume based on the mass of stone content obtain after sieving and assuming a rock density of 2.65 g/cm3.BD = M (weight of dried soil)/ (V (volume of the steel cylinder) – V (volume of the coarse fragments)). We will update the methods section to better reflect what we did.
RC1: Line 119
Were bulk densities measured from organic samples? You calculated C stocks of organic horizons in the Result section.
Author's response: Bulk densities for the organic horizon were not measured, but C stocks were estimated from the masses collected in an area of 0.25x0.25m which were then upscaled to m2.
RC1: Line 144-149
Based on Table 1, it seems the depth 0 cm started from the top of mineral soils. It looks like the formula only calculates SOC stock of the mineral soils. Were SOC stocks of organic horizons calculated using the same formula?
Author's response: Indeed, we did not include SOC stocks of the organic horizon to the mineral soil after calculating the two pools separately, we then added up for the final SOC stock. SOC stocks of the organic horizons were calculated differently to the mineral soil their carbon concentration was multiplied by the weight [g] of the respective sample from an area of 0.25x0.25m and then upscaled to m2. We will clarify this in the method section.
RC1: Line 151-155
If I understand correctly, forest age and soil depth are both fixed effect predictors. Is this right? Where can I find the outcomes of mixed-effect models from Tables & figures and the Result section?
Author's response: Correct, forest age and soil depth are modelled as fixed effects to test for their effect on SOC stocks. You are also correct, and we apologize for the oversight of not adding the proper reporting of statistical modelling. We will add these results in the updated version of the manuscript.
RC1: Results:
3.1 and 3.2
It is not common that the results start with two sections based on supplementary materials. Maybe some information in the supplementary materials can be incorporated into figures and tables.
Author's response: Thank you for your suggestion and we will reframe our results section to first describe key parameters, then the results of the ANOVA and finally the results of the modelling. We will further incorporate the vegetation composition into the main text.
RC1: Line 159 and later
It is a suggestion that it may be not necessary to present F values and P values at the same time whenever you make comparisons. Maybe only P value is enough; alternatively, even omit P values if you mention that the significant difference means P < 0.05. Instead, consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Author's response: We think that it is important to show the statistics value (F-value) and its corresponding p-value. Comparing the F statistic value with the theoretical value serves as threshold for our sample size and test framework. As such we will keep both for completeness. However, to keep the text clear, F-values will only be reported in a new supplementary Table and not in the text.
RC1: Line 159 and 167
Table S2 and Table S3?
Author's response: Indeed, there was a mistake in referencing the supplementary Tables. We will change this accordingly.
RC1: Line 166
It seems that pinecones and branches do not match the names in Table S3. Branch = wood and twigs? Pinecones = spruce cones?
Author's response: Thanks. you are correct and we will consistently using the term spruce cones in Table S3.
RC1: Line 175 and later
It is not necessary to mention P adj <0.95 all the time.
Author's response: We agree and will remove it where it is unnecessary.
RC1: Line185-188
The P value seems not to be 0.801 but (nearly-) significant if you look at the values of combined SCO stocks of organic horizons. (i.e. 1.7±0.2, 1.3±0.2, and 0.8±0.1).
Author's response: This simply occurred because we used the individual values for the statistical test and not the combined one. We will change this in the next version of the manuscript and use the combined values of the SOC stock to test whether forest age has a significant effect or not on total (mineral soil plus organic horizon) SOC stock.
RC1: Discussion:
Line 244 “… a significant lower SOC stock at 15 to 25 cm…”
Is it from Table 1? It will be easier for readers if significant differences could be found directly from tables and figures.
Author's response: This is the outcome of the mixed model analysis, and we did not provide those results yet as figures or tables. We will revised this in the manuscript.
RC1: Paragraph 1 until Line 250:
I suggested making it clear that you are talking about mineral soils (next paragraph for organic horizons).
Author's response: We agree and will make it clear when we are referring to organic or mineral soils to facilitate understanding. We will rephrase it in the next version of the manuscript.
RC1: Line 262 and after:
I think it might be helpful to make it clear when we can say something is non-significantly, significant, or unchanged. This does not mean that we need to draw a massive line between P < 0.05 and P > 0.05, but we should more or less be sure when we can say that A is higher than B. For example, it is not significant when P = 0.669 (line 245), but it is a (slightly) increased SOC sequestration when P = 0.229 (line 15, 195 and 262). I know that the latter might be a key conclusion for this manuscript, but it would be better to present it in a more convincing way.
Author's response: Thank you for this comment, and we fully concur with this. We will review the instances where we may have been inconsistent in our reporting. We will indeed state when something is significant only when p<0.05 and avoid qualitative statements that are not warranted by the results.
RC1: Line 274 and after “…C:N ratios of Oi horizons increased…”
Table 1 should be Table 3?
Author's response: Apologies, it should be Table 3, we will fix this typo.
RC1: Please be aware in lines 203-204 you said C:N ratios were not significantly different between forest ages, but you mentioned here C:N ratios increased with age. Values of 42.4±10.8 indicated a very large variation within the samples of 130-year-old soils. This is probably the reason why it is not statistically significant.
Author's response: We agree and apologies for the inconsistencies. We will review the instances where we may have been inconsistent in our reporting. We will indeed state when something is significant only when p<0.05 and avoid qualitative statements that are not warranted by the results.
RC1: Line 287 and after
In general, organic matter input from the root is comprised of root litter and root exudates, from which the exudate is considered as the more important belowground input (Sokol et al. 2018 doi: 10.1111/nph.15361). Root exudates contain a lot of easily-decomposable (high-quality) compounds, which promote the microbial transformation of these compounds and the formation of microbial-derived compounds as stabilized SOC with low C:N ratios (Cotrufo et al. 2013, you cited). It is possible that the intensive (or effective) utilization of root exudates promotes the formation of stable microbial-derived organic matter, and therefore, the C:N ratios were still low. Therefore, my suggestion is that try to discuss it in a more convincing way with the consideration of the transformation of root exudates
Author's response: Thank you for suggestion of a potential mechanism. Indeed, it is possible that root-exudates could be a potential source of OM with low C:N ratio and thereby be an alternative hypothesis to explain our results. We will include this in the discussion.
RC1: Figure and Table:
In general, there is no information of significant difference in figures and table. This makes it difficult to link figures and tables to the Result and Discussion sections. Please consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Author's response: Indeed, you are correct as these were missing. We will add proper statistical analyses reporting in the corresponding tables and figures.
RC1: Table 1
When combined organic and mineral soils, SOC was 11.6 + 0.8 =12.4 for 40-year-old soil and 11.0 + 1.3 = 12.3 for 130-year soil, respectively. In line 194, the 40-year-old soil and 130-year-old soil had 7% and 11% higher SOC stocks than pasture soil, respectively. Is this a mistake? I think both organic and mineral soils should be included when compare total SOC stocks.
Author's response: This is true. We first wanted to separately show the data subsequently combine it as we think that it is valuable that mineral soil stocks remain identical, but only SOC stocks increase when considering the organic horizons. We will rephrase this.
RC1: Table 2
63.5 + 02.8, remove ‘0’.
Author's response: Apologies: we will remove the 0 in Table 2.
Citation: https://doi.org/10.5194/egusphere-2023-645-AC1
-
AC1: 'Reply on RC1', Tatjana Carina Speckert, 19 Jul 2023
-
RC2: 'Comment on egusphere-2023-645', Anonymous Referee #2, 22 Jun 2023
Speckert et al. Soil organic carbon stocks did not change after 130 years of afforestation on a former Swiss Alpine pasture
Review notes
The study presented here covers a topic that is interesting and relevant since many pastures in alpine regions are abandoned and expected to become forests in the future – which could feed back to climate change. Knowing the impact of this land-use conversion is needed. The topic and proposed research questions however are not very novel and do not contribute much new knowledge.
There are two things I value in the current manuscript:
- The long gradient in time (if the chronosequence would have been replicated) adding more information on the long term impacts of afforestation
- The study is well executed (- the design) and the manuscript is well written
My major concern:
- The study is based on pseudoreplications so the conclusions in terms of time since afforestation are not supported by the data, they could be coincidence or site effects. In my opinion this is a major issue that is hard to resolve.
Specific comments:
Line
Comment
19-22
As roots … -> unclear sentence
22-25
Is this novel enough?
37-38
Afforestation is a promising measure -> is this also true for alpine regions where albedo effects might have a negative impact on climate change
49-50
Be explicit and call it “context-dependency”
51
Initial carbon stock is a key factor -> yet you do not take this into account in your research
54-55
Transition line 54 to 55 not very fluent (10-20 cm is not that deep? Was that the point?)
58-59
Age is underexplored -> I think this is the main strength of your study that you have a large gradient in time going back 130 years
75-77
Clear hypothesis. Is (iii) really needed? It is accepted as common knowledge I would say
Introduction
Overall very well written introduction with nice literature review.
85
Afforested -> Encroached? Natural regeneration? Planted to avoid erosion? Not explicitly stated in the M&M. If planted, at what planting distances? What is was the basal area (the density) of the stands at the time of sampling. This is important information that is needed to interpret the results
M&M
There are no replications of the different ages. The replications are made within the same forest stands resulting in pseudoreplications. This means you cannot say anything about the effect of age / time since afforestation. It can just be differences in between individual stands that do not correspond with the age effect. This should be very explicitly discussed that you make assumptions and in the discussion (and in the end of the abstract) you should be very careful with generalizing your interpreted results.
98
Five individual plots … -> not true, only for the pasture. This is misleading.
100
Your forests are quite small and your plots are taken as pseudoreplications within these small stands -> weakness of the study
100
Add also the area for the 40y old stand
110
Here it becomes clear that there are only 3 replications for the forest stands
111
I appreciate the effort to make soil profile pits to have a more detailed understanding of the belowground ecosystem (as well as a complete sampling of the roots)
113
Roots were counted on three profile walls -> per pit? Per replication? Unclear
115
In the end the results were pooled -> be already transparent about this in the M&M
130
Typo 45mm? So how many samples for the roots do you have per forest stand? 3 replications of pooled samples? In the figures it seems more? Unclear
130
Can you discuss the results of pooling these samples in the M&M section
163
I miss information about the density (do you have basal area)?
194
Not very clear “age gradient” in the results. This indicates that the differences between stands might just be random (site effects) that do not correspond with the age of the forest
203
between
224
You have a very high initial carbon stock -> could you explain more how come?
228
Careful with phrasing -> here you cannot say it is the effect of 40 years of afforestation for sure
232
Again very case specific
247-248
Can you really say it is forest age?
305
Here you speak of forest encroachment but you did not make this clear in the M&M. In that case how did the stand evolve?
515
Many outliers. Which ones are from the sides and which ones from the levels? Or are they pooled in that way as well. Be clear about the experimental design
-
AC2: 'Reply on RC2', Tatjana Carina Speckert, 19 Jul 2023
Review notes
RC2: The study presented here covers a topic that is interesting and relevant since many pastures in alpine regions are abandoned and expected to become forests in the future – which could feed back to climate change. Knowing the impact of this land-use conversion is needed. The topic and proposed research questions however are not very novel and do not contribute much new knowledge.
Author's response: Thank you for your assessment. We are aware that we work at a single site planted with a single tree species and is therefore rather case specific. However, finding a study site with such a long-term afforestation chrono-sequence with four forest stands of different age classes and a reference pasture area on a homogenous slope is unique. Moreover, the location (subalpine region) is highly relevant as studies concerning the organic carbon stock in the Alps are scarce (Thuille and Schulze, 2006, https://doi.org/10.1111/j.1365-2486.2005.01078.x; Zimmermann et al., 2010, https://doi.org/10.1016/j.agee.2010.06.010) and there is an ongoing process of subalpine pastures abandonment and forest encroachment, which policy makers consider as beneficial in terms of carbon storage. However, recent studies (Clemmensen et al., 2014, https://doi.org/10.1111/nph.13208; Friggens et al., 2020, https://doi.org/10.1111/gcb.15229) show evidence that priming, mycorrhizal nutrient mining associated with ectomycorrhizal tree species can result in a net carbon loss from the mineral soil. We therefore used this afforestation chrono-sequence as a natural experiment to test whether there is an increased carbon sequestration with increasing forest age.
RC2: Line 19-22
As roots … -> unclear sentence
Author's response: We will change this sentence to make it clearer: “ Due to the high root C:N ratio (pasture 63.5 ± 2.8 and forests between 54.7 ± 3.9 and 61.2 ± 2.9), particulate root-derived organic matter seems to have a rather small effect on the forest soil C:N ratio as well as on the SOC stock accumulation in the mineral soil.
RC2: Line 22-25
Is this novel enough?
Author's response: As far as we know there are not many studies reporting SOC stocks in a decadal to centennial timescale after afforestation, particularly on former grasslands in mountainous regions. The majority of afforestation chrono- sequence studies focus on SOC stocks of stand ages less than 50 years (Hou et al., 2020 and references therein; https://doi.org/10.1007/s13595-020-00997-3).
RC2: Line 37-38
Afforestation is a promising measure -> is this also true for alpine regions where albedo effects might have a negative impact on climate change
Author's response: Changes in albedo could be induced by afforestation but studies such as that of Sofiadis et al., (2022; https://doi.org/10.5194/gmd-15-595-2022) and Tang et al., (2018;https://doi.org/10.3390/rs10040529) showed a summer surface cooling in forested areas in comparison to open lands. But we acknowledge that there are also other findings where the accumulation in SOC stocks is outbalanced by increased radiation following afforestation (Schwaab et al., 2015; https://doi.org/10.5194/bg-12-467-2015).
RC2: Line 49-50
Be explicit and call it “context-dependency”
Author's response: Thank you for the suggestion, we have done so.
RC2: Line 51
Initial carbon stock is a key factor -> yet you do not take this into account in your research
Author's response: That is correct, we did not measure the initial carbon stock so we will remove this point.
RC2: Line 54-55
Transition line 54 to 55 not very fluent (10-20 cm is not that deep? Was that the point?)
Author's response: We will remove qualifiers as too or not too deep and only refer to the soil horizons being analysed.
RC2: Line 58-59
Age is underexplored -> I think this is the main strength of your study that you have a large gradient in time going back 130 years
Author's response: Thank you for your assessment and we agree and will make sure that is more explicitly mentioned and discussed in the paper.
RC2: Line 75-77
Clear hypothesis. Is (iii) really needed? It is accepted as common knowledge I would say
Author's response: We do agree that in general an increase in SOC stock in the organic horizons is common knowledge. Yet, as mentioned in the previous comment, our study analysed a longer temporal frame and therefore we thought it would be interesting to see if this is still the case. In fact, we find that the highest SOC stock in the organic horizon of the 55-year-old forest, seeming to suggest that SOC peaks at some point during the reforestation process and then lowers again.
RC2: Introduction
Overall very well written introduction with nice literature review.
Author's response: Thank you very much!
RC2: Line 85
Afforested -> Encroached? Natural regeneration? Planted to avoid erosion? Not explicitly stated in the M&M. If planted, at what planting distances? What is was the basal area (the density) of the stands at the time of sampling. This is important information that is needed to interpret the results
Author's response: In our case, afforestation was due to plantations in the early 1950 as protection against avalanches.
The tree density of 58 ± 5% in the 40-year-old forest, 48 ± 3% in the 55-year-old forest, and 65 ± 5% in the 130-year-old forest. We will add this information in the next version of the manuscript.RC2: M&M
There are no replications of the different ages. The replications are made within the same forest stands resulting in pseudoreplications. This means you cannot say anything about the effect of age / time since afforestation. It can just be differences in between individual stands that do not correspond with the age effect. This should be very explicitly discussed that you make assumptions and in the discussion (and in the end of the abstract) you should be very careful with generalizing your interpreted results.
Author's response: We fully agree with your point and that we do have repeated measures of the same stand instead of repeated measures of multiple stands of the same age. We will carefully revise the manuscript to avoid overinterpreting our results.
RC2: Line 98
Five individual plots … -> not true, only for the pasture. This is misleading.
Author's response: Thanks for pointing this out, we will correct it in the updated version.
RC2: Line 100
Your forests are quite small and your plots are taken as pseudoreplications within these small stands -> weakness of the study
Author's response: Thank you for your comment. Indeed, these are important limitations of our study. Yet we believe that the results still provide interesting reference data on what happens in these types of locations, over a long afforestation period and advance interesting new hypotheses to be tested further. For example, is there really a peak in SOC at mid-afforestation period? What do these variations within stand tell us about the different soil stocks. We believe that descriptive and site-specific studies have an added value also to contribute to larger initiatives as for example larger databases, etc.
RC2: Line 100
Add also the area for the 40y old stand
Author's response: We will add that the area of the 40-year-old forest ~55 A.
RC2: Line 110
Here it becomes clear that there are only 3 replications for the forest stands
Author's response: We will make sure that this is consistent throughout the manuscript
RC2: Line 111
I appreciate the effort to make soil profile pits to have a more detailed understanding of the belowground ecosystem (as well as a complete sampling of the roots)
Author's response: Thanks!
RC2: Line 113
Roots were counted on three profile walls -> per pit? Per replication? Unclear
Author's response: The roots were counted in each soil pit on the back wall, on the left, and on the right wall. Therefore, we have 5 replicates in the pasture and 3 replicates in each forest stand age. We will clarify this in the M&M section.
RC2: Line 115
In the end the results were pooled -> be already transparent about this in the M&M
Author's response: We will make sure this is clear.
RC2: Line 130
Typo 45mm? So how many samples for the roots do you have per forest stand? 3 replications of pooled samples? In the figures it seems more? Unclear
Author's response: Roots were combined as the number of roots <2mm was too less to mill and to analyse. The number of root samples (0-5mm) was 40 in the pasture, 25 in the 40-year-old forest, 25 in the 55-year-old forest, and 26 in the 130-year-old forest. The number of replications is the number of soil pits, e.g., 5 in pasture and 3 in each forest stand. The variability in the number of root samples is not due to pooling it is simply the natural variability in the field with some soil horizons having more, some having less roots. But you are right, and we will clarify this in the M&M section and add the number of roots for the individual forest ages and the number of the combined (0-5mm) root samples and not only the total number of root samples.
RC2: Line 130
Can you discuss the results of pooling these samples in the M&M section
Author's response: That is a good point as pooling samples might be problematic due to missing information. In our study, however, we are interested in the difference in the organic carbon concentration in roots between pasture and forest areas. Combining the root samples (0-5mm) gives a more representative as well as a consistent and comparable result between these two ecosystems. And we do not expect that this pooling will have much effect on the results, yet in the discussion we return to this point when needed to interpret the results.
RC2: Line 163
I miss information about the density (do you have basal area)?
Author's response: Yes, we did not add it yet and will do so in the next version. We counted number of trees and understorey shrubs in an area of 10x10m around each soil pit.
RC2: Line 194
Not very clear “age gradient” in the results. This indicates that the differences between stands might just be random (site effects) that do not correspond with the age of the forest
Author's response: Thank you for explaining your interpretation. Our interpretation is slightly more nuanced. Indeed, when considering the carbon stock in the mineral soil, we find no differences between that of each forest stand age to that of the pasture. But when considering the C stock of the organic horizons we find that forest age 55 has significantly higher stock amounts than 40. We will make it clearer in the text.
RC2: Line 203
between
Author's response: Thanks for finding this typo, we have fixed it.
RC2: Line 224
You have a very high initial carbon stock -> could you explain more how come?
Author's response: Actually, this is not a very high initial carbon stock for alpine areas. Budge et al. (2011;https://doi.org/10.5194/bg-8-1911-2011) reported SOC stocks between 5.50 to 10.21 kg m-2 in alpine grasslands in the Swiss Alps. Garcia-Pausas et al., (2007; https://doi.org/10.1007/s10533-007-9071-9) reported mean SOC stocks of 15.3 ± 0.90 kg m-2 in mountainous grasslands in the Pyrenees and Kühnel et al., (2019; https://doi.org/10.1016/j.agee.2019.04.036) reported an SOC stock of 9.6 ±1.2 kg m-2 in mountainous grasslands in Bavaria.
RC2: Line 228
Careful with phrasing -> here you cannot say it is the effect of 40 years of afforestation for sure
Author's response: That is correct, thanks. We will remove the qualitative interpretations.
RC2: Line 232
Again very case specific
Author's response: We do agree that this is specific to our forest areas, and we will discuss it in context of what we know about alpine and temperate forests.
RC2: Line 247-248
Can you really say it is forest age?
Author's response: Yes, as the other measured soil properties such as soil pH and/or bulk densities did not change as well compared to the study of Hiltbrunner et al.(2013). The only variable that changed is the time. Although, we are fully aware that 10 years might be too short to see an effect in soil organic carbon storage.
RC2: Line 305
Here you speak of forest encroachment but you did not make this clear in the M&M. In that case how did the stand evolve?
Author's response: Thank you for bringing this up; indeed, we used the terms interchangeably and will review the text to make sure that the terminology is adequate.
RC2: Line 515
Many outliers. Which ones are from the sides and which ones from the levels? Or are they pooled in that way as well. Be clear about the experimental design
Author's response: Yes, we found many outliers in the root count. The roots were counted horizontally and on the left and right wall of the soil profile and resulted in an overall root frequency per individual plot. The data was not pooled afterwards. The variation in the root frequency lies in the natural variability in the field.
RC1: There are two things I value in the current manuscript:
- The long gradient in time (if the chronosequence would have been replicated) adding more information on the long term impacts of afforestation
- The study is well executed (- the design) and the manuscript is well written
Author's response: Thank you
RC2: My major concern:
- The study is based on pseudoreplications so the conclusions in terms of time since afforestation are not supported by the data, they could be coincidence or site effects. In my opinion this is a major issue that is hard to resolve.
Author's response: We do understand you point of “pseudoreplication” and we are aware of these limitations in our experimental design and as you state difficult to resolve. However, the study site is characterized by a homogenous soil depth and carbon distributions across the slope (Hiltbrunner et al., 2013, https://doi.org/10.1007/s10533-013-9832-6) and thus, we are confident that the detected changes therein should be caused by the different forest ages. Our sampling protocol further ensured a minimum distance of ~5m between individual plots of the same forest age class. Given the heterogeneity of soil properties and processes therein (Hiltbrunner et al., 2013, https://doi.org/10.1007/s10533-013-9832-6) we are confident of the independence of 3 to 5 observations we made within the same forest ages, and we therefore treat these observations as true statistical replicates, rather than pseudoreplicates. Further, in light of what we know these small scale and local studies can contribute with reporting results that are local and fundamental to studies such as meta-analyses.
Citation: https://doi.org/10.5194/egusphere-2023-645-AC2
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-645', Anonymous Referee #1, 19 Jun 2023
The manuscript shows the effects of a 130-year afforestation of alpine pasture on SOC stocks in the Swiss Alps. Authors found that afforestation did not necessarily change SOC stocks but clearly altered SOC dynamics. In general, the manuscript gives clear conclusions and has high scientific value. To improve the manuscript quality, the way of presenting data in figures, tables, and the Result section could be improved. In addition, the conclusion regarding words such as “increased”, “no influence”, and “moderately affected” can be connected with statistical significance (P values) in a more consistent and convincing way. Finally, several places showing inconsistent contents among figures & tables, results, and discussions should be addressed.
Abstract:
Line 13-14 “SOC 20-year-old forest” and later
Please be consistent with the usage of subscripts.
Line 22 “moderately affected the SOC stock”
I am not sure if “moderately affected” is a precise way to describe your results because SOC stocks remained constant in mineral soils but has a slight increase (P = 0.229) in organic horizons. When combining things together, I find a “slight increase in overall SOC stock” in line 15, a “moderately affected the SOC stock” in line 22, and “there is no SOC sequestration on ...” in line 25. I would suggest finding a better way to present them in order to avoid any potential confusion.
Introduction：
It might be better to divide the Introduction into several paragraphs. For example, different paragraphs might focus on topics such as the significance of studies on relevant topics, the effects of afforestation on SOM, current knowledge gaps in the relevant study area, and so on.
Line 43-44 “OM accumulates in living biomass”. It is a bit confusing to mention living biomass. Do you mean tree biomass? It could also be soil microbial biomass.
Line 75-76
To be prudent, it is suggested to specify “litter-derived carbon input” as something like aboveground or foliage litter because root litter is also a part of the litter.
Material and methods:
Line 115-123 and Line 143
How is SOC stock calculated? Is it corrected by the volume of stones? Based on the content of Line 115-123, it seems that soils collected by steel cylinders contained stones, which were removed later while passing through the 2mm sieve. I guess the bulk density should be corrected by stone volume. Is it possible to make it clearer?
Line 119
Were bulk densities measured from organic samples? You calculated C stocks of organic horizons in the Result section.
Line 144-149
Based on Table 1, it seems the depth 0 cm started from the top of mineral soils. It looks like the formula only calculates SOC stock of the mineral soils. Were SOC stocks of organic horizons calculated using the same formula?
Line 151-155
If I understand correctly, forest age and soil depth are both fixed effect predictors. Is this right? Where can I find the outcomes of mixed-effect models from Tables & figures and the Result section?
Results:
3.1 and 3.2
It is not common that the results start with two sections based on supplementary materials. Maybe some information in the supplementary materials can be incorporated into figures and tables.
Line 159 and later
It is a suggestion that it may be not necessary to present F values and P values at the same time whenever you make comparisons. Maybe only P value is enough; alternatively, even omit P values if you mention that the significant difference means P < 0.05. Instead, consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Line 159 and 167
Table S2 and Table S3?
Line 166
It seems that pinecones and branches do not match the names in Table S3. Branch = wood and twigs? Pinecones = spruce cones?
Line 175 and later
It is not necessary to mention P adj <0.95 all the time.
Line185-188
The P value seems not to be 0.801 but (nearly-) significant if you look at the values of combined SCO stocks of organic horizons. (i.e. 1.7±0.2, 1.3±0.2, and 0.8±0.1).
Discussion:
Line 244 “… a significant lower SOC stock at 15 to 25 cm…”
Is it from Table 1? It will be easier for readers if significant differences could be found directly from tables and figures.
Paragraph 1 until Line 250:
I suggested making it clear that you are talking about mineral soils (next paragraph for organic horizons).
Line 262 and after:
I think it might be helpful to make it clear when we can say something is non-significantly, significant, or unchanged. This does not mean that we need to draw a massive line between P < 0.05 and P > 0.05, but we should more or less be sure when we can say that A is higher than B. For example, it is not significant when P = 0.669 (line 245), but it is a (slightly) increased SOC sequestration when P = 0.229 (line 15, 195 and 262). I know that the latter might be a key conclusion for this manuscript, but it would be better to present it in a more convincing way.
Line 274 and after “…C:N ratios of Oi horizons increased…”
Table 1 should be Table 3?
Please be aware in lines 203-204 you said C:N ratios were not significantly different between forest ages, but you mentioned here C:N ratios increased with age. Values of 42.4±10.8 indicated a very large variation within the samples of 130-year-old soils. This is probably the reason why it is not statistically significant.
Line 287 and after
In general, organic matter input from the root is comprised of root litter and root exudates, from which the exudate is considered as the more important belowground input (Sokol et al. 2018 doi: 10.1111/nph.15361). Root exudates contain a lot of easily-decomposable (high-quality) compounds, which promote the microbial transformation of these compounds and the formation of microbial-derived compounds as stabilized SOC with low C:N ratios (Cotrufo et al. 2013, you cited). It is possible that the intensive (or effective) utilization of root exudates promotes the formation of stable microbial-derived organic matter, and therefore, the C:N ratios were still low. Therefore, my suggestion is that try to discuss it in a more convincing way with the consideration of the transformation of root exudates
Figure and Table:
In general, there is no information of significant difference in figures and table. This makes it difficult to link figures and tables to the Result and Discussion sections. Please consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Table 1
When combined organic and mineral soils, SOC was 11.6 + 0.8 =12.4 for 40-year-old soil and 11.0 + 1.3 = 12.3 for 130-year soil, respectively. In line 194, the 40-year-old soil and 130-year-old soil had 7% and 11% higher SOC stocks than pasture soil, respectively. Is this a mistake? I think both organic and mineral soils should be included when compare total SOC stocks.
Table 2
63.5 + 02.8, remove ‘0’.
Citation: https://doi.org/10.5194/egusphere-2023-645-RC1 -
AC1: 'Reply on RC1', Tatjana Carina Speckert, 19 Jul 2023
RC1: The manuscript shows the effects of a 130-year afforestation of alpine pasture on SOC stocks in the Swiss Alps. Authors found that afforestation did not necessarily change SOC stocks but clearly altered SOC dynamics. In general, the manuscript gives clear conclusions and has high scientific value. To improve the manuscript quality, the way of presenting data in figures, tables, and the Result section could be improved. In addition, the conclusion regarding words such as “increased”, “no influence”, and “moderately affected” can be connected with statistical significance (P values) in a more consistent and convincing way. Finally, several places showing inconsistent contents among figures & tables, results, and discussions should be addressed.
Author's response: Thank you for the helpful comments and feedback! To improve the presentation of the figures, we will add a map of the location within Switzerland, and we will increase the points in Figure 1. In Figure 2 and 3 we will enlarge the font size and correct the corresponding caption. In the result section, we will avoid terms like “moderately”, “slightly” etc. in case there is no statistical significance, and we will correct inconsistencies between statements and the corresponding p-values.
RC1: Abstract:
Line 13-14 “SOC 20-year-old forest” and later
Author's respose: Thanks for pointing out the inconsistency of the formatting and we will fix it throughout the revised text.
RC1: Line 22 “moderately affected the SOC stock”
I am not sure if “moderately affected” is a precise way to describe your results because SOC stocks remained constant in mineral soils but has a slight increase (P = 0.229) in organic horizons. When combining things together, I find a “slight increase in overall SOC stock” in line 15, a “moderately affected the SOC stock” in line 22, and “there is no SOC sequestration on ...” in line 25. I would suggest finding a better way to present them in order to avoid any potential confusion.
Author's response: We agree that using terminology such as “moderately” or “slightly” is not precise. We removed all the qualitative judgements and only state that there is no SOC stock change with different forest age.
RC1: Introduction：
It might be better to divide the Introduction into several paragraphs. For example, different paragraphs might focus on topics such as the significance of studies on relevant topics, the effects of afforestation on SOM, current knowledge gaps in the relevant study area, and so on.
Author's response: We agree with the suggestion and will divide the introduction into paragraphs organized by topic.
RC1: Line 43-44 “OM accumulates in living biomass”. It is a bit confusing to mention living biomass. Do you mean tree biomass? It could also be soil microbial biomass.
Author's response: Indeed, we meant tree biomass instead of living biomass, and will change the text accordingly.
RC1: Line 75-76
To be prudent, it is suggested to specify “litter-derived carbon input” as something like aboveground or foliage litter because root litter is also a part of the litter.
Author's response: We fully agree and will change the wording to “needle litter-derived carbon input”.
RC1: Material and methods:
Line 115-123 and Line 143
How is SOC stock calculated? Is it corrected by the volume of stones? Based on the content of Line 115-123, it seems that soils collected by steel cylinders contained stones, which were removed later while passing through the 2mm sieve. I guess the bulk density should be corrected by stone volume. Is it possible to make it clearer?
Author's response: Thank you for your questions. We calculated SOC stock as described in the M&M section. We correct SOC stock to stone content to the final calculation of bulk volume based on the mass of stone content obtain after sieving and assuming a rock density of 2.65 g/cm3.BD = M (weight of dried soil)/ (V (volume of the steel cylinder) – V (volume of the coarse fragments)). We will update the methods section to better reflect what we did.
RC1: Line 119
Were bulk densities measured from organic samples? You calculated C stocks of organic horizons in the Result section.
Author's response: Bulk densities for the organic horizon were not measured, but C stocks were estimated from the masses collected in an area of 0.25x0.25m which were then upscaled to m2.
RC1: Line 144-149
Based on Table 1, it seems the depth 0 cm started from the top of mineral soils. It looks like the formula only calculates SOC stock of the mineral soils. Were SOC stocks of organic horizons calculated using the same formula?
Author's response: Indeed, we did not include SOC stocks of the organic horizon to the mineral soil after calculating the two pools separately, we then added up for the final SOC stock. SOC stocks of the organic horizons were calculated differently to the mineral soil their carbon concentration was multiplied by the weight [g] of the respective sample from an area of 0.25x0.25m and then upscaled to m2. We will clarify this in the method section.
RC1: Line 151-155
If I understand correctly, forest age and soil depth are both fixed effect predictors. Is this right? Where can I find the outcomes of mixed-effect models from Tables & figures and the Result section?
Author's response: Correct, forest age and soil depth are modelled as fixed effects to test for their effect on SOC stocks. You are also correct, and we apologize for the oversight of not adding the proper reporting of statistical modelling. We will add these results in the updated version of the manuscript.
RC1: Results:
3.1 and 3.2
It is not common that the results start with two sections based on supplementary materials. Maybe some information in the supplementary materials can be incorporated into figures and tables.
Author's response: Thank you for your suggestion and we will reframe our results section to first describe key parameters, then the results of the ANOVA and finally the results of the modelling. We will further incorporate the vegetation composition into the main text.
RC1: Line 159 and later
It is a suggestion that it may be not necessary to present F values and P values at the same time whenever you make comparisons. Maybe only P value is enough; alternatively, even omit P values if you mention that the significant difference means P < 0.05. Instead, consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Author's response: We think that it is important to show the statistics value (F-value) and its corresponding p-value. Comparing the F statistic value with the theoretical value serves as threshold for our sample size and test framework. As such we will keep both for completeness. However, to keep the text clear, F-values will only be reported in a new supplementary Table and not in the text.
RC1: Line 159 and 167
Table S2 and Table S3?
Author's response: Indeed, there was a mistake in referencing the supplementary Tables. We will change this accordingly.
RC1: Line 166
It seems that pinecones and branches do not match the names in Table S3. Branch = wood and twigs? Pinecones = spruce cones?
Author's response: Thanks. you are correct and we will consistently using the term spruce cones in Table S3.
RC1: Line 175 and later
It is not necessary to mention P adj <0.95 all the time.
Author's response: We agree and will remove it where it is unnecessary.
RC1: Line185-188
The P value seems not to be 0.801 but (nearly-) significant if you look at the values of combined SCO stocks of organic horizons. (i.e. 1.7±0.2, 1.3±0.2, and 0.8±0.1).
Author's response: This simply occurred because we used the individual values for the statistical test and not the combined one. We will change this in the next version of the manuscript and use the combined values of the SOC stock to test whether forest age has a significant effect or not on total (mineral soil plus organic horizon) SOC stock.
RC1: Discussion:
Line 244 “… a significant lower SOC stock at 15 to 25 cm…”
Is it from Table 1? It will be easier for readers if significant differences could be found directly from tables and figures.
Author's response: This is the outcome of the mixed model analysis, and we did not provide those results yet as figures or tables. We will revised this in the manuscript.
RC1: Paragraph 1 until Line 250:
I suggested making it clear that you are talking about mineral soils (next paragraph for organic horizons).
Author's response: We agree and will make it clear when we are referring to organic or mineral soils to facilitate understanding. We will rephrase it in the next version of the manuscript.
RC1: Line 262 and after:
I think it might be helpful to make it clear when we can say something is non-significantly, significant, or unchanged. This does not mean that we need to draw a massive line between P < 0.05 and P > 0.05, but we should more or less be sure when we can say that A is higher than B. For example, it is not significant when P = 0.669 (line 245), but it is a (slightly) increased SOC sequestration when P = 0.229 (line 15, 195 and 262). I know that the latter might be a key conclusion for this manuscript, but it would be better to present it in a more convincing way.
Author's response: Thank you for this comment, and we fully concur with this. We will review the instances where we may have been inconsistent in our reporting. We will indeed state when something is significant only when p<0.05 and avoid qualitative statements that are not warranted by the results.
RC1: Line 274 and after “…C:N ratios of Oi horizons increased…”
Table 1 should be Table 3?
Author's response: Apologies, it should be Table 3, we will fix this typo.
RC1: Please be aware in lines 203-204 you said C:N ratios were not significantly different between forest ages, but you mentioned here C:N ratios increased with age. Values of 42.4±10.8 indicated a very large variation within the samples of 130-year-old soils. This is probably the reason why it is not statistically significant.
Author's response: We agree and apologies for the inconsistencies. We will review the instances where we may have been inconsistent in our reporting. We will indeed state when something is significant only when p<0.05 and avoid qualitative statements that are not warranted by the results.
RC1: Line 287 and after
In general, organic matter input from the root is comprised of root litter and root exudates, from which the exudate is considered as the more important belowground input (Sokol et al. 2018 doi: 10.1111/nph.15361). Root exudates contain a lot of easily-decomposable (high-quality) compounds, which promote the microbial transformation of these compounds and the formation of microbial-derived compounds as stabilized SOC with low C:N ratios (Cotrufo et al. 2013, you cited). It is possible that the intensive (or effective) utilization of root exudates promotes the formation of stable microbial-derived organic matter, and therefore, the C:N ratios were still low. Therefore, my suggestion is that try to discuss it in a more convincing way with the consideration of the transformation of root exudates
Author's response: Thank you for suggestion of a potential mechanism. Indeed, it is possible that root-exudates could be a potential source of OM with low C:N ratio and thereby be an alternative hypothesis to explain our results. We will include this in the discussion.
RC1: Figure and Table:
In general, there is no information of significant difference in figures and table. This makes it difficult to link figures and tables to the Result and Discussion sections. Please consider presenting significant differences in Figures and Table by adding P values or marking ‘a’, ‘b’, ‘c’….
Author's response: Indeed, you are correct as these were missing. We will add proper statistical analyses reporting in the corresponding tables and figures.
RC1: Table 1
When combined organic and mineral soils, SOC was 11.6 + 0.8 =12.4 for 40-year-old soil and 11.0 + 1.3 = 12.3 for 130-year soil, respectively. In line 194, the 40-year-old soil and 130-year-old soil had 7% and 11% higher SOC stocks than pasture soil, respectively. Is this a mistake? I think both organic and mineral soils should be included when compare total SOC stocks.
Author's response: This is true. We first wanted to separately show the data subsequently combine it as we think that it is valuable that mineral soil stocks remain identical, but only SOC stocks increase when considering the organic horizons. We will rephrase this.
RC1: Table 2
63.5 + 02.8, remove ‘0’.
Author's response: Apologies: we will remove the 0 in Table 2.
Citation: https://doi.org/10.5194/egusphere-2023-645-AC1
-
AC1: 'Reply on RC1', Tatjana Carina Speckert, 19 Jul 2023
-
RC2: 'Comment on egusphere-2023-645', Anonymous Referee #2, 22 Jun 2023
Speckert et al. Soil organic carbon stocks did not change after 130 years of afforestation on a former Swiss Alpine pasture
Review notes
The study presented here covers a topic that is interesting and relevant since many pastures in alpine regions are abandoned and expected to become forests in the future – which could feed back to climate change. Knowing the impact of this land-use conversion is needed. The topic and proposed research questions however are not very novel and do not contribute much new knowledge.
There are two things I value in the current manuscript:
- The long gradient in time (if the chronosequence would have been replicated) adding more information on the long term impacts of afforestation
- The study is well executed (- the design) and the manuscript is well written
My major concern:
- The study is based on pseudoreplications so the conclusions in terms of time since afforestation are not supported by the data, they could be coincidence or site effects. In my opinion this is a major issue that is hard to resolve.
Specific comments:
Line
Comment
19-22
As roots … -> unclear sentence
22-25
Is this novel enough?
37-38
Afforestation is a promising measure -> is this also true for alpine regions where albedo effects might have a negative impact on climate change
49-50
Be explicit and call it “context-dependency”
51
Initial carbon stock is a key factor -> yet you do not take this into account in your research
54-55
Transition line 54 to 55 not very fluent (10-20 cm is not that deep? Was that the point?)
58-59
Age is underexplored -> I think this is the main strength of your study that you have a large gradient in time going back 130 years
75-77
Clear hypothesis. Is (iii) really needed? It is accepted as common knowledge I would say
Introduction
Overall very well written introduction with nice literature review.
85
Afforested -> Encroached? Natural regeneration? Planted to avoid erosion? Not explicitly stated in the M&M. If planted, at what planting distances? What is was the basal area (the density) of the stands at the time of sampling. This is important information that is needed to interpret the results
M&M
There are no replications of the different ages. The replications are made within the same forest stands resulting in pseudoreplications. This means you cannot say anything about the effect of age / time since afforestation. It can just be differences in between individual stands that do not correspond with the age effect. This should be very explicitly discussed that you make assumptions and in the discussion (and in the end of the abstract) you should be very careful with generalizing your interpreted results.
98
Five individual plots … -> not true, only for the pasture. This is misleading.
100
Your forests are quite small and your plots are taken as pseudoreplications within these small stands -> weakness of the study
100
Add also the area for the 40y old stand
110
Here it becomes clear that there are only 3 replications for the forest stands
111
I appreciate the effort to make soil profile pits to have a more detailed understanding of the belowground ecosystem (as well as a complete sampling of the roots)
113
Roots were counted on three profile walls -> per pit? Per replication? Unclear
115
In the end the results were pooled -> be already transparent about this in the M&M
130
Typo 45mm? So how many samples for the roots do you have per forest stand? 3 replications of pooled samples? In the figures it seems more? Unclear
130
Can you discuss the results of pooling these samples in the M&M section
163
I miss information about the density (do you have basal area)?
194
Not very clear “age gradient” in the results. This indicates that the differences between stands might just be random (site effects) that do not correspond with the age of the forest
203
between
224
You have a very high initial carbon stock -> could you explain more how come?
228
Careful with phrasing -> here you cannot say it is the effect of 40 years of afforestation for sure
232
Again very case specific
247-248
Can you really say it is forest age?
305
Here you speak of forest encroachment but you did not make this clear in the M&M. In that case how did the stand evolve?
515
Many outliers. Which ones are from the sides and which ones from the levels? Or are they pooled in that way as well. Be clear about the experimental design
-
AC2: 'Reply on RC2', Tatjana Carina Speckert, 19 Jul 2023
Review notes
RC2: The study presented here covers a topic that is interesting and relevant since many pastures in alpine regions are abandoned and expected to become forests in the future – which could feed back to climate change. Knowing the impact of this land-use conversion is needed. The topic and proposed research questions however are not very novel and do not contribute much new knowledge.
Author's response: Thank you for your assessment. We are aware that we work at a single site planted with a single tree species and is therefore rather case specific. However, finding a study site with such a long-term afforestation chrono-sequence with four forest stands of different age classes and a reference pasture area on a homogenous slope is unique. Moreover, the location (subalpine region) is highly relevant as studies concerning the organic carbon stock in the Alps are scarce (Thuille and Schulze, 2006, https://doi.org/10.1111/j.1365-2486.2005.01078.x; Zimmermann et al., 2010, https://doi.org/10.1016/j.agee.2010.06.010) and there is an ongoing process of subalpine pastures abandonment and forest encroachment, which policy makers consider as beneficial in terms of carbon storage. However, recent studies (Clemmensen et al., 2014, https://doi.org/10.1111/nph.13208; Friggens et al., 2020, https://doi.org/10.1111/gcb.15229) show evidence that priming, mycorrhizal nutrient mining associated with ectomycorrhizal tree species can result in a net carbon loss from the mineral soil. We therefore used this afforestation chrono-sequence as a natural experiment to test whether there is an increased carbon sequestration with increasing forest age.
RC2: Line 19-22
As roots … -> unclear sentence
Author's response: We will change this sentence to make it clearer: “ Due to the high root C:N ratio (pasture 63.5 ± 2.8 and forests between 54.7 ± 3.9 and 61.2 ± 2.9), particulate root-derived organic matter seems to have a rather small effect on the forest soil C:N ratio as well as on the SOC stock accumulation in the mineral soil.
RC2: Line 22-25
Is this novel enough?
Author's response: As far as we know there are not many studies reporting SOC stocks in a decadal to centennial timescale after afforestation, particularly on former grasslands in mountainous regions. The majority of afforestation chrono- sequence studies focus on SOC stocks of stand ages less than 50 years (Hou et al., 2020 and references therein; https://doi.org/10.1007/s13595-020-00997-3).
RC2: Line 37-38
Afforestation is a promising measure -> is this also true for alpine regions where albedo effects might have a negative impact on climate change
Author's response: Changes in albedo could be induced by afforestation but studies such as that of Sofiadis et al., (2022; https://doi.org/10.5194/gmd-15-595-2022) and Tang et al., (2018;https://doi.org/10.3390/rs10040529) showed a summer surface cooling in forested areas in comparison to open lands. But we acknowledge that there are also other findings where the accumulation in SOC stocks is outbalanced by increased radiation following afforestation (Schwaab et al., 2015; https://doi.org/10.5194/bg-12-467-2015).
RC2: Line 49-50
Be explicit and call it “context-dependency”
Author's response: Thank you for the suggestion, we have done so.
RC2: Line 51
Initial carbon stock is a key factor -> yet you do not take this into account in your research
Author's response: That is correct, we did not measure the initial carbon stock so we will remove this point.
RC2: Line 54-55
Transition line 54 to 55 not very fluent (10-20 cm is not that deep? Was that the point?)
Author's response: We will remove qualifiers as too or not too deep and only refer to the soil horizons being analysed.
RC2: Line 58-59
Age is underexplored -> I think this is the main strength of your study that you have a large gradient in time going back 130 years
Author's response: Thank you for your assessment and we agree and will make sure that is more explicitly mentioned and discussed in the paper.
RC2: Line 75-77
Clear hypothesis. Is (iii) really needed? It is accepted as common knowledge I would say
Author's response: We do agree that in general an increase in SOC stock in the organic horizons is common knowledge. Yet, as mentioned in the previous comment, our study analysed a longer temporal frame and therefore we thought it would be interesting to see if this is still the case. In fact, we find that the highest SOC stock in the organic horizon of the 55-year-old forest, seeming to suggest that SOC peaks at some point during the reforestation process and then lowers again.
RC2: Introduction
Overall very well written introduction with nice literature review.
Author's response: Thank you very much!
RC2: Line 85
Afforested -> Encroached? Natural regeneration? Planted to avoid erosion? Not explicitly stated in the M&M. If planted, at what planting distances? What is was the basal area (the density) of the stands at the time of sampling. This is important information that is needed to interpret the results
Author's response: In our case, afforestation was due to plantations in the early 1950 as protection against avalanches.
The tree density of 58 ± 5% in the 40-year-old forest, 48 ± 3% in the 55-year-old forest, and 65 ± 5% in the 130-year-old forest. We will add this information in the next version of the manuscript.RC2: M&M
There are no replications of the different ages. The replications are made within the same forest stands resulting in pseudoreplications. This means you cannot say anything about the effect of age / time since afforestation. It can just be differences in between individual stands that do not correspond with the age effect. This should be very explicitly discussed that you make assumptions and in the discussion (and in the end of the abstract) you should be very careful with generalizing your interpreted results.
Author's response: We fully agree with your point and that we do have repeated measures of the same stand instead of repeated measures of multiple stands of the same age. We will carefully revise the manuscript to avoid overinterpreting our results.
RC2: Line 98
Five individual plots … -> not true, only for the pasture. This is misleading.
Author's response: Thanks for pointing this out, we will correct it in the updated version.
RC2: Line 100
Your forests are quite small and your plots are taken as pseudoreplications within these small stands -> weakness of the study
Author's response: Thank you for your comment. Indeed, these are important limitations of our study. Yet we believe that the results still provide interesting reference data on what happens in these types of locations, over a long afforestation period and advance interesting new hypotheses to be tested further. For example, is there really a peak in SOC at mid-afforestation period? What do these variations within stand tell us about the different soil stocks. We believe that descriptive and site-specific studies have an added value also to contribute to larger initiatives as for example larger databases, etc.
RC2: Line 100
Add also the area for the 40y old stand
Author's response: We will add that the area of the 40-year-old forest ~55 A.
RC2: Line 110
Here it becomes clear that there are only 3 replications for the forest stands
Author's response: We will make sure that this is consistent throughout the manuscript
RC2: Line 111
I appreciate the effort to make soil profile pits to have a more detailed understanding of the belowground ecosystem (as well as a complete sampling of the roots)
Author's response: Thanks!
RC2: Line 113
Roots were counted on three profile walls -> per pit? Per replication? Unclear
Author's response: The roots were counted in each soil pit on the back wall, on the left, and on the right wall. Therefore, we have 5 replicates in the pasture and 3 replicates in each forest stand age. We will clarify this in the M&M section.
RC2: Line 115
In the end the results were pooled -> be already transparent about this in the M&M
Author's response: We will make sure this is clear.
RC2: Line 130
Typo 45mm? So how many samples for the roots do you have per forest stand? 3 replications of pooled samples? In the figures it seems more? Unclear
Author's response: Roots were combined as the number of roots <2mm was too less to mill and to analyse. The number of root samples (0-5mm) was 40 in the pasture, 25 in the 40-year-old forest, 25 in the 55-year-old forest, and 26 in the 130-year-old forest. The number of replications is the number of soil pits, e.g., 5 in pasture and 3 in each forest stand. The variability in the number of root samples is not due to pooling it is simply the natural variability in the field with some soil horizons having more, some having less roots. But you are right, and we will clarify this in the M&M section and add the number of roots for the individual forest ages and the number of the combined (0-5mm) root samples and not only the total number of root samples.
RC2: Line 130
Can you discuss the results of pooling these samples in the M&M section
Author's response: That is a good point as pooling samples might be problematic due to missing information. In our study, however, we are interested in the difference in the organic carbon concentration in roots between pasture and forest areas. Combining the root samples (0-5mm) gives a more representative as well as a consistent and comparable result between these two ecosystems. And we do not expect that this pooling will have much effect on the results, yet in the discussion we return to this point when needed to interpret the results.
RC2: Line 163
I miss information about the density (do you have basal area)?
Author's response: Yes, we did not add it yet and will do so in the next version. We counted number of trees and understorey shrubs in an area of 10x10m around each soil pit.
RC2: Line 194
Not very clear “age gradient” in the results. This indicates that the differences between stands might just be random (site effects) that do not correspond with the age of the forest
Author's response: Thank you for explaining your interpretation. Our interpretation is slightly more nuanced. Indeed, when considering the carbon stock in the mineral soil, we find no differences between that of each forest stand age to that of the pasture. But when considering the C stock of the organic horizons we find that forest age 55 has significantly higher stock amounts than 40. We will make it clearer in the text.
RC2: Line 203
between
Author's response: Thanks for finding this typo, we have fixed it.
RC2: Line 224
You have a very high initial carbon stock -> could you explain more how come?
Author's response: Actually, this is not a very high initial carbon stock for alpine areas. Budge et al. (2011;https://doi.org/10.5194/bg-8-1911-2011) reported SOC stocks between 5.50 to 10.21 kg m-2 in alpine grasslands in the Swiss Alps. Garcia-Pausas et al., (2007; https://doi.org/10.1007/s10533-007-9071-9) reported mean SOC stocks of 15.3 ± 0.90 kg m-2 in mountainous grasslands in the Pyrenees and Kühnel et al., (2019; https://doi.org/10.1016/j.agee.2019.04.036) reported an SOC stock of 9.6 ±1.2 kg m-2 in mountainous grasslands in Bavaria.
RC2: Line 228
Careful with phrasing -> here you cannot say it is the effect of 40 years of afforestation for sure
Author's response: That is correct, thanks. We will remove the qualitative interpretations.
RC2: Line 232
Again very case specific
Author's response: We do agree that this is specific to our forest areas, and we will discuss it in context of what we know about alpine and temperate forests.
RC2: Line 247-248
Can you really say it is forest age?
Author's response: Yes, as the other measured soil properties such as soil pH and/or bulk densities did not change as well compared to the study of Hiltbrunner et al.(2013). The only variable that changed is the time. Although, we are fully aware that 10 years might be too short to see an effect in soil organic carbon storage.
RC2: Line 305
Here you speak of forest encroachment but you did not make this clear in the M&M. In that case how did the stand evolve?
Author's response: Thank you for bringing this up; indeed, we used the terms interchangeably and will review the text to make sure that the terminology is adequate.
RC2: Line 515
Many outliers. Which ones are from the sides and which ones from the levels? Or are they pooled in that way as well. Be clear about the experimental design
Author's response: Yes, we found many outliers in the root count. The roots were counted horizontally and on the left and right wall of the soil profile and resulted in an overall root frequency per individual plot. The data was not pooled afterwards. The variation in the root frequency lies in the natural variability in the field.
RC1: There are two things I value in the current manuscript:
- The long gradient in time (if the chronosequence would have been replicated) adding more information on the long term impacts of afforestation
- The study is well executed (- the design) and the manuscript is well written
Author's response: Thank you
RC2: My major concern:
- The study is based on pseudoreplications so the conclusions in terms of time since afforestation are not supported by the data, they could be coincidence or site effects. In my opinion this is a major issue that is hard to resolve.
Author's response: We do understand you point of “pseudoreplication” and we are aware of these limitations in our experimental design and as you state difficult to resolve. However, the study site is characterized by a homogenous soil depth and carbon distributions across the slope (Hiltbrunner et al., 2013, https://doi.org/10.1007/s10533-013-9832-6) and thus, we are confident that the detected changes therein should be caused by the different forest ages. Our sampling protocol further ensured a minimum distance of ~5m between individual plots of the same forest age class. Given the heterogeneity of soil properties and processes therein (Hiltbrunner et al., 2013, https://doi.org/10.1007/s10533-013-9832-6) we are confident of the independence of 3 to 5 observations we made within the same forest ages, and we therefore treat these observations as true statistical replicates, rather than pseudoreplicates. Further, in light of what we know these small scale and local studies can contribute with reporting results that are local and fundamental to studies such as meta-analyses.
Citation: https://doi.org/10.5194/egusphere-2023-645-AC2
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 415 | 130 | 16 | 561 | 43 | 8 | 8 |
- HTML: 415
- PDF: 130
- XML: 16
- Total: 561
- Supplement: 43
- BibTeX: 8
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
Tatjana Carina Speckert
Jeannine Suremann
Konstantin Gavazov
Maria Joao Santos
Frank Hagedorn
Guido Lars Bruno Wiesenberg
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1057 KB) - Metadata XML
-
Supplement
(187 KB) - BibTeX
- EndNote
- Final revised paper