Management Intensity Controls Nitrogen-Use-Efficiency and Flows in Grasslands—A ¹⁵N Tracing Experiment

Round 1

Reviewer 1 Report

The authors say they reduced the amount of text to the max, but I am still not happy with the language. I feel it is still lacking logic rigor.
- Example l. 35-38: The sentence quoted below says ‘biodiversity is vital for biodiversity’. Please compare
≫ Furthermore, they are 35 characterized by high plant biodiversity, especially under still common traditional extensive 36 management [5], which is vital for the maintenance of ecosystem services such as biodiversity 37 and socioeconomic services [3,4]. ≪
- Example l. 38-41: Textbook-like ‘explanations’ continue to slow down the story. In an agronomy journal it is not necessary to establish a line of argument like ‘productivity > forage > dairy- and cattle farming > economic value’.
Please compare
≫ While 38 high SOC and TN contents are a crucial factor in the provision of ecological soil functions such as 39 nutrient retention and groundwater protection, they also facilitate high productivity [6,7], thus 40 economic value via forage production for dairy- and cattle farming. ≪
- Example l.53-54: example for redundancies
≫ … slurry application rates under intensive grassland management in 53 the pre-alpine region have been as high as up to 305 kg N ha-1 yr-1 [14]. ≪
- Example l.117-118: just plain sloppyness
≫ … fertilization events with liquid cattle 117 surface application). ≪
Liquid cattle application would really make this an extremely hard to compare study.
- Example l. 307-308: Some words are apparently interspersed at random. Why ’thus’? Please compare https://www.merriam-webster.com/dictionary/thus
I doubt the authors would produce such a sentence in their native language.
≫ Total recovery of applied 15N fertilizer in the soil up to 60 cm depth and plant biomass ranged 307 between 48 – 50 % and thus was neither affected by management nor by climate change.≪
I have to say that I am really tired to either ignore this stuff, or pick it out, line by line and demand it getting fixed. I am sure the authors are totally capable to detect such issues themselves.
Also, the revised MS I received is a .pdf with track changes made visible.
First, I don’t want to have read around all the old issues of the text and piece the new solutions together myself.
After all these language issues, I will pick up at least one (!) scientific issue from my first review. In ‘Conclusion’ and as a general motivation for the study the authors keep stressing the risk of N-mining on a decadal scale and the likely detrimental consequences for soil ecosystem services. The N-loss is assumed to be 86 kg N ha-1 year-1 in the intensive mgmt. In the six years of the experiment this would have amounted to more than half a ton of soil N lost. If this loss was real, a grassland site at any altitude, that is under intensive mgmt for 30 years, would have lost c- 15% of
its soil N stock. Why are no such numbers presented or why is their absence not discussed?
The authors seemingly misunderstood this comment in in my previous review and focus on altitude and climate (that have no effect in their study) instead. Compare:
“But also the applied management practices already are in place for decades, yet there is no indication of a detrimental SOM decline.”
>> We do not agree with regard to intensive management. Intensive management has hardly been used at high elevation grasslands like that investigated here, where extensive management has been strongly dominating. At high elevation only recently there is a trend towards intensification due to expanded growing seasons, and possibly also due to new legal frameworks such as the German fertilizer ordinance. Furthermore, a detrimental decline of SOM can only be seen over long time spans and such studies are hardly available for such sites and soils.
To cut a long story short, I am not happy with the new manuscript. The authors politely mended the issues highlighted as examples in my first review, but in essence the manuscript remained the same.

Author Response

The authors say they reduced the amount of text to the max, but I am still not happy with the language. I feel it is still lacking logic rigor.
- Example l. 35-38: The sentence quoted below says ‘biodiversity is vital for biodiversity’. Please compare
≫ Furthermore, they are 35 characterized by high plant biodiversity, especially under still common traditional extensive 36 management [5], which is vital for the maintenance of ecosystem services such as biodiversity 37 and socioeconomic services [3,4]. ≪

>> We had adjusted this sentence according to another reviewers’ comment, who asked us to mention the ecosystem services we refer to. In order to comply with this comment, we adjusted the sentence in the revised manuscript: “Furthermore, the traditional extensive management [5] remains common and is vital for the maintenance of ecosystem services such as biodiversity and socioeconomic services [3,4].”

- Example l. 38-41: Textbook-like ‘explanations’ continue to slow down the story. In an agronomy journal it is not necessary to establish a line of argument like ‘productivity > forage > dairy- and cattle farming > economic value’.
Please compare
≫ While 38 high SOC and TN contents are a crucial factor in the provision of ecological soil functions such as 39 nutrient retention and groundwater protection, they also facilitate high productivity [6,7], thus 40 economic value via forage production for dairy- and cattle farming. ≪

>> We omitted this causality from the sentence and changed it to: “High SOC and TN contents are a crucial factor in the provisioning of ecological soil functions such as nutrient retention and groundwater protection, while also facilitating high plant productivity [6,7].”

- Example l.53-54: example for redundancies
≫ … slurry application rates under intensive grassland management in 53 the pre-alpine region have been as high as up to 305 kg N ha-1 yr-1 [14]. ≪

>>  We assume the redundancy refers to the “as high as up to”, since the previous amounts of fertilizer application were not mentioned elsewhere. We changed the sentences as follows:

“Beforehand, slurry application rates in the pre-alpine region have been up to 305 kg N ha-1 yr-1 [18].”

- Example l.117-118: just plain sloppyness
≫ … fertilization events with liquid cattle 117 surface application). ≪
Liquid cattle application would really make this an extremely hard to compare study.

>> Thank you for the thorough reading, indeed a word was missing here. We added “slurry” in the revised manuscript.

- Example l. 307-308: Some words are apparently interspersed at random. Why ’thus’? Please compare https://www.merriam-webster.com/dictionary/thus
I doubt the authors would produce such a sentence in their native language.

≫ Total recovery of applied 15N fertilizer in the soil up to 60 cm depth and plant biomass ranged 307 between 48 – 50 % and thus was neither affected by management nor by climate change.≪

>> The use of “thus” in this sentence is in fact wrong, or at least misleading. We wanted to express that the differences, which were present between management treatments for recovery in soil and plant biomass were not present for the overall recovery. This causality is in this case however of minor importance. Therefore, we omitted “thus” from the revised manuscript.

I have to say that I am really tired to either ignore this stuff, or pick it out, line by line and demand it getting fixed. I am sure the authors are totally capable to detect such issues themselves.

>> In order to improve the language quality, a native speaker and fellow scientist working on these grasslands carefully read and improved the English. While it is difficult for non-native speakers to achieve the language quality by that of native speakers, we do hope that the manuscript now fulfills the language requirements of the journal.

Also, the revised MS I received is a .pdf with track changes made visible.
First, I don’t want to have read around all the old issues of the text and piece the new solutions together myself.

>> Thank you for that comment. It is a requirement of the journal to submit the revised manuscript including track changes: “Any revisions should be clearly highlighted, for example using the "Track Changes" function in Microsoft Word”.

After all these language issues, I will pick up at least one (!) scientific issue from my first review. In ‘Conclusion’ and as a general motivation for the study the authors keep stressing the risk of N-mining on a decadal scale and the likely detrimental consequences for soil ecosystem services. The N-loss is assumed to be 86 kg N ha-1 year-1 in the intensive mgmt. In the six years of the experiment this would have amounted to more than half a ton of soil N lost. If this loss was real, a grassland site at any altitude, that is under intensive mgmt for 30 years, would have lost c- 15% of
its soil N stock. Why are no such numbers presented or why is their absence not discussed?

The authors seemingly misunderstood this comment in in my previous review and focus on altitude and climate (that have no effect in their study) instead. Compare:
“But also the applied management practices already are in place for decades, yet there is no indication of a detrimental SOM decline.”
>> We do not agree with regard to intensive management. Intensive management has hardly been used at high elevation grasslands like that investigated here, where extensive management has been strongly dominating. At high elevation only recently there is a trend towards intensification due to expanded growing seasons, and possibly also due to new legal frameworks such as the German fertilizer ordinance. Furthermore, a detrimental decline of SOM can only be seen over long time spans and such studies are hardly available for such sites and soils.

>> It was not a misunderstanding, but we do not want to generalize our findings to other soils and altitude. And we also want to generalize over longer time spans such as 30 years only with great care for the reasons given below.

We agree with some of the reviewers statements: It is correct that at lower elevations, intensive management is in place for decades. While a decline of ~500kg N ha-1 (a change by less than 3 % of the N stocks of 18 t ha-1 in six years) will not be detectable by TN stock measurements, a 30 year change of the annual effect size reported here should be measurable. However, a generalization of our findings “for 30 years” and for “any altitude” appears too speculative to us for two reasons. The first reason is that we investigated intensification effects in the first years after management change, while the annual effect size might be different at decadal timescales, e.g., get smaller due to exhaustion of easily mineralizable N. Second, at lower altitudes, the N contents are much smaller and thus management-induced changes in TN stocks could be different as well. Hence, we see this study as a case study on a typical soil that is found at high altitude, and on a recent change in management at this altitude. We do not want to get overly speculative by too much generalization.  But, as in fact management-induced changes in TN stocks at decadal time scales are hardly reported, we just started to sample paired sites of decadal-scale management differences (extensive vs. intensive). Unfortunately in our target region, such pairs were – for the given reasons - only found at lower elevation than that of the current study. However, this study will likely shed light on the magnitude of N mining at longer time scales.

To cut a long story short, I am not happy with the new manuscript. The authors politely mended the issues highlighted as examples in my first review, but in essence the manuscript remained the same.

>> While we do not agree with the reviewer in every respect and have tried to highlight where and why we disagree, we appreciate their comments, as they were helpful to improve the manuscript. We hope they find the latest round of revisions satisfactory.

Reviewer 2 Report

In my opinion this improved version of your manusript is good for reading and understanding the presented scientific problem.

Author Response

Thank you for this positive feedback. We further improved the English language as it was recommended by the reviewer.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Review of MS agronomy-708454-peer-review-v1.pdf

Management intensity controls N-use-efficiency and flows in grasslands- a 15N tracing experiment

Marcus Zistl-Schlingmann, Steve Kwatcho Kengdo, Ralf Kiese and Michael Dannenmann

Content

Large montane-grassland lysimeters are exposed to warming + drying and to 2 levels of management intensity. Using 15N tracer an N-balance for the grassland is established. Authors hypothesize that grassland productivity and N exports increase as a result of climate change. Also, the intensive management is expected to counteract a potential soil N mining. The study finds the N-balance to be negative for all treatment combinations. Climate has no effect, but intensive management increases the N-loss. Aboveground yield increases by 20% with 1.6° warming. It is concluded that, on a decadal scale, the observed N-mining will deplete SON and will thus be a threat to soil functions in the long run.

Review

I question the Conclusion and ask the authors to respond to this: The lack of a climate treatment response of N makes it unlikely that the loss is temperature driven. But also the applied management practices already are in place for decades, yet there is no indication of a detrimental SOM decline. Instead, it is likely that the estimated N-loss in both managements (50-100 kg ha^-1 yr^-1) is part of a (soil) OM-loss, equivalent to a loss of c. 0.5-1 ton C ha^-1 yr^-1. This is a quite undramatic, interannual difference in a SOC-rich grassland, that may well have a positive sign next year. Therefore, the available data do not support the assumption of the predicted long-term N-loss, but instead suggest a specific weather effect of the experimental year 2017, representing a short-term fluctuation within an apparent mid-term steady state.

The experimental design is the reason for limited use of study from a mechanistic perspective: A) Only two treatment levels, therefore no dose/response analysis possible. B) Quanitfied parameters of climate change treatment are warmth and soil water content. Design does not allow to distinguish between their interacting effects. C) The one-year-only analysis does not allow to analyze how weather effects interact with management and climate.

Also, it is unclear whether the intensive/extensive management constitutes a continuation of their respective management-history for all plots. If not, one group is in a state of dynamic change, while the other one is under ± steady state conditions. The fact that the experiment was already going on for six years before the test mitigates the problem somewhat.

The language needs improvement, mostly in the sense of becoming more concise. Often, sentences are too long, due to redundancies/repetitions. The text needs to become 30-50% shorter. Example:

295 ff  The key characteristics of the N fluxes in the investigated montane grassland with a Haplic Cambisol soil were the high plant productivity and associated plant N exports that even exceeded total N inputs by fertilization.

Sections

Mat & Meth

Please also indicate the difference in growing season temperature and length of growing period/snow free period.

Results

From a distance the Results section looks like a Discussion. Please avoid the ‘putting into perspective’, as this should happen in Discussion. Instead, be much shorter and focus on data that are relevant for the Discussion.

Non-comprehensive list of specific comments that serve as examples of typical issues:

18        ‘explicitly’ reword

21        ‘deteriorate’ reword

22        ‘suggest a risk of soil N mining’

1) microbial soil N mining?

2) isn’t the risk rather the N-leaking into non-agronomic ecosystems?

3) in my understanding ‘microbial N mining’ describes the process of microbes accessing previously unaccessed N sources in the context of the decomposition of organic material. Please explain why this poses a risk or consider the use of a different word (depletion of … ?)

36        both references have nothing to do with ecosystem services

40-42   Syntax problem. Please reformulate

42        Omit dispensable ‘progressing’

44        In my experience topography is the factor that keeps agriculture from intensifying the management. Topography will not change with warming. But undoubtedly the climate warming will reduce thermal growth limitations.

44-45   This contrasting makes me curious whether and how you define your ‘pre-alpine’ later on, because ‘montane’ is ‘pre-alpine’ in the first place. I am confused, because I have the impression that here and throughout the MS the term ‘pre-alpine’ was used as if it was an elevational zone, which it is not.

46        consider replacing ‘speed’ with ‘rate’.

76        consider avoiding ‘holistic’ if you simply want to say (eg) comprehensive or mechanistic.

91        860 m

100-102          Confused again: What are ‘replicated sites’? Omit? Suggested wording:

In 2011 a total of 12 large intact soil cores (1m2, 1.5 m height) from within 5 km distance from the HE site were excavated without disturbance of plants and soil and transported to the HE site [34].

103      Suggestion: ‘Lysimeters were fully installed and operated since mid to the end of 2011 and …’

105      Omit ‘16’

106-109          1) Did the grassland lysimeters have an intensive and extensive management, respectively, already before the experiment? If not one group experiences a change, the other one a continuation of management. Please explain.

2) Tables show 3 harvests for Ext. treatment, but here you say it has only 1-2 cuts per year. Why is that?

163-175          The N balance is essential for this study and the contribution of the biological N fixation (BNF) is assumed to be very low. I find the explanation of BNF-estimation not satisfying: The BNF input is estimated based on a publication focusing specifically on the potentially much smaller contribution of asymbiotic (as opposed to symbiotic) BNF. There is no mention of the potentially extremely wide range of BNF, depending on climate, cutting management, fertilization and species composition. All these factors vary in the experiment and have the potential to completely change the balance. It is therefore necessary to be specific about how BNF was calculated.

180      Please reword. You did not compare treatments, but yields, etc., right?

180-181 and Table 1  Comparing yields for Intensive (I) vs. Extensive (E) treatment, when they happen to coincide does not provide a meaningful information since it suggests higher productivity with extensive management.

209      How is this possible: Single measurements have no SE!?

211      Within one climate treatment?

213      What is the purpose of Table 2? Omit?

Reviewer 2 Report

 (Management intensity controls N-use-efficiency and flows in grasslands – a ¹⁵N tracing experiment) submitted to Agronomy. This study attempts to follow N partitioning and losses in a montane grassland following fertilization with a ¹⁵N spiked manure-based slurry. Plant biomass production, N content and N recovery as indicated by isotopic content was measured in lysimeters. In addition, the authors looked at climate change by using two elevations to provide different temperature regimes and moving lysimeter soils and plants from one site to the other. Although, the experimental design has basic flaws in relation to climate change, the overall methodology appears sound. The paper is well written and may be of interest to agronomists interested in slurry applications as a way to either reduce farm wastes or to fertilize crops.

The main problem with this paper is the results are not particularly interesting with few significant or substantive differences to highlight. Negative results are in themselves not the problem as scientific knowledge advances as we knock down hypotheses in our quest to find truth. This journal may very well publish these result in order to provide the scientific community with a data set from which to build other studies or to formulate different hypotheses. However, negative results are valuable only if the experiment was well designed and carefully executed.

This study has a major flaw when assessing the effects of climate change. The argument is that by moving the contents of a lysimeter from higher to lower elevations they are simulating an approximate 2 ^oC shift in temperature and reduced precipitation. Leaving aside other uncontrolled variables, the authors made a fatal mistake in assuming temperature and moisture inputs are the only factors that need consideration. According to basic global warming concepts the reason temperatures increase is due to rising greenhouse gases and in particular CO₂. Carbon dioxide is a critical input to photosynthesis and as such will modify numerous physiological processes including C gain itself. The study should have included an elevated CO₂ treatment along with the increased temperature. I realize that elevating CO₂ is not trivial in a system such as this, but it is hard to argue that the design can say anything meaningful about our future world without taking into consideration the impact CO₂ levels can have on any ecosystem. The authors need to at least acknowledge the design flaw and maybe model impacts of elevated CO₂. The results did not show any effect of climate change on the N economy, and so this is not a major component of the paper. Climate change is discussed, however, and so issues that led to non-significant results (like CO₂) must be acknowledged.     

Another experimental problems was labelling (with ¹⁵N) mineral sources of N and not the slurry components themselves. While I understand labelling the manure would be nearly impossible, the ¹⁵N tracer is following mineral N flows and not organic N. It may well be that slurry organic matter mineralization might occur rapidly, but the authors do not actually test the validity of substituting mineral N inputs for tracing organic N inputs. In addition, since gaseous loss of N was not measured in this study, there appears to be a great deal of N flux unaccounted for, which weakens and already less than impressive set of results.

Specific comments.

1. Lines 68 and 69 seem to be hyperbole.

2. Line 97. How certain are we that precipitation will decline in a warmer world? Not all model indicate as such and I would like to know if you have long-term observations to support this contention or at least well-validated models that do.

3. Lines 145-147. This is not an adequate description of stable isotope analysis. What model of IRMS was utilized, what kind sample replication was employed, and what is the long-term precision for d15N does the lab report?

4. Figure 2 and elsewhere. Can you clarify what you mean by “export of plant N”? I think I understand what you are measuring but this could be confusing to many as plants generally do not export N.

5. Figure 3. 15N, the 15 should be superscripted in the axis legend.

6. Line 264. You cannot say the values are “slightly lower” if there were no significant differences. You can only say that they were not different.

7. Figures 2 and 3. The use of the two asterisks for different test is confusing.

8. Line 34-37. I am not sure your data set can support this statement with confidence.

9. Line 374 and elsewhere. It seems that much of this study assumes that N is at least partially limiting for plant productivity. Could a lack of N limitation in this system exlain some of your findings?

Reviewer 3 Report

The paper is well written and presents a well-conducted study of the fate of nitrogen in montane grasslands (2 field sites at different elevation) under different types of agricultural land use, intensive vs. extensive farming.I have only a few minor comments and suggestions.

Line 18: 'explicitly' is maybe not the best word to use here as it often indicate intent. Maybe 'significantly' is better.

Line 63 and other places. The subscript in N2O (and other subscripts and superscripts throughout) does not have the right formatting. I'm not sure if that is a problem only with my version or this font, but it should be checked.   

Figure 1: It is confusing that the colours red and blue are used in all of a, b, c as they signify temperature and precipitation in a and b but different elevation in c. If you use different colours in c, it will make the presentation nicer. 

Table 1 & 2: The use of lower case and capital letters here is very confusing, what is signified by A, a, B, and b? Would it be better with a column header and tick marks? Or maybe provide onem or two examples as to how this is calculated.

Figure 4: I agree that the trend for the intensive use field sites are not linear and look like it is exponnetial, but I don't think thant you can say that the trends for the extensive use field sites is linear and not exponential. To me it looks like it is exponential, the HE site has a high value at the topmost sample which fites poorly on a linear trend and the LE site has a very high uncertainty in the the top most sample, so it can fit with almost any type of trend. Linear and exponential fits (and goodness of fit) to the data should be presented and also a discussion of ehat an exponential versus a linear tred indicates.