the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Oct 2023
Nitrifier denitrification potentially dominates N2O production in a sandy soil – results from different fertilization and irigation regimes in potato cropping in Germany
Abstract. Spatial and temporal distribution of water and nitrogen supply affects soil-borne nitrous oxide (N2O) emissions. In this study, the effects of different irrigation technologies (no irrigation, sprinkler irrigation and drip irrigation) and nitrogen (N) application types (broadcasted and dissolved in irrigation water) on N2O emissions and the potentially underlying, genetically determined microbial processes were investigated over an entire season in potato cropping. N2O fluxes were highest during the first half of the season and mostly affected by the applied water volume rather than the N application types. The comparison of the different water application types revealed that nitrifier denitrification might potentially be the dominant source of N2O emissions, especially under sprinkler irrigation. The type of N fertilizer supply, broadcasted application or dissolved in irrigation water, showed only minor differences in the potential microbial community functionality. N2O production in both treatments was most likely also dominated by nitrifier denitrification, while the process of denitrification might be feasible too. Even though the current agronomic management measures generally meet the crop demand of water and N, it might be recommendable to adapt the time of application considering that potatoes mainly require N at later growth stages which could also reduce N2O emissions at the same time.
- Preprint
(833 KB) - Metadata XML
-
Supplement
(968 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-2277', Anonymous Referee #1, 13 Nov 2023
The authors investigate the influence of fertilization and irrigation on N2O emissions and on the abundance of functional genes. They present a comprehensive measurement program, although I would have chosen a different approach for studies on the question posed in the title. I therefore recommend changing the title and focusing on what is actually shown. The cited preliminary study Storch et al. 2023 could certainly serve as a model.
I think it is necessary for the authors to critically check all their statements again to see whether the suggestion of causality can really be maintained or whether only the correlation should be described. The assumptions made should also be discussed more in the context of existing experiments on the underlying processes. The previous isotope work in potato crops assumes denitrification, which, however, is particularly strong in the furrows and even stronger in the lanes, both of which have not been investigated if I understand correctly.
I would like to illustrate this with a few points below.
L16: In my opinion, the spatial distribution of soil moisture is not adequately described, especially the differences between ridges and furrows in potato crops is mentioned, but is not the subject of the study, at least not the results.
L18: I would rather speak here of genes potentially involved than of processes
L78ff: Please note the size of the plots.
L81: The amount of precipitation measured during the study period is at least as important as the annual precipitation.
L96: A continuous color scheme would be helpful, e.g. light and dark brown for ZI-ZN and ZI-N light and dark chalk blue for SI-ZN and SI-N light and dark green for DI-ZN and DI-N , a dark color (because irrigated) for F (e.g. purple or red) and black like black fallow for F-ZC. this can then also be used for the table.
L106 14.4+1.1 is unequal 15
L115ff: I haven't checked all the citations but here are two that are missing from the references Flessa et al 1998 and Linn&Doran 1984
L117: When was the N2O emission measured before or after the irrigation and if after how much time has passed in between?
Soil moisture results are not shown anywhere, not even in the supplement
L119: Are the samples only taken on the ridge?
L140: What is meant by development?
L152: I miss yield data. At least for the assessment of N2O emissions (different terms are used here: production emissions, fluxes, I would standardise this, I think it always means the same thing), it makes a big difference whether irrigation produces more yield.
167 Figure1: In any case, the uncertainties (e.g. SD) are missing here. From my experience of the spatial variability of N2O emissions, all these trends could also come from one variant. It would also be nice to mark the times of irrigation and the variants should be easier to recognise with colours.
L223 Figure 2: Perhaps the authors can make it easier for readers by marking the clusters they see (as is often done in principal component analyses), but for me it is quite mixed.
L237: Under the right conditions, a large proportion of N2O production can also be chemical. It should at least be discussed here why the authors think that it is not in their case.
L243: Higher soil moisture does not necessarily lead to anaerobic conditions. This is a complex interplay between O2 consumption and transport. Especially in well-structured soils, irrigation can have a only small and very short-term effect.
L279 Figure 3: The graphic is difficult to read due to the very small numbers. Even if the representation in circles is nice, I would prefer small tables. The purpose of the picture in each subgraph at the top right is not clear to me.
L382: Are the differences significant enough to be mentioned in the conclusion? Especially when you consider that the measurements were only taken on the ridges.
L384ff: In this generality it does not help for future work, so it should definitely be sharpened up.
L391: I assume that JK is supposed to be JMK
Citation: https://doi.org/10.5194/egusphere-2023-2277-RC1 -
AC1: 'Reply on RC1', Laura Storch, 18 Dec 2023
First, we thank the reviewer for her/his overall evaluation and valuable comments on our manuscript. To the best of our abilities, we answered to the reviewers’ comments and made the necessary adjustments in the manuscript. Please find below the list of reviewer comments with our corresponding answer.
Response to the report of reviewer #1
General comment R1#1
The authors investigate the influence of fertilization and irrigation on N2O emissions and on the abundance of functional genes. They present a comprehensive measurement program, although I would have chosen a different approach for studies on the question posed in the title. I therefore recommend changing the title and focusing on what is actually shown. The cited preliminary study Storch et al. 2023 could certainly serve as a model.
Answer: Thank you for this comment. We see that the title could mislead the expectations of the readers. Therefore, we will change it as follows: "The volume of irrigation water rather than nitrogen fertilizer application affected the quantities of functional genes related to N2O production in potato cropping” We think that this title is now more related to the applied methods and actual findings of this study.
general comment R1#2
I think it is necessary for the authors to critically check all their statements again to see whether the suggestion of causality can really be maintained or whether only the correlation should be described.
Answer: Thank you for this comment. We carefully checked the entire manuscript whether our statements could be misunderstood as causalities instead of correlations. Due to the performed, statistically analyses our results are based on correlations, while some of them are statistically significant. Therefore, we used terms such as “might enhance/favor/feasible/…”, “probably”, “seemed to be”, “potentially”, “it can be assumed”, “suggested that”, “indicated a potential”, “could/might have been” or “the correlation pattern show/indicate/suggest”. However, we found some places in the text, where corrections are useful and hence will revise the respective text parts where necessary.
general comment R1#3
The assumptions made should also be discussed more in the context of existing experiments on the underlying processes. The previous isotope work in potato crops assumes denitrification, which, however, is particularly strong in the furrows and even stronger in the lanes, both of which have not been investigated if I understand correctly.
Answer: Regarding this comment, we think the title of the manuscript was a little bit misleading the expectations of the readers (see answer to general comment R1#1). The title might give the impression that we elucidated the N2O production pathways, which can indeed be determined more precisely by isotopic approaches. However, in this study we focused on the assessment of different agronomic management measures and their direct effects on the microbial community within the rhizosphere or more precise on the quantities of functional genes related to the N cycle. We are aware, that N2O flux rates from furrows and lanes are commonly considered when investigating the effects of surface irrigation, as the flux rates from furrows and lanes are generally stronger. But the main objective of our project was to elucidate whether fertigation can increase crop yields while simultaneously reducing external inputs of water and nutrients (by placing them constantly in small amounts directly to the roots) and hence mitigate N2O production compared to “standard” irrigation techniques such as sprinkler irrigation with broadcasted nitrogen application. In fertigated systems, the application of water and nitrogen is carried out punctually with emitters while water and nitrogen disperse in the soil with a bulb pattern beneath the emitters in the ridge. Of course, N2O production could have happened in the furrows and lanes, but the common areas of water application for both types of irrigation (sprinkler and drip irrigation) were the ridges where potatoes were planted. The observed values in this study partly exceeding the common range of N2O flux rates found in earlier studies on sandy soils and potato cropping (Trost et al., 2014; Mathivanan et al., 2021; Thilakarathna et al., 2022), but they are lower compared to a previous study performed at the same study site (Storch et al., 2023).However, we further would like to highlight, that we compared and discussed our results in the context of relevant existing studies that are comparable with our study considering the methodological approach particularly in terms of quantification of functional genes of the N cycle using qPCR. Additionally, we paid attention to take studies into consideration with comparable climatic conditions and soil types, as well as comparable cropping systems and management measures in terms of comparable amounts and types of irrigation water and nitrogen fertilizer application.
Mentioned reference: Trost et al., 2014: Nitrous oxide emissions from potato cropping under drip-fertigation in eastern Germany. Arch. Agron. Soil Sci., 60, 1519–1531. // Mathivanan et al., 2021: New N2O emission factors for crop residues and fertiliser inputs to agricultural soils in Germany. Agric. Ecosyst. Environ., 322, 107640. // Thilakarathna et al., 2022: Nitrous oxide emissions and productivity of irrigated potato: Effects of nitrogen fertilization options. Agron. J., 115, 161–180. // Storch et al., 2023: Nitrogen fertilization and irrigation types do not affect the overall N2O production potential of a sandy soil, but the microbial community structure and the quantity of functional genes related to the N cycle. Appl. Soil Ecol., 192, 105083.
specific comment R1#1
L16: In my opinion, the spatial distribution of soil moisture is not adequately described, especially the differences between ridges and furrows in potato crops is mentioned, but is not the subject of the study, at least not the results.
Answer: Thank you for this comment. Maybe we have a different perception of the term “spatial” here. From the reviewer’s point of view, “spatial” means the differentiation of potato fields in ridges, furrows and lanes, while we refer “spatial” to the type of water and fertilizer application as described in Table 1. We separated the spatial distribution in homogeneous (such as sprinkling and broadcasted N application which cover the full area) vs. punctual in grid knot pattern (such as drip irrigation or fertigation). For clarification, we will change the respective text part as follows “Spatial and temporal distribution of water and nitrogen supply affects soil-borne nitrous oxide (N2O) emissions. In this study, the effects of different irrigation technologies (no irrigation, area-covering sprinkler irrigation and punctual drip irrigation) and nitrogen (N) application types (no fertilizer, broadcasted and punctual with irrigation water) on N2O flux rates and the quantities of functional genes related to the N cycle were investigated over an entire season in potato cropping.” We would like to emphasize again, that the focus of the study was to assess the effects of different agronomic management measures, particularly the intended positive effects of fertigation on the genetically determined functional potential of the microbial community within the rhizosphere (see above answer to general comment R1#3). Therefore, we did not differentiate between ridges, furrows and lanes, but rather did all analyses only at the ridge. Although we used the water filled pore space values (WFPS) within/for the statistically analyses, we indeed missed to give information on the WFPS during the cropping season. We will add the missing information at respective text parts in the main document and in the supplement as Figure S1.
specific comment R1#2
L18: I would rather speak here of genes potentially involved than of processes
Answer: We will change the sentence from “In this study, the effects of different irrigation technologies (no irrigation, sprinkler irrigation and drip irrigation) and nitrogen (N) application types (broadcasted and dissolved in irrigation water) on N2O emissions and the potentially underlying, genetically determined microbial processes were investigated over an entire season in potato cropping” to “In this study, the effects of different irrigation technologies (no irrigation, area-covering sprinkler irrigation and punctual drip irrigation) and nitrogen (N) application types (no fertilizer, broadcasted and punctual with irrigation water) on N2O flux rates and the quantities of functional genes related to the N cycle were investigated over an entire season in potato cropping.”
specific comment R1#3
L78ff: Please note the size of the plots.
Answer: Thank you for this comment, the plot size was indeed missing. The plot size was 6m x 8m. We will add this information at the respective text part chapter 2.1.
specific comment R1#4
L81: The amount of precipitation measured during the study period is at least as important as the annual precipitation.
Answer: We agree with this comment. We presented the air and soil temperature as well as the precipitation for each sampling day in Table S1. However, we missed to give information on the water filled pore spaces during the cropping season. We will add the missing information at respective text parts in the main document and in the supplement as Figure S1.
specific comment R1#5
L96: A continuous color scheme would be helpful, e.g. light and dark brown for ZI-ZN and ZI-N, light and dark chalk blue for SI-ZN and SI-N, light and dark green for DI-ZN and DI-N, a dark color (because irrigated) for F (e.g. purple or red) and black like black fallow for F-ZC. this can then also be used for the table.
Answer: We agree with this comment. The existing continuous color scheme was revised (and hopefully strengthened) as follows: dark and light blue ((#1874CD, (#87CEFA) for ZI-ZN and ZI-N, dark and light green (#228B22, #90EE90) for SI-ZN and SI-N, dark and light orange (#CD600, #FFA500) for DI-ZN and DI-N, and dark and light purple (#8B3A62, #CD8CD95) for F and F-ZC. Additionally, we adapted the selected color scheme for Fig. 1 as well as for Fig. S1 and S2. As far as we know, there is no journal where the format specification for tables allows a color code. Hence the color code was by now not used in the provided tables. If the journal allows a color code within tables, we can make the recommended adjustments.
specific comment R1#6
L106 14.4+1.1 is unequal 15.
Answer: That´s true. The chemical composition of the applied fertilizer consisted of 15.5 % total N. Therefore, the sum out of 14.4 % nitrate N and 1.1 % ammonium N is equal to 15.5 % of total N. We will correct this at the respective text part.
specific comment R1#7
L115ff: I haven't checked all the citations but here are two that are missing from the references Flessa et al 1998 and Linn & Doran 1984
Answer: Thank you for pointing this out. We added the missing references in the reference list and carefully checked all cited literature again.
specific comment R1#8
L117: When was the N2O emission measured before or after the irrigation and if after how much time has passed in between?
Answer: The gas measurements were carried out approx. 3 hours after irrigation and fertilization events. To determine a sampling strategy that identifies key periods which best reflects our system and to minimize potential errors associated with measurements performed at a frequency lower than on a daily basis, we carried out several 24 hours lasting experiments (not reported). Within these experiments gas samples were taken every three hours directly after irrigation and/or fertilization over 24 hours and subsequently daily for the rest of the week. The results showed an increase of the N2O flux approx. three hours after fertilization and stayed constant without statistically significant differences over the rest of the week. These experiments led to three conclusions: (1) gas measurements should be carried out in the mid-morning hours, (2) gas measurements should be carried out approx. three hours after irrigation and fertilization events, and (3) a weekly basis for gas measurements seems to be appropriate for our systems. As our sandy study site is generally characterized by low N2O flux rates (Trost et al. 2013, Storch et al., 2023), the chosen weekly basis for gas measurements is in accordance with recommendations from Charteris et al. (2020).
Mentioned references: Trost et al. 2013: Irrigation, soil organic carbon and N2O emissions. A review. Agron. Sustain. Dev., 33, 733-749. // Storch et al. 2023: Nitrogen fertilization and irrigation types do not affect the overall N2O production potential of a sandy soil, but the microbial community structure and the quantity of functional genes related to the N cycle. Appl. Soil Ecol., 192, 105083. // Charteris et al. 2020: Global Research Alliance N2O chamber methodology guidelines: Recommendations for deployment and accounting for sources of variability. J Environ Qual 49, 1092-1109.
specific comment R1#9
Soil moisture results are not shown anywhere, not even in the supplement.
Answer: We will add the missing information in an additional figure in the supplement (see answer to specific comment R1#4).
specific comment R1#10
L119: Are the samples only taken on the ridge?
Answer: Yes, samples were only taken on the ridge (see answer to general comment R1#3).
specific comment R1#11
L140: What is meant by development?
Answer: N2O flux rates are characterized by fluctuations over the season; hence we refer this expression to the seasonal development of N2O flux rates.
specific comment R1#12
L152: I miss yield data. At least for the assessment of N2O emissions (different terms are used here: production emissions, fluxes, I would standardize this, I think it always means the same thing), it makes a big difference whether irrigation produces more yield.
Answer: We will add a table (Table 2) showing the cumulative area-related N2O emissions, the obtained potato crop yields and hence the yield-related N2O emissions. N2O fluxes and N2O emissions are different things. N2O fluxes (or better N2O flux rates) are calculated based on the linear regression using the slope of temporal change of N2O content of gas samples taken with the closed chamber method (Flessa et al. 1998) at a respective time point. They are given in µg N m-2 h-1. The term N2O emissions refers to the cumulative N2O flux rates over the entire season related either to the cultivated area (kg N2O ha-1) or to the received crop yields (kg N2O t-1). Cumulative area-related N2O emissions are calculated by linear interpolation of the measured N2O flux rates. The yield-related N2O emissions are estimated by dividing the calculated cumulative area-related N2O emissions by the obtained crop yields.
Mentioned reference: Flessa et al. 1998: Nitrous oxide and methane fluxes from organic soils under agriculture. Eur. J. Soil Sci. 49, 327–335.
specific comment R1#13
167 Figure1: In any case, the uncertainties (e.g. SD) are missing here. From my experience of the spatial variability of N2O emissions, all these trends could also come from one variant. It would also be nice to mark the times of irrigation and the variants should be easier to recognize with colors.
Answer: Thank you for pointing this out. We agree that a color code will be nice here. Accordingly, we will mark the different treatments with the same color code used in Fig. 2 (see specific comment R1#5). We also agree that the uncertainties in terms of providing mean values (MV) with standard deviation (SD) are missing. The data shown in Fig. 1 illustrated the median N2O flux rates of three field replicates per treatment. However, to provide all important information we will modify the visualization in Fig. 1 by showing boxplots of the raw data including the median values of the recorded N2O flux rates per sampling day, the minimal and maximal values and outliner (antennae) as well as the upper and lower quartiles (containing 50% of the recorded data) within the box. The MV and SD values will be added in Table S3. Finally, during the data evaluation process, we have tried different data visualization types including the option to mark the irrigation and fertilization times points in the N2O flux rates figure separately for each treatment. However, this combined illustration was confusing and not easy to understand for the reader. Therefore, we decided to provide the application dates and amounts of water and nitrogen in Table S1. To interlink the requested information, we will add the following sentence to the caption of Figure 1: “Information on the application dates and amounts of irrigation water and nitrogen fertilizer is provided in Table S1.”
specific comment R1#14
L223 Figure 2: Perhaps the authors can make it easier for readers by marking the clusters they see (as is often done in principal component analyses), but for me it is quite mixed.
Answer: We agree. The data shown in the NDMS ordination plot could be identified as mixed. For better visualization, we will add a circle around the mentioned time-depended clustering in the left-hand corner.
specific comment R1#15
L237: Under the right conditions, a large proportion of N2O production can also be chemical. It should at least be discussed here why the authors think that it is not in their case.
Answer: We are aware that N2O can also be generated by chemical processes. This chemical N2O production has been reported to be favored in acidic soils with a pH value below 4.5 (Hu et al. 2015, Heil et al. 2016, Chalk & Smith 2020). The chemical process can almost be ruled out for our study site that is characterized by a pH value of 6.75. Nevertheless, we took this comment into consideration at the respective text part in the discussion and will add the following sentence: “Apart from the microbial mediated N2O production, chemical reactions could also contribute to total N2O emissions, especially under acidic soil conditions (pH < 4.5) (Hu et al. ,2015; Heil et al., 2016; Chalk and Smith, 2020). However, the chemical process can almost be ruled out for our study site which is characterized by a pH value of 6.75.”
Mentioned references: Hu et al. 2015: Microbial regulation of terrestrial nitrous oxide formation: understanding the biological pathways for prediction of emission rates. FEMS Microbiol Rev 39, 249. // Heil et al. 2016: A review of chemical reactions of nitrification intermediates and their role in nitrogen cycling and nitrogen trace gas formation in soil. Eur J Soil Sci 67, 23–39. // Chalk & Smith 2020: The role of agroecosystems in chemical pathways of N2O production. Agric Ecosyst Environ 290, 106783.
specific comment R1#16
L243: Higher soil moisture does not necessarily lead to anaerobic conditions. This is a complex interplay between O2 consumption and transport. Especially in well-structured soils, irrigation can have only small and very short-term effect.
Answer: We fully agree with this comment. While we think that the sentences from lines 243 to 247 illustrate the generally accepted and often reported correlation between the soil oxygen level and the soil water content which is affected by irrigation and the related water filled pore space, the occurring processes are much more complex and the statement is too generalized. To emphasize the temporal character of the irrigation effects, we will change the sentence as follows: “In this regard, irrigation in general leads to a temporal and spatial increase in the soil water content and hence anaerobic conditions, which might stimulate the microbial N2O production via the denitrification pathway (Butterbach-Bahl et al., 2013; Trost et al., 2013; Hu et al., 2015; Kuang et al., 2021), although it has to be considered that these effects can be of short terms.”. Furthermore, in line 258ff, we returned to this fact: “Contrary to the hypothesis that higher water volumes result in a stimulation of the process of denitrification, the detected correlation patterns of the investigated genes showed that denitrification was unlikely the underlying pathway of N2O production under SI-ZN (Fig. 3, supplementary Table S6). It is more probable that environments with fluctuating aerobic-anaerobic conditions under sprinkler irrigation systems promote N2O production by nitrifier denitrification (Wrage-Mönnig et al., 2018).” Thus, we are putting the commonly accepted theories into perspective.
Mentioned references: Butterbach-Bahl et al. 2013: Nitrous oxide emissions from soils: How well do we understand the process and their control? Phil. Trans., 5, 389-395. // Trost et al. 2013: Irrigation, soil organic carbon and N2O emissions. A review. Agron. Sustain. Dev., 33, 733-749. // Hu et al. 2015: Microbial regulation of terrestrial nitrous oxide formation: Understanding the biological pathways for prediction of emission rates. FEMS Microbiol. Rev., 39, 729–749. // Kuang et al. 2021: A global meta-analysis of nitrous oxide emission from drip-irrigated cropping system. Glob. Chang. Biol., 27, 3244-3256. // Wrage-Mönnig et al. 2018: The role of nitrifier denitrification in the production of nitrous oxide revisited. Soil Biol. Biochem., 123, A3-A16.
specific comment R1#17
L279 Figure 3: The graphic is difficult to read due to the very small numbers. Even if the representation in circles is nice, I would prefer small tables. The purpose of the picture in each subgraph at the top right is not clear to me.
Answer: We agree that this figure is a bit difficult to read as the single graphs, particularly the correlation values are quite small. As the presentation of the correlation pattern can be best presented in the circle variant, we will modify the figure as follows. First, for a better resolution, we separated this figure into two figures; Fig. 3 contains the correlation pattern for the irrigated but unfertilized plots (SI-ZN, DI-ZN) and the control plot (ZI-ZN), while Fig. 4 contains the correlation pattern for the fertilized plots (ZI-N, SI-N, DI-N, F and F-ZC) compared to the control plot (ZI-ZN). Second, we will replace the subgraph in the top right corner by letters (Fig, 3A-C, Fig. 4A-F) indicating the different treatments. Third, we will enlarge the size of the correlation values, hoping for a better readability. Finally, due to the complexity of correlation patterns among the determined quantities of functional genes related to the N cycle and between the quantities of functional genes and the N2O flux rates, both figures only contain the most important correlation values, while all correlation values are additionally provided in the existing supplement in Table S5-S12.
specific comment R1#18
L382: Are the differences significant enough to be mentioned in the conclusion? Especially when you consider that the measurements were only taken on the ridges.
Answer: We think the differences are significant enough. Our conclusions are based on various statistical analyses that we have carried out. We see no limitation in the fact that we did our analyses in the ridge area. According to the main objective of this project (see response to general comment R1#3), four treatments received their water and/or fertilizer via a drip irrigation system on top of the ridge. Thus, we had to perform all analyses in the ridge area.
specific comment R1#19
L384ff: In this generality it does not help for future work, so it should definitely be sharpened up.
Answer: We agree with this reviewer comment. To sharpen up the conclusions, we will modify them as follows: “The applied irrigation and fertilization technologies in this study led to different N2O flux rates over the entire crop-ping season with highest flux rates during the first half of the season accompanied by a low NUE and ANR of the provided N fertilizer by the potato crops. Regarding the potentially underlying genetically determined N2O production pathway, this study indicates that the nitrifier denitrification process might be of great importance. Further research is required to adjust the amount and time of water and N fertilizer application based on crop demand for nutrients and their related nitrogen use efficiency and apparent nitrogen recovery during the different crop growth stages in order to mitigate N2O release from agriculture. This could be achieved by a constant monitoring of the N2O flux rates using largely automated gas measurements with a high temporal resolution to determine N2O emissions more accurately. Moreover, research is needed combining the quantification of functional genes of the N cycle with the occurring taxonomic profiles of the microbial communities and further with the occurring N2O pathways determined by isotopic approaches. Considering this, management measures of cropping systems can be improved regarding the nutrient use efficiency and apparent nutrient recovery rates of cultivated crops accompanied with a targeted control of the occurring microbial-mediated regulation mechanisms in nutrient cycles and hence mitigate N2O emissions from agricultural systems.”
specific comment R1#20
L391: I assume that JK is supposed to be JMK
Answer: This is right, we added the missing M
Citation: https://doi.org/10.5194/egusphere-2023-2277-AC1
-
AC1: 'Reply on RC1', Laura Storch, 18 Dec 2023
-
RC2: 'Comment on egusphere-2023-2277', Anonymous Referee #2, 25 Nov 2023
This study assesses the effects of different irrigation techniques (no irrigation, sprinkler irrigation and drip irrigation) and nitrogen (N) application types (broadcasted and dissolved in irrigation water) on N2O emissions and the potentially underlying, genetically determined microbial processes in a potato crop field. The title is indicating that nitrifier denitrification is identified as an important process for N2O emission. This is surely an interesting topic, and language quality of the manuscript is fine.
Two hypotheses are tested: First, the authors hypothesize that N2O flux rates differ between irrigation treatments due to differences in soil water content affecting denitrification, and second, that the addition of several small N doses due to fertigation will lead to lower N2O emission compared to broadcast N application due to better N use efficiency of the potato crops.
N2O emissions were measured on a weekly basis only. Due to the notorious temporal variability of N2O emissions with e.g., sharp short-lived peaks after irrigation or fertilization, this sampling design does not allow to accurately quantify cumulative emissions for the single treatments, and even less to differentiate N2O emissions between treatments such as e.g., fertigation and broadcast fertilization.
N2O measurements are combined with analysis of a set of nitrifier and denitrifier gene abundance. Generally it can be very valuable to link molecular analysis of functional N cycle gene abundance with N2O flux measurements. But to address the objectives of this study (in particular for N2O source process attributions as indicated already in the title) there are better methods, e.g., based on isotopic analysis of N2O and 15N labelling of process substrates. Since links between gene abundance, related processes and microbial activity are not trivial, and might be e.g. more pronounced for nitrifier genes than for denitrifier genes, the authors interpretations on origin of N2O appear overly speculative in my view. Furthermore, information on how the authors tried to bridge the huge spatial scales between mg-scale soil DNA analysis and chamber measurements is missing, information on soil moisture is hardly provided, and NUE seems to be not reported. Hence, it overall appears that this study is not able to satisfactorily address its goals, because of the chosen measurement design and due to the chosen methods.
Some specific comments:
L 35 The authors write “Current agricultural systems are characterized by a low nitrogen use efficiency (NUE), resulting inter alia in the loss of large amounts of actually available N through nitrate leaching and/or the generation of N2O (Wang et al., 2020; Wang et al., 2021; Menegat et al., 2022).” Statement is too general in my view, not all current agricultural systems show low N use efficiency. And cause and effect can be vice versa, low N use efficiency due to high losses also possible.
L 53 the denitrification pathway (NO3− to N2O/N2). This seems oversimplified, would write at least stepwise reduction of nitrate or nitrite to NO, N2O, N2.
L 60 given that nirk/nirS is explained here, nitrite should be mentioned earlier (see also comment to L 53).
Fig. 1 N2O emissions: The weekly measurements are not suitable to quantify N2O emissions across the treatments. Furthermore, only median values without uncertainty are given, and irrigation / fertilization events or climate information are not shown. And the figure setup does not allow to easily see differences between the treatments.
chapter 3.1 The entire discussion on causes of N2O differences across treatments reads odd given the weekly temporal resolution of measurements, which prevented accurate quantification of N2O emissions for the treatments.
Chapter 3.2: The statements on source processes of N2O seem pretty speculative and need to be reduced to their simple correlative nature.
Fig. 3: I like the figure design generally, but numbers and arrows are too small.
Citation: https://doi.org/10.5194/egusphere-2023-2277-RC2 -
AC2: 'Reply on RC2', Laura Storch, 18 Dec 2023
First, we thank the reviewer for her/his overall evaluation and valuable comments on our manuscript. To the best of our knowledge and abilities we answered to the reviewers’ comments and made the necessary adjustments in the manuscript. Please find below the list of reviewer comments and our answers.
general comment R2#1
N2O emissions were measured on a weekly basis only. Due to the notorious temporal variability of N2O emissions with e.g., sharp short-lived peaks after irrigation or fertilization, this sampling design does not allow to accurately quantify cumulative emissions for the single treatments, and even less to differentiate N2O emissions between treatments such as e.g., fertigation and broadcast fertilization.
Answer: We are aware that N2O flux rates underlie temporal fluctuations, particularly regarding expectable N2O flux rate peaks shortly after irrigation and/or fertilizations events, which might be undetected when N2O flux rates are measured on a weekly basis. To determine a sampling strategy that identifies key periods which best reflects our system and to minimize potential errors associated with measurements performed at a frequency lower than on a daily basis, we carried out several 24 hours lasting experiments (not reported). Within these experiments gas samples were taken every three hours directly after irrigation and/or fertilization over 24 hours and subsequently daily for the rest of the week. The results showed an increase of the N2O flux approx. three hours after fertilization and stayed constant without statistically significant differences over the rest of the week. These experiments led to three conclusions: (1) gas measurements should be carried out in the mid-morning hours, (2) gas measurements should be carried out approx. three hours after irrigation and fertilization events, and (3) a weekly basis for gas measurements seems to be appropriate for our systems. As our sandy study site is generally characterized by low N2O flux rates (Trost et al. 2013, Storch et al., 2023), the chosen weekly basis for gas measurements is in accordance with recommendations from Charteris et al. (2020). While a high sampling frequency might be desirable in general, measurements on a weekly basis are applied in many studies (e.g. Ruser et al., 2017; Shao et al., 2023; Thilakarathna et al., 2022; Kuang et al., 2023; Petersen et al., 2023, Storch et al. 2023, to name a few currently published and state of the art studies). Furthermore, the aim of this study was to evaluate the quantities of functional genes of the N cycle and correlate them to the N2O flux rates as N2O formation is mainly the result of the microbial activity at a given time point. Therefore, we see the provided cumulative N2O emissions as an additional information here. Taken this into consideration, we revised the relevant parts of our manuscript and found that the title (see answer to general comment R2#2) and the mentioned objective and subsequent the hypotheses have not been formulated concisely enough. While the objective clearly stated the focus of this study in term of elucidating the genetically determined microbial-mediated pathways of N2O formation, the wording of the hypotheses did not fully reflect this objective. Therefore, we will rephase the hypotheses as follows: “It was firstly hypothesized, that N2O flux rates differ between the treatments expecting lowest flux rates under ZI-ZN and higher N2O flux rates under sprinkler and drip irrigation (SI-ZN, DI-ZN) due to a higher soil water content and therefore an enhanced denitrification process reflected by a higher copy number and distinct correlation pattern of genes related to the denitrification process. Secondly, regarding the different N fertilization regimes, it was hypothesized that the application of several small N doses in irrigation water by fertigation (F) will lead to lower N2O flux rates compared to the broadcasted N applications (SI-N, DI-N) due to better N use efficiency of the potato crops accompanied with general lower and more equally contributed quantities of the investigated functional gene resulting in more diverse potential of N2O production pathways.” We think that the hypotheses now better express our focus on the potential occurring microbial processes.
Mentioned references: Trost et al. 2013: Irrigation, soil organic carbon and N2O emissions. A review. Agron. Sustain. Dev., 33, 733-749. // Storch et al. 2023: Nitrogen fertilization and irrigation types do not affect the overall N2O production potential of a sandy soil, but the microbial community structure and the quantity of functional genes related to the N cycle. Appl. Soil Ecol., 192, 105083. // Charteris et al. 2020: Global Research Alliance N2O chamber methodology guidelines: Recommendations for deployment and accounting for sources of variability. J Environ Qual 49, 1092-1109 // Ruser et al. 2017: Nitrous oxide emissions from winter oilseed rape cultivation. Agric Ecosyst Environ 249, 57 69. // Shao et al. 2023: Impacts of monoculture cropland to alley cropping agroforestry conversion on soil N2O emissions. GCB Bioenergy. 15, 58-71 // Thilakarathna et al. 2022: Nitrous oxide emissions and productivity of irrigated potato: Effects of nitrogen fertilization options. Agron J 115, 161-180 // Kuang et al. 2023: Soil profile N2O efflux from a cotton field in arid Northwestern China in response to irrigation and nitrogen management. Front Environ Sci 11, 1123423 // Petersen et al. 2023: Higher N2O emissions from organic compared to synthetic N fertilisers on sandy soils in a cool temperate climate. Agric Ecosyst Environ 358, 108718 //
general comment R2#2
N2O measurements are combined with analysis of a set of nitrifier and denitrifier gene abundance. Generally, it can be very valuable to link molecular analysis of functional N cycle gene abundance with N2O flux measurements. But to address the objectives of this study (in particular for N2O source process attributions as indicated already in the title) there are better methods, e.g., based on isotopic analysis of N2O and 15N labelling of process substrates. Since links between gene abundance, related processes and microbial activity are not trivial, and might be e.g. more pronounced for nitrifier genes than for denitrifier genes, the authors interpretations on origin of N2O appear overly speculative in my view. Furthermore, information on how the authors tried to bridge the huge spatial scales between mg-scale soil DNA analysis and chamber measurements is missing, information on soil moisture is hardly provided, and NUE seems to be not reported. Hence, it overall appears that this study is not able to satisfactorily address its goals, because of the chosen measurement design and due to the chosen methods.
Answer: We agree that the title can be misleading the expectations of the readers as it might give the impression that we elucidated the N2O production pathways, which can indeed be determined more precise by isotopic approaches. Therefore, we will change the title as follows: “The volume of irrigation water rather than nitrogen fertilizer application affected the quantities of functional genes related to N2O production in potato cropping” This title would better reflect the applied methods and actual findings of this study. Given the objective of our study (see general comment R2#1) we consider our methodology being sound since qPCR is a highly valuable and widely used method, in particular regarding the determination of the gene copy numbers per gram soil for genes encoding enzymes of the N cycle (e.g. Kuypers et al., 2018; Prosser et al., 2020; Kumar et al., 2020; You et al., 2022; Yang et al., 2022). A simple request on Google Scholar using the terms “qPCR N cycle” revealed approx. 19.000 results for the years 2020-2023. On the other hand, isotope approaches have also clear disadvantages, beyond other e.g. natural abundances of N2O isotopic species are not as direct as the 15N-labelling approach with regards to the actual pathways of N2O production (Yu et al. 2020). We disagree that our interpretations appear overly speculative. We used approved statistical methods to analyze whether our results and correlation patterns are significant, and we interpreted our findings based on the results of the statistical analyses. To emphasize that we refer to correlations and not causalities. Therefore, we used terms such as “might enhance/favor/feasible/…”, “the correlation patterns show/indicate/suggest”, “probably”, “seemed to be”, “potentially”, “it can be assumed”, “suggested that”, “indicated a potential”, or “could/might have been”. Revising the manuscript after having received the reviewers’ comments, we found some formulations in the text where a correction is useful and hence will revise the respective text parts. Regarding the missing “bridge” between microbiological analyses and gas measurements, three soil cores per plot were taken in the rhizosphere of the potato crops with a geological drill while the replicates from the same plot were pooled into one composite sample as explained in chapter 2.3. For each soil sample per plot, three DNA extractions were carried out. To our mind, this is the commonly approved way of taking soil samples to ensure an appropriate sampling design to address the study objectives while combining gas measurements with microbiological analyses. Finally, information of the water filled pore space as well as the nitrogen use efficiency (NUE) and the apparent nitrogen recovery (ANR) are indeed missing; hence we will add the missing information in the manuscript.
Mentioned references: Kuypers et al., 2018: The microbial nitrogen-cycling network. Nat. Rev. Microbiol., 16, 263-276. // Prosser et al., 2020: Nitrous oxide production by ammonia oxidizers: physiological diversity, nice differentiation and potential mitigation strategies. Glob. Chang. Biol. 26, 103-118 // Kumar et al., 2020: Molecular and ecological perspectives of nitrous oxide producing microbial communities in agro-ecosystems. Rev. Environ. Sci. Biotechnol., 19, 717–750 // You et al., 2022: Global meta-analysis of terrestrial nitrous oxide emissions and associated functional genes under nitrogen addition. Soil Biol. Biochem., 165, 108523 // Yang et al., 2022: Response of N2O emission and denitrification genes to different inorganic and organic amendments. Sci. Rep., 12, 3940 // Yu et al. 2020: What can we learn from N2O isotope data? - Analytics, processes and modelling. Rapid Commun Mass Spectrom 34, e8858.
specific comment R2#1
L 35 The authors write “Current agricultural systems are characterized by a low nitrogen use efficiency (NUE), resulting inter alia in the loss of large amounts of actually available N through nitrate leaching and/or the generation of N2O (Wang et al., 2020; Wang et al., 2021; Menegat et al., 2022).” Statement is too general in my view, not all current agricultural systems show low N use efficiency. And cause and effect can be vice versa, low N use efficiency due to high losses also possible.
Answer: Low nitrogen use efficiencies in agricultural systems have consistently been reported by several recent meta-analyses and reviews (see for example Anas et al., 2020; Tian et al., 2020; Wang et al., 2020; Wang et al., 2021; Menegat et al., 2022; You et al., 2023). There is an oversupply of synthetic fertilizer into agricultural systems all over the world while the supplied fertilizer is insufficiently available for and/or used by the cultivated crops. In order to include exceptions and both directions of cause and effect, we will modify the sentence as follows: “The vast majority of current agricultural systems is characterized by a low nitrogen use efficiency (NUE) or low apparent nitrogen recoveries (ANR), associated inter alia with the loss of large amounts of actually available N through nitrate leaching and/or the generation of N2O (Milroy et al., 2019; Wang et al., 2020; Anas et al, 2020; Wang et al., 2021; Menegat et al., 2022; You et al. 2023).”
Mentioned references: Anas et al., 2020: Fate of nitrogen in agriculture and environment: agronomic, eco‑physiological and molecular approaches to improve nitrogen use efficiency. Biol Res 53, 47 // Tian et al., 2020: A comprehensive quantification of global nitrous oxide sources and sinks. Nature, 586, 248-256 // Wang et al., 2020: Optimum nitrogen rate to maintain sustainable potato production and improve nitrogen use efficiency at a regional scale in China. A meta-analysis. Agron. Sustain. Dev., 40, 37 // Wang et al., 2021: Nitrogen fertiliser-induced changes in N2O emissions are attributed more to ammonia-oxidising bacteria rather than archaea as revealed using 1-octyne and acetylene inhibitors in two arable soils. Biol. Fertil. Soils, 52(8), 1163-1171 // Menegat et al., 2022: Greenhouse gas emissions from global production and use of nitrogen synthetic fertilisers in agriculture. Sci. Rep., 12, 14490 // You et al. 2023: Global mean nitrogen recovery efficiency in croplands can be enhanced by optimal nutrient, crop and soil management practices. Nat. Commun. 14, 5747// Milroy et al. (2019): Defining upper limits of nitrogen uptake and nitrogen use efficiency of potato in response to crop N supply. Field Crops Res., 239, 38-46
specific comment R2#2
L 53 the denitrification pathway (NO3− to N2O/N2). This seems oversimplified, would write at least stepwise reduction of nitrate or nitrite to NO, N2O, N2.
Answer: Done.
specific comment R2#3
L 60 given that nirk/nirS is explained here, nitrite should be mentioned earlier (see also comment to L 53).
Answer: Done.
specific comment R2#4
Fig. 1 N2O emissions: The weekly measurements are not suitable to quantify N2O emissions across the treatments. Furthermore, only median values without uncertainty are given, and irrigation / fertilization events or climate information are not shown. And the figure setup does not allow to easily see differences between the treatments.
Answer: Regarding the comment, that weekly N2O flux rate measurements are not suitable to quantify cumulative N2O emissions, please refer to our answer to general comment R2#1. However, we agree that the uncertainties in terms of providing mean values (MV) with standard deviation (SD) are missing. The data shown in Fig. 1 illustrated the median N2O flux rates of three field replicates per treatment. To provide all important information we will modify the visualization in Fig. 1 by showing now boxplots of the raw data that include the median values of the recorded N2O flux rates per sampling day, the minimal and maximal values and outliner (antennae) as well as the upper and lower quartiles (containing 50% of the recorded data) within the box. The MV and SD values will be added in Table S3. Finally, during the data evaluation process, we have tried different data visualization types including the option to mark the irrigation and fertilization times points in the N2O flux rates figure separately for each treatment. However, this combined illustration was confusing and not easy to understand for the reader. Therefore, the authors decided to provide the application dates and amounts of water and nitrogen in Table S1. To interlink the requested information, we added the following sentence to the caption of Fig. 1: “Information on the application dates and amounts of irrigation water and nitrogen fertilizer is provided in Table S1.”
specific comment R2#5
chapter 3.1 The entire discussion on causes of N2O differences across treatments reads odd given the weekly temporal resolution of measurements, which prevented accurate quantification of N2O emissions for the treatments.
Answer: Regarding the comment related to the weekly N2O flux rate measurements, please refer to our answer to general comment R2#1. N2O fluxes and N2O emissions are different things. Additionally, we would like to emphasize that N2O fluxes (or better N2O flux rates) are calculated based on the linear regression using the slope of temporal change of N2O content of gas samples taken with the closed chamber method (Flessa et al. 1998) at a respective time point. They are given in µg N m-2 h-1. The term N2O emissions refers to the cumulative N2O flux rates over the entire season related either to the cultivated area (kg N2O ha-1) or to the received crop yields (kg N2O t-1). Cumulative area-related N2O emissions are calculated by linear interpolation of the measured N2O flux rates. The yield-related N2O emissions are estimated by dividing the calculated cumulative area-related N2O emissions by the obtained crop yields. However, the aim of this study was to evaluate the quantities of functional genes of the N cycle and correlate them to the N2O flux rates as N2O formation is mainly the result of the microbial activity at a given time point. Therefore, we see the provided cumulative N2O emissions as an additional information here.
Mentioned references: Flessa et al. 1998: Nitrous oxide and methane fluxes from organic soils under agriculture. Eur. J. Soil Sci. 49, 327–335.
specific comment R2#6
Chapter 3.2: The statements on source processes of N2O seem pretty speculative and need to be reduced to their simple correlative nature.
Answer: We carefully checked the entire manuscript whether our interpretations could be misunderstood as causalities instead of correlations and, where necessary, we corrected respective text parts. As mentioned above (see answer to general comment R2#2) we used approved statistical methods to elucidate whether our results and correlation patterns are significant. Due to the performed statistically analyses our results are based on correlations, of course, while some of them are statistically significant. Based on the current state of knowledge provided by other studies we have tried to find interpretations of the recorded data, which we do not consider to be speculative but rather scientifically sound based on statistically significant findings compared to results reported by other studies.
specific comment R2#7
Fig. 3: I like the figure design generally, but numbers and arrows are too small.
Answer: We agree with the reviewer that this figure is a bit difficult to read as the single graphs, particularly the correlation values are quite small. As the presentation of the correlation pattern can be best presented in the circle variant, we will modify (have modified) the figure as follows. First, for a better resolution, we separated this figure into two figures; Fig. 3 contains the correlation pattern for the irrigated but unfertilized plots (SI-ZN, DI-ZN) and the control plot (ZI-ZN), while Fig. 4 contains the correlation pattern for the fertilized plots (ZI-N, SI-N, DI-N, F and F-ZC) compared to the control plot (ZI-ZN). Second, we will replace the subgraph in the top right corner by letters (Fig, 3A-C, Fig. 4A-F) indicating the different treatments. Third, we (will) enlarge(d) the size of the correlation values, hoping for a better readability. Finally, due to the complexity of correlation patterns among the determined quantities of functional genes related to the N cycle and between the quantities of functional genes and the N2O flux rates, both figures only contain the most important correlation values, while all correlation values are additionally provided in the existing supplement in Table S5-S12.
Citation: https://doi.org/10.5194/egusphere-2023-2277-AC2
-
AC2: 'Reply on RC2', Laura Storch, 18 Dec 2023
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 226 | 72 | 25 | 323 | 35 | 17 | 16 |
- HTML: 226
- PDF: 72
- XML: 25
- Total: 323
- Supplement: 35
- BibTeX: 17
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1