the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Diachronic assessment of soil organic C and N dynamics under long-term no-till cropping systems in the tropical upland of Cambodia
Abstract. No-till (NT) cropping systems have been proposed as a potential strategy to combat soil degradation and global warming by storing soil organic carbon (SOC) and nitrogen (N). Yet, there are ongoing debates about the real benefits of NT systems and factors influencing SOC and N accumulation. Assessing the dynamics of SOC and N on the long-term is needed to fill knowledge gaps and provide robust scientific evidence for potential additional SOC storage. We quantified the changes in SOC and N stocks and fractions down to 100 cm depth from three 13-year-old experiments in a tropical red Oxisol in Cambodia, comparing conventional tillage (CT) to NT monocropping and NT crop rotation systems using a diachronic and equivalent soil mass (ESM) approach. The three experiments comprised maize-, soybean-, and cassava-based cropping system trials, hereafter called MaiEx, SoyEx, and CasEx, respectively. Soil samples were collected in 2021, 10 years after the first sampling in 2011, at 7 depths: 0–5, 5–10, 10–20, 20–40, 40–60, 60–80, and 80–100 cm. Over the 10-year period (2011–2021), significant impacts on SOC stock and its vertical distribution differed among the NT systems and in the three experiments. In MaiEx and CasEx, the soils under all the NT systems significantly (P > 0.05) accumulated SOC stock across the soil depths, with the accumulation ranging from 6.97 to 14.71 Mg C ha-1 in the whole profile (0–100 cm). In SoyEx, significant increase in SOC stock was limited to the top 0–20 cm under NT monocropping, whereas NT crop rotation systems had significantly accumulating SOC stock from 0 to 80 cm depths. When considering 0–100 cm as a single stratum, the annual SOC cumulative rate in NT systems ranged from 0.86–1.47, 0.65–1.00, and 0.70–1.07 Mg C ha-1 yr-1 in MaiEx, SoyEx, and CasEx, respectively. In the top 0–10 cm, NT systems significantly increased C concentration in particulate organic matter (POM) by 115 %, 118 %, in MaiEx and SoyEx, respectively, and by 37 % in CasEx although not significantly. Similarly, at 0–10 cm depth, NT systems significantly enhanced C concentration in the mineral-associated organic matter (MAOM) by 33 %, 21 %, in MaiEx and SoyEx, respectively. Significant increase of C in MAOM was also observed from 0 to 40 cm in CasEx. In contrast, total N stock in NT systems increased in the surface 0–5 cm depth but decreased below 10 cm and in the whole profile (0–100 cm), particularly under NT monocropping with an annual loss rate of -0.10 and -0.17 Mg N ha-1 yr-1 in SoyEx and CasEx, respectively. Although NT systems increased N concentration in POM in the top 0–10 cm of MaiEx and SoyEx, a decreasing trend was observed below 10 cm depth. The N concentration in POM under NT systems in CasEx also decreased with soil depth. From 2011 to 2021, N concentration in MAOM under NT systems remained stable in MaiEx and SoyEx in the top 0–5 cm, but significant decreases in MaiEx and CasEx below 5 cm.
Our findings suggest that adopting NT cropping systems with diverse crop and cover crop species and high biomass C inputs in the long-term leads to SOC accumulation not only in the surface but also in deeper layers, by increasing both the C pools in the POM and MAOM size fractions, even on the cassava-based system, which is believed to be an annual crop that could cause serious soil fertility depletion. This study highlights the potential of NT cropping systems to store SOC over time, but raises questions about soil N dynamics.
- Preprint
(2138 KB) - Metadata XML
-
Supplement
(463 KB) - BibTeX
- EndNote
Status: open (extended)
-
RC1: 'Comment on egusphere-2024-541', Anonymous Referee #1, 15 Apr 2024
reply
I have read “Diachronic assessment of soil organic C and N dynamics under long-term no-till cropping systems in the tropical upland of Cambodia.” The manuscript describes a study where soil organic carbon (SOC) mass fractions (“concentrations”), stocks, and particulate/mineral pools were measured at a ten-year interval across different crop types (maize, soybean, and casava) and management practices (monocrop with conventional till without cover crop, monocrop with no till with cover crop, and two phases of crop rotations with no till and cover crop). The authors report that SOC concentration, SOC stocks, and SOC pools generally increased near the surface under no-till treatments, but generally showed fewer changes at depth. N concentrations and stocks generally increased near the surface but decreased at depth under no-till. Space-for-time sampling (“synchronic”) generally underestimated SOC stock changes relative to longitudinal sampling (“diachronic”). The authors conclude that no-till cover crop systems may help to increase SOC, but more work is needed to understand N dynamics.
The dataset is comprehensive, the topic is of general interest, and manuscript is generally well-written. However, I have some suggestions for improvement.
General comments:
The manuscript is too long, primarily because of the details provided in the results text. The figures and tables contain the details that the reader can reference, and the text should serve to bring attention to the important findings. Therefore, it is not necessary to repeat in the text every result shown in the figures and tables. I would also recommend combining the maize, soybean, and casava results into one section (for both the results and the discussion), first pointing out the similarities in the results (there are many), then pointing out the differences. This would not only reduce the amount of text but also improve the flow.
Related to the last point, the authors could reduce the number of tables and figures in the main text by focusing on specific questions. Right now, there are five tables and six figures, and many of the figures are multipaneled. Tables 2 and 5 seem like they could easily be moved to the supplemental. If the manuscript’s focus is “storage,” “stocks,” and “accumulation,” as indicated in the first three sentences of the abstract, then it also seems like it is unnecessary to report the C and N concentrations in the main text (i.e., Figures 1 and 2 could be moved to supplemental).
Under each crop type (maize, soybean, and casava), the conventional till system does not have a cover crop whereas the no-till systems have cover crops. Therefore, the effects of no-till and cover crops are confounded. This should be corrected throughout the manuscript when referring to “no-till” effects, including in the title.
I found the descriptions of the experiments and the treatments to be insufficient. I had to reference several of the previous publications from this study to understand that 1) the maize, soybean, and casava treatments were “separate” experiments at the same site and, 2) NT2 and NT3 are the two phases of a rotation, and 3) the site has been in various agricultural uses since 1937. These are important details that must be explained sufficiently in the text. There are still some questions that I have about the experimental design that I list in the specific comments below.
The authors use the equivalent soil mass (ESM) approach to calculate changes in SOC and N stocks, which is robust assuming that there are no losses or gains of mineral soil. However, the authors also mention that erosion is an issue in this area, which would cause ESM (and the fixed-depth approach) to be inaccurate. The authors should address this issue, either by providing assurance that significant erosion does not occur at the site, or providing an approximation of erosion rates (and thus the uncertainty of stock change estimates). Also, the C and N concentrations and POM/MAOM values were not calculated using ESM, but these could also be affected by changes in bulk density. I recommend using ESM for these values in addition to the stocks.
Specific comments:
L35-38: The results given at the end of the abstract indicate that the NT system contained cover crops, but that is not mentioned here. A better description of the systems is necessary.
L40-41: It is not necessary to list all the depth increments here, as they are not reported in the abstract.
L48-50, and throughout the manuscript: If the results are not significant, then do not calculate the percent change.
L61: This implies that diversity was explicitly tested in this experiment, which it was not. If diversity is of interest, then it should be tested in the analysis.
L76, elsewhere: The word “inappropriate” could be changed to something more specific like “intensive.”
L81-82: Please provide a citation.
L96: The fate of eroded SOC is difficult to assess. See Sanderman and Chappell 2013 (doi:10.1111/gcb.12030).
L99-100: Please provide a citation.
L101: The term “SOC content” is ambiguous, as it is sometimes used to describe stocks and sometimes used to describe mass fractions (“concentration”). Please be explicit.
L112: NT is part of CA, should “NT” can be removed from this sentence.
L121-122: The introduction should have more information about this. Some of this could be moved from the discussion, e.g., L793-796.
L153-168: This seems out of place. It should come earlier in the discussion because it is part of the “big picture.”
L171-173: What is the hypothesis regarding the POM and MAOM?
L173-174: This isn’t so much a hypothesis as it is a test bias.
L180: What is the slope of the site?
L201-207: The treatment abbreviations for the systems (especially the NT systems) could be improved to better convey similarities and differences. For example, CTM (conventional till monoculture), NTM (no-till monoculture), NTR1 (no-till rotation phase 1), and NTR2 (no-till rotation phase 2).
L205-207: This makes it sound like the NT2-Sb, NT3-Sb, NT2-Cs, and NT3-Cs do not have cover crops, but I don’t believe that is correct.
L221-222: Why weren’t the root inputs estimated, for example, by using a root-to-shoot ratio?
L253: Pyrophosphate is a more common term; otherwise thermophosphate as one word.
Table 1: While this table is very comprehensive, it is very difficult to interpret because there are so many different types of crops. I would recommend making it easier for the reader to understand by putting the crops into categories. For example, main crop could be bolded, cash crop could be underlined, and cover crop could be italicized.
Table 2: It seems like this can be supplemental information that is referenced in the text where appropriate.
L282: A 20-year-old coffee plantation with otherwise identical history to the experimental treatments does not seem like appropriate “reference vegetation.” Reference vegetation would be an old growth forest (e.g., 100 years old) or remnant forest (never cultivated). Readers might assume that the “reference vegetation” contains the maximum SOC stock possible, which is highly unlikely in this case. Moreover, the SOC stock in the “reference vegetation” was only measured to 20 cm deep. I suggest that the reference vegetation data and Figure 3 are removed.
L286: Can the authors explain the rationale for using a different soil sampling strategy in 2021 compared to 2011?
L289: What is the model of the “soil column cylinder auger?”
L297, 301: It appears that SOC and N was measured differently between 2011 and 2021 (different lab, different instrument). Did the authors run any of the samples on both instruments/labs to check for consistency. Or could this difference explain the unexpected decrease in N reported in 2021? For example, a small change in the N calibration between instruments could result in apparent N changes through time.
L351: “computed” could be “conducted.”
L367-371, 512-517: This preamble is unnecessary.
Figure 1. Figure 2: In general, error bars are necessary for graphs. Also, the description does not mention that text in bold brackets in the 2011 panels (presumably) represents a significant decrease.
Table 3, Table 4: Measurements should have units for error (e.g., standard error).
L599: The discussion would be stronger if it started with the takeaway messages and significance of the study.
L619-630: These are detailed results that should not be presented in the discussion.
L665-679: This would fit better with the discussion of POM and MAOM.
L680: This seems like a separate idea from the prior part of the paragraph
L707-711: This should be in the results section.
L714: A major reason why the synchronic method produced lower SOC stock changes than the diachronic results is because the control treatment had increased SOC stocks over time (although it may not have been statistically significant, it was still numerically higher, which affected the calculation).
Figure 6: Statistically significant changes (i.e., different from zero) should be noted, for example with an asterisk. Also, it would be helpful to draw a horizonal line on each graph to represent zero.
L755-764: Could N uptake and/or N priming from the cover crops have resulted in N loss?
L765-791: This section doesn’t add much in addition to the previous section, especially since the comparison the “RV” was only done to 20 cm and the RV vegetation is not likely representative of the pre-agricultural SOC stocks. I recommend removing this.
L796-799: This level of detail should be confined to the results.
L825-839: This is too much detail for the conclusions. Provide a high-level overview of the results and implications.
L846: In addition to “question about N dynamics,” it also raises questions about the sustainability of these systems in general. If the systems are continually losing N, they may not be sustainable. Moreover, SOC has a fixed ranged of C:N, so at some point, C will no longer accumulate if N is being lost.
Citation: https://doi.org/10.5194/egusphere-2024-541-RC1
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
205 | 59 | 9 | 273 | 27 | 4 | 4 |
- HTML: 205
- PDF: 59
- XML: 9
- Total: 273
- Supplement: 27
- BibTeX: 4
- EndNote: 4
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1