Advancing Soil Organic Carbon and Total Nitrogen Modelling in Peatlands: The Impact of Environmental Variable Resolution and vis-NIR Spectroscopy Integration
Round 1
Reviewer 1 Report
Dear Authors;
After my detailed evaluation on the paper;
The paper well organized. However, it needs some revision to improve. You can see my detailed suggestions in the paper file.
Comments for author File: 
Comments.pdf
Author Response
Author's Reply to the Review Report (Reviewer 1)
Dear Authors;
After my detailed evaluation on the paper;
The paper well organized. However, it needs some revision to improve. You can see my detailed suggestions in the paper file.
Comments retrieved exactly as it was in the attached file.
Line 15: Remove “therefore”.
Answer (A): It was removed!
Line 17: A total of 262 soil samples were collected from soil surface (X depth).
A: We rewrote the sentence and better explained the total number of soil samples. Please, check the reviewed version for:
“A total of 57 soil cores, comprising 262 samples from various horizons (< 2 m), were collected and analysed for SOC and TN content using traditional methods and ASD Fieldspec® 4.”
Line 21: After that you should share some results!
Answer (A): There is a limit of 200 words in the abstract section. Therefore, we summarised the main results. Please, check the reviewed version for:
“By employing the Cubist modelling approach, the results demonstrated that incorporating high-resolution LiDAR data with vis-NIR spectra improved predictions of SOC (RMSE: 4.60%, RPIQ: 9.00) and TN (RMSE: 3.06 g kg-1, RPIQ: 7.05).”
Line 23: You should add clearly conclusion!
A: There is a limit of 200 words in the abstract section. Therefore, we summarised the main results. Please, check the reviewed version for:
“In conclusion, the integration of LiDAR and soil spectroscopy holds significant potential for enhancing soil mapping and promoting sustainable soil management.”
Line 53: You should add some referances !
A: We really appreciated your suggestion. A citation was added. Please, check the reviewed version for:
“Essentially, the spectral information obtained from vis-NIR spectroscopy can be utilised to model soil properties because the spectral bands or wavelengths, which act as multiple independent variables, have an impact on the soil properties, which are the dependent variables [5]”
Line 61: You should utilize the referance below.
https://www.sciencedirect.com/science/article/pii/S1002016021600929
A: We appreciated your suggestion, but there are already enough and proper citations. Please, check the reviewed version for:
We do not agree with or support reviewers asking to cite their own work in order to increase their research indicators. It is almost clear to us who reviewed our manuscript just based on the reviewer's request.
“For instance, Cubist regression is a popular and high-performing machine learning technique for vis-NIR spectroscopy [5,10,11].”
Line 99: You should add soem detailed information about mean temperature and precipitation !
A: Thank you for your comment. We added these information. Please, check the reviewed version for:
“The main type of wetland in the study area is peatland, with an average groundwater level of up to 0.3 m depth, a mean annual temperature of 9.5°C, and a mean annual precipitation of 550 mm”
Line 126: You should add relevant reference for conventional laboratory analysis.
A: Thank you for your comment. We added these information. Please, check the reviewed version for:
“The determination of total carbon (C) and nitrogen (N) content was carried out using the elemental analysis (CNS analyser TruSpec, LECO Ltd., Mönchengladbach, Germany) through dry combustion at 1250°C [22,23].”
Line 136: How many times did measure each soil samples.
A: Thank you for your comment. We added these information. Please, check the reviewed version for:
“All vis-NIR spectra were collected using a standard contract probe in three replications. Each spectrum was an average of 10 scans taken at the centre of the samples, with a radius of 5 mm.”
Line 149: Finally, Do you have 2052 spectral points in the prediction models?
A: We are really thankful for your question because it made us to realise that the sentence was written wrong. Therefore, we correct it and added the right information. Please, check the reviewed version for:
“The spectral interval was narrowed to 400-2380 nm to exclude noisy regions of the vis-NIR spectra, resulting in 1981 spectral bands.”
Line 150: Did you apply SNV? to reduce multicolineratiy? You should utilize some paper to better understanding for pre-processing methods for spectra? You can see some relevant paper below!
https://www.sciencedirect.com/science/article/pii/S0341816222005008
A: The dataset in our study is the same as Mendes et al. (2022) used and evaluated the different pre-processing methods such as Savitzky-Golay, Standard Normal Variate, and Detrend. These authors concluded that the best fitted model was simply using a spline correction method and no further pre-processing procedure was necessary. For this reason, we did not perform further pre-processing in the vis-NIR data. Moreover, the aim of our study is not to test different pre-processing and machine learning algorithms as these topics have been already well-described in the current literature. Therefore, this information would not bring any novelty to our current paper.
Mendes et al. (2022): https://doi.org/10.1016/j.jenvman.2022.115383
Line 151: Why did not resample spectral points to reduce dimensinaly.
A: Please, see the explanation given in the last question.
Line 174: remove “i.e.,” and write “209 samples”.
A: Thank you for your comment. We added these information. Please, check the reviewed version for:
“The 262 observations were divided using the conditioned Latin Hypercube Sampling (cLHS; [27]): 80% (209 observations) for training the prediction model and 20% (53 observations) for testing accuracy.”
Line 181: you should the setting number for each parameters.
A: Thank you for your comment. I think the reviewer might not be familiar with the R package “caret”, but we can explain it. The default hyperparameter settings, such as the neighbours, number of committees, and rules were selected using the training data and a 10-fold cross-validation method implemented through the R package “caret”, which automatically selects the best fitted calibration model based on the lowest RMSE and highest R2.
Line 211: add references!
A: Thank you for your comment. We added these information. Please, check the reviewed version for:
“Nonetheless, useful information about organic and inorganic materials in soils can be related to chromophores [31] and iron minerals [32] in the vis spectral region (400 – 700 nm).”
Line 213: add references!
A: Thank you for your comment. We added these information. Please, check the reviewed version for:
“The overtones of hydroxyl (OH-), carbonate (CO32-) and sulfate (SO42-) groups, and combinations of fundamental features of carbon dioxide (CO2) and water (H2O) can be associated with absorption features in the NIR spectral region (400 – 700 nm) [31,32].”
Line 219: Figure 4. Why did not share spectral curves fo testing dataset?
A: We added the spectral curves from the testing data. Please, take a look into the Fig. 4.
Lines 267-275: This is not relevant because your study not focus on the effects of pre-processing methods on prediction performance!
A: Thank you for your suggestion. We removed the entire paragraph as suggested.
Lines 276-282: you should add detailed discussion on the prediciton performance of Vis-NIR for SOC and TN!
A: These information were already in the manuscript. It seems that the reviewer dismissed it.
Please, check the reviewed version for:
“In this study, the model evaluation was carried out using an independent test dataset of 53 observations, which avoids overly optimistic conclusions. As a benchmark to determine the potential contribution of the environmental variables derived from LiDAR and SRTM, the prediction of SOC and TN was performed using the vis-NIR spectral data alone. The results are in line with previous studies for SOC and TN [22,37–39]. For example, SOC predictions were found to be successful, with R2 ranging from 0.64 – 0.96. TN is also predicted quite well, with R2 ranging from 0.48 – 0.94.”
Reviewer 2 Report
The authors have presented an interesting paper which evaluated the effect of the advandcing soil organic carbon and total nitrogen modelling in Peatlands, the impact of environmental variable resolution and vis-NIR spectroscopy integration. The topic of this manuscript is very interesting, since in the last decades the excessive chemical fertilization provoked many damages in the environment. Following, I have included some comments aimed to enhance the paper:
1. I suggest to the authors to add a new section detailing the state of the art. In this section, authors have to describe the relevant related work in which explain the use of amended in soils and their effects, authors have to identify the innovation of their study with other existing, and to cite also some results of the efficacy of the use of organic residues.
2. Can the authors include at the end of the introduction, more details of the objectives of their study, sine they are comparing different organic residues.
3. Consider extending the conclusions and adding a Future works paragraph. The summary and Conclusions, it is better to combine them in only section of conclusions.
Finally, we consider this work very interesting, explaining the changes in soil chemical composition with the application of organics residues. Even though the authors have not cited the effects on environment protection (and I consider that it is very important), I think that this study is very important when we applied chemical or organic residues or the contribution of the two techniques for the protection of soils. The chemical changes on soil can help extremely (in other regions with similar soils and climate) in the preservation of environment, the reduction of pollution and the protection of soil.
Author Response
Author's Reply to the Review Report (Reviewer 2)
The authors have presented an interesting paper which evaluated the effect of the advandcing soil organic carbon and total nitrogen modelling in Peatlands, the impact of environmental variable resolution and vis-NIR spectroscopy integration. The topic of this manuscript is very interesting, since in the last decades the excessive chemical fertilization provoked many damages in the environment. Following, I have included some comments aimed to enhance the paper:
- 1. I suggest to the authors to add a new section detailing the state of the art. In this section, authors have to describe the relevant related work in which explain the use of amended in soils and their effects, authors have to identify the innovation of their study with other existing, and to cite also some results of the efficacy of the use of organic residues.
A: We really appreciated the reviewer’s suggestion. However, we already did that.
Please, check again the reviewed version of our paper.
- Can the authors include at the end of the introduction, more details of the objectives of their study, sine they are comparing different organic residues.
A: Thank you for your suggestion! We think there is a misunderstanding here. Our manuscript is not about different organic residues. All comments here have nothing to do with our manuscript.
- Consider extending the conclusions and adding a Future works paragraph. The summary and Conclusions, it is better to combine them in only section of conclusions.
A: Thank you for your suggestion. We did that. Please, check the reviewed version of our paper.
