the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Can soil spectroscopy contribute to soil organic carbon monitoring on agricultural soils?
Abstract. Carbon sequestration in soils under agricultural use can contribute to climate change mitigation. However, the spatial-temporal monitoring of soil organic carbon (SOC) requires more efficient data acquisition. The use of soil Vis-NIR spectroscopy is a promising research field in this context. However, the interpretation of the recorded spectral signal with regards to SOC is not trivial due to the complexity of the soil matrix, and factors affecting the measurements under field conditions. A model-building process is required to relate the spectral signal to the SOC content. For this study, spectral on-the-go proximal measurements and soil sampling were conducted on a long-term field experiment (LTE) located in the state of Saxony-Anhalt, Germany. SOC values ranged between 14–25 g kg−1 due to different fertilization treatments. Partial least squares regression (PLSR) models were built on behalf of spectral laboratory and field measurements conducted with two spectrometers and preprocessed by various methods. A data correction of the field data was done with three different approaches: linear transformation, piecewise direct standardization (PDS), and external parameter orthogonalization (EPO). The models were then thoroughly interpreted with regards to spectral wavelength importance using regression coefficients (RC) and variable importance in projection scores (VIP). The detailed wavelength importance analysis disclosed the challenge of using soil spectroscopy for SOC monitoring. The use of spectrometers with a differing spectral resolution for soil Vis-NIR measurements under varying soil conditions revealed shifts in wavelength importance. Still, some wavelengths related to SOC were extracted (560 nm, 1330 nm, 1400 nm, 1720 nm, and 1900 nm) by various preprocessing methods and were highly important in models trained on both, laboratory, and field measurements. Furthermore, we showed, that the correction of spectral field data with spectral laboratory measurements improved the predictive performance of the models built on behalf of the proximal on-the-go sensing measurements.
- Preprint
(4940 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2022-273', Anonymous Referee #1, 10 Jul 2022
This paper aims to evaluate the ability to use soil vis-NIR spectroscopy to monitor SOC at the field scale. Two instruments (ASD and Veris) are used to acquire spectral data under laboratory and field conditions. Different spectral processing methods are combined with PLS regression to derive the best prediction models. In addition, spectral field data are corrected with the laboratory-based counterparts, which improves the performance of models built with field measurements. The topic is interesting and fits within the scope of SOIL. However, I think the paper needs a major restructuring and more clarity in the objectives and results section.
From the title, abstract (L5-10), and the introduction section (L75-80), I can see that this paper aims to evaluate the possibility of using vis-NIR to detect the spatial-temporal changes in SOC at the field scale. This is an interesting research question, because spatially continuous SOC monitoring in an efficient manner is important for studying SOC sequestration and formulating climate mitigation strategies. However, as I advanced in reading through the results and discussion sections, it seems to me that the current analysis actually focuses on ‘assessing the ability to use vis-NIR to predict SOC’’. Many studies on a field, regional or continental scale actually have done this, using PLS (current study), Memory-based machine learning, or Cubist models combing with different spectral processing techniques. Based on this, it is not clear to me what the main objective of this paper is, and thus the novelty of this paper.
If the authors focus on ‘spatially continuous SOC monitoring based on the long-term field experiment’, I think readers may want to know whether we can accurately detect changes in SOC using field vis-NIR spectroscopy (Veris), and what are the main factors that affect the accuracy of predictions. I am assuming that if the changes in SOC are small (e.g., based on a short-term fertilization experiment), then vis-NIR may be unable to capture it because of the higher prediction error based on the field spectral measurements (Figure 4 Veris). Furthermore, the addition of chemical fertilizers (L100) may affect the soil mineral matrix, e.g., by releasing the Fe minerals, which will strongly affect the spectral absorption features in the electronic transition region (425, 480, 513, 650, 903, 1000 nm). So, the treatments (fertilizer addition, and crop rotation… as described in L95) may lead to different interpretations in terms of model performance parameters, RC, and VIP. In my opinion, authors should focus on this perspective and pay more attention to what vis-NIR can do in detecting changes in SOC at spatial and temporal scales. If authors were to add new analysis in this regard, you may delete some previous analysis, i.e., comparing the model performance of 4 spectral processing methods (just pick the best one), because 1) multiple studies have shown that the effect of pre-processing on the accuracy of SOC predictions is very limited (Baldock et al., 2013; Dotto et al., 2018; Igne et al., 2010), and 2) too much information may distract from the key message of this paper.
In a summary, the manuscript is not appropriate for publication in SOIL in its present form. I suggest authors carefully consider the objectives of this paper and then reformulate the title, results, and conclusions.
Some minor comments:
Abstract Only 3 sentences are about the results and conclusions while concentrating too much on the description of the context and methods. Need to revise.
L35-42 in the introduction, the authors described the continuous SOC monitoring experiments in detail, so I think this is relevant to your research objectives. but…(see general comment).
L60 be more specific, preferably by providing examples.
L158 I suggest adding the ‘Ratio of Performance to Deviation (RPD)’ as an additional model performance parameter. This is a robust, widely used indicator to judge spectral models. (Change et al., 2001; Viscarra Rossel et a., 2012 EJSS).
L195-197 R2 < 0.5, PRD < 1.4? it seems an un-reliable model, please check with RPD values.
L330 revise this conclusion after 1) calculating RPD (previous comment) to judge model performance; 2) adding analysis regarding “monitoring change” (see general comment)
L250-270 consider that add statistical analysis to compare the differences in model performance between different spectral processing methods and spectral correction methods (L306)
L330, L345 be careful with these conclusions: given the low model performance of Veris field IR (Fig. 4 Veris field-SGCR, R2< 0.5), small changes in SOC may be undetectable using vis-NIR.
Figure cation: explain the elements of boxplot, e.g., min, max, Q1 Q3 outliers...
Figure 3, 5: consider combining the plots that belong to the same categories, e.g., A with B, C with D (using colors to distinguish them).
Figure 9, 10: round the wavelength numbers
Figures 7, 8, 9, 10. If I understand correctly, Figure 9 (median values) only carried part of the information that was already given by Figure 7 (median + range)? If yes, why not combine them? Also for Figure 8 and Figure 10.
Citation: https://doi.org/10.5194/egusphere-2022-273-RC1 -
AC1: 'Reply on RC1', Javier Reyes, 11 Nov 2022
Replies are included in bold text
This paper aims to evaluate the ability to use soil vis-NIR spectroscopy to monitor SOC at the field scale. Two instruments (ASD and Veris) are used to acquire spectral data under laboratory and field conditions. Different spectral processing methods are combined with PLS regression to derive the best prediction models. In addition, spectral field data are corrected with the laboratory-based counterparts, which improves the performance of models built with field measurements. The topic is interesting and fits within the scope of SOIL. However, I think the paper needs a major restructuring and more clarity in the objectives and results section.
From the title, abstract (L5-10), and the introduction section (L75-80), I can see that this paper aims to evaluate the possibility of using vis-NIR to detect the spatial-temporal changes in SOC at the field scale. This is an interesting research question because spatially continuous SOC monitoring in an efficient manner is important for studying SOC sequestration and formulating climate mitigation strategies. However, as I advanced in reading through the results and discussion sections, it seems to me that the current analysis actually focuses on ‘assessing the ability to use vis-NIR to predict SOC’’. Many studies on a field, regional or continental scale actually have done this, using PLS (current study), Memory-based machine learning, or Cubist models combing with different spectral processing techniques. Based on this, it is not clear to me what the main objective of this paper is, and thus the novelty of this paper.
Thank you for your comments. Our work aimed to evaluate the potential of using on-the-go VisNIR data to predict SOC at the field scale for SOC monitoring. Although many studies have evaluated the potential of VisNIR to predict SOC, few studies have shown the application of the on-the-go data which could be used to predict the spatial and temporal variation on the field with appropriate resolution, and, the application of data correction to the on-the-go data is even more uncommon (more examples are found in punctual field measurements or remote sensing data).
If the authors focus on ‘spatially continuous SOC monitoring based on the long-term field experiment’, I think readers may want to know whether we can accurately detect changes in SOC using field vis-NIR spectroscopy (Veris), and what are the main factors that affect the accuracy of predictions. I am assuming that if the changes in SOC are small (e.g., based on a short-term fertilization experiment), then vis-NIR may be unable to capture it because of the higher prediction error based on the field spectral measurements (Figure 4 Veris). Furthermore, the addition of chemical fertilizers (L100) may affect the soil mineral matrix, e.g., by releasing the Fe minerals, which will strongly affect the spectral absorption features in the electronic transition region (425, 480, 513, 650, 903, 1000 nm). So, the treatments (fertilizer addition, and crop rotation… as described in L95) may lead to different interpretations in terms of model performance parameters, RC, and VIP. In my opinion, authors should focus on this perspective and pay more attention to what vis-NIR can do in detecting changes in SOC at spatial and temporal scales. If authors were to add new analysis in this regard, you may delete some previous analysis, i.e., comparing the model performance of 4 spectral processing methods (just pick the best one), because 1) multiple studies have shown that the effect of pre-processing on the accuracy of SOC predictions is very limited (Baldock et al., 2013; Dotto et al., 2018; Igne et al., 2010), and 2) too much information may distract from the key message of this paper.
The manuscript will be adapted to put more emphasis on the potential of on-the-go measurements to obtain reliable data for SOC prediction and how the data could be corrected based on additional laboratory measurements, which is the novelty of the work. We tested different methods of preprocessing methods, as we wanted to achieve the best model performance and identify how it could change when comparing devices and field/laboratory conditions, and also for the methods of field data correction.
In our work, we tested deeply the potential of the on-the-go spectrometer, and for that reason with compared it with conventional laboratory measurements to identify how far it was in terms of model performance. We agree with the referee that further work is needed in terms of evaluating spatial and temporal changes in SOC variation, which could be done in future works as the next steps, considering that temporal changes in SOC in the field need to be evaluated in longer periods to be manifested. The effect of fertilizers noted by the referee will be included as part of the discussion.
In a summary, the manuscript is not appropriate for publication in SOIL in its present form. I suggest authors carefully consider the objectives of this paper and then reformulate the title, results, and conclusions.
Some minor comments:
Abstract Only 3 sentences are about the results and conclusions while concentrating too much on the description of the context and methods. Need to revise.
The abstract will be adapted accordingly in the new version.
L35-42 in the introduction, the authors described the continuous SOC monitoring experiments in detail, so I think this is relevant to your research objectives. but…(see general comment).
The reply is included in the general comment.
L60 be more specific, preferably by providing examples.
The references regarding this aspect were included (Lee et al., 2009; Sarathjith et al., 2016)
L158 I suggest adding the ‘Ratio of Performance to Deviation (RPD)’ as an additional model performance parameter. This is a robust, widely used indicator to judge spectral models. (Change et al., 2001; Viscarra Rossel et a., 2012 EJSS).
RPD will be added to the results.
L195-197 R2 < 0.5, PRD < 1.4? it seems an un-reliable model, please check with RPD values.
RPD will be part of the revised version. The statement was based on the models with the best performance (which was one of the reasons to test different methods), where we found an R2 < 0.75, and now calculated an RPD < 2 in the Veris Field-gapDer with EPO correction.
L330 revise this conclusion after 1) calculating RPD (previous comment) to judge model performance; 2) adding analysis regarding “monitoring change” (see general comment).
These comments were previously replied.
L250-270 consider that add statistical analysis to compare the differences in model performance between different spectral processing methods and spectral correction methods (L306)
Thank you for the suggestion. We prefer to not include an additional analysis to compare models, as we wanted to find the best models and not focus the study on methods comparison.
L330, L345 be careful with these conclusions: given the low model performance of Veris field IR (Fig. 4 Veris field-SGCR, R2< 0.5), small changes in SOC may be undetectable using vis-NIR.
The statement was done based on the Veris field corrected data which provided acceptable results on best models as we responded on a previous comment.
Figure cation: explain the elements of boxplot, e.g., min, max, Q1 Q3 outliers...
We prefer to not add a description to the elements of the boxplot, as we understand it is a standardized type of plot.
Figure 3, 5: consider combining the plots that belong to the same categories, e.g., A with B, C with D (using colors to distinguish them).
We prefer to maintain the figures separated to have better visibility of the results.
Figure 9, 10: round the wavelength numbers
The wavelength numbers will be rounded.
Figures 7, 8, 9, 10. If I understand correctly, Figure 9 (median values) only carried part of the information that was already given by Figure 7 (median + range)? If yes, why not combine them? Also for Figure 8 and Figure 10.
As figures 9 and 10 are showing local peaks of important wavelength, they will be maintained. Figures 7-8 will be included as supplemental material, as the ranges vary relatively low.
Citation: https://doi.org/10.5194/egusphere-2022-273-AC1
-
AC1: 'Reply on RC1', Javier Reyes, 11 Nov 2022
-
RC2: 'Comment on egusphere-2022-273', Anonymous Referee #2, 02 Oct 2022
The manuscript “Can soil spectroscopy contribute to soil organic carbon monitoring on agricultural soils?” deals with an interesting topic and aims at evaluating the capability of soil spectroscopy in predicting SOC contents under laboratory and field conditions. Two instruments and different pre-processing methods were tested. The script is well-written, the figures are well displayed and the results are clear. The analysis about predictive bands under Lab and field conditions are comprehensive and informative.
However, there are some concerns related to:
Regarding the objective of this study, the title seems to contain two aspects of research questions: (1) the capability of soil spectroscopy (in SOC prediction), and (2) how to deal with disturbing factors that come from field conditions (for on-the-go measurements). To me, the first aspect is not new because the soil spectroscopy has been widely proved its capability for SOC estimation, especially under laboratory condition. The interesting point here might be whether it is capable for dealing with the relatively small variation of SOC content in this study (14-25 g kg-1), otherwise I do not consider there is novelty in this regard. As for the second aspect, for agricultural fields, the results might be influenced by:
- the presence of some residues
- soil roughness left by plough, harrowing or other tillage operations
- moisture (depending e.g. on the rain events or wet air)
- curst
The manuscript did not provide soil information from these aspects, thus the readers are not aware of how different the soils are between field conditions and laboratory ‘standard’ soil samples, e.g., by providing some photos of field condition or providing some descriptive indices of soil roughness or moisture. It may help readers understand how much it is effective of these pre-processing methods when mitigating noisy signals coming from field conditions.
The effects of field disturbing factors should be better discussed if we want to know whether such on-the-go measurements are consistent if days/months past and the field conditions changed. Specifically, for such small variation of SOC range, whether the disturbing factors result in error that beyond the limitation of detection, i.e., whether the change of spectral signals caused by disturbing factors results in wrong estimation of SOC contents. Here it involves spatial and temporal variations of the heterogeneity of disturbing factors, and in my opinion these issues are interesting to address for the on-the-go spectral measurements. More specifically, the different pre-processing methods were well compared to derive the best prediction models, however, the readers do not get the information about to what extent these methods can deal with different types/levels of disturbing factors, and to what extent the solutions are transferable, e.g., if we switch a dataset derived from another field, is the best processing method in this study still the best one? Which processing method is sensitive to SOC-variation induced spectrum feature and immune to disturbing factors induced spectrum noise?
In my opinion, to answer the question of the title, the authors might need to focus more on field conditions and (quantifying) their effects on soil spectrum in-situ, and whether/what/to what extent pre-processing methods can mitigate such perturbing effects.
In summary, the manuscript may have not sufficiently or comprehensively answered the question of the title, and accordingly may need to improve: 1. The clarity of objectives; 2. The structure of the results: paying less attention to Lab-based results, paying more attention to Field-based results and quantifying the disturbing factors of field conditions, and even, quantifying the variation range of such conditions. However, I understand it might be hard to achieve this goal based on current experimental design, and I look forward to follow-up studies if possible.
If the main content and structure are going to be remained, I suggest the title should be modified into a more specific one to focus on the pre-processing methods.
Some minor suggestions: (1) Adding RPD or RPIQ as additional indicators apart from R2 (see Bellon-Maurel et al., (2010). https://doi.org/10.1016/j.trac.2010.05.006) (2) Giving more context about perturbing factors and relevant pre-processing approaches in introduction. (3) These studies may be relevant: https://doi.org/10.1016/j.geoderma.2021.115432
https://doi.org/10.1016/j.catena.2012.01.007
https://doi.org/10.1016/j.geoderma.2009.06.002
etc.
Citation: https://doi.org/10.5194/egusphere-2022-273-RC2 -
AC2: 'Reply on RC2', Javier Reyes, 11 Nov 2022
Replies are included in bold text
The manuscript “Can soil spectroscopy contribute to soil organic carbon monitoring on agricultural soils?” deals with an interesting topic and aims at evaluating the capability of soil spectroscopy in predicting SOC contents under laboratory and field conditions. Two instruments and different pre-processing methods were tested. The script is well-written, the figures are well displayed and the results are clear. The analysis about predictive bands under Lab and field conditions are comprehensive and informative.
However, there are some concerns related to:
Regarding the objective of this study, the title seems to contain two aspects of research questions: (1) the capability of soil spectroscopy (in SOC prediction), and (2) how to deal with disturbing factors that come from field conditions (for on-the-go measurements). To me, the first aspect is not new because the soil spectroscopy has been widely proved its capability for SOC estimation, especially under laboratory condition. The interesting point here might be whether it is capable for dealing with the relatively small variation of SOC content in this study (14-25 g kg-1), otherwise I do not consider there is novelty in this regard. As for the second aspect, for agricultural fields, the results might be influenced by:
- the presence of some residues
- soil roughness left by plough, harrowing or other tillage operations
- moisture (depending e.g. on the rain events or wet air)
- curst
The manuscript did not provide soil information from these aspects, thus the readers are not aware of how different the soils are between field conditions and laboratory ‘standard’ soil samples, e.g., by providing some photos of field condition or providing some descriptive indices of soil roughness or moisture. It may help readers understand how much it is effective of these pre-processing methods when mitigating noisy signals coming from field conditions.
Thank you for your comments. Our work aimed to evaluate the potential of using on-the-go VisNIR data to predict SOC at the field scale for SOC monitoring. The title and content of the manuscript will be adapted to put more emphasis on the potential of on-the-go measurements to obtain reliable data for SOC prediction and how the data could be corrected, which is the novelty of the work. As it was noted by the referee, different field conditions are affecting the measurements thus we apply a correction of the data based on laboratory measurements to reduce this effect. Additional information and discussion on the field conditions will be added to the manuscript.
The effects of field disturbing factors should be better discussed if we want to know whether such on-the-go measurements are consistent if days/months past and the field conditions changed. Specifically, for such small variation of SOC range, whether the disturbing factors result in error that beyond the limitation of detection, i.e., whether the change of spectral signals caused by disturbing factors results in wrong estimation of SOC contents. Here it involves spatial and temporal variations of the heterogeneity of disturbing factors, and in my opinion these issues are interesting to address for the on-the-go spectral measurements. More specifically, the different pre-processing methods were well compared to derive the best prediction models, however, the readers do not get the information about to what extent these methods can deal with different types/levels of disturbing factors, and to what extent the solutions are transferable, e.g., if we switch a dataset derived from another field, is the best processing method in this study still the best one? Which processing method is sensitive to SOC-variation induced spectrum feature and immune to disturbing factors induced spectrum noise?
Our work was the first step to test the potential of the on-the-go spectrometer, and for that reason with compared it with conventional laboratory measurements to identify how far it was in terms of model performance in our case. The field data correction was done in a way of reducing the effect of the field disturbing factor in the data. As we mentioned in the discussion several factors will affect the transferability, including the devices and methods to obtain the data.
In my opinion, to answer the question of the title, the authors might need to focus more on field conditions and (quantifying) their effects on soil spectrum in-situ, and whether/what/to what extent pre-processing methods can mitigate such perturbing effects.
In summary, the manuscript may have not sufficiently or comprehensively answered the question of the title, and accordingly may need to improve: 1. The clarity of objectives; 2. The structure of the results: paying less attention to Lab-based results, paying more attention to Field-based results and quantifying the disturbing factors of field conditions, and even, quantifying the variation range of such conditions. However, I understand it might be hard to achieve this goal based on current experimental design, and I look forward to follow-up studies if possible.
If the main content and structure are going to be remained, I suggest the title should be modified into a more specific one to focus on the pre-processing methods.
Regarding the last three paragraphs, we will do adaptations to the title and structure of the manuscript as we mentioned in previous replies.
Some minor suggestions: (1) Adding RPD or RPIQ as additional indicators apart from R2 (see Bellon-Maurel et al., (2010). https://doi.org/10.1016/j.trac.2010.05.006) (2) Giving more context about perturbing factors and relevant pre-processing approaches in introduction. (3) These studies may be relevant:
https://doi.org/10.1016/j.geoderma.2021.115432
https://doi.org/10.1016/j.catena.2012.01.007
https://doi.org/10.1016/j.geoderma.2009.06.002
Regarding the minor suggestions:
- RPD will be added to the results (which was also suggested by another referee).
- We will add more information in the introduction.
- Thank you for recommending additional references. It will be considered in the adaptation of the manuscript.
Citation: https://doi.org/10.5194/egusphere-2022-273-AC2
-
AC2: 'Reply on RC2', Javier Reyes, 11 Nov 2022
Status: closed
-
RC1: 'Comment on egusphere-2022-273', Anonymous Referee #1, 10 Jul 2022
This paper aims to evaluate the ability to use soil vis-NIR spectroscopy to monitor SOC at the field scale. Two instruments (ASD and Veris) are used to acquire spectral data under laboratory and field conditions. Different spectral processing methods are combined with PLS regression to derive the best prediction models. In addition, spectral field data are corrected with the laboratory-based counterparts, which improves the performance of models built with field measurements. The topic is interesting and fits within the scope of SOIL. However, I think the paper needs a major restructuring and more clarity in the objectives and results section.
From the title, abstract (L5-10), and the introduction section (L75-80), I can see that this paper aims to evaluate the possibility of using vis-NIR to detect the spatial-temporal changes in SOC at the field scale. This is an interesting research question, because spatially continuous SOC monitoring in an efficient manner is important for studying SOC sequestration and formulating climate mitigation strategies. However, as I advanced in reading through the results and discussion sections, it seems to me that the current analysis actually focuses on ‘assessing the ability to use vis-NIR to predict SOC’’. Many studies on a field, regional or continental scale actually have done this, using PLS (current study), Memory-based machine learning, or Cubist models combing with different spectral processing techniques. Based on this, it is not clear to me what the main objective of this paper is, and thus the novelty of this paper.
If the authors focus on ‘spatially continuous SOC monitoring based on the long-term field experiment’, I think readers may want to know whether we can accurately detect changes in SOC using field vis-NIR spectroscopy (Veris), and what are the main factors that affect the accuracy of predictions. I am assuming that if the changes in SOC are small (e.g., based on a short-term fertilization experiment), then vis-NIR may be unable to capture it because of the higher prediction error based on the field spectral measurements (Figure 4 Veris). Furthermore, the addition of chemical fertilizers (L100) may affect the soil mineral matrix, e.g., by releasing the Fe minerals, which will strongly affect the spectral absorption features in the electronic transition region (425, 480, 513, 650, 903, 1000 nm). So, the treatments (fertilizer addition, and crop rotation… as described in L95) may lead to different interpretations in terms of model performance parameters, RC, and VIP. In my opinion, authors should focus on this perspective and pay more attention to what vis-NIR can do in detecting changes in SOC at spatial and temporal scales. If authors were to add new analysis in this regard, you may delete some previous analysis, i.e., comparing the model performance of 4 spectral processing methods (just pick the best one), because 1) multiple studies have shown that the effect of pre-processing on the accuracy of SOC predictions is very limited (Baldock et al., 2013; Dotto et al., 2018; Igne et al., 2010), and 2) too much information may distract from the key message of this paper.
In a summary, the manuscript is not appropriate for publication in SOIL in its present form. I suggest authors carefully consider the objectives of this paper and then reformulate the title, results, and conclusions.
Some minor comments:
Abstract Only 3 sentences are about the results and conclusions while concentrating too much on the description of the context and methods. Need to revise.
L35-42 in the introduction, the authors described the continuous SOC monitoring experiments in detail, so I think this is relevant to your research objectives. but…(see general comment).
L60 be more specific, preferably by providing examples.
L158 I suggest adding the ‘Ratio of Performance to Deviation (RPD)’ as an additional model performance parameter. This is a robust, widely used indicator to judge spectral models. (Change et al., 2001; Viscarra Rossel et a., 2012 EJSS).
L195-197 R2 < 0.5, PRD < 1.4? it seems an un-reliable model, please check with RPD values.
L330 revise this conclusion after 1) calculating RPD (previous comment) to judge model performance; 2) adding analysis regarding “monitoring change” (see general comment)
L250-270 consider that add statistical analysis to compare the differences in model performance between different spectral processing methods and spectral correction methods (L306)
L330, L345 be careful with these conclusions: given the low model performance of Veris field IR (Fig. 4 Veris field-SGCR, R2< 0.5), small changes in SOC may be undetectable using vis-NIR.
Figure cation: explain the elements of boxplot, e.g., min, max, Q1 Q3 outliers...
Figure 3, 5: consider combining the plots that belong to the same categories, e.g., A with B, C with D (using colors to distinguish them).
Figure 9, 10: round the wavelength numbers
Figures 7, 8, 9, 10. If I understand correctly, Figure 9 (median values) only carried part of the information that was already given by Figure 7 (median + range)? If yes, why not combine them? Also for Figure 8 and Figure 10.
Citation: https://doi.org/10.5194/egusphere-2022-273-RC1 -
AC1: 'Reply on RC1', Javier Reyes, 11 Nov 2022
Replies are included in bold text
This paper aims to evaluate the ability to use soil vis-NIR spectroscopy to monitor SOC at the field scale. Two instruments (ASD and Veris) are used to acquire spectral data under laboratory and field conditions. Different spectral processing methods are combined with PLS regression to derive the best prediction models. In addition, spectral field data are corrected with the laboratory-based counterparts, which improves the performance of models built with field measurements. The topic is interesting and fits within the scope of SOIL. However, I think the paper needs a major restructuring and more clarity in the objectives and results section.
From the title, abstract (L5-10), and the introduction section (L75-80), I can see that this paper aims to evaluate the possibility of using vis-NIR to detect the spatial-temporal changes in SOC at the field scale. This is an interesting research question because spatially continuous SOC monitoring in an efficient manner is important for studying SOC sequestration and formulating climate mitigation strategies. However, as I advanced in reading through the results and discussion sections, it seems to me that the current analysis actually focuses on ‘assessing the ability to use vis-NIR to predict SOC’’. Many studies on a field, regional or continental scale actually have done this, using PLS (current study), Memory-based machine learning, or Cubist models combing with different spectral processing techniques. Based on this, it is not clear to me what the main objective of this paper is, and thus the novelty of this paper.
Thank you for your comments. Our work aimed to evaluate the potential of using on-the-go VisNIR data to predict SOC at the field scale for SOC monitoring. Although many studies have evaluated the potential of VisNIR to predict SOC, few studies have shown the application of the on-the-go data which could be used to predict the spatial and temporal variation on the field with appropriate resolution, and, the application of data correction to the on-the-go data is even more uncommon (more examples are found in punctual field measurements or remote sensing data).
If the authors focus on ‘spatially continuous SOC monitoring based on the long-term field experiment’, I think readers may want to know whether we can accurately detect changes in SOC using field vis-NIR spectroscopy (Veris), and what are the main factors that affect the accuracy of predictions. I am assuming that if the changes in SOC are small (e.g., based on a short-term fertilization experiment), then vis-NIR may be unable to capture it because of the higher prediction error based on the field spectral measurements (Figure 4 Veris). Furthermore, the addition of chemical fertilizers (L100) may affect the soil mineral matrix, e.g., by releasing the Fe minerals, which will strongly affect the spectral absorption features in the electronic transition region (425, 480, 513, 650, 903, 1000 nm). So, the treatments (fertilizer addition, and crop rotation… as described in L95) may lead to different interpretations in terms of model performance parameters, RC, and VIP. In my opinion, authors should focus on this perspective and pay more attention to what vis-NIR can do in detecting changes in SOC at spatial and temporal scales. If authors were to add new analysis in this regard, you may delete some previous analysis, i.e., comparing the model performance of 4 spectral processing methods (just pick the best one), because 1) multiple studies have shown that the effect of pre-processing on the accuracy of SOC predictions is very limited (Baldock et al., 2013; Dotto et al., 2018; Igne et al., 2010), and 2) too much information may distract from the key message of this paper.
The manuscript will be adapted to put more emphasis on the potential of on-the-go measurements to obtain reliable data for SOC prediction and how the data could be corrected based on additional laboratory measurements, which is the novelty of the work. We tested different methods of preprocessing methods, as we wanted to achieve the best model performance and identify how it could change when comparing devices and field/laboratory conditions, and also for the methods of field data correction.
In our work, we tested deeply the potential of the on-the-go spectrometer, and for that reason with compared it with conventional laboratory measurements to identify how far it was in terms of model performance. We agree with the referee that further work is needed in terms of evaluating spatial and temporal changes in SOC variation, which could be done in future works as the next steps, considering that temporal changes in SOC in the field need to be evaluated in longer periods to be manifested. The effect of fertilizers noted by the referee will be included as part of the discussion.
In a summary, the manuscript is not appropriate for publication in SOIL in its present form. I suggest authors carefully consider the objectives of this paper and then reformulate the title, results, and conclusions.
Some minor comments:
Abstract Only 3 sentences are about the results and conclusions while concentrating too much on the description of the context and methods. Need to revise.
The abstract will be adapted accordingly in the new version.
L35-42 in the introduction, the authors described the continuous SOC monitoring experiments in detail, so I think this is relevant to your research objectives. but…(see general comment).
The reply is included in the general comment.
L60 be more specific, preferably by providing examples.
The references regarding this aspect were included (Lee et al., 2009; Sarathjith et al., 2016)
L158 I suggest adding the ‘Ratio of Performance to Deviation (RPD)’ as an additional model performance parameter. This is a robust, widely used indicator to judge spectral models. (Change et al., 2001; Viscarra Rossel et a., 2012 EJSS).
RPD will be added to the results.
L195-197 R2 < 0.5, PRD < 1.4? it seems an un-reliable model, please check with RPD values.
RPD will be part of the revised version. The statement was based on the models with the best performance (which was one of the reasons to test different methods), where we found an R2 < 0.75, and now calculated an RPD < 2 in the Veris Field-gapDer with EPO correction.
L330 revise this conclusion after 1) calculating RPD (previous comment) to judge model performance; 2) adding analysis regarding “monitoring change” (see general comment).
These comments were previously replied.
L250-270 consider that add statistical analysis to compare the differences in model performance between different spectral processing methods and spectral correction methods (L306)
Thank you for the suggestion. We prefer to not include an additional analysis to compare models, as we wanted to find the best models and not focus the study on methods comparison.
L330, L345 be careful with these conclusions: given the low model performance of Veris field IR (Fig. 4 Veris field-SGCR, R2< 0.5), small changes in SOC may be undetectable using vis-NIR.
The statement was done based on the Veris field corrected data which provided acceptable results on best models as we responded on a previous comment.
Figure cation: explain the elements of boxplot, e.g., min, max, Q1 Q3 outliers...
We prefer to not add a description to the elements of the boxplot, as we understand it is a standardized type of plot.
Figure 3, 5: consider combining the plots that belong to the same categories, e.g., A with B, C with D (using colors to distinguish them).
We prefer to maintain the figures separated to have better visibility of the results.
Figure 9, 10: round the wavelength numbers
The wavelength numbers will be rounded.
Figures 7, 8, 9, 10. If I understand correctly, Figure 9 (median values) only carried part of the information that was already given by Figure 7 (median + range)? If yes, why not combine them? Also for Figure 8 and Figure 10.
As figures 9 and 10 are showing local peaks of important wavelength, they will be maintained. Figures 7-8 will be included as supplemental material, as the ranges vary relatively low.
Citation: https://doi.org/10.5194/egusphere-2022-273-AC1
-
AC1: 'Reply on RC1', Javier Reyes, 11 Nov 2022
-
RC2: 'Comment on egusphere-2022-273', Anonymous Referee #2, 02 Oct 2022
The manuscript “Can soil spectroscopy contribute to soil organic carbon monitoring on agricultural soils?” deals with an interesting topic and aims at evaluating the capability of soil spectroscopy in predicting SOC contents under laboratory and field conditions. Two instruments and different pre-processing methods were tested. The script is well-written, the figures are well displayed and the results are clear. The analysis about predictive bands under Lab and field conditions are comprehensive and informative.
However, there are some concerns related to:
Regarding the objective of this study, the title seems to contain two aspects of research questions: (1) the capability of soil spectroscopy (in SOC prediction), and (2) how to deal with disturbing factors that come from field conditions (for on-the-go measurements). To me, the first aspect is not new because the soil spectroscopy has been widely proved its capability for SOC estimation, especially under laboratory condition. The interesting point here might be whether it is capable for dealing with the relatively small variation of SOC content in this study (14-25 g kg-1), otherwise I do not consider there is novelty in this regard. As for the second aspect, for agricultural fields, the results might be influenced by:
- the presence of some residues
- soil roughness left by plough, harrowing or other tillage operations
- moisture (depending e.g. on the rain events or wet air)
- curst
The manuscript did not provide soil information from these aspects, thus the readers are not aware of how different the soils are between field conditions and laboratory ‘standard’ soil samples, e.g., by providing some photos of field condition or providing some descriptive indices of soil roughness or moisture. It may help readers understand how much it is effective of these pre-processing methods when mitigating noisy signals coming from field conditions.
The effects of field disturbing factors should be better discussed if we want to know whether such on-the-go measurements are consistent if days/months past and the field conditions changed. Specifically, for such small variation of SOC range, whether the disturbing factors result in error that beyond the limitation of detection, i.e., whether the change of spectral signals caused by disturbing factors results in wrong estimation of SOC contents. Here it involves spatial and temporal variations of the heterogeneity of disturbing factors, and in my opinion these issues are interesting to address for the on-the-go spectral measurements. More specifically, the different pre-processing methods were well compared to derive the best prediction models, however, the readers do not get the information about to what extent these methods can deal with different types/levels of disturbing factors, and to what extent the solutions are transferable, e.g., if we switch a dataset derived from another field, is the best processing method in this study still the best one? Which processing method is sensitive to SOC-variation induced spectrum feature and immune to disturbing factors induced spectrum noise?
In my opinion, to answer the question of the title, the authors might need to focus more on field conditions and (quantifying) their effects on soil spectrum in-situ, and whether/what/to what extent pre-processing methods can mitigate such perturbing effects.
In summary, the manuscript may have not sufficiently or comprehensively answered the question of the title, and accordingly may need to improve: 1. The clarity of objectives; 2. The structure of the results: paying less attention to Lab-based results, paying more attention to Field-based results and quantifying the disturbing factors of field conditions, and even, quantifying the variation range of such conditions. However, I understand it might be hard to achieve this goal based on current experimental design, and I look forward to follow-up studies if possible.
If the main content and structure are going to be remained, I suggest the title should be modified into a more specific one to focus on the pre-processing methods.
Some minor suggestions: (1) Adding RPD or RPIQ as additional indicators apart from R2 (see Bellon-Maurel et al., (2010). https://doi.org/10.1016/j.trac.2010.05.006) (2) Giving more context about perturbing factors and relevant pre-processing approaches in introduction. (3) These studies may be relevant: https://doi.org/10.1016/j.geoderma.2021.115432
https://doi.org/10.1016/j.catena.2012.01.007
https://doi.org/10.1016/j.geoderma.2009.06.002
etc.
Citation: https://doi.org/10.5194/egusphere-2022-273-RC2 -
AC2: 'Reply on RC2', Javier Reyes, 11 Nov 2022
Replies are included in bold text
The manuscript “Can soil spectroscopy contribute to soil organic carbon monitoring on agricultural soils?” deals with an interesting topic and aims at evaluating the capability of soil spectroscopy in predicting SOC contents under laboratory and field conditions. Two instruments and different pre-processing methods were tested. The script is well-written, the figures are well displayed and the results are clear. The analysis about predictive bands under Lab and field conditions are comprehensive and informative.
However, there are some concerns related to:
Regarding the objective of this study, the title seems to contain two aspects of research questions: (1) the capability of soil spectroscopy (in SOC prediction), and (2) how to deal with disturbing factors that come from field conditions (for on-the-go measurements). To me, the first aspect is not new because the soil spectroscopy has been widely proved its capability for SOC estimation, especially under laboratory condition. The interesting point here might be whether it is capable for dealing with the relatively small variation of SOC content in this study (14-25 g kg-1), otherwise I do not consider there is novelty in this regard. As for the second aspect, for agricultural fields, the results might be influenced by:
- the presence of some residues
- soil roughness left by plough, harrowing or other tillage operations
- moisture (depending e.g. on the rain events or wet air)
- curst
The manuscript did not provide soil information from these aspects, thus the readers are not aware of how different the soils are between field conditions and laboratory ‘standard’ soil samples, e.g., by providing some photos of field condition or providing some descriptive indices of soil roughness or moisture. It may help readers understand how much it is effective of these pre-processing methods when mitigating noisy signals coming from field conditions.
Thank you for your comments. Our work aimed to evaluate the potential of using on-the-go VisNIR data to predict SOC at the field scale for SOC monitoring. The title and content of the manuscript will be adapted to put more emphasis on the potential of on-the-go measurements to obtain reliable data for SOC prediction and how the data could be corrected, which is the novelty of the work. As it was noted by the referee, different field conditions are affecting the measurements thus we apply a correction of the data based on laboratory measurements to reduce this effect. Additional information and discussion on the field conditions will be added to the manuscript.
The effects of field disturbing factors should be better discussed if we want to know whether such on-the-go measurements are consistent if days/months past and the field conditions changed. Specifically, for such small variation of SOC range, whether the disturbing factors result in error that beyond the limitation of detection, i.e., whether the change of spectral signals caused by disturbing factors results in wrong estimation of SOC contents. Here it involves spatial and temporal variations of the heterogeneity of disturbing factors, and in my opinion these issues are interesting to address for the on-the-go spectral measurements. More specifically, the different pre-processing methods were well compared to derive the best prediction models, however, the readers do not get the information about to what extent these methods can deal with different types/levels of disturbing factors, and to what extent the solutions are transferable, e.g., if we switch a dataset derived from another field, is the best processing method in this study still the best one? Which processing method is sensitive to SOC-variation induced spectrum feature and immune to disturbing factors induced spectrum noise?
Our work was the first step to test the potential of the on-the-go spectrometer, and for that reason with compared it with conventional laboratory measurements to identify how far it was in terms of model performance in our case. The field data correction was done in a way of reducing the effect of the field disturbing factor in the data. As we mentioned in the discussion several factors will affect the transferability, including the devices and methods to obtain the data.
In my opinion, to answer the question of the title, the authors might need to focus more on field conditions and (quantifying) their effects on soil spectrum in-situ, and whether/what/to what extent pre-processing methods can mitigate such perturbing effects.
In summary, the manuscript may have not sufficiently or comprehensively answered the question of the title, and accordingly may need to improve: 1. The clarity of objectives; 2. The structure of the results: paying less attention to Lab-based results, paying more attention to Field-based results and quantifying the disturbing factors of field conditions, and even, quantifying the variation range of such conditions. However, I understand it might be hard to achieve this goal based on current experimental design, and I look forward to follow-up studies if possible.
If the main content and structure are going to be remained, I suggest the title should be modified into a more specific one to focus on the pre-processing methods.
Regarding the last three paragraphs, we will do adaptations to the title and structure of the manuscript as we mentioned in previous replies.
Some minor suggestions: (1) Adding RPD or RPIQ as additional indicators apart from R2 (see Bellon-Maurel et al., (2010). https://doi.org/10.1016/j.trac.2010.05.006) (2) Giving more context about perturbing factors and relevant pre-processing approaches in introduction. (3) These studies may be relevant:
https://doi.org/10.1016/j.geoderma.2021.115432
https://doi.org/10.1016/j.catena.2012.01.007
https://doi.org/10.1016/j.geoderma.2009.06.002
Regarding the minor suggestions:
- RPD will be added to the results (which was also suggested by another referee).
- We will add more information in the introduction.
- Thank you for recommending additional references. It will be considered in the adaptation of the manuscript.
Citation: https://doi.org/10.5194/egusphere-2022-273-AC2
-
AC2: 'Reply on RC2', Javier Reyes, 11 Nov 2022
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
597 | 403 | 53 | 1,053 | 33 | 25 |
- HTML: 597
- PDF: 403
- XML: 53
- Total: 1,053
- BibTeX: 33
- EndNote: 25
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1