Relating Profile Wall Root-Length Density Estimates to Monolith Root-Length Density Measurements of Cover Crops
, Roman Kemper 2, Phillip Müller 3, Timo Kautz 4, Thomas F. Döring 2 and Miriam Athmann 5
Round 1
Reviewer 1 Report
| Line | Comment |
| 42 | replace 'named as profile wall' with 'named the profile wall' |
| 49 | replace 'Gäth and Meuser [24] later on, reported a conversion' with 'Gäth and Meuser [24] later reported a conversion' |
| 51 | should vulgaris be italicised? |
| 61 | replace 'obtained with PW method' with 'obtained with the PW method' |
| 62 | replace 'with MO method' with 'with the MO method' |
| 64 | replace 'profile, in' with 'profile; in' |
| 69 | suggest delete 'in calibrating both methods' |
| 83 | replace 'are' with 'were' |
| 88 | make spacing between digits and units consistent in '(7 ° 17 'East, 50 ° 48' North)' |
| 88-98 | need to be consistent with units e.g. mm or cm and spacing between the digit and the unit |
| 96 | replace 'were' with 'was' |
| 97 | insert comma between after and sowing |
| 103 | Rep.' not needed in table |
| Table 1 | should olei-formis be italicised? |
| Table 1 | need scientific names of the mixed crops |
| 106 | be consistent with decimal places e.g. 1.2 and 1.0 |
| 111 and 114 | need to be consistent with units e.g. mm or cm and spacing between the digit and the unit |
| 112 | insert spaces either side of the multiplication sign |
| 114 | does root length density need a unit? |
| 123 | replace 'considered together as one' with 'combined' |
| 125 | insert The at the start of the sentence |
| 127 | lower case i after the colon |
| 127-128 | are you having spaces between the number and the unit or not? Check for consistency throughout the paper |
| 131 | compare format of degree C with lines 91-92 |
| 133 | litre vs liter |
| 138 | insert s at end of content |
| 139 | put mesh size 0.5 mm in brackets |
| 142 | insert flatbed scanner after 12000XL |
| 143 | Format of reference |
| 157 | Should not start sentences with numbers |
| 163 | replace 'cumulative (RL)' with 'cumulative data (RL)' |
| 174-175 | Suggest delete 'Brassicas were in between.' |
| 198 | suggest transpose 'there was' |
| Table 3 | Need consistency with capital case in linear |
| 227-228 | Was this only with the absolute values? |
| 230-231 | Could this be due to root fineness? |
| Table 4 | Need consistency with capital case in linear |
| 270-274 | Delete repeated text |
| 275 | Check spacing in (0.71mm,0.63mm, 0.5mm) |
| 296-299 | Data presented seem to confirm this. The deeper rooting profile of the brassicas can be used to account for the poorer correlations and 'noisier' data |
| The could be more discussion/consideration of the rooting profile with respect to the modelling, e.g. it might be less appropriate for deeper rooted crops | |
| 329 | insert hyphen 'in-depth' |
| 340 | if the relationship is significant, then it does show something |
| 350-352 | perhaps it is just less appropriate for deeper rooted crops |
| 358 | replace 'their' with 'the' |
| 377 | replace 'is' with 'was' |
| References | Check for consistencies e.g. journal abbreviations, sentence case vs title case, |
| 426 | Check author initials |
| Supp. | Need titles on all table columns |
Author Response
Dear reviewer,
Thank you very much for your feedback. Below is a table with the answers to some of your comments. All the other comments not addressed in the table, were applied in the text as your suggestion.
Yours sincerely,
Tábata Aline Bublitz
| Table 1 - need scientific names of the mixed crops | Scientific names of the mixed crops were not informed because they were informed already when listing single crops in the same table. Since they are from the same species already mentioned, they were not written again, also due to limited space in the table. |
| 230-231 - Could it be due to root fineness? | All crops presented very fine roots not only in the topsoil, but also in the subsoil, and agreement in the subsoil was higher than in the topsoil for all crops, so I would not say that fine roots are the reason for little agreement between methods. When counting the roots on the profile wall, I noticed that in highly rooted areas (topsoil), one root can hide behind another or they can be very close together, difficulting their visual differentiation. While fine roots can get easily clumped, it does not happen exclusively to them. Most of roots found in the subsoil were also very fine, however often very far from each other, making the counting more accurate. In root washing from monoliths, the root clumping of the topsoil does not exist because root length density is measured by scanning roots floating in water, separated with tweezers so they do not clump together. This makes one method very accurate and the other not, thus the little agreement between both. |
|
Data presented seem to confirm this. The deeper rooting profile of the brassicas can be used to account for the poorer correlations and 'noisier' data. The could be more discussion/consideration of the rooting profile with respect to the modelling, e.g. it might be less appropriate for deeper rooted crops |
Added the following statement to the discussion at line 291: "The highly scattered data of brassicas is also a good indicator that LR might be, as well, less appropriate for deep rooting crops (Figure 1 A)." |
| 350-352 - perhaps it is just less appropriate for deeper rooted crops |
Inserted the following complement in line 349: "especially in deep rooted crops" |
Reviewer 2 Report
A good research topic, however, the expriment should be improved. For the field research, two years and one site were not sufficent. Especially for the
regression analysis, enough data with multiple environment is necessary. Furthermore, for the different crop speices, different layer, the calibration
value were different. Even for the different soil type. therefore，So many issue should be solved. therefor， the novelty and the major interesting finding were not enough for present research.
Author Response
Dear Reviewer,
Thank you very much for your feedback. I agree with your concern on the time and amount of sites used in this experiment, and I am aware of the limitations of our approach, however, considering the large effort required for determination of root-length density with different methods, it is not feasible now, to collect more data. Moreover, the large number of species and two experimental years with a complete dataset throughout the whole soil profile with both methods and three field repetitions is already a solid dataset. This dataset presented as well, a good degree of robustness, since it was solid enough to run these calibration attempts for a valid statement for our specific site and our species.
I will, definetly keep the your feedback in mind and might repeat the calibration with data from other crop environments when available and report on a new paper.
Best regards,
Tabata Bublitz
Reviewer 3 Report
The manuscript presents an attempt to find correlation between profile wall and monolith method to measure root length density. Authors employed regression models to compare methods but in general it seems that they didn’t found reasonable dependency. Overall, the manuscript is well written and scientific approach is ok. The results are not breathtaking but might increase pressure for further development of crops measurement methods and point solution for their compatibility.
One minor comment is that authors should keep graphs order consistent and in fig. 3 it should be same as in fig 2 and 1: Brassica, Grass, Legumes
Author Response
Dear Reviewer,
Thank you very much for your comments.
I have adjusted the order of the crops inside the graphs as suggested.
Best regards
Tabata Bublitz
Round 2
Reviewer 2 Report
As previous suggestion indicated, no much improvement was done.
Author Response
Dear Reviewer,
Thank you again for your feedback. Root studies is an area with lots of different methods, from which certain methods are mostly used individually. The experimental effort on determining the root length density of a large number of plots and repetitions in the field is avoided by most of the researchers working in this area. The monolith method, specifically, requires specific equipment for root washing and well-trained personnel for each step involved. Many modellers have recently seen the necessity to have big datasets with reliable root length data to use in future predictions, and other studies have highlighted the problem of having to use methods such as the monolith extraction in their experiments because there are no other methods with similar level of accuracy.
The need for a reliable dataset has motivated us to perform the monolith extraction in 30 plots over two years, where the extraction was performed down to 1-meter soil depth (around 12 monoliths per plot) in order to have a variety of root data that can in the future be used for modelling of root growth. It is important to mention that, the performance of a profile wall counting during the same period, although not in all the same plots, can be considered a novelty in this area and probably will motivate further researchers to invest time in trying to perfect a method that has a huge potential in facilitating experiments regarding root length estimation of cover crops.
I agree that this maybe was not clearly stated in the introduction of our paper, therefore I tried to add more arguments defending the execution of such experiment. Also, I tried to highlight the fact that this paper was written to fill in a gap in which regression models were not yet fully employed in the comparison of monolith and profile wall method, and the few conversion attempts were done using conversion factors only, which were used following a temporal plant development. A root length investigation in different plant stages is often not possible because of the intense labour involved, and using a mean conversion factor was shown in other studies to not be reliable. Thus, one of the possibilities, for root length estimations that do not consider plant development stages, is to try and apply another type of (more complex) conversion tool.
Since we observed a change of root length density with depth, we considered trying to adapt the tested models also to consider this factor in the comparisons. Basically, we attempted at proposing another way to compare both methods, but this time with a rarely studied (in root length) type of crop with a huge importance: cover crops.
It is certainly in our plans, to perform further testing of the regression models here developed, in unseen data (root length density measurements using monoliths are already being performed in another experiment, and might be used to increase the robustness of the functions presented in the present paper).
Please find resubmitted a version of the paper with the mentioned improvements in the introduction chapter.
Kind regards,
Tabata Bublitz
