Influence of Environmental Factors on Forest Understorey Species in Northern Mexico

Round 1

Reviewer 1 Report

I judged that not only did the authors respond sincerely to my suggestions and questions, but also made sufficient revisions in the revised version. Therefore, I think that the paper structure and content should remain mostly the same.

Actually, there are still some inadequacies in the analysis and other aspects of the paper, but this may be the limit with this observation data, and I don't think that the results obtained from further statistical analysis will necessarily achieve the goals that the authors aim to achieve. In the sense of properly describing the authors' carefully observed data, I think it is better to leave it as it is.
However, I think the only thing that needs to be corrected is that the figures are still not clear. As for the figures, the text is sometimes squashed (a problem with the vertical and horizontal scales?). In addition, the font is too small in some cases, making it difficult to see, so please make sure to fix this. For tables, the font of the text should be consistent. For example, only Table 3 seems to use a different font.
This kind of font unification is the minimum correction that should be made before submission, and in some journals, these discrepancies alone are not enough to be considered for review. It is better to keep this in mind so that you will not be pointed out like this in the future.

Author Response

We deeply appreciate the careful review you have made on the whole manuscript and the suggestions that certainly helped to improve it

Author Response File: Author Response.docx

Reviewer 2 Report

General comments

 

  I have finished reviewing the paper entitled “Influence of environmental factors on forest understory species in northern Mexico”. The paper provides valuable information for the better understanding the forest structure of Pinus-Quercus forests in Mexico. One of the concerns is that the authors do not consider human impacts. Check the minor comments below. In addition, GLMM (generalized linear model analysis) may be recommended in data analyses, since the field survey was conducted in five different sites.

 

Minor comments

 

Lines 84-85

  What types of forests are included in uneven-aged Pinus-Quercus forests? Why are those forests uneven-aged? Due to natural disturbances and/or human impacts (e.g., logging)?

 

Study area

  Pinus-Quercus forests are natural forests? Human impacts are not considered (Lines 349-350, 356-357), but the anthropogenic impacts would affect vegetation structure and soil properties. Actually, the authors mention the impacts of logging on species distribution (Lines 357-359). The explanation of the human interference to Pinus-Quercus forests should be added.

  Please consider to add the information of the height of vegetation (large trees). Due to the lack of such information, I cannot imagine the stand structure of Pinus-Quercus forests.

 

Lines 226-228

  According to Table 2, the position (gap or canopy) seems to affect the presence/absence of Penstemon miniatus var. townsendianus and Agastache pallida. The authors also mention “five show a preference for gaps and two were more frequent under the canopy” in the Discussion section (Lines 418-420)”.

 

Line 249 (Sat)

  Sand?

 

Lines 251-253

  I do not understand what you intend here. Why did you compare AIC between different species?

 

Author Response

We appreciate your review of the manuscript, as well as your recommendations to correct and improve it. All of them have been considered in the new version

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

In this paper, the authors examine the environmental factors that influence forest floor vegetation in five typical temperate forests in Mexico by conducting intensive survey of forest floor vegetation (only presence/absence though) with a variety of environmental factors. The author's survey area is large and has a sufficient number of replications, so it is thought that there is enough data to examine the relationship between these factors. For these reasons, I believe that this paper is a suitable submission for Forests.
However, there are some flaws with the paper, such as the lack of description of the survey method, especially the survey and analysis method of environmental factors that are important in this study, and the limited time period of August-September. In addition, the structure of the introduction is logically uncomfortable and should be substantially rewritten. As for the results, only some of the statistical analysis results are available, but I suppose that authors should be carefully described first.
In addition to these major revisions, there are a few minor revisions. The latter are written directly in the pdf, and I strongly recommend that you refer to it for correction.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

General comments

The authors of the manuscript Environmental influence on forest understory species in northern Mexico explored the effects of soil characteristics on species-specific presence in five pine-oak forests of the Sierra Madre Occidental.

I recognise a big effort in surveying and censuing understory vegetation in 2000 units within the five locations; however, I find the manuscript confusing, with important information that is missing. In fact, I am struggling to understand the experimental design, the statistical approach and the robustness of the outcomes. I may have missed something, but if I understood correctly, the authors performed statistical analyses to infer the effect of 27 soil variables taken from a single point within 20 x 20 m subplots to the presence of each species -coded as 0-1 (absence/presence)- in 100 units per subplot, considering only the 22 most abundant species. Thus, the final sample size is 2000 observations for each species grouped by 100, which share the same soil characteristics. If this is true, I think the approach is not valid because such precise value (presence of a particular species in a specific unit) cannot be related to a soil characteristic that it is assigned to another 99 units, without considering that these 100 units belong to the same subplot. It would make more sense to join presence records per subplot and then have count data following e.g. a Poisson distribution. Another potential approach would be to consider the hierarchical structure that takes into account the autocorrelation among units within subplots and sites (units nested in subplot and subplot nested in site) with random factors.  Even doing so, I doubt that the analyses could be run for some of the 22 most abundant species with 27 exploratory variables that in turn have not been tested for collinearity. Although authors mentioned something about correlation analyses, no results are shown.

Another concern regarding the design is that forests are considered equivalent, however, nothing is said about their similarities or differences. How can we be sure that site is not a confounding factor? We have no information about them. I also have doubts regarding the use of canopy/gap as a factor, as authors said that only were considered those units under canopy or totally exposed, and ignored those units partially under canopy. Ignoring these units may imply an important biased towards shade-tolerant or light demanding species, neglecting a lot of information.

Finally, there are not previous hypotheses regarding which factors might be more important for given species, or a group of them, or in each site. Accordingly, I find that the objective of the study is quite descriptive, and accordingly, Results and Discussion as a mere list of examples without a clear message. Thus, I hope the authors will be able to address pertinent changes to clarify all these aspects together with some specific comments below.

 

Specific comments

L 43. What did the authors mean with ‘geographical factors’?

L 67. Which ‘variables’ did the authors mean here?

L69-80. These two paragraphs are vague and general without focusing the topic. The introduction would gain if the authors would focus on the understorey and its role in the ecosystem functioning, and pose hypotheses about the specific soil characteristics influencing species or group of species there. It seems obvious that plants respond to abiotic factors, including soil characteristics. Thus, I do not see any novelty in just checking which one affect more or less, unless the authors set specific questions and hypotheses.

L 92. Did all sites share the same vegetation? In addition, what about other factors (exposition, slope, tree density, dominant tree species, etc.)? How far they are?

L 109. What do ‘ejidos’ and’colonia’ mean?

L 122. During sampling time, was there a schedule to identify appropriately all species? I mean, during August and September, could all species be easily identified?

L123.’Registration of shrub and herbaceous species was done by direct observation and through digital photographs’. In my opinion, this should be said in the previous section (sampling sites), when it was said, that occurrence of species was recorded, as first they need to be identified.

L 129-134. It would be nice to see the data showing that there were no difference. Actually, it was not explained the protocol for soil sampling. How many of the 20 soil samples were collected in gaps and how many under canopy? Was it a systematically sampling? Was the sampling point in the centre of the plot? In a corner? Moreover, these soil characteristic ranges do not mean anything without indicating the sites where they were recorded, or whether they were obtained under canopy or in gaps.  

Fig1. It took me a while to understand it completely. I would use one symbol to all sampling points, and add the name close to each location.

Table 2 and Table S1. I have no idea what this chi-squared and p-value mean. In fact, figures and tables should be self-explanatory, and in most of them in this study lack important information. In Table 2, does it make sense to average number of occurrence in gaps and under canopy per species?

L 138-142. Where are the results from the correlation analysis? Nothing is said about which variables were selected, or even whether there was a selection.

In general, I find this section messy, unorganised and confusing. For instance, I do not understand what the authors do to test the effect of canopy/gap, or how the model selection method was achieved.

L 149. What does this ‘5-fold’ mean?

L 155-159. A brief explanation should be given about this analysis for those not familiar with it.

L 194. What does CEC mean? And why was this characteristic chosen among the rest to show some results?

L 209-215. It is a repeated paragraph.

Table 3. Where are the other species from the 22 selected? Moreover, the sample size should be indicated for each species. I think that some results cannot be trusted as the sample size was really small (e.g. n=3 for Chimaphila maculata).

Fig. 3. These graphs are little informative. Why were those species selected?

Table 4. Again, why did the authors show the results from these species and not others? This table seems not relevant. I would place it in Supplementary Information.

L 259. No other factors have been tested, a part from gap vs canopy. How can the authors say that they are the main regulating factors? Compared to what else?

Could you explain the difference between the sentence in 261-263 and the consecutive one (L263-264). They sound quite repetitive.

L 268-269. This result encompasses tropical forest species, which sounds not to comparable to the results here.

L 317. I did not know the authors used also climatic variables in the analyses. This is the first mention.

L 342. There is no mention about density ranges in this area. Thus, readers not familiar with this type of communities cannot have an idea about how open they are. Moreover, I do not completely agree with the discussion here. The authors justified the lack of canopy effect because light is not a limiting factor. However, density may also affect soil water availability as higher density determine micro-climate conditions that usually diminishes direct radiation and evapotranspiration. A more elaborated discussion could be addressed regarding this point that would enrich the discussion.

Author Response

Please see the attachment

Author Response File: Author Response.docx