Review Reports

Round 1

Reviewer 1 Report

The manuscript presents interesting soil and vegetation data across a narrow gradient in N deposition in the Appalachians.

This is information that will be a good addition to the literature. 

I have made quite a few comments within the attached pdf document.  I hope  you find them helpful.

Overall an interesting study with information relevant to understanding the recovery of high elevation forests from acidic deposition.  The gradient in elevation (160 m) and N deposition (73 kg/ha) are used in the analysis, precipitation is also found to be significant and should be included in Table 1.  Overall this represents a fairly narrow gradient.  I must also note that all three of these ‘significant’ variables are often confounded.

Review comments:

Table 1.  These data seem to be part of the site description. If so, include citation.  Include method for pH measurement, H₂O or CaCl₂.

I like Figure 4.  However, it is very difficult to read the elements listed on the x and y axes and discern the relationship between the soil and plant concentrations.  Perhaps it would be better to only include elements that are presented and interpreted in the manuscript.  One of the dangers of elemental analyses is that presentation of all data rapidly moves beyond the original hypothesis proposed for the study.

Line 138

Soil collection method is not clear.  0-15 cm beginning where?  At the Oa horizon?  Mineral horizon, called a B, is really a mix of all mineral layers below the O.  This has an impact on soil chemistry.  Please clarify by describing the horizons sampled in this ‘horizon’.  It appears that the depth of mineral soil is quite variable and therefore the sample includes various proportions of different soil horizons.  Because this is not really a taxonomic horizon it is actually a soil layer defined by depth.  You should probably change the designation.

Some of the terminology needs clarification.  For example, ‘cation depletion’ suggests that cation movement out of the soil horizon is being measured when actually certain chemical constituents decline with increasing deposition along the gradient.

Ln 304.  ‘symptoms of cation depletion’.  Perhaps a better term is deficiency.  Depletion of cations from tree foliage would suggest the measurement of throughfall chemistry.

Be sure to keep information in the proper section.  There is no need to repeat methods in the results section or results in the discussion section. 

The units used in Table 4, deposition slope and elevation slope are not defined in the methods statistics section.

Please see the attached document for more specific comments.

Comments for author File: Comments.pdf

Author Response

Please see the attachment. 

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript assessed the potential impact of excess N deposition on soil nutrient availability in high elevation spruce forests of West Virginia/Virginia.  It did this both by measuring soil and foliar nutrients of spruce and two hardwood species along a gradient of N deposition.  Findings support earlier work in numerous site in eastern North America that excess N can decrease availability of soil nutrient cations.  Of note is that there was interspecific variability in nutrient relationships.

Although I do not find a great deal of novelty in this study, I think that it was generally well conceived and carried out.  There were several studies on effects of atmospheric deposition on spruce forests, but far fewer recently.  As the authors point of via referencing, despite current declines in N deposition, effects of excess N have a bit of a legacy in this part of the country.  In addition, use of hardwood species to the analysis adds an interesting dimension.  Accordingly, I would like to see this eventually published.

There are a few areas of suggested change.

1.  The title is a bit misleading because it does not mention what is truly novel about the study--inclusion of prominent hardwood species.

2.  The Introduction can be improved with additional relevant literature, some older, some more recent.  The Eagar and Adams 1992 book remains an important reference.  Although the McGowan Mt. site is not in the Fernow Experimental Forest, it certainly is close by.  Work there by Peterjohn and Gilliam would inform some of the background, especially with respect to interspecific controls on N cycling (Peterjohn et al. 2015--FEM) and N-mediated loss of soil cations (Gilliam et al. 2016--Atmo Environ, 2018--FEM).

3.  I would like to see the authors address the potential confounding factor of varying importance of spruce in these stands, all the way from ~20% to 75%.  That is, can the variation of how much spruce comprises canopy dominance influence a stand's response to N?

4.  I find it confusing why the authors use total N dep, rather than annual means, as an independent factor in their data analysis.  I would rather each site be characterized by annual means.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper presents a study on soil and foliar nutrient concentrations of three tree species in a gradient of N deposition. The topic is of high actuality. However, the study has a number of weak points which affect the strength of the conclusions. First, the N deposition gradient consists of four sites with the highest N deposition being only 23% higher than the lowest deposition. For gradient studies this is both a very small gradient and a very low number of sites. For each predictor there should be at least 10 plot replicates. Then the observed gradient in soil chemistry more likely caused by spatial heterogeneity than N deposition as N deposition is acidifying the soil and the highest pH was observed at the site with highest N deposition. It is stated in the paper that limestone influenced two of the four sites which makes an important geological gradient. It is thus highly questionable that the conclusions drawn are related to N deposition. I will therefore not go into details of the conclusion.

The data were analyzed using a normal ANOVA. As the tree and soil sampling was repeated within the plots, the data were clustered and needed a mixed regression. Not including the clustering estimates the number of replicates too high and may result in false positive regression results.

The soil samples were dried at 105°C which is an unusually high temperature. It is, however, not known how much this affected the results of the chemical analyses. Otherwise the results of the study could be presented as an ecological study describing tree nutrition in four different plots.

Detailed comments

Line 166: How much ground tissue was digested?

Line 176: Not the predictors for the regression must meet the assumptions of normality but the residuals. A residual analysis is essential after a statistical analysis.

Author Response

Please see the attachment.

Author Response File: Author Response.docx