Boreal Forest Multifunctionality Is Promoted by Low Soil Organic Matter Content and High Regional Bacterial Biodiversity in Northeastern Canada

Round 1

Reviewer 1 Report

This paper is a well written and researched; I’ve found it too long and should possibly be shorthened

I have just a few concerns about statistics (one could be amajor flow):

Sampling: it is not really clear if there was replication or pseudoreplication: please explain in detail the way of selecting subsamples; this may be a weak point for the whole statistical analysis

May the introduction of one or more random factors improve the GLM model?

In addition, it is not clear 8at least to me), what are the variables measured directly by the AA

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Giguère-Tremblay and colleagues present a timely study on soil microbial diversity and ecosystem function relationships and the abiotic drivers of these relationships in Boreal forests in Canada. I believe the research topic is important, the paper is well written, and the tables and figures are high quality.  I appreciate the authors approach in considering multiple ecosystem functions both above and below-ground including nutrient cycling, enzyme production, respiration, NPP.    

However, I have some significant concerns about some of the metrics used: 

First 1) Why was species richness used for fungal alpha diversity and the exponential of shannon used for bacterial alpha diversity? These indices can create quite different estimates of diversity based on species evenness within a sample. Thus when comparing differences in your study in the patterns of bacterial and fungal alpha diversity it is unclear if the difference is reflective of the microbial group or the difference between alpha diversity metrics. This choice needs to be clearly justified or better I would recommend re-doing the alpha diversity analyses with the same metric for bacteria and fungi. 

Second 2) I have never seen total DNA concentration in a sample used as an estimate for "secondary productivity". I see several problems with this proxy. First plants and autrotrophic bacterial DNA (and maybe more) will be present in your DNA extractions prior to PCR with targeted primers and these are not secondary consumers (heterotrophic). Also, and more importantly, total concentrations of DNA or sequence reads are not reflective of actual biomass. There are many reasons for this including primer biases, single vs multinucleate taxa (particularly for fungi), dead or dormant taxa that are present in DNA samples, and the very small amount of soil from which DNA is extracted. If there are relevant studies that have used this approach please cite them but I do not see this as a justifiable proxy.

In general, I would see the substrate induced respiration assay as your closest metric to secondary productivity because you have 1) selected for heterotrophs by using the cellulose treatment and 2) measured the active/living portion of the soil community. SIR measurements have been used as proxies for microbial biomass C extensively in the literature.  

Because the alpha diversity and secondary productivity metrics are included throughout the remainder of the analyses in this paper, including the PCA and SEM, it is difficult to assess the the latter analyses and conclusions drawn with them.  I would suggest to have these two issues addressed and the PCAs/SEMs re-run. 

Minor Comments 

The authors state that they used a Gaussian distribution for their GLMs of alpha/beta diversity on the PCA axes however the alpha diversity data do not appear normally distributed (Fig 2 and 3) especially fungal alpha diversity and bacterial beta diversity seem left and right skewed respectively. Did the authors test for normality? Is another family distribution better or a data transformation?

Line 111 Please clarify what role(s) you are referring to here: to determine the respective role of local and regional heterogeneity of fungal and bacterial biodiversity. 

Line 81-85 I don't follow your description of the link between alpha diversity and multi-functionality. How does this differ from a standard Biodiversity-ecosystem function relationship?  

Line 48 Which citation corresponds to this statement? please clarify in the text "In addition, only the role of fungal biodiversity was found to be significant in boreal systems"

Lines 77-78 please provide some examples of potential bacterial multidimensionality dimension(s) 

Line 82- define "locally"

Line 92- composition of microbial diversity?

Figure 1 it is difficult to see the different overlapping functions on PC1

Line 245 Please more thoroughly describe the Byrnes & al. (2014) multifunctionality index and how it is calculated as this is a major component of your analysis. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

about my previous comments and Author reply - the whole ms is far too long and should be shortened: it is up to the Authors deciding where and how - replication is still puzzling me: I am confused by the forest polygon sampling paragraph, which remains unclear, at least to me (mixing the subsamples collected in the field is pseudoreplication; alternatively, what is the level of replication for each site?) - AA= Authors

Author Response

Please see the attachment.

Author Response File: Author Response.docx