Changes within a South Carolina Coastal Wetland Forest in the Face of Rising Sea Level

Round 1

Reviewer 1 Report

See attached file for comments to the authors along with edits/suggestions.

Comments for author File: Comments.pdf

Author Response

I have read the referenced manuscript be considered for Forests. The authors have examined a tidal freshwater forested wetland along the South Carolina coast by determining change in productivity over time along a salinity gradient. There is also ancillary work examining understory plant species occupying the wetland. Given the susceptibility of these wetlands to sea level rise, studies such as this are needed and suitable for a journal such as Forests. I would support publication of this manuscript after the authors address the issues below. Thanks for the support and suggestions.

While not the focus- the understory plant survey is an interesting element, but the authors could possibly do more with their results. Given the focus on a salinity gradient, I was curious if there were patterns in understory plant occurrence that aligned with it. Perhaps the occurrence of more salinity tolerant species? Reporting this kind of result is contingent on how these plant surveys were conducted which is not clear and more information is needed. At a minimum, methodology about where and how plant surveys (same as collection trips?) were conducted is needed. See the added material in the Methods section. The objective of Dr. Stalter’s plant surveys over the past 50 years has been to collect as many species on each collection trip as possible. We added a couple of sentences describing how he collected plants for this project . It is mentioned that site preferences and abundances were considered but I do not see these data in the paper or the supplementary table. This sentence was deleted. Also, these were conducted between 2015 and 2019. Yes Were there any temporal trends detected? No, too short of a time period. This would all fit with the theme of forest changes over time. Finally, there is no mention about plant results in the Discussion section. Given this is likely a very dynamic part of the forest community, I would strongly suggest the authors expand their handling of this part of the study. Added to Discussion 3 sentences about hearbaceous community.

Another clarification needed in the patterns related to productivity over time. These results are interesting, but it wasn’t clear to me how much of the declining ANPP over time (where >1ppt) is due to tree mortality versus inhibited growth? Given the gradual decrease in site salinity over the course of the study period and data in Fig. 3 I suspect the former, but this should be clarified by the authors in the results and discussion. This is true and we added a short comment in the Results saying woody growth is of surviving trees.

I have provided some more specific edits/comments below:

L2 The term ‘forest’ is used twice in the title- perhaps delete the first use. First forest deleted as suggested

L56 Change ‘evidence of’ to ‘baseline data to assess’. Done

L76 Verify this is an added freshwater site (not a new site). Wording was changed to ensure reader knew that the added site was in same area.

L108 On the aerial photo in Figure 1, it appears to show a road extending over where marine exchange occurs. Is this the case? I am curious if tidal exchange is restricted by a culvert and if it might contribute to an impoundment and some of the salinity dynamics? Also, perhaps ‘tidal exchange’ is a better descriptor instead of ‘marine exchange’. You are correct about this and we added a sentence in the Study Area description to note this.  Also changed wording to tidal exchange in Figure legend, and elsewhere.

L121 Provide depth and details about the wells. We added the following to 2.2 Water Depth and Saminity section:  The wells were slotted, 5-cm PVC pipes screened and inserted to a 1-m depth and capped. Wells were first pumped of residual water and allowed to backfill with pore water.

L126 Again, as indicated above, I think providing Year 1 data about mean forest composition at each salinity zone would be appropriate and helpful for interpretations.  A new table was added showing tree composition at beginning and end of study.  A paragraph was also added in results.

L175 Later in the discussion, the importance of using the 1 ppt threshold is mentioned. However, without any previous background, this salinity level seems arbitrary. Either mention why it was used herein the methods or explain its importance in the Introduction. The importance of this threshold limit was moved to the Statistical Methods section.

 

Reviewer 2 Report

No comments

Author Response

Reviewer had no comments

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

Conner et al. present an interesting study of the impact of sea-level-rise on forest productivity in a South Carolina coastal wetland forest. Such coastal wetland ecosystems are highly productive but also highly sensitive to climate change via SLR. Thus, investigations such as this are highly relevant as they can help illuminate coastal ecosystem responses to climate change. The authors report that productivity declined with increasing salinity, and documented 165 plant species at the study site. The study is however somewhat limited as it is lacking in robust statistical analysis to determine differences in forest growth along a salinity gradient. Overall, the study is rather local in its focus and does not sufficiently place the findings in a broader regional/national/global context.

Detailed comments

Abstract

Line 11. not clear what a "wetland drainage" is. Would "drained wetland" be more appropriate?

Line 18 and elsewhere. The key focus of this study is examining how SLR, coupled with increasing soil salinity, is impacting forest productivity. Yet, at their site, salinity is declining. The reason for this is insufficiently explained.

 

Introduction

Line 32. Can this be supported by a reference? Undoubtably, TFFW are locally important in terms of carbon sequestration, but are they globally important? Reference to estimates of the global area (and possibly a mention of regions with lots of TFWW) would be useful.

Line 35. Slightly unclear, consider rewording

Line 37. New point, consider beginning new paragraph here.

Line 49/50. Missing link sentence. Need to link TFFW, SLR and floristic composition i.e. SLR leads to changes in community composition, which can affect ecosystem function….

Line 58. "the present study area" has not been mentioned yet so the reader doesn't know where it is

Line 66. Add “in south Carolina” after ecosystem.

 

Methods

Line 76. “Because…..” – consider rewording this sentence to indicate that the fourth site served as a baseline, non-saline site.

Line 96. Is this the average low temp or the absolute low temp?

Line 101. Fig. 1. The placement of plots needs further explanation. The fresh, low, and high plots seem paired, being more or less adjacent to each other. The mid site is more geographically separated. This is not necessarily a problem, but more explanation of the salinity regime across the site is required.

Line 104. The location of the sensors should be indicated on Fig. 1

Line 114. Please provide the coordinates of this weather station.

Line 136. This section is lacking clarity. If the goal is to assess changes in forest plant species composition over time, why not just do vegetation surveys at each plots for each year, or at intervals?

 

Results

Line 175 and elsewhere. While the results presented are compelling, overall the study is lacking a quantitative analysis of differences in productivity along a salinity gradient.

Line 186. This approach seems too local. To contribute the global discourse on changes in community structure with SLR, re-surveys, or quantitative analysis of differences in species composition across a salinity gradient would seem much more appropriate that once off collection of species richness data from the entire site.

 

Discussion

Line 237. As mentioned previously, this data is difficult to interpret given the lack of robust statistical analysis.

Line 268. This is the first mention of tree regeneration. Was it measured in the study? If so, it would be very useful to include that information.

Line 270. The reasons why salinity is increasing at the site needs further discussion.

 

Conclusions

The topic discussed and the data has the potential to make an excellent contribution to the international literature. However, statistical analysis is lacking and the approach to the vegetation survey is too local and not focused on specific research question with broader implications.

Author Response

Please see attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Discussion section is disjointed. Needs better structure aand flow to show progression of related ideas. Suggest organizing as first a summary of findings at this site, bringing in the information about other sites, and then following with how this relates to climate change. As is, it's kind of mixed through. This is true for the introduction as well. A sentence or two framing the overarching message might help, with the details and supporting information following and arranged by topic.    Suggest adding a graph showing salinity vs NPP for all sites and relationship, how well do these points show a trend. Or salinity vs woody growth or leaf litter fall.    The sampling of all species- is there information on whether this is changing, were there species there in the past that aren't there now? How does this list compare to other sites? This could be fit into the overall narrative better, it seems unrelated now.    Introduction talks about sea level intrusion and salinity going up, but the data shows salinity decreasing. Can you show this is due to increased precipitation?    Fix grammar, what was impacted: line 60 "These abandoned rice fields and coastal forested wetlands on the edges of Hobcaw Barony were severely damaged by Hurricane Hugo in September 1989 impacted the forested wetlands with a powerful storm surge."   Contradictory values around line 95: "Air and water temperatures drop below 0 °C for only a few days in December and January with the lowest temperature being 2.5 °C."   Overly complicated sentences such as line 216: "Though Strawberry Swamp does not experience the effects of daily tidal fluctuations that TFFW often do, it is subject to the same press stress brought on by marine-derived salinity, and ameliorated by local upland drainage (from rain events) that tends to flush the porewaters of salts to result in lowered porewater salinities across the relatively small gradient in Strawberry Swamp [53]."  

Author Response

Please see attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have addressed all the comments and the manuscript is much improved as a results. However, three issues still remain –

Experimental design: it has not been sufficiently explained why fresh, low and high salinity plot replicates are adjacent to each other whereas mid salinity plots are quite spatially separated. This is problematic as adjacent sites may be confounded by other factors.

The analysis takes a regression-based approach rather than an ANOVA based approach. Again, this questions why plots were not arranged along a continuous salinity gradient, or, alternatively, more replicate plots are each salinity level were not measured to facilitate an ANOVA based analysis (two replicate plots at each zone is insufficient). Moreover, Fig 3 is lacking detail (see below comment)

The vegetation study: In the initial review, I question the suitability of the vegetation data in its current form for an international journal. The authors have quite rightly pointed out that “the flora of the forest is much broader and potentially more valuable to future researchers. Inclusive floristic inventories such as the recently published paper on the vegetation in the 3 nearby abandoned rice fields that we reference on line 60 has more valuable data than a few quadrats.”. This is true, but as highlighted by the reference mentioned, this kind of data only becomes valuable to a non-local audience when re-survey data is recorded. When resurveys are collected in 10 yrs+, this vegetation data will be of interest to an international audience.  

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

1. Experimental design: it has not been sufficiently explained why fresh, low and high salinity plot replicates are adjacent to each other whereas mid salinity plots are quite spatially separated. This is problematic as adjacent sites may be confounded by other factors.

The paired plot design for productivity studies has been used by the authors for the past 40 years. When this project was initially started in 2013, we established the plots along the existing salinity gradient in the Strawberry watershed and published a paper in Forest Ecology and Management in 2017 (see Liu et al. in references). In the Mid-saline area, the plots are not side by side as the others are but are still in the same salinity regime. This was done because of site variations (large openings) in the Mid-saline area and the desire to have plots of the same forest community structure as possible as was the case for the other plots. Since the design has already been accepted and published in our previous paper, we referred readers back to that paper to show we were utilizing their same design.

Unless I am missing a recent update to the manuscript, no reference to a Liu et al Forest Ecology and Management paper is provided. A mesocosm study by Liu et al, Soil Biol Biogeochem is provided, but from what I can see that does not describe the same experimental setup. While I accept that this design has been published in Forest Ecology and Management, my reservations remain about the overall approach (see below). At a minimum, the detail provided above regarding open areas and community structure needs to be added to the manuscript.

2. The analysis takes a regression-based approach rather than an ANOVA based approach. Again, this questions why plots were not arranged along a continuous salinity gradient, or, alternatively, more replicate plots are each salinity level were not measured to facilitate an ANOVA based analysis (two replicate plots at each zone is insufficient). Moreover, Fig 3 is lacking detail.

The regressions illustrate the plots span a range of salinities across the site, and thus salinity is continuous. Salinity can vary from year to year even with sea level rise from storm surge, rainfall events, or drought. We removed the designation of plots as low-, mid-, and high-salt in the paper to remove confusion on the idea of a constant salinity at each spot, and replaced Figure 4 with a regression through time for plots with salinity above 1 ppt.

From what I can see, the designation of plots remains on Fig. 1 and Fig 2.  The issue remains that the design neither allows for ANOVA type approach (3+ reps at each salinity level) or an regression type analysis (ideally 4+ levels of salinity). Clearly the salinity fluctuates at each site, but the way the design is set up it generates confusion (sites grouped by salinity in Fig 1 and 2 but not Fig 3 and 4. Moreover, multiannual data from Fig 3 and 4 is not independent as it comes from repeated measures of the same plots. Usually this is accounted for by including a random effect for site or a repeated measures anova type analysis, but this has not been done in this case.  

3. The vegetation study: In the initial review, I question the suitability of the vegetation data in its current form for an international journal. The authors have quite rightly pointed out that “the flora of the forest is much broader and potentially more valuable to future researchers. Inclusive floristic inventories such as the recently published paper on the vegetation in the 3 nearby abandoned rice fields that we reference on line 60 has more valuable data than a few quadrats.” This is true, but as highlighted by the reference mentioned, this kind of data only becomes valuable to a non-local audience when re-survey data is recorded. When resurveys are collected in 10 yrs+, this vegetation data will be of interest to an international audience. As the reviewer said “this kind of data only becomes valuable to a non-local audience when re-survey data is recorded.” We agree and thus the need of having a baseline inventory in place so future researchers can refer to when re-surveys are made. The baseline survey is needed in order to know what taxa are extirpated as sea level rises and water salinity increases in the area in the future. The flora of the forest is much broader and potentially more valuable to a future reviewer as quadrats would provide only a sampling of the vascular plant species in the area. Inclusive floristic inventories such as the recently published paper we quoted are more valuable than data from a few quadrats that would probably only include the few dominant taxa.

I agree that having a baseline inventory in place is important so future researchers can refer to when re-surveys are made. The key point is that, currently, such a list of flora would be more appropriate in a regional journal – future studies can refer to this list, re-survey, and report on changes in a broad focus international journal such as this. The drivers of change is what is of interest to the international audience, rather than a static list of species from a single site.