Recovery of Carbon and Vegetation Diversity 23 Years after Fire in a Tropical Dryland Forest of Indonesia

Round 1

Reviewer 1 Report

line 2-3, it should be written as a title case

line 12-13, FORDA has been changed/merged to BRIN

line 48, please rewrite for clear subject ('this')

line 72, unclear transects figures/note

line 81, Figure S2 has no axes titles?

line 126,unclear sentence

line 219, inconsistency in using the terms of: 'burned', 'unburned', 'burnt', 'unburnt', 'primary forest', 'post-fire forest', 'unburned forest'.  Please use similar terms for the whole article.

line 227, this study is not experimental research, so the term of 'treatments' seems to be incorrect.

line 267, please put 'Mean' carbon fraction......on Table 3 title

line 282, please rewrite and spell check!

line 323, please put notes of stars symbol for figures in Table 5

line 328-329, please see comments on line 219 

line 353-354, please rewrite

line 358-359, it is on the contrary of figures in Table 2

line 382, need to add the influence of fire adaptive traits of each species to discuss why certain species survive and adaptive to the burned condition

line 419-420, please see comments on line 219

line 421-422, please rewrite and spell check

Figure S1: Please add axes titles

Figure S1: Please spell check on the chart

Figure S3: please add axes titles 

Comments for author File: Comments.pdf

Author Response

First of all, we deeply appreciate your helpful comments. Our replies to the reviewer’s inquiries and revised points are as follows.

Comment 1:

line 2-3, it should be written as a title case

Response: Thank you for your correction. We have revised it

Comment 2:

line 12-13, FORDA has been changed/merged to BRIN

Response: Thank you for your correction. We have revised it.

Comment 3:

line 48, please rewrite for clear subject ('this')

Response: Thank you for your valuable comment. We have revised the sentence.

Comment 4:

line 72, unclear transects figures/note

Response: Thank you for your valuable comment. We have sharpened note of figures.

Comment 5:

line 81, Figure S2 has no axes titles?

Response: Thank you for the nice reminder. We have added axes titles on figure

Comment 6:

line 126,unclear sentence

Response: Thank you very much for the comment. We have revised the sentence.

Comment 7:

line 219, inconsistency in using the terms of: 'burned', 'unburned','burnt', 'unburnt', 'primary forest', 'post-fire forest', 'unburnedforest'. Please use similar terms for the whole article.

Response: Thank you for the reminder. We have revised it

Comment 8:

line 227, this study is not experimental research, so the term of'treatments' seems to be incorrect.

Response: Thank you for the reminder. We have revised it

Comment 9:

line 267, please put 'Mean' carbon fraction......on Table 3 title

Response: Thank you for suggestion. We have revised it

Comment 10:

line 282, please rewrite and spell check!

Response: Thank you very much for the comment. We have rewritten.

Comment 11:

line 323, please put notes of stars symbol for figures in Table 5

Response: Thank you for suggestion. We have added it

Comment 12:

line 328-329, please see comments on line 219

Response: Thank you for the reminder. We have decided use “burned” and “unburned” for consistency

Comment 13:

line 353-354, please rewrite

Response: Thank you for suggestion. We have rewritten

Comment 14:

line 358-359, it is on the contrary of figures in Table 2

Response: Thank you for your valuable comment. In table 2 we show the value of carbon fraction/carbon content from the results of sample analysis, while in this line (358-359), we explain the value of soil carbon stock/storage. Soil carbon storage value is influenced by soil depth, bulk density and carbon content. Although the soil's carbon-organic content declined linearly with depth, the deeper soil layers' total carbon storage capacity was larger than the surface soil layers due to increase in soil bulk density.

Comment 15:

line 382, need to add the influence of fire adaptive traits of eachspecies to discuss why certain species survive and adaptive to the burned condition

Response: Thank you for suggestion. We have added sentences (line 394-401) to concern how the species survive and adaptive to the burned condition

Comment 16:

line 419-420, please see comments on line 219

Response: Thank you for the reminder. We have revised it

Comment 17:

line 421-422, please rewrite and spell check

Response: Thank you for suggestion. We have rewritten the sentence

 Comment 18:

Figure S1: Please add axes titles

Figure S1: Please spell check on the chart

Response: As the reviewer suggested, We have revised it

Comment 19:

Figure S3: please add axes titles

Response: Thank you for suggestion. We have added axes titles

Author Response File: Author Response.pdf

Reviewer 2 Report

I was very pleased to read the article "Recovery of Carbon and Vegetation Diversity 23 years after fire in a tropical Dryland Forest of Indonesia". It is a very well written article. It is a very interesting read, everything is presented in a clear and readable way.

The article addresses an important topic and fits in with current research trends.

The article is properly structured. It contains all elements required in a good scientific article. The conducted very extensive research is extensively presented, the research methodology is well chosen and described, it does not raise any doubts. The results are presented in a series of figures and tables in an easy-to-read manner, and I have no editorial concerns.

I have one comment that I think will improve the already high quality of the article. What I miss in the research presented is an indication of its wider purpose. The authors write in lines 63-64: "This study aimed to analyze the recovery of biodiversity and carbon stock in secondary tropical forests 23 years after fire in a tropical dryland forest of Indonesia." Please expand, especially in the Introduction chapter, on what this analysis serves in a broader context, especially from the perspective of Environmental Sustainability. The authors have only very briefly addressed this, e.g. in lines 51-53. Please expand on this. It will also be a good justification of the research undertaken. In the current version it looks as if the Authors did the analysis for the purpose of analysis itself.

Author Response

First of all, we deeply appreciate your helpful comments. Our replies to the reviewer’s inquiries and revised points are as follows.

Comment :

I have one comment that I think will improve the already highquality of the article. What I miss in the research presented is anindication of its wider purpose. The authors write in lines 63-64:"This study aimed to analyze the recovery of biodiversity andcarbon stock in secondary tropical forests 23 years after fire in atropical dryland forest of Indonesia." Please expand, especially inthe Introduction chapter, on what this analysis serves in a broadercontext, especially from the perspective of EnvironmentalSustainability. The authors have only very briefly addressed this,e.g. in lines 51-53. Please expand on this. It will also be a goodjustification of the research undertaken. In the current version itlooks as if the Authors did the analysis for the purpose of analysisitself.

Response: Thank you for suggestion. We have added sentences (line 57-63) to concern the broadercontext from the perspective of Environmental Sustainability. Forest disturbance regimes have lately intensified in many parts of the world, and future climatic changes are likely to intensify this trend in the next decades. These changes are making it more difficult to achieve the basic goals of forest ecosystem management, which are to deliver ecological services to society while preserving forest biological diver-sity. Understanding the process of biodiversity recovery in disturbed regions will be useful for identifying management strategies to address ecosystem change, such as when, what species, and how to intervene to restore forest ecosystem function.

Author Response File: Author Response.pdf

Reviewer 3 Report

This is an interesting and potentially important paper, but it has a major defect: it is based on just three transects, all within 200m of each other (Figure 1).  Each transect has a segment that was burnt 23 years ago, and a segment that was not burnt, so arguably there are six study sites (3 burnt, 3 unburnt).  But in fact it was the same fire that burnt the three burnt segments, and the same geography that protected the three unburnt segments, so it could be argued that the study lacks credible replication.

Furthermore, we are told that the unburnt sites were all managed as part of a “nature leisure park”, and I wonder if that contributed in some way to the outcome that they did not burn.  We should be told much more about that.  For example, was there a deliberate program of fuel reduction within the leisure park?  Were there active efforts to control the fire as it approached the leisure park?  Was the site of the leisure park selected because of some intrinsic features (e.g. geology, forest type or forest maturity) that may have reflected propensity to burn, carbon profile, biodiversity and manner of subsequent regeneration?  These are important points because they may have affected the manner of regeneration as well as the intrinsic base values.  For example, is it possible that the dominance of Shorea laevis in the regeneration reflects an abundance of that species before it was even burnt?

Some of the differences observed in the current study could be due to differences in intrinsic base values as mentioned above.  A broader study (with more transects distributed much more widely) would be needed to dispel that concern.  Perhaps it’s too late to do the broader study, but if this is to be published it is essential to tell us what is known about any pre-fire differences between the burnt and unburnt segments of the three transects.  Are there photographs, satellite images or botanical assessments of the areas from before the fire?  If so, we should be told about them so we can assess whether the burnt and unburnt segments are truly comparable.

The same issue has implications also for the statistical analysis.  I fear that this has been done with the assumption that each of the 48 plots is independent (n=48), though I’d be happy to be corrected.  Section 2.4 is completely inadequate as it doesn’t tell us how it was done.  There are only six study segments and n=6 (arguably only two).  I suspect that the analysis needs to be redone, and it certainly needs to be explained more carefully.

I have a number of minor comments as follows:

Line 220, hotter and drier.

Fig.5, Table 3, etc. Do these graphs and tables refer to burnt or unburnt?  As fire is the main variable of interest, all graphs and tables should distinguish burnt from unburnt.

Lines 314-317, don’t just tell us that these indices differed significantly between burnt and unburnt: it’s more important to tell us the direction of the difference, i.e. were the burnt sites more or less diverse than the unburnt sites?

Table 5, don’t just show us F and P values: what were the predicted means?

Lines 326-329, this sort of information should be in results not discussion.

Lines 342-343, can’t say this is “usually” the case when it’s based on just one reference (Slik et al. 2008), referring to an unstated number of sites.  If Slik et al’s work was based on multiple sites, that should be stated.

Lines 408-412, the recovery of “evenness” may be worth reporting in the results, but I doubt that it deserves mention in the discussion, as evenness is an artificial construct, not well understood by the average reader.

I’m sorry to be so negative, because I can see that a lot of hard work has gone into this study and I think the results may be worth reporting.  Perhaps the best way forward is to recast it as a case study, and be very circumspect about what it says about the recovery of vegetation and carbon stocks in general after fire: make it very clear that it refers to just one small area with unknown comparability between parts that were burnt or unburnt.  Make it very clear that this cannot be considered a definitive study of the effects of fire on recovery of vegetation and carbon stocks, for a broader study with more replication is needed.   

Comments for author File: Comments.pdf

Author Response

First of all, we deeply appreciate your helpful comments. Our replies to the reviewer’s inquiries and revised points are as follows.

General Comment :

This is an interesting and potentially important paper, but it has a major defect: it is based on just three transects, all within 200m of each other (Figure 1). Each transect has a segment that was burnt 23 years ago, and a segment that was not burnt, so arguably there are six study sites (3 burnt, 3 unburnt). But in fact it was the same fire that burnt the three burnt segments, and the same geography that protected the three unburnt segments, so it could be argued that the study lacks credible replication.

Response: Thank you for for your valuable comment. This study focuses on one of the most significant fires that happened in Indonesia in 1998. We chose Bangkirai Hill as the site of our study because it has places that were affected by 1998 fires and some area unburned. Burned and unburned forest are separated by a river or 2- to 3-meter-wide border. Three 20 m x 320 m long transects were laid to cross from unburned forest into burned forest, with 160 m of transect in each forest conditions. On each transect, eight plots per forest condition were established, spaced ± 3 m apart, resulting in 3 sites to represent Fire 1998 in Bangkirai Hill, 48 plots total represent edge effect from fire break (24 plots in burned and 24 plots in unburned areas).

Furthermore, we are told that the unburnt sites were all managed as part of a “nature leisure park”, and I wonder if that contributed in some way to the outcome that they did not burn. We should be told much more about that. For example, was there a deliberate program of fuel reduction within the leisure park? Were there active efforts to control the fire as it approached the leisure park? Was the site of the leisure park selected because of some intrinsic features (e.g. geology, forest type or forest maturity) that may have reflected propensity to burn, carbon profile, biodiversity and manner of subsequent regeneration? These are important points because they may have affected the manner of regeneration as well as the intrinsic base values. For example, is it possible that the dominance of Shorea laevis in the regeneration reflects an abundance of that species before it was even burnt?

Response: Thank you for your valuable comment. Burned and unburned transects are located at the same forest management location, namely nature leisure parks Bangkirai Hills. There is no intervention in the area after the fire, so there is a natural regeneration process. Burned and unburned areas are located in adjacent locations only separated by rivers or firebreaks so that they have the same forest conditions before the fire. We have added photographs in Figure 2.

Some of the differences observed in the current study could be due to differences in intrinsic base values as mentioned above. A broader study (with more transects distributed much more widely) would be needed to dispel that concern. Perhaps it’s too late to do the broader study, but if this is to be published it is essential to tell us what is known about any pre-fire differences between the burnt and unburnt segments of the three transects. Are there photographs, satellite images or botanical assessments of the areas from before the fire? If so, we should be told about them so we can assess whether the burnt and unburnt segments are truly comparable.

Response: Thank you for suggestion. To make truly comparable, 20 m wide x 320 m long transects were laid to cross from unburned forest into burned forest, with 160 m of transect in each forest conditions. Burned and unburned areas are located in adjacent locations only separated by rivers or firebreaks so that they have the same forest conditions before the fire. Several other studies in this site has been done, including Simbolon et al. (2003) Impacts of dry season and forest fire 1997-1998 episodes on mixed dipterocarp forest at bukit bangkirai, east kalimantan. We have added hotspot map in supplement (Figure S2), the conditions prior to this fire can be represented in unburned areas in adjacent sites, the same conditions recorded by Simbolon et al. (2003). The 1998 fire in Bangkirai Hill caused 500-583 standing dead trees per ha and the number of life trees decreased to 488 – 1000 per ha, basal area in burned area 5.92 – 20.87 m2/ha with Shannon's Diversity Index of <2. We have added sentence (line 88-93) to explain the forest condition after fire. The photograph of forest condition 23 yrs after fire and unburned forest can be seen in Figure 2.

The same issue has implications also for the statistical analysis. I fear that this has been done with the assumption that each of the 48 plots is independent (n=48), though I’d be happy to be corrected. Section 2.4 is completely inadequate as it doesn’t tell us how it was done. There are only six study segments and n=6 (arguably only two). I suspect that the analysis needs to be redone, and it certainly needs to be explained more carefully

Response: Thank you for suggestion. We have reanalyzed as a reviewer suggestion, we used transect as unit sampel (n=6).

I’m sorry to be so negative, because I can see that a lot of hard work has gone into this study and I think the results may be worth reporting. Perhaps the best way forward is to recast it as a case study, and be very circumspect about what it says about the recovery of vegetation and carbon stocks in general after fire: make it very clear that it refers to just one small area with unknown comparability between parts that were burnt or unburnt. Make it very clear that this cannot be considered a definitive study of the effects of fire on recovery of vegetation and carbon stocks, for a broader study with more replication is needed.

Response: Thank you for your valuable comment. Yes, we agree, this research is a case study of the 1998 fire incident in Bangkirai hills. We point out that our study represents only a case study of the effects of fire on recovery of vegetation and carbon stocks in dryland forests of Indonesia, a broadscale level experiment will be required to better understand the dynamics of forset carbon recovery at the landscape level. We have added sentence (line 499-502) to make it very clear.

Minor Comment :

Line 220, hotter and drier

Response: Thank you for your correction. We have revised it

Fig.5, Table 3, etc. Do these graphs and tables refer to burnt or unburnt? As fire is the main variable of interest, all graphs and tables should distinguish burnt from unburnt

Response: Thank you for suggestion. We have revised Fig.5, Table 3, as a reviewer suggestion

Lines 314-317, don’t just tell us that these indices differed significantly between burnt and unburnt: it’s more important to tell us the direction of the difference, i.e. were the burnt sites more or less diverse than the unburnt sites?

Response: Thank you for your valuable comment. We have added the sentence (line 472 - 476) to incorporate your suggestion

Table 5, don’t just show us F and P values: what were the predicted means?

Response: Thank you for the nice reminder. We have revised the Table 5

Lines 326-329, this sort of information should be in results not discussion

Response: Thank you for suggestion. We think this information related with next paragraph that described carbon stock recovery.

Lines 342-343, can’t say this is “usually” the case when it’s based on just one reference (Slik et al. 2008), referring to an unstated number of sites. If Slik et al’s work was based on multiple sites, that should be stated.

Response: Thank you for your nice reminder. We have revised it to incorporate your suggestion

Lines 408-412, the recovery of “evenness” may be worth reporting in the results, but I doubt that it deserves mention in the discussion, as evenness is an artificial construct, not well understood by the average reader.

Response: Thank you for suggestion. We have added sentences (line 472-476)

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Thank you for addressing my earlier questions.  I still have a few more comments and questions as in attached report.  Some grammatical issues arose, and the whole ms would benefit from a thorough edit.  I hope that can be arranged.  When those things are done it will be a valuable paper.  

Comments for author File: Comments.pdf

Author Response

Thank you for the second round of Reviewers’ comments. A valuable paper, getting the manuscript into the form in which the results are most relevant to and best understood by readers is a high priority for my co-authors and myself. We appreciate the time and effort the reviewers have invested in assisting with that.

The specific changes made in response to the reviewer comments are detailed below:

Line 21. Change “is” to “was” (or “was insufficient in this case”) to emphasise that you are talking about your results from this study, not a general property of forests in Kalimantan.

Response: Thank you for your correction. We have revised it

Line 27. Change “is required” to “would be required”, for the same reason as above.

Response: Thank you for your correction. We have revised it

Line 36. Delete full stop after “forests”.

Response: Thank you for your correction. We have revised it

Lines 82-83. Need to add full stop after “[11, 24, 25]”, and revise syntax or grammar in what becomes the next sentence (I’m not sure what it’s trying to say?).

Response: Thank you for your correction. We have revised it

Line 90. Change “caused” to “produced”. [The fires caused mortality, and produced the observed density of dead trees.]

Response: Thank you for your correction. We have revised it

Lines 90-91. Need full stop after 1000 per ha. Rewrite what becomes the next sentence about basal area and Shannon’s Diversity Index. I’m not sure what it’s meant to be saying, and suspect I may need more than one new sentence to say how basal area and the diversity index differ between burnt and unburnt forest.

Response: Thank you for your correction and suggestion. We have revised it

Line 128. Change “class” to “classes”.

Response: Thank you for your correction. We have revised it

Line 245. Change “different to” to “less than”. It is always important to describe the nature of a difference (e.g. more or less, not just different).

Response: Thank you for your suggestion. We have revised it

Line 301. Change “number” to “numbers”.

Response: Thank you for your correction. We have revised it

Table 5, thanks for changing it, that’s good. Perhaps change “tree” to “large tree”, as I assume you are not counting saplings and poles in that category?

Response: Thank you for your suggestion. We have revised it

Lines 336-337. Revise sentence beginning “Yet..” to be more informative. For example, could say “Yet, the species richness index was significantly lower in the 23-year regrowth than in unburnt forest, in both saplings and large trees.” Also, am I right in thinking that “trees” means large trees rather than total trees? Perhaps spell that out throughout.

Response: Thank you for your suggestion. We have revised it

Line 388. Change “contribution” to “contributions”.

Response: Thank you for your correction. We have revised it

Line 392. Delete full stop after “biomass”.

Response: Thank you for your correction. We have revised it

Line 404. Change “influencing” to “influence”.

Response: Thank you for your correction. We have revised it

Lines 406-407. You have shown that tree mortality was inversely related both to tree diameter and bark thickness, but I’m not sure that you’ve shown that either of those metrics was a better predictor than the other? If I’m right, you could change the sentence to read:

Response: Thank you for your suggestion. We have revised it

Line 418. Are these comparisons between areas of similar size?

Response: Thank you for your question. The area is approximately the same, our area is 0.96 ha, while the reference area is 1 ha. We've added the area units in the sentence.

Lines 434-435. Perhaps change the new sentence to “However, species richness of saplings and old trees had not yet recovered to levels found in unburned forests.”

Response: Thank you for your suggestion. We have revised it

Lines 436-437. Add space after “levels”. Change “has recovered to unburned forests” to “had recovered to levels found in unburned forests”.

Response: Thank you for your correction and suggestion. We have revised it

Lines 463-464. Thanks for adding this sentence. Perhaps revise slightly to “….Indonesia, and a broadscale experiment would be required….”

Response: Thank you for your suggestion. We have revised it

Figure S4. This “histogram of time recovery” needs further explanation.

Response: Thank you for your suggestion. We have added sentence as a note to explain this histogram

Author Response File: Author Response.pdf