Second-Entry Burns Reduce Mid-Canopy Fuels and Create Resilient Forest Structure in Yosemite National Park, California

Round 1

Reviewer 1 Report

Review of “Repeat fire targets mid-canopy fuels and increases forest health”

 

This study uses airborne light detection and ranging data (LIDAR) to measure forest structure in Yosemite National Park, CA, USA. Data from the Sentinel satellite is also used to calculate the normalized differenced vegetation index (NDVI), a proxy for vegetation greenness. The authors then assess how forest type, temperature, precipitation, the number of fires, and fire return intervals are related to LIDAR derived forest structure metrics, both for multiple metrics as a multivariate response, and for individual metrics as a univariate response. 

The paper had two main interesting findings, which are best stated in the middle of a paragraph on lines 265--268:

• “We found that mid-canopy strata, in particular 8-16m, were most impacted by multiple fires, supporting restoration actions that seek to remove smaller diameter trees.” 

• “Importantly, two entries were required to achieve these structural and density changes, but additional fire did not significantly impact canopy structure.”

Both findings are addressed in the abstract, but only the first finding is addressed in the title. Likewise, the discussion also focuses on the first finding. I actually think the second finding—specifically that there was little structure change after two fires—is the most novel result of the paper. The results have implications for management outside Yosemite National Park, in that it take two fires to move forest structure to a state that seems to be at an equilibrium through subsequent burns. I would suggest emphasizing and discussing this result in more detail, and consider if it could be integrated into the title. 

 

The general mythological approach is robust, but I have concerns about multicollinearity among predictor variables in the modeling. Specifically, I would expect temperature and precipitation to be associate with forest type. I would also expect that burn class and median fire return interval departure would be highly correlated. If these sets of variables are correlated with each other, they should not be used in the same statistical model. This statistical issue must be address for the paper to be acceptable for publication. Addressing this problem could occur through either reduce the number of variables in the predictive models, or demonstrate that predictor variables are not highly correlated. 

 

I have one other major comment on the methodology, which I would strongly recommend that the authors address. The methods describe an NMDS ordination, but it is not shown. Because the authors are describing a multivariate response, showing that ordination is key to explaining how forest structure differs in response to your predictor variables. Consider replacing the conceptual figure (Figure 1) with the ordination results. Specifically, display the NMDS results with forest type, number of fires, FRID,  temperature and precipitation plotted over them. Also consider plotting NDVI, and ratio of prescribed to wildland fires on the ordination. Conceptual figures are a great place to start, but the authors have created an ordination that show the same type of information using observational data. The following citation has an ordination of similar lidar forest structure metrics, which perhaps could serve as a visual guide for the authors: 

Cansler, C.A., Kane, V.R., Bartl-Geller, B.N., Churchill, D.J., Hessburg, P.F., Povak, N.A., Lutz, J.A., Kane, J. and Larson, A.J., 2022. Postfire treatments alter forest canopy structure up to three decades after fire. Forest Ecology and Management, 505, p.119872.

 

Finally, the discussion would be more impactful if the authors made more direct comparisons to previous studies of forest structure both in Yosemite National Park, and in the central Sierra Nevada. There has been previous work with both lidar and with field data examining the effects of fire on forest structure development. In particular, studies led by V. Kane (citations 30 and 39), and work generally on forest structure development in the Sierra Nevada (citations below and the Sierra Nevada ecosystem project report by Millar et al in 1996, etc.) deserves to be discussed in more detail. How does this study—which uses data covering the extent of four forest types in Yosemite National Park—depend our understanding beyond what was covered in previous studies? Adding another paragraph on the expected forest stand structures of each of the forest types you assessed, and how your results match up or further develop expectations would make this a more impactful paper. 

Lutz, J.A., Van Wagtendonk, J.W. and Franklin, J.F., 2009. Twentieth-century decline of large-diameter trees in Yosemite National Park, California, USA. Forest Ecology and Management, 257(11), pp.2296-2307.

Collins, B.M., Everett, R.G. and Stephens, S.L., 2011. Impacts of fire exclusion and recent managed fire on forest structure in old growth Sierra Nevada mixed‐conifer forests. Ecosphere, 2(4), pp.1-14.

Stephens, S.L., Lydersen, J.M., Collins, B.M., Fry, D.L. and Meyer, M.D., 2015. Historical and current landscape‐scale ponderosa pine and mixed conifer forest structure in the Southern Sierra Nevada. Ecosphere, 6(5), pp.1-63.

Lydersen, J.M., Collins, B.M., Miller, J.D., Fry, D.L. and Stephens, S.L., 2016. Relating fire-caused change in forest structure to remotely sensed estimates of fire severity. Fire Ecology, 12(3), pp.99-116.

Stephens, S.L., Stevens, J.T., Collins, B.M., York, R.A. and Lydersen, J.M., 2018. Historical and modern landscape forest structure in fir (Abies)-dominated mixed conifer forests in the northern Sierra Nevada, USA. Fire Ecology, 14(2), pp.1-14.

In relation to forest structure and NDVI:

Collins, B.M., Moghaddas, J.J. and Stephens, S.L., 2007. Initial changes in forest structure and understory plant communities following fuel reduction activities in a Sierra Nevada mixed conifer forest. Forest Ecology and Management, 239(1-3), pp.102-111.

 

Additional line-by-line comments

 

 

L126: state the date range for the NPS fire records here. Also, acknowledge and discuss uncertainty with older records. Were polygons mapped back to 1930? Were smaller fires recorded? Minimum mapping unit?

 

169: TAO needs to be defined. The citation for the method used in your data is: 

Jeronimo, S.M., Kane, V.R., Churchill, D.J., McGaughey, R.J. and Franklin, J.F., 2018. Applying LiDAR individual tree detection to management of structurally diverse forest landscapes. Journal of Forestry, 116(4), pp.336-346.

Use this paper to better described the methods used for individual tree detection (and thus TAO detection). Add text noting limitations to tree detection with LIDAR.  

 

171: What alpha level was used to determine significance? 

 

178: A plot showing the total number of samples in a given number of fires, per forest type would allow your readers to understand if and how sample sizes decreased in >3 fires class. This information is needed to interpret the box plot. There seems to be some difference in the >3 category in the box plots, but without sample size information it is difficult to determine if this is a true difference or if it reflects the small sample. Thoughts on that class and the box plots in Figure 2 and in the SI?

 

185: State that these are results of the PERMANOVA, or that the response was multivariate forest structure. Also, the pseudo-F values from the PERMANOVA can be viewed as a measure of effect size. Consider reporting them in order of decreasing importance (e.g., temperature, forest typ, burn class, topographic context (??), FRID, and precipitation. 

The “topographic context” is a surprise here. This variable is not defined in table S1, nor described as a predictor in the PERMANOVA. It is also not explaining a lot of variance. Consider dropping this variable, or describing it in more detail.

 

221-225: Additional examples of how after two fires responses become less predictable. This is an interesting theme throughout. 

 

267: Consider developing a paragraph about the need for two entries to achieve structural changes, and implications for forest and fire management, maintenance of forests structures that are resilient to fire, etc. 

295: here and elsewhere where you discuss burn severity, you should cite and incorporate Lutz et al. 2009, in addition to the Collins et al. 2007 paper you currently cite. 

Lutz, J.A., Van Wagtendonk, J.W., Thode, A.E., Miller, J.D. and Franklin, J.F., 2009. Climate, lightning ignitions, and fire severity in Yosemite National Park, California, USA. International Journal of Wildland Fire, 18(7), pp.765-774.

 

Author Response

Thank you for thoughtful review and suggestions. We hope we have addressed your concerns and believe the manuscript is much improved. We have included specific responses to your review in the attached document. 

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript presents a good topic, it follows the scope of the journal and would be interesting for the readers. Here are same minour comments which should be corrected for publication.

-    -  Abstract needs to be reworked: There is no comment about the novelty of this work. You should briefly state the principal results, and major conclusions.

-        -Add new references using the same approach in the introduction section.

-        - Add new references and comparisons with other research at the end of your discussion

Good luck

Author Response

Thank you for review and suggestions. We hope we have addressed your concerns and believe the manuscript is much improved. We have included specific responses to your review in the attached document. 

Author Response File: Author Response.docx

Reviewer 3 Report

 

Forests – Reviewer

Article

Repeat fire targets mid-canopy fuels and increases forest health

 

Line 2: Title

The title needs to specify the place where the fire helps in the health of the forest, because, for example in tropical forests, repeated fire causes the loss of several species of flora and fauna, being quite harmful.

I suggest:

"Repeat fire targets mid-canopy fuels and increases forest health in the Yosemite National Park, California"

Lines 182-183: Materials and Methods – Table 1

What is the meaning of the abbreviations "MAT" and "MAP" in Table 1?

The explanation must be inserted in the legend of table 1.

Lines 302-325:

I suggest creating a topic of "Final Thoughts" or "Conclusion"

It would be interesting to include images of the forest mapped by Lidar, demonstrating the mapping process with the sensor to better elucidate such methodological processes for Forests readers.

 

 

 

Author Response

Thank you for your review and suggestions. We hope we have addressed your concerns and believe the manuscript is much improved. We have included specific responses to your review in the attached document. 

Author Response File: Author Response.docx