Monitoring Approach for Tropical Coniferous Forest Degradation Using Remote Sensing and Field Data

Round 1

Reviewer 1 Report

The study attempted new combination of approaches to as forest change dynamics and associated carbon stock change.

In the first part of the study a new combination of approached is proposed to assess the forest change dynamics, however it require some explanations for further clarity.

1. Author provided forest definition and degradation parameter in Figure 2 but has not used/linked these parameters in their assessment.  

2. According to the (Figure 3: Step 3) Stable forest class (1990-2018) is picked to further process for degradation assessment which is not logical.

3. The study results heavily rely on classification of forest change classes’ (deforestation, degradation, restored forest). Google Earth Images are being used for reference, validation and also for determining the importance spectral indices in degradation assessment while the high resolution GE images are only available for recent years. A temporal analysis using proposed spectral indices can be performed in five yearly window for entire time period (1990 – 2018) to better understand the change dynamics.

The second part uses conventional method for linking field based biomass measurement to extrapolate with remote sensing assessments. The analysis across different pools and protected area provide good insights relevant for forest emissions understanding and protected area managements.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript with the title “Monitoring approach for tropical forest degradation using remote sensing and data field” presents a study that evaluates deforestation and forest degradation using freely available data and cloud based-platform GEE. The paper takes into a working topic and presents an interesting approach. However, there are some issues that need to be addressed in order to the paper will be accepted for publication.

Here are my comments:

General

The graphical abstract should be outside the manuscript. Please remove it.

I believe there is no agreement between the title of the manuscript “Monitoring approach for tropical forest degradation….” and the study area. The authors stated that the study area is focused on pine forests. According to FAO, Pine forests occurring in the tropics are often held to be mere extensions of temperate-zone forests into the tropical zone, and it is maintained that the only true tropical pines in the whole world are P. merkusii in the Asia region, and P. hondurensis in Mexico to Nicaragua and in the West Indies. I would suggest changing the title to “Monitoring approach for forest degradation using remote sensing data”

Another concern, Is this methodology suitable for tropical forests since it is based on forest defined by the Dominican Republic?

The definition of forest degradation is the core, and subject of debate for several years, of the paper. There are two types of definitions, those made by institutions (i.e. FAO, UNEDP, UNFCCC, ITTO, etc) that have a strong theoretical/political component and those proposed by authors in literature base on particular needs, data type, vegetation type, resources etc. (Simula 2009; Danielsen et al. 2011; Hein et al. 2018; Olander, Galik, and Kissinger 2012; Romero-Sanchez and Ponce-Hernandez 2017; Thompson et al. 2013)

There is, however, a consensus that forest degradation represents a human-induced negative impact on carbon stocks, with measured forest variables (GOFC-GOLD 2015).

From the definition used by the authors apparently the natural disturbances (that could be part of the forest dynamic) are considered as degradation. Could you please elaborate a little bit more about this issue?

References

Danielsen, Finn, Margaret Skutsch, Neil D. Burgess, Per Moestrup Jensen, Herizo Andrianandrasana, Bhaskar Karky, Richard Lewis, et al. 2011. “At the Heart of REDD+: A Role for Local People in Monitoring Forests?” Conservation Letters 4 (2): 158–67. https://doi.org/10.1111/j.1755-263X.2010.00159.x.

GOFC-GOLD. 2015. A Sourcebook of Methods and Procedures for Monitoring and Reporting Anthropogenic Greenhouse Gas Emissions and Removals Associated with Deforestation, Gains and Losses of Carbon Stocks in Forests Remaining Forests, and Forestation. GOFC-GOLD Report Version COP18-1. Wageningen: Land Cover Project Office, Wageningen University, The Netherlands. http://www.gofcgold.wur.nl/redd/sourcebook/GOFC-GOLD_Sourcebook.pdf.

Hein, Jonas, Alejandro Guarin, Ezra Frommé, and Pieter Pauw. 2018. “Deforestation and the Paris Climate Agreement: An Assessment of REDD + in the National Climate Action Plans.” Forest Policy and Economics 90 (November 2017): 7–11. https://doi.org/https://doi.org/10.1016/j.forpol.2018.01.005.

Olander, Lydia P, Christopher S Galik, and Gabrielle A Kissinger. 2012. “Operationalizing REDD+: Scope of Reduced Emissions from Deforestation and Forest Degradation.” Current Opinion in Environmental Sustainability 4 (6): 661–69. https://doi.org/10.1016/j.cosust.2012.07.003.

Romero-Sanchez, Martin Enrique, and Raul Ponce-Hernandez. 2017. “Assessing and Monitoring Forest Degradation in a Deciduous Tropical Forest in Mexico via Remote Sensing Indicators.” Forests 8 (9): 302. https://doi.org/10.3390/f8090302.

Simula, Markku. 2009. “Towards Definin of Forest Degradation: Comparative Analysis of Existing Definitions.” 154. Forest Resources Assessment Working Paper No. 154. Rome:UNFAO.

Thompson, Ian D, Manuel R Guariguata, Kimiko Okabe, Carlos Bahamondez, Robert Nasi, Victoria Heymell, and Cesar Sabogal. 2013. “An Operational Framework for Defining and Monitoring Forest Degradatio.” Ecology and Society 18 (2). https://doi.org/10.5751/ES-05443-180220.

About the imagery pre-processing, did you perform Gap-filling for the Landsat ETM+ imagery?

More information about the National forest inventory is needed. For instance, what is the size of the sample plot? If the authors claim that are permanent sample plots; when they were established?

Do they match the imagery dates? How many re-measurements?  Are 51 plots enough to cover the study area?

Does the sample size and shape match the pixel size and shape from Landsat imagery? how do you deal with the mismatching? In case of having it

In lines 210-211 the authors stated: “The sampling design adopted for the NFI corresponds to a systematic and stratified sampling, with plots evenly distributed (equidistant apart from each other) within each forest stratum”. What was the distance between each plot? How many strata did you have? It would be possible to add a table that shows the number of plots per stratum.

According to the URL provided https://www.forestcarbonpartnership.org/country/dominican-republic the national monitoring system for REDD+ started in 2012

Section 2.6 carbon stock change magnitude

It is no clear to me how the estimations were made. First, I do understand how the magnitude can be estimated from a Landsat time series, however, there is no description of such a process. At least it is not a time series analysis per se because you are analyzing two years (1990 and 2018). Then, it becomes a bitemporal analysis based on “n” classes. Second, is it not clear how the change in carbon stocks are quantified. What are the bases of equation 1? How to relate simple spectral responses (Vegetation indices) to four compartments of carbon (especially belowground, litter and deadwood). What assumptions are made? And more importantly where is the support of those assumptions?

There is a big problem with the intrinsic forest dynamic of the study area that is not considered. For instance, there is a long time between 1990 and 2018 and I am not sure if the approach is capturing the “events” that happened in between that could lead to gains or losses of carbon stocks. For the purposes of REDD+, the main issue is to set a “baseline” that could be measurable, verifiable, and reportable (MRV systems) with a low degree of uncertainty. The other issue is the use of optical imagery to estimate carbon stocks and changes. Landsat only “sees” the canopy, which can remain the same for a long time (in terms of %canopy cover), and relate the spectral response to a volume, Aboveground biomass/carbon value respect to a measurement on the field. If the spectral response does not change over the time, the assumption is that carbon or volume content is the same, but is not necessarily true because the forest is growing or suffering natural disturbances that are not noted if the analysis is made without the proper assumptions. Please see (Gómez et al. 2014)

Gómez, Cristina, Joanne C. White, Michael A. Wulder, and Pablo Alejandro. 2014. “Historical Forest Biomass Dynamics Modelled with Landsat Spectral Trajectories.” ISPRS Journal of Photogrammetry and Remote Sensing 93 (July): 14–28. https://doi.org/10.1016/j.isprsjprs.2014.03.008.

After all the discussion above (I am sorry for the wording) the questions are What are your assumptions for do the analysis in the way you did? and how to support the decision made?

Particular

Line 87-88. Please provide a reference to support the statement.

Line 245. Is Stratum referring to Land cover classes?

Line 248. What supervised classifier?

Line 259. Eliminate “mixture”

Line 273 What is “LTS”?

Line 402. Why 328,777 ha if in methods section the study area described as pine forest is 450,000 ha?

Line 437 should be in the methods section.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Very well developed and presented study.

Please, adjust Figure 3 title removing from it the explanation. Put this explanation ina separate paragraph immediately after the figure.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have adequately addressed the issues highlighted in the previous review.

Author Response

Response to Reviewer 1 Comments

The authors appreciate the overall support and constructive feedback provided by the reviewer, which will improve the clarity of the paper.

Thank you.

Reviewer 2 Report

Many thanks to the authors for addressing most of my revisions. The manuscript has been substantially improved. However, small issues still remain.

1) About the graphical abstract, my understanding (following the guide for authors) is that the graphical abstract goes in a separate file, and when the paper is finally published it will go aside from the manuscript. Let the editor decide in this matter.

2) I stand for my point number two. Besides the author's answer does not justify why is it calling "tropical forest" to pine forests (Pinus occidentalis). Again, the title as it is creates false expectations for readers.  I suggest to change it. Perhaps if you add to the title "Tropical coniferous forests" will make more sense.

For the rest of the revisions, I consider they were addressed properly. I acknowledge the authors for a fine job. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

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.