Wall-to-Wall Mapping of Forest Biomass and Wood Volume Increment in Italy

Round 1

Reviewer 1 Report

After getting acquainted with the manuscript entitled “Wall-to-wall mapping of forest biomass and wood volume increment in Italy”, there were many questions and critical comments.

1. How current are the data used to measure biomass and CO2, monitor the state of forest ecosystems, etc. obtained 17 years ago (data from the National Forest Inventories (NFI) 2005) and overlayed on maps (ESA biomass map and JRC biomass map 2010)? How will these data help to make decisions on forest management if some of the forest areas (not all but some of them) may have disappeared over the last 17 years? Forest fires, tree cutting, and other factors (such as critical age, pest outbreaks, drought, etc.) all will have the potential to cause their demise.

2. “According to the European Forest Types classification system [46,47], seven of the 14 EU Forest Types classes occur in Italy.” (Lines 142-143).

How do these seven classes compare with the 12 types listed in Table 1?

3. “The CAI variations shown in Figure 3 follow the main patterns of the forest 294 GSV, but differences due to forest types can be also appreciated.” (Limes 294-295)

The data presented in the paper do not allow estimation of these differences, as it is unclear how the layout of different types of forests with different regions of Italy relates. These data cannot be extracted from either tables or figures.

4. Discussion section needs to be strengthened. It only captures the distinctions between the various types of maps. The explanation for why there are these discrepancies should be presented. Inaccuracies in the modeling? Inaccuracies in national forest inventories in different parts of the country? Errors in interpreting and analysing satellite information?

5. “…with the only exception in JRC-BIO for Puglia region…” (Lines 322)

May be in ESA-BIO? And what about ITA-BIO vs ESA-BIO for Sardegna region?

6. Please adjust drawing numbering and include correct numbers and references in the text.

Author Response

See attached file

Author Response File: Author Response.docx

Reviewer 2 Report

GENERAL COMMENTS

This paper presents the methodology developed to create high resolution forest biomass and CAI maps of Italy. A CAI map is developed using the NPP map of Chirici et al. (2015) validating with Italian NFI field data. Also, the biomass map is compared with maps developed by European Spatial Agency and the European Joint Research Centre.

In general, this is a well written paper. Materials and methods are sound and correctly applied. Discussion and conclusions are coherent with the results obtained but discussion must be expanded. I am not able to appreciate the English grammar, but the manuscript is in general well understandable.

From my point of view, the main concern of the paper is the CAI model. As the authors well know, a good model should be unbiased, so that an even distribution of underestimates and overestimates leads to an overall bias of zero. However, the CAI model developed is clearly biased (although precise) if we consider good the NFI estimations. I encourage the authors to try to correct the bias (if possible) or at least to carry out a deeper discussion of its possible causes and how to correct it.  The reasons given on pages 339-341 seem poor to me.

I have other minor suggestions or clarifications mainly concerning organization or presentation of the information (see below).

INTRODUTION

Line 73: what is SAR data ??(i.e., SAR (synthetic aperture radar) data?

MATERIALS AND METHODS

Since author want to illustrate the methodology used to develop forest biomass and CAI high resolution maps in Italy, I consider very useful to present a workflow with the main process followed to generate both types of maps.

How BEF and WBD were assigned to a CLC class? Please specify in Table 1 the source of BEF and WBD (i.e. Federici et al., 2008).

Lines 232-233 and equation 4: Why is the bias subtracted and not added? Is there a systematic overestimation? Please explain.

Line 210: The right reference must be [49] ??. Please correct it.

DISCUSSION AND CONCLUSIONS

Pag 338. What is MBE? (mean bias estimated??). Please use the same nomenclature as in lines 227-229.

REFERENCES

Line 4, reference 37: This is a wrong reference. The right is “Novo-Fernández, A., Barrio-Anta, M., Recondo, C., Cámara-Obregón, A., López-Sánchez, C.A., 2019. Integration of National Forest Inventory and nationwide airborne laser scanning data to improve forest yield predictions in North-Western Spain. Remote Sens. 11(14), 1693. https://doi.org/10.3390/rs11141693”

Author Response

see attached file

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors:

This manuscript describes an approach to obtain the map of above-ground biomass (AGB) and current annual volume increment (CAI) for three different forest types. The research content of this article is meaningful. The structure of the article is well organized, but the results are too short without comprehensive description of the graphs and tables listed in the results part. Meanwhile, the discussion part need to further organized and connect this study to the exist similar researches. More suggestions and questions are as following:

1.       Research background in the abstract part should be simplified and add more introduction of research model and method. For the results in the abstract, I suggest to compare and analyze the specific values such as SE of evaluation indicators.

2.       The results should add more description of the graph and tables in the results part.

3.       It is unreasonable to compare the maps mapped in this study with that acquired in 2010.

4.       The conclusion of this paper is only compared with the forest biomass map of FRC-BIO and ESA BIO, it should be discussed and compared with other results using the same or similar methods.

5.       Each parameter of the CAI modelling should be described with the formula, and the data from which each variable is obtained should be explained.

6.       I suggest that Italy's forest coverage and forest area should be included in the study area.

7.       Reference [13] is omitted in the text.

8.       Unit m is omitted at line 109, similar errors at line 151 and 166.

9.       The "environmental predictors" in line 149 cannot include the "satellite images" in brackets.

10.   All variables in the table should be added with units in the header.

11.   The title number of this manuscript content is wrong of “2.2 Data” in line 157, 178 and 196.

12.   The title number of figure is wrong in “3. Results”.

Author Response

See attached file

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I have carefully read the authors' replies and the changes made to the paper. All comments and suggestions were considered and the manuscript was carefully edited. Thanks to the authors for their efforts.

Reviewer 2 Report

Authors have adequately considered or refuted my suggestions and I now find the article ready for publication

Reviewer 3 Report

Except that the reference [13] is not quoted in the text, other questions have been revised.