CO Editor

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment | : Report abuse

- Printer-friendly version

- Supplement

RC1: 'Referee's Comments', Anonymous Referee #1, 21 Mar 2017
- AC1: 'Response RC01', Guilherme Mataveli, 14 Jun 2017
RC2: 'Comments on “Analysis of fire dynamics in the Brazilian savannas” by Guilherme Augusto Verola Mataveli et al.', Anonymous Referee #2, 12 Apr 2017
- AC2: 'Response RC02', Guilherme Mataveli, 14 Jun 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (further review by Editor and Referees) (28 Jun 2017) by Ricardo Trigo

AR by Guilherme Mataveli on behalf of the Authors (09 Aug 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (15 Aug 2017) by Ricardo Trigo

RR by Anonymous Referee #2 (06 Sep 2017)

Suggestions for revision or reasons for rejection

I thank the authors for the new version of the paper which is improved when compared to the anterior one. Though some new questions arise in this new version and there is a need of major review to make it suitable for publication.
1) Although land-use and land cover data maps are mentioned in the Materials and Methods section, it is not clear how they were applied in the analysis. For instance, the study still lacks a discussion relating these LULC maps and the climatic, fire, and vegetation variables. In addition, why do you show a map of MODIS RGB (Figure 1)? Why not the two LULC maps used in the study? Why not a combination of both?
2) How were the results from Table 1 compared with fire? I suggest including the study of fire occurrence per LULC type for a complete description of the fire incidence associated with LULC.
3) In addition, why using only LULC data from one year if the land cover may change from year to year? Why not annual MODIS Land cover and land use product?
4) There are too many typos in citations/references. Please, make a careful review of the text.
5) Modis vegetation indices: Which MODIS collection is used, the collection 5 or the new collection 6?
6) Figure 3. It is not a good idea to show VCI and PRECIPITATION in the same y axis since they have different units and ranges.
7) Hotspots are usually described in counts. Figure 3 shows hotspots in km2, so it would be nice to show how counts were converted to km2. On the other hand, Figure 4 shows hotspots without unit.
8) In fact, Figure 3 and Figure 4 (left panels) give the same information. My feeling is that you can get to the same conclusions without using figure 3. So, I suggest removing it.
9) Figure 4 (right panels), I would like to know if the obtained trends were found to be statistically significant. It would be nice to quantify both trends and significances.
10) I think that right panels from Figure 4 should be replaced by each respective intraanual boxplots.
11) Vegetation data: Please clarify which MODIS product was used, TERRA or AQUA or both? What about the spatial resolution?
12) MODIS vegetation indices: did you use any method to filter low-quality values, due to cloud contamination? Please see (Bi et al., 2016)
13) The vegetation and burned area data set have different spatial resolutions, how do the authors compare them? It should be in data and methodology.
14) P8L25-32 P9L1-5 It is hard to conclude all those points from figure 2. I suggest removing figure 2, because all interannual variability may be better observed in Figure 4.
15) Please only show pixels with significant correlations in Figure 10 (same as figure 8).
16) Please merge the results from figures 8 and 9.
17) Please, it is advisable to consider the results of (Nogueira et al., 2017) in the introduction and discussions.
18) VCI values from figure 6: It is well established from previous studies that VCI decreases due to meteorological conditions and can be used for drought monitoring. Nevertheless, VCI also decreases after fire occurrence. So I would like to know how this point was taken into account in the discussion of Figure 6. For instance, how can we conclude that VCI minimum values are due to meteorological conditions (lack of precipitation), due to fire occurrence or due to the synergic effect of both?
Moreover, since vegetation reflectance in the near infrared band is affected primarily by leaf structure, green vegetation exhibits high reflectance values. As the leaf matures the cells enlarge, crowing together, and reducing the intercellular space, consequently decreasing the reflectance. Accordingly, NDVI (VCI) values can decrease due to senescence, to drought or to fire. The authors must at least explore this point in the discussion. In this case, it is advisable to consider the VCI values during the pre-fire season.
19) The purpose of the conclusion is to show the real contribution of the research
(i.e., how the research enhances overall knowledge of the field). I think this section needs to be rewritten.

Bi J, Myneni R, Lyapustin A, Wang Y, Park T, Chi C, Yan K, Knyazikhin Y. 2016. Amazon Forests’ Response to Droughts: A Perspective from the MAIAC Product. Remote Sensing. Multidisciplinary Digital Publishing Institute 8(4): 356.

Nogueira JMP, Rambal S, Mouillot F. 2017. Spatial Pattern of the Seasonal Drought / Burned Area Relationship across Brazilian Biomes : Sensitivity to Drought Metrics and Global Remote-Sensing Fire Products. DOI: 10.3390/cli5020042.

Hide

ED: Reconsider after major revisions (further review by Editor and Referees) (07 Sep 2017) by Ricardo Trigo

AR by Guilherme Mataveli on behalf of the Authors (19 Oct 2017) Manuscript

ED: Reconsider after major revisions (further review by editor and referees) (23 Oct 2017) by Ricardo Trigo

ED: Referee Nomination & Report Request started (01 Nov 2017) by Ricardo Trigo

RR by Anonymous Referee #2 (13 Nov 2017)

Suggestions for revision or reasons for rejection

Authors have done a good review. Now, there is a need adequate the text according to my first two comments and some minor review to make it suitable for publication:
1) Fire risk models are usually based on previously (pre-fire) information about lack of precipitation, vegetation condition, humidity, temperature. So I do not agree that VCI information simultaneous with fire occurrence could help in fire risk models. In this case, I think VCI could at least be an indicator of the area affected by the fire. However, you agree that you cannot conclude that VCI minimum values are due to meteorological conditions (lack of precipitation), due to fire occurrence or due to the synergic effect of both. In such context, you do not have any evidence about it and cannot affirm that VCI is good or not for fire risk models/burned area information. So, you should remove the text (in both abstract and conclusions): ‘qualifying the index as an input for fire risk models’.
2) P16L25-27. I suggest brief discussions about the implication of the obtained trends, concerning you are using only 14 years.

Minors comments:
3) Please check the citation and reference:
JOLLY et al., 2015
Joly, W. M., Cochrane, M. A., Freeborn, P. H., Holden, Z. A., Brown, T. J., Williamson, G. J., and Bowman, D. M. J. S.: 10 Climate-induced variations in global wildfire danger from 1979 to 2013, Nature Comms, 6, doi:10.1038/ncomms8537, 2015.

4) P4l26 – MCD45A1 insert reference:
D.P. Roy, Y. Jin, P.E. Lewis, C.O. Justice. Prototyping a global algorithm for systematic fire-affected area mapping using MODIS time series data. Remote Sens. Environ., 97 (2005), pp. 137-162, 10.1016/j.rse.2005.04.007

5) P5L1314- MOD13A3 insert reference:
K. Didan. (2015). MOD13A3 MODIS/Terra vegetation Indices Monthly L3 Global 1km SIN Grid V006. NASA EOSDIS Land Processes DAAC. https://doi.org/10.5067/modis/mod13a3.006

6) P6L2-MCD12Q1 insert reference:
Friedl, M.A., D. Sulla-Menashe, B. Tan, A. Schneider, N. Ramankutty, A. Sibley and X. Huang (2010), MODIS Collection 5 global land cover: Algorithm refinements and characterization of new datasets, 2001-2012, Collection 5.1 IGBP Land Cover, Boston University, Boston, MA, USA.

7) These references are not cited:
Prentice, I. C., Roos, C. I., Scott, A. C., Swetnam, T. W., van der Werf, G. R., and Pyne, S. J.: Fire in the Earth System, Science, 324, 5926, doi:10.1126/science.1163886, 2009.
Straschnoy, J. V., CRC Press, Boca Raton, United States of America,125-148, 2013.

8) Please insert e), f), g), h) in figure 2 and its caption. Also provide information about boxplot (i.e. how the median, outliers and quartiles are shown). Please also insert in each figure the respective y axis label.

9) I am not sure if Moreira de Araújo et al. (2012), Libonati et al. (2015a), Libonati et al. (2015b), Moreira de Araújo and Ferreira (2015), and Libonati et al. (2016) have used the same burned area dataset (MCD45). Please specify.

Hide

ED: Publish subject to minor revisions (review by editor) (16 Nov 2017) by Ricardo Trigo

AR by Guilherme Mataveli on behalf of the Authors (22 Nov 2017) Author's response Manuscript

ED: Publish as is (23 Nov 2017) by Ricardo Trigo

AR by Guilherme Mataveli on behalf of the Authors (28 Nov 2017)