The Coexistence of Trees, Shrubs, and Grasses Creates a Complex Picture of Land Surface Phenology in Dry Tropical Ecosystems

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript is generally well-written. I have only two minor comments.

First, I suggest including representative on-site photographs taken by the PhenoCam, along with examples of the selected ROIs, to provide readers with a clearer understanding of the vegetation characteristics and coverage at each site.

Second, the time-series data at the open woodland cerrado site exhibit notable discrepancies among different vegetation indices. Please provide further explanation or discussion on the potential causes of these differences.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This study compares MOLDS satellite data and near-surface phenocam observations to evaluate vegetation phenology across four seasonally dry tropical sites in Brazil. The authors found limited correlation between vegetation indices derived from MODIS and phenocam, especially in areas with high evergreen cover or heterogeneous vegetation components. This research enhances our understanding of complex phenological patterns within mixed tree-shrub-grass systems in seasonally dry tropics. However, there are still several concerns that need to be further addressed before the recommendation of acceptance for this paper.

Main comments:

The authors note in both the Abstract and Discussion that the correlation between vegetation indices derived from MODIS and phenocam is lower at sites characterized by high heterogeneity and a greater proportion of evergreen vegetation. However, no phenocam images are presented in the manuscript, which makes it difficult for readers to understand what the phenocams were actually observing. I recommend that the authors include representative phenocam images from all four study sites, annotated with the different ROIs. More importantly, only three types of ROIs are presented, grass, deciduous, and community. It is unclear how the authors concluded that poorer correlations are associated with a higher proportion of evergreen vegetation. In the current version of the manuscript, only line 188 mentions that the community ROI includes evergreen strategies at the Cerrado shrubland site. However, according to the results, it is the other two Cerrado sites that exhibit poor correlations.

Other comments:

1. Figure 1, “woordland” should be “woodland”
2. Line 178, please check the form of the citation.
3. Lines 186–188 appear to be a continuation of the previous paragraph that introduces the ROIs. There is no need to separate the description of grassy ROIs into a standalone paragraph, as it disrupts the flow and consistency of the section.
4. Line 242-244. Please move the maximum cross-correlation function (CCF) here, instead of introducing it in the Results section.
5. Line 320-322. According to Fig. 3d, the R2 between MODIS GCC and phenocam deciduous GCC is only 0.02. Please double-check it.
6. Figure 3. In panel d, the correlations between MODIS GCC and phenocam GCC (deciduous and community) are positive, but those between MODIS EVI and phenocam GCC are negative. How to explain it? Is this because of the differences in spatial resolutions of the two MODIS products? This would be an interesting point for further discussion.
7. Line 335, “degree of similarity”, do you mean R2? Clarify it.
8. Line 416, “land-surface phenology” or “land surface phenology”. Please maintain consistency throughout. I suggest using the latter, as it is more commonly adopted in the literature.
9. Line 425-427, please see my main comment.
10. Line 445-453, This conclusion should be drawn with caution, as it may not be fully supported by the current data. It is possible that the PhenoCam field of view (FOV) does not adequately represent the broader MODIS pixel area. Although only 20% of the FOV is classified as grass, the majority of the area covered by the corresponding MODIS pixel could in fact consist primarily of grassland.
11. Line 463, it should be “MODIS time series” instead of “MODIS EO”.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

see attached file

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx