Assessing Intraspecific Variation of Tree Species Based on Sentinel-2 Vegetation Indices Across Space and Time

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript has certain research value. This manuscript demonstrates that the use of Sentinel-2 can effectively monitor forest variation and provide meaningful references for studying management strategy.

1、In the 1. Introduction part, “[14] revealed a relationship of intraspecific genetic variation visible in airborne imaging spectroscopy and pinpointed the strongest relationships to spectral regions representative of leaf water content, phenols, pigments, and waxes. These traits reflect genotype-specific phenotypic responses [14].” I suggest deleting duplicate references.

2、In the 2.5. Assessing intraspecific variation part, “All analyses and visualizations were conducted in R, based on vegetation index time series processed using the FORCE framework [47].” What is FORCE framework?

3、In the Table 1. Two-step data filtering for provisioning species-specific Sentinel-2 time series part, I suggest changing the table format to a standard three line table.

4、In the Figure 5. The proportions of the components of intraspecific variation of each vegetation index and species within the vegetation period part, How are temporal, spatial, and spatiotemporal components calculated?

5、In the 4.1. Spatial and temporal patterns of intraspecific variation part, “Interestingly, several local maxima in standard deviation for different species and indices” I suggest placing it after Figure 6 and connecting it with “were observed.”

6、In the 5. Conclusions part, I suggest rephrasing the textual description, adding some textual descriptions of the 3. Results part.

Comments on the Quality of English Language

The English could be improved.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study proposes an innovative method to quantify intraspecific variation within tree species using Sentinel-2 time series data and Swiss national forest resource inventory data. It holds significant value for forest ecology and forest management in the context of climate change. The study is scientifically designed but requires further refinement in certain areas.

1. The Introduction (lines 20-89) is not sufficiently well-written and lacks depth, failing to adequately establish the necessity and sufficiency of this research.
2. The selection criteria for the five vegetation indices (e.g., NDVI, EVI, etc.) need to be clarified: Are they based on species-specific sensitivity testing?
3. Figure 2 on page 6 is not clear enough; please revise it.
4. The definition of "pure species plots" needs to be explained (e.g., minimum canopy cover threshold, standards for excluding mixed forests). Edge effects or microenvironmental variation within the plots may affect the results.
5. The small figure in Figure 4 on page 9 is recommended to be numbered.
6. Lines 296-480: The significant temporal variation in deciduous species aligns with phenological characteristics, but does this variation only reflect seasonal cycles? Does it include the impacts of extreme climate events (e.g., drought)? A discussion on this would be beneficial.
7. Lines 296-480: It is suggested to discuss the applicability of this method in high-diversity areas (e.g., tropical forests) to enhance its universal value.
8. Lines 482-492: It is recommended to summarize the points separately.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

1、In the 2.2. Sentinel-2 time series processing part, “and stored the images in a data cube structure. We also co-registered all Sentinel-2 images with monthly Landsat Collection 2 composites based on near-infrared images from 2014 to2021 to improve the geometric consistency of the full time series [31,32]” I suggest placing it in front of Figure 1 and connecting it with 'We resampled all bands to a 10m spatial resolution using cubic convolution'.

2、In the 2.4. Tree species-specific Sentinel-2 time series part, “procedure reduced the dataset for the year 2020 in step two from 610 to 200 plots. These plots contained 763 pixels with tree trunk positions, which were used for the analyses. A total of 380 pixels represent evergreen species, and 383 pixels deciduous species. The dataset is openly available [51].” I suggest placing it in front of Figure 2 and connecting it with ' The winter images were important to represent species phenology, and as we removed snow and poorly illuminated areas, vegetation can still be characterized using these images. This'.

3、In the 3. Results part, “within the vegetation period than outside of it. During the start of the growing season, the intraspecific variation of the deciduous species was among the lowest in the whole year.” I suggest placing it in front of Figure 3 and connecting it with 'The intra specific variations of CCI, CIre NDVI, and NDMI was lower'.

4、In the 3. Results part, “The associated percentages for the spatial, temporal and spatiotemporal components of intraspecific variation are shown in Figure 5 (Further details in Figures Supplementary material S2.1– S.26)…As for the EVI Castanea sativa had the largest amplitude for the NDMI.” I suggest placing it in front of Figure 5 and Figure 6.

5、In the 4. Discussion part, I suggest placing Figure 5 and Figure 6 after “Intraspecific variation due to temporal differences (Figure 5) plays a significant role in deciduous tree species……Moreover, analyzing stability across vegetation indices can help identify species resilience and provide benchmarks for comparison with other species”.

 

Comments on the Quality of English Language

The English could be improved to more clearly express the research. 

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf