A Region-Growing Segmentation Approach to Delineating Timberline from Satellite-Derived Tree Fractional Cover Products

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper introduces a region-growing segmentation approach for delineating timberlines from satellite-derived tree fractional cover products. It is well written, but I have some suggestions for improvement:

The introduction could be strengthened. Many forest fractional cover products are available, and these could be discussed as well. Additionally, what is the difference between tree and forest fractional cover?

Please explain how the optimal values of fCovermean and fCoverstd were determined.

Could you discuss the potential of Dynamic World, which provides class probabilities at 10 m resolution—similar to fractional tree cover?

The quantitative evaluation of the results should be enhanced, particularly when comparing them with other products.

Figure 2 appears unusual. The x-axis ranges from 61.5 to 69, while the y-axis ranges from 66 to 67.5. However, the lengths of the x- and y-axes in the figure are similar, which may distort the representation.

Author Response

please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study proposes an automated method for delineating the timberline based on region-growing segmentation and satellite-derived fractional tree cover (fCover) data and tested in the West Siberian Plain and the Alaskan Arctic. The approach is innovative and the experimental design is relatively comprehensive. However, the manuscript has several key issues that should be addressed before publication：

General Comments

1. Section 2.2 does not provide the specific kernel sizes and number of iterations used in the morphological operations (i.e., bridge and fill). For the sake of reproducibility, the authors are strongly encouraged to report these parameter settings in detail.It would be desirable to assess the impact of these parameters on the delineated forest line boundaries.

2. The current validation is restricted to six sites in Alaska. To support the claimed cross-regional applicability of the proposed method, quantitative validation should be extended to include sites in the West Siberian Plain. Additional test locations in Siberia and reporting of distance-based accuracy metrics are recommended to enhance the credibility and robustness of the results.

3. The validation compares CGLS tree cover data from 2019 with high-resolution imagery from 2023–2024. This temporal mismatch may introduce uncertainty due to vegetation dynamics such as forest growth or disturbance events. The authors should acknowledge and discuss the potential impact of this mismatch on validation results.

4. The conclusion that “regional-scale forest encroachment is less significant than previously” is based on a comparison with the CAVM stand line (circa 2003). However, the CAVM stand line denotes the northernmost limit of individual trees, while the present study delineates the boundary of continuous forest cover. These represent different ecological thresholds, and direct comparison may be misleading. The authors should clarify the ecological definitions and avoid overinterpreting differences.

5. The manuscript tends to conflate the terms “timberline,” “treeline,” and “taiga–tundraecotone (TTE)” (e.g., in Figures 1 and 4.1), which could lead to conceptual confusion. Section 4.1 in particular would benefit from clearer definitions and more precise use of terminology.

6. Normally, the length of Remote sensing should more than 18 pages, and author should add some new result or some detail and let it suitable to publish.

Specific comments

1. Section 2.2 needs to be supplemented with the main basis for determining criteria for seed selection.

2. In Figure 4, the red dashed line representing the delineated timberline from this study is difficult to distinguish from the white fCover=0.3 line. Consider using a higher-contrast color scheme or different line styles to improve visual clarity.

3. Figure 6 should include a scale bar and a north arrow to enhance spatial interpretation and geographic referencing.

Author Response

please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have revised this paper according to my suggestions. 

Author Response

Thanks!

Reviewer 2 Report

Comments and Suggestions for Authors

The comments I mentioned has been revised and I also have some suggestions about the manuscript.

1. the unit in table was missed, such as table 2.
2. I think the input and output used different symbols or different color to avoid confusion in figure 3(a). 

Author Response

Thanks, we have put the unit in the title of Table 2 and used different colors for the input and output in Figure 3a