A Robust Framework for Bamboo Forest AGB Estimation by Integrating Geostatistical Prediction and Ensemble Learning

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study proposes a two-stage framework that integrates multi-source remote sensing data. Through EBKRP spatial interpolation and stacked ensemble learning, it achieves high-precision estimation of aboveground biomass in complex mountain bamboo forests. The overall design is relatively reasonable, but there are some issues that need to be modified and improved.

2.1 Research Area

The text mentions that the surface fragmentation index is 0.68, but it lacks a clear definition and explanation of the calculation method, which affects the repeatability of the results. If the result is calculated by others, please refer to relevant references for explanation.

2.2.3GEDI Data

From the initial dataset of 70,619 footprints, this filtering process generated 55,649 high-quality footprints suitable for analysis. What is the basis for the filtering criteria and how is Sensitivity calculated specifically? Please add the data retention rate under different thresholds and the sensitivity analysis of model performance.

3.1 Accuracy of Spatially Extrapolated LiDAR Metrics

Variables such as h_min_canopy, n_toc_photons, and the reflectance related rv and rg series had R² values concentrated between 0.26 and 0.47. Is the low precision of these parameters related to specific terrain areas? For instance, steep slopes may mask terrain-dependent errors. Please supplement the spatial correlation analysis between terrain factors (slope/direction) and low-precision parameters.

4.2Influence of Algorithm Selection on Bamboo AGB Estimation

It is mentioned in the text that "recent literature that used single models for similar forest types". Please specify what type it is and discuss the potential applicability of the model in similar bamboo species.

4.3.1 Ecological Drivers and Non-linear Mechanisms

The discussion mentioned that the GIS coverage analysis shows that approximately 73% of the high AGB areas are concentrated in regions with an altitude of 800 to 1,300 meters, a slope of less than 20°, and an annual precipitation of more than 1200mm. However, the average annual precipitation in the study area is 838.7 mm. Please explain in which areas the annual precipitation exceeding 1200mm is mainly concentrated.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

Thank you very much for your insightful and constructive comments. We have carefully revised the manuscript according to all of your suggestions.

A detailed point-by-point response has been prepared and uploaded as a separate file ("Response to Reviewer 1 Comments.pdf"). 

Thank you again for your time and consideration.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

A robust framework for the bamboo AGB is constructed by the authors of this manuscript. This system methodology is somewhat meaningful in terms of using ICESAT-2 and GEDI data to retrieve some special vegetation AGB. Here are some revision suggestions as follows:

1. In the first paragraph of the introduction section, I think the authors should refer to the meaning of the bamboo AGB inversion.
2. Lines 65-77, since the authors refer to using LiDAR and the optical sensors data can enhance the accuracy of AGB estimation, I suggest the authors’ references to the papers should list some optical data in content as proof.
3. In the 2.2.3 section, I suggest the authors add the acquire time information of GEDI data in this part.
4. Line 348, the metric “M” of the equation maybe is repeatedly expressed.
5. I suggest the authors add some outlook comment in the conclusion section

Author Response

Dear Reviewer,

Thank you very much for your insightful and constructive comments. We have carefully revised the manuscript according to all of your suggestions.

A detailed point-by-point response has been prepared and uploaded as a separate file ("Response to Reviewer 2 Comments.pdf"). As instructed by the submission system, please see the attachment.

Thank you again for your time and consideration.

Author Response File: Author Response.pdf