Towards Forest Condition Assessment: Evaluating Small-Footprint Full-Waveform Airborne Laser Scanning Data for Deriving Forest Structural and Compositional Metrics

Round 1

Reviewer 1 Report

 

This study use airborne full-waveform Lidar data to assess conservation status of semi-natural, and plantation forest in southern England. It's a followon from a previous 2016 paper, which compared discrete return and full-waveform lidar, and uses the same field and lidar data as that paper.

 

This is a good paper and an interesting topic. However, the two figures are of insufficient quality and need to be improved (see my lengthy comment below.) Given the similarity with the previous paper, there is too much repitition in the Material and Methods section.  I have some other minor suggestions and comments, listed below. Overall I would recommend the paper for acceptance if the figures are improved.

 

Comments:

 

Fig 1 – what do the thick black lines and faded grey lines correspond to. Does the edge of the Figure correspond to the edge of your study area? Study area boundaries are not clearly shown.

 

My chief complaint: your Fig 1 and Fig 2 don't match up, making it very hard to see correspondance between decidous and conifer and FCS values. You need to change either one or both figures. Does your Fig 1 show all the study area? The boundary lines on your Fig 1 and Fig 2 seem to be indicating different things, which doesn't help. It should be easy to compare the two Figs with each other. You say conifer areas have lower values but it is very difficult comparing the 2 figs to see this. The coordinate ticks should be the same values on both figs to improve ease of comparison.

 

Ln117

Recommendations to enhance breeding bird diversity in managed plantation forests determined using LiDAR DOI: 10.1002/eap.2678

might be a good additional citation at this point.

 

 

Ln 172 plantation

 

Ln 178 Quercus petraea

 

Section 2.3 Field study This section is overwhelmingly repitition from your previous 2016 paper. Maybe shorten by citing paper, and summarizing, and mentioning differences (if any.)

 

Ln226-228 This reads like a list of both native and naturalized species : scots pine (native to Uk but not S. England) and sweet chestnut (non-native. ) Maybe change to 'native or naturalized' or alternatively 'native or long-established non-native.'

 

Results:

 

In Fig S1 you could indicate coniferous, deciduous and mixed plots through use of different symbols.

 

In caption for Fig S1 you should specify this is for leaf on and leaf off combined.

 

Ln406-407 I think you have the supplementary table numbers wrong here.

 

Ln507 NBias

 

 

Ln 527-528 I think you should say what was the mean FCS value+-error for semi-natural deciduous and coniferous, rather than just stating that deciduous was higher. Was the difference significant?

Author Response

Please see attached Word document

Author Response File: Author Response.docx

Reviewer 2 Report

In this paper, 27 forest metrics were estimated from small-area full-waveform airborne laser scanning (ALS) data using random forest (RF) and automatic variable selection (Boruta) methods. The overall best model was obtained by comparing the RF models obtained from the upper, lower and combined upper/lower leaf datasets. The results of this study show that full-band, multi-temporal ALS holds a large amount of potential information that is very useful for estimating forest characteristics, both within and below the main forest canopy, and for continuous mapping over a large area. FCS maps are closely aligned with forest types and stand boundaries, suggesting that ALS data provide a viable method for mapping and monitoring forest conditions. Having high-resolution forest vegetation maps with acceptable consistency for field validation will greatly advance the understanding of forest ecology and improve conservation efforts. This article is rich in content and clear in context, but it still has the following deficiencies:

Three pivotal factors to assess the efficacy of airborne LiDAR on forest structure analysis is LAI, scanning patterns and algorithms. First, LAI determines the permeability of airborne laser beams related to the forest gap fraction, which results in the integrity degree of the middle (sub canopy tree) and lower layers (shrubs and fallen dead trees) of forest canopies. Second, the scanning pattern always involves aircraft flight height, side-lab of the flight lines, instrument specifications of scanning sensors with different scanning angular resolutions, footprint size, beam divergence and other technical details, which also effects the parameter characterization of the final target forest. Third, algorithms also contain CHM or DSM-based methods or point clouds-based methods, different methods also lead to the fluctuations in the final parameter retrieval. Hence, the above factor is worthy mentioned in the manuscript, a paper listed below may be afforded you some suggestions on the algorithms for forest property calculation from LiDAR data. “Individual Tree Crown Segmentation and Crown Width Extraction From a Heightmap Derived From Aerial Laser Scanning Data Using a Deep Learning Framework”. You could mention it in the Introduction section or Discussion section. Meanwhile, the above opinions could be elaborated in your manuscript for a rich content presentation and comprehensively quantified analysis.

Author Response

Reviewer #2

In this paper, 27 forest metrics were estimated from small-area full-waveform airborne laser scanning (ALS) data using random forest (RF) and automatic variable selection (Boruta) methods. The overall best model was obtained by comparing the RF models obtained from the upper, lower and combined upper/lower leaf datasets. The results of this study show that full-band, multi-temporal ALS holds a large amount of potential information that is very useful for estimating forest characteristics, both within and below the main forest canopy, and for continuous mapping over a large area. FCS maps are closely aligned with forest types and stand boundaries, suggesting that ALS data provide a viable method for mapping and monitoring forest conditions. Having high-resolution forest vegetation maps with acceptable consistency for field validation will greatly advance the understanding of forest ecology and improve conservation efforts. This article is rich in content and clear in context, but it still has the following deficiencies:

We thank you for these comments – and address your specific points below.

Three pivotal factors to assess the efficacy of airborne LiDAR on forest structure analysis is LAI, scanning patterns and algorithms. First, LAI determines the permeability of airborne laser beams related to the forest gap fraction, which results in the integrity degree of the middle (sub canopy tree) and lower layers (shrubs and fallen dead trees) of forest canopies. Second, the scanning pattern always involves aircraft flight height, side-lab of the flight lines, instrument specifications of scanning sensors with different scanning angular resolutions, footprint size, beam divergence and other technical details, which also effects the parameter characterization of the final target forest. Third, algorithms also contain CHM or DSM-based methods or point clouds-based methods, different methods also lead to the fluctuations in the final parameter retrieval. Hence, the above factor is worthy mentioned in the manuscript, a paper listed below may be afforded you some suggestions on the algorithms for forest property calculation from LiDAR data. “Individual Tree Crown Segmentation and Crown Width Extraction From a Heightmap Derived From Aerial Laser Scanning Data Using a Deep Learning Framework”. You could mention it in the Introduction section or Discussion section. Meanwhile, the above opinions could be elaborated in your manuscript for a rich content presentation and comprehensively quantified analysis.

Additional text has been added to the discussion (towards the end of section 4.4) acknowledging these potential issues relative to the transferability of approaches to other contexts.

Reviewer 3 Report

please find the attachment

Comments for author File: Comments.pdf

Author Response

Please see attached Word document

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

 

As a whole, I feel that the revisions and additions to the manuscript add value to both its readability and scientific merit. The authors did a nice job of forest condition assessment based on the airborne laser scanning technique. The revisions made to the current draft make it better suited for publication.

 

 

Reviewer 3 Report

Accept in the present form