Round 1

Reviewer 1 Report

Dear author,

This paper adresses the needs of tracking land changes related to anthropic activities and ecological processes. A new index is proposed, that aims to be replicable to various geographical contexts. The research is well introduced and supported by litterature. 

Here are some comments to improve the publication :

- I'm missing an overview figure that present the methodology, the three indices, and the finalities. If this could be added, it would ease the understanding of the research.

- L56: As any link been made between the "wilderness mapping" and the "Intact forest Landscapes" initiative?

- Figure 1 quality is low. Maybe a real exemple could be shown (from the GIS data to the index result?). This would also help (by seeing the data integrated) identifying how easy would the replicability be on the reader area (do I also have similar data to reproduce the method?). Here only table 1 lists data without precising their format, quality, source... this part is missing and is my only one major comment.

- Figure 2: please enlarge the legend size (valable for all figures)

- Figure 4: add a third image with a satellite view to ease the index quality with regards to ground truth?

- The current study only use GIS data, that are often out to date in some ways. Would it be interesting to discuss the interest of integrating remote sensing to validate the output and discuss the quality of the input. If your indice reports that an area is wild but that the ground truth data show a strong anthorpic activity, this can quickly slow down the use of the proposed method. L439 justify the choice of not including remote sensing but it still could be useful for at least validation.

Apart from that, the research is well presented and paper structure is clear.

 

 

 

Author Response

Reviewer 1:

This paper adresses the needs of tracking land changes related to anthropic activities and ecological processes. A new index is proposed, that aims to be replicable to various geographical contexts. The research is well introduced and supported by litterature. 

Thank you for this assessment!

Here are some comments to improve the publication :

- I'm missing an overview figure that present the methodology, the three indices, and the finalities. If this could be added, it would ease the understanding of the research.  

Figure intended as graphical abstract is included as Fig2 (Caption: Figure 2. An overview of the procedure of calculating LUI with its components.)

- L56: As any link been made between the "wilderness mapping" and the "Intact forest Landscapes" initiative?

We have compared with INON mapping which is a Norwegian wilderness mapping. This has been clarified in the text. New text: The comparison between LUI and the official index of interference-free areas, i.e. the Norwegian wilderness map (INON; Norwegian Environment Agency, 2019) in chapter 3.4 shows considerable complementarity between the two indices.  

We have also clarified the situation versus Intact Forest Landscape mapping in the text. New text: We have not included remote sensing as an opportunity to put additional data into the LUI index. Remote sensing advances in continuous mapping of built-up areas, such as Normalized Difference Built-up Index (NDBI) (Zha et al. 2003) and ecological intactness such as the Intact Forest Landscapes (Potapov et al. 2008, Hanssen et al. 2013), can give independent and gradient-like inputs.

We have not done a comparison with the Intact Forest Landscape product. The data set “forest 2020” from this source has a very poor coverage for Norway. It is only one polygon in Southern Norway that consist of mountain (no forest), bogs and forest at the border between Sweden. We have therefore not made any attempt for comparison.

  IFL

The forest change map 2001-2020 yields, however, interesting results, but must be interpreted and validated in a local setting before any comparison can be done. We have assessed this outside our scope.

- Figure 1 quality is low. Maybe a real exemple could be shown (from the GIS data to the index result?). This would also help (by seeing the data integrated) identifying how easy would the replicability be on the reader area (do I also have similar data to reproduce the method?).

Figure has been redrawn according to the idea of the reviewer, Thanks for this idea, it was a good one!

Here only table 1 lists data without precising their format, quality, source... this part is missing and is my only one major comment. 

All data are from official Norwegian topographic map infrastructure and are updated fairly regularly and are of high reasonably uniform quality. The reference to this has been clarified in the table heading. New table heading: Table 1. Landscape elements included in each component of the human land-use (LUI) index. The data are included in the official topographic map of Norway and data downloadable in different GIS format from https://www.geonorge.no/ as part of the Norwegian official map strategy (Norwegian Ministry of Local Government and Regional Development 2018). Data structure and quality are defined in the Norwegian SOSI standard (Norwegian Mapping Authority, 2017). The dataset “regulated lakes” is obtained from The Norwegian Water Resources and Energy Directorate:  https://gis3.nve.no/metadata/tema/Magasin.html.

- Figure 2: please enlarge the legend size (valable for all figures) 

Done!

- Figure 4: add a third image with a satellite view to ease the index quality with regards to ground truth? 

Done!

- The current study only use GIS data, that are often out to date in some ways. Would it be interesting to discuss the interest of integrating remote sensing to validate the output and discuss the quality of the input. If your indice reports that an area is wild but that the ground truth data show a strong anthorpic activity, this can quickly slow down the use of the proposed method. L439 justify the choice of not including remote sensing but it still could be useful for at least validation.  

The text has been specified. But ground truth data have high sensibility to scale. Small areas within build areas may show elements of wilderness or at least have high ecological value. This has been described. The index show that such areas has build up areas in their surroundings and the detailed scale makes it possible to assess this effect. We have added two sentences to the text that now reads:  Remote sensing advances in continuous mapping of built-up areas, such as Normalized Difference Built-up Index (NDBI) (Zha et al. 2003) and ecological intactness such as the Intact Forest Landscapes (Potapov et al. 2008, Hanssen et al. 2013), can give independent and gradient-like inputs. However, this will also add uncertainty to the index as all remote sensed products need quality checks, as well as interpretation or classification. A robust index based on official data will have other qualities than a remote sensed product and can be considered a lot more trustworthy as a source of information for instance in an environmental management setting. In countries that lack reliable map databases, the use of remote sensing will be of vital importance to improve or validate data input.

New reference has been added. We regard a further discussion of input of remotely sensed data in such areas, as well as the question if it should be an aim to have one index that covers everything, or if it is a better strategy to construct several indices with different foci as a tool to assess ecological status and quality, as a question for further research and that such a discussion here would be lengthy and out of the scope for the article.

Apart from that, the research is well presented and paper structure is clear.

Author Response File: Author Response.docx

Reviewer 2 Report

There is an urgent need for robust and reproducible methods for quantifying changes in landscape patterns. This is a well-written manuscript to present, exemplify and discuss a gradient-based index of land use intensity. Significantly, this work explores different use of the index and compare the results with other indices used nationally in Norway and internationally. The indices and procedures used to calculate the index are well designed. I would like to recommend it for publication after some minor revisions.

 

The Abstract could be more orderly and focused on the major findings of the research (i.e., the results of comparison between the LUI and other indices) and their significance. 

 

As for the discussion part on the attractive feature of the LUI, it should be noted that its requirements for quality controlled and regularly updated data in official databases also presents a greater challenge to the regions and countries without sufficient data, compared to other indices based on more general data and coarser scale.

 

Finally, it’s not so normal to have several references in the conclusion part.

Author Response

Reviewer 2:

There is an urgent need for robust and reproducible methods for quantifying changes in landscape patterns. This is a well-written manuscript to present, exemplify and discuss a gradient-based index of land use intensity. Significantly, this work explores different use of the index and compare the results with other indices used nationally in Norway and internationally. The indices and procedures used to calculate the index are well designed. I would like to recommend it for publication after some minor revisions.

Thank you for this assessment

The Abstract could be more orderly and focused on the major findings of the research (i.e., the results of comparison between the LUI and other indices) and their significance. 

Abstract has been adjusted to mitigate this. The mid part of the abstract with corrections show like this:

 We hope this is an improvement.

As for the discussion part on the attractive feature of the LUI, it should be noted that its requirements for quality controlled and regularly updated data in official databases also presents a greater challenge to the regions and countries without sufficient data, compared to other indices based on more general data and coarser scale. 

We have added one sentence:  In countries that lack reliable map databases, the use of remote sensing will be of vital importance to improve or validate data input.

Finally, it’s not so normal to have several references in the conclusion part. 

References has been removed from the conclusion. Some text has been added in the discussion to justify these references.

Author Response File: Author Response.docx