Comparison of CORINE Land Cover Data with National Statistics and the Possibility to Record This Data on a Local Scale—Case Studies from Slovakia

Round 1

Reviewer 1 Report

This paper is like a project report more than a scientific paper.

No scientific methodology is included in this paper, it is just a comparison between the CORINE data and the NS. 

The proposed higher level of classification for CORINE data were proposed in previous work not a novel method of the authors in this paper.

Author Response

(Reviewer 1)

The authors thank the reviewer for comments that helped improve quality of the paper.

The research design was clarified and a more extensive description of the methods was added.

This paper is like a project report more than a scientific paper.

In the introduction, we specified the purpose of the scientific article, lines  131-137.

We bring information on projects CORINE Land Cover, our firstly used methodology of detailed land cover mapping  and national statistics  because of chosen recommended topic of Special issue CORINE Land Cover „Advantages and disadvantages of the CORINE data as compared to other data on land cover and use and their possible integration for the study of landscape dynamics at different scales“

No scientific methodology is included in this paper, it is just a comparison between the CORINE data and the NS. 

The used method of identification and delimitation of CLC5 classes was supplemented and explained with reference to the original work  of the co-authors of  this manuscript (Oťaheľ et al.- 31) in lines 212-251.

We have supplemented methodology also with the information on data sources for detailed research of land cover, lines 165-168 and comparison of data accuracy, lines 268-276.

We also bring new detailed classification of 163 CLC5 types from our proposed methodology into LU categories usable for precise LC/LU comparisons in local studies.

The proposed higher level of classification for CORINE data were proposed in previous work not a novel method of the authors in this paper.

Modification of the CLC method and detailing of the CLC 5  nomenclature, realized by co-authors of the paper,  was verified on several examples from Slovakia. The case studies presented in this article firstly document the possibilities of applying the method for Central European lowland, basin and mountain landscape .

The paper builds on the long-term author's team works in the field of CORINE land cover. Several authors of the paper collaborated on its methodology since the 90s. They have used their many years of experiences in the theoretical design of a modification of the methodology for detailed CLC5 research (Oťaheľ et al., 2017). In years 2018-2019 , authors collaborated to verify and refine the proposed methodological procedure and firstly present their outputs now.

Our study also contributes to a more detailed recording of the current state of the country, its changes and to the prediction of its further development at the local level.

We apologize for omission of  these facts in the first version of the manuscript.

Reviewer 2 Report

The authors compared official CLC data and national statistics (NS) of LU in a case study of Slovakia between the years 2000 and 2018 at national, county and local levels. The general methodology is appropriate, and the results were well presented and discussed. However,  the scientific problems in this research are not clear. It is undoubted that there are differences between different LULC data sets because of different LULC classification system and different data collection methods. Therefore, the results of this research are not universal.

Author Response

(Reviewer 2)

The authors thank the reviewer for comments that helped improve quality of the paper.

The authors compared official CLC data and national statistics (NS) of LU in a case study of Slovakia between the years 2000 and 2018 at national, county and local levels. The general methodology is appropriate, and the results were well presented and discussed. However,  the scientific problems in this research are not clear. It is undoubted that there are differences between different LULC data sets because of different LULC classification system and different data collection methods. Therefore, the results of this research are not universal.

The authors thank the reviewer for comments that helped improve quality of the paper.

The research design was clarified and a more extensive description of the methods was added.

In the introduction, we specified the purpose of the scientific article, lines  131-137.

The used method of identification and delimitation of CLC5 classes was supplemented and explained with reference to the original source of the authors of this article (Oťaheľ et al.- 31) in lines 212-251.

We have supplemented methodology also with the information on data sources for detailed research of land cover, lines 165-168 and comparison of data accuracy, lines 268-276.

We also bring new detailed classification of 163 CLC5 types from our proposed methodology into LU categories usable for precise LC/LU comparisons in local studies.

The authors are aware that the modification of the CLC and the detailing of the CLC5 nomenclature is specific to the territory of Slovakia and Central Europe.

Results were complemented by additional information on detailed land cover classification and its utilization.

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript entitled “Comparison of CORINE Land Cover data with national statistics and the possibility to accurate this data on a local scale – case studies from Slovakia” presents a comparison from different resolutions of CLC data between years 2000 to 2018 at different levels (national, regional and local). In general, the paper is well structured, with good English writing. However, before considering this article suitable for publication on Remote Sensing, the authors should clarify some methods points and sentences that seem wrong.

The main concern is related to topic 2.3. “Land cover interpretation at a scale fo 1:100000”.  It s not clear how the authors get the CLC5 data. Given the importance of this subsection, the authors should describe better how they proceeded to get the detailed LU. The paper cited [30] has wrong DOI, directing to another paper, which makes it difficult to visualize the steps to get CLC5 data.

About the misunderstanding sentences, please check line 73-74 (looks like a subtopic title), line 130-141 (does not has a link with the state-of-art), line 194-195 (cite Table 3 to reference LU numbers), lines 328 and 347 (avoid personal comments, like “There is no doubt” and “We do not doubt”).

Author Response

(Reviewer 3)

The authors thank the reviewer for comments that helped improve quality of the paper.

The manuscript entitled “Comparison of CORINE Land Cover data with national statistics and the possibility to accurate this data on a local scale – case studies from Slovakia” presents a comparison from different resolutions of CLC data between years 2000 to 2018 at different levels (national, regional and local). In general, the paper is well structured, with good English writing. However, before considering this article suitable for publication on Remote Sensing, the authors should clarify some methods points and sentences that seem wrong.

The authors thank the reviewer for comments that helped improve quality of the paper.

The research design was clarified and a more extensive description of the methods was added.

In the introduction, we specified the purpose of the scientific article, lines  131-137.

Some senteces were corected.

The main concern is related to topic 2.3. “Land cover interpretation at a scale fo 1:100000”.  It s not clear how the authors get the CLC5 data. Given the importance of this subsection, the authors should describe better how they proceeded to get the detailed LU.

The used method of identification and delimitation of CLC5 classes was supplemented and explained in lines 212-251.

We have supplemented methodology also with the information on data sources for detailed research of land cover, lines 165-168 and comparison of data accuracy, lines 268-276.

The paper cited [30] has wrong DOI, directing to another paper, which makes it difficult to visualize the steps to get CLC5 data.

The paper [30] was published in Slovak language and has no DOI

About the misunderstanding sentences, please check line 73-74 (looks like a subtopic title)

Comment was accepted. Lines were added to following paragraph.

, line 130-141 (does not has a link with the state-of-art),

Goals were more specified

In the introduction, we specified also the purpose of the scientific article, lines  131-137

The used method of identification and delimitation of CLC5 classes (topic 2.3) was supplemented and explained with reference to the original source of the authors of this article (Oťaheľ et al.- 31) lines 212-236.

line 194-195 (cite Table 3 to reference LU numbers),

We cited Table 3

 lines 328 and 347 (avoid personal comments, like “There is no doubt” and “We do not doubt”).

Personal commets were deleted.

Author Response File: Author Response.docx

Reviewer 4 Report

I reviewed the manuscript “Comparison of CORINE Land Cover data with national statistics and the possibility to accurate this data on a local scale – case studies from Slovakia”. This Manuscript presents an interesting research about the comparison of the development of LU changes in Slovakia and utilizing CLC methodology for  for identifying LC classes on case studies at local scales. Using more advanced geo-statistical methods could improve the manuscript, however I recommend this paper to be published.

Author Response

(Reviewer 4)

The authors thank the reviewer for comments that helped improve quality of the paper.

I reviewed the manuscript “Comparison of CORINE Land Cover data with national statistics and the possibility to accurate this data on a local scale – case studies from Slovakia”. This Manuscript presents an interesting research about the comparison of the development of LU changes in Slovakia and utilizing CLC methodology for  for identifying LC classes on case studies at local scales. Using more advanced geo-statistical methods could improve the manuscript, however I recommend this paper to be published.

The authors thank the reviewer for comments that helped improve quality of the paper.

The research design was clarified and a more extensive description of the methods was added.

In the introduction, we specified the purpose of the scientific article, lines  131-137

The used method of identification and delimitation of CLC5 classes was supplemented and explained in lines 212-251.

We have supplemented methodology also with the information on data sources for detailed research of land cover, lines 165-168 and comparison of data accuracy in GIS, lines 268-276.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors compared the CORINE CLC data with the National  Land Use Data for several years (2000, 2006, 2012, and 2018) at different levels without evaluating the CORINE CLC3 data, hence no other country can depend on their conclusions in other research work. 

It is clear that there is a lot of work has been done for this project, however the presentation of the scientific methodology is not clear enough.

The proposed higher level of classification for CORINE data were proposed in previous work, and applied in this research work. However the results accuracy were not evaluated. it is obvious that there are differences between the different hierarchical levels, but which one is more accurate or similar to the real situation. 

Author Response

Review Report Form 1

Author thanks the reviewer for suggestions for improving the article.

Comments and Suggestions for Authors

The authors compared the CORINE CLC data with the National  Land Use Data for several years (2000, 2006, 2012, and 2018) at different levels without evaluating the CORINE CLC3 data, hence no other country can depend on their conclusions in other research work. 

We have added a comment to development of LC changes between 2000 and 2018 in Slovakia and comparison of CLC3 and NS data.  Lines 326 – 352.

Other comparisons are in the sub-chapters 3.2 for county and 3.3 for municipality level

It is clear that there is a lot of work has been done for this project, however the presentation of the scientific methodology is not clear enough.

The manuscript was revised again.The methodology was supplemented and refined. – Lines 158 – 160, 258 – 282, 298 – 307. Subtitles have also been inserted for better orientation in this chapter.

The proposed higher level of classification for CORINE data were proposed in previous work, and applied in this research work. However the results accuracy were not evaluated. it is obvious that there are differences between the different hierarchical levels, but which one is more accurate or similar to the real situation. 

 The main purpose of the article was to point out that not all NS data report the real state of the LU. CLC3 data register the real state, but their shortcoming is that they are interpreted on a regional scale. We tried to solve this shortcoming by modifying the CLC5 data. We agree with the reviewer that examples from Slovakia may not be universal, but may point to similar problems in other countries.

The higher level of classification (CLC5) was proposed in previous work of co-authors (Oťaheľ et al., 2017) but our paper firstly describe its usability for aerial image interpretation and its results were validated during field research.

The use of CLC5 methodology delimits LC classes reflecting detailed LU at local level and uses available aerial images in the visible spectrum. The methodology can record individual adjacent, often contrasting classes of CLC not only in urbanized (e.g. discontinuous built-up areas, accompanying vegetation) but also in fragmented rural (detailed division of heterogeneous agricultural areas) and semi-natural landscape (transitional shrubs and young forests). Thus, it is possible to capture landscape variability without the need to use heterogeneous classes. This is important when comparing actual LU evidence between NS and CLC5.

English sentences in the paper were corrected.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors have revised the manuscript a little. The general methodology is appropriate, and the results were well presented and discussed. However, this research is short of innovativeness.

Author Response

Review Report Form 2

Author thanks the reviewer for suggestions for improving the article.

Comments and Suggestions for Authors

The authors have revised the manuscript a little. The general methodology is appropriate, and the results were well presented and discussed. However, this research is short of innovativeness.

The whole manuscript was revised again.The methodology was supplemented and refined again. – Lines 158 – 160, 258 – 282, 298 – 307. Subtitles have also been inserted for better orientation in this chapter.

The main purpose of the article was to point out that not all NS data report the real state of the LU. CLC3 data register the real state, but their shortcoming is that they are interpreted on a regional scale. We tried to solve this shortcoming by using our CLC5 methodology and data usable on a local scale.

The higher level of classification (CLC5) was proposed in previous work of co-authors (Oťaheľ et al., 2017) but our paper firstly describe its usability for aerial image interpretation and its results were validated during field research

The use of CLC5 methodology delimits LC classes reflecting detailed LU at local level and uses available aerial images in the visible spectrum. The methodology can record individual adjacent, often contrasting classes of CLC not only in urbanized (e.g. discontinuous built-up areas, accompanying vegetation) but also in fragmented rural (detailed division of heterogeneous agricultural areas) and semi-natural landscape (transitional shrubs and young forests). Thus, it is possible to capture landscape variability without the need to use heterogeneous classes. This is important when comparing actual LU evidence between NS and CLC5.

Author Response File: Author Response.docx

Reviewer 3 Report

Please, recheck reference 31. The crossref is not correct.

Please, again, avoid terms like "This was also for authors of paper the motive and intention of the modification". A scientific paper must sound universal, finding solutions that can be applied to other researchers.

Author Response

Review Report Form 3

Author thanks the reviewer for suggestions for improving the article.

Comments and Suggestions for Authors

Please, recheck reference 31. The crossref is not correct.

It was a mistake. CrossRef has been deleted. The journal Geografický časopis had not been assigned a doi for articles that year

Please, again, avoid terms like "This was also for authors of paper the motive and intention of the modification". A scientific paper must sound universal, finding solutions that can be applied to other researchers.

Personal comment was deleted. – Line 135

The methodology was supplemented and refined. – Lines 158 – 160, 258 – 282, 298 – 307. Subtitles have also been inserted for better orientation in this chapter.

The use of CLC5 methodology delimits LC classes reflecting detailed LU at local level and uses available aerial images in the visible spectrum. The methodology can record individual adjacent, often contrasting classes of CLC not only in urbanized (e.g. discontinuous built-up areas, accompanying vegetation) but also in fragmented rural (detailed division of heterogeneous agricultural areas) and semi-natural landscape (transitional shrubs and young forests). Thus, it is possible to capture landscape variability without the need to use heterogeneous classes. This is important when comparing actual LU evidence between NS and CLC5.

Authors thank the reviewer for suggestions for improving the article.

English sentences in the paper were corrected with minor changes.

Author Response File: Author Response.docx