Data Fusion and Accuracy Analysis of Multi-Source Land Use/Land Cover Datasets along Coastal Areas of the Maritime Silk Road

Round 1

Reviewer 1 Report

Line 101: what is the source of the study area-Maritime Silk Road? When googling it, there are different maps of Maritime Silk Road.

Line 142: Why did you resample to 300 meters? Not in 30 or 500 meters.

Table 2: What is the number?

Figure 2: prove in higher quality

Figure 2: how did you adjust the different land classes in three data sources? 10 types and 22 types,...

Did you reclassify? What are the sources?

Please simplify the flowchart.

Line 175 to line 185: it is not clear. Please explain clearly, if you did reclassify the three data sources and then agreement analysis or vice versa?

In the method, please add the complete information about the ground-based data obtained visually by google earth.

Line 242: this should be results

Line 261 to line 273: I think some parts should move to method

Line 265: why 5896 sampling points? Sampling points is a good word?

Line 268: it is not clear how did you give a label to 5896 data? It is based on visual analysis?  If your 8 land use categories clearly distinguishable by visual analysis?   

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

For this article "Data fusion and accuracy analysis of multi-source 2 land use/land cover datasets along coastal areas of the 3 Maritime Silk Road" by Wan Hou and Xiyong Hou, the authors proceed to a high-precision land use/land cover classification mapping derived from remote sensing, based on the coastal areas of the Maritime Silk Road.

It is a very well constructed and scientifically sound article. This research is innovative. The introduction is very well conducted. The objectives are clear. The text is generally well written. Table 1 is very useful to follow your comments. The method is well detailed and summarized in Figure 2, with some minor modifications that can be made to improve the quality of the paper. In particular:

- Insist in abstract and introduction on the choice of maritime coasts in particular
- Line 46-47, you can add the following more recent reference: Stewart, I. D. and Oke, T. R., Local Climate Zones for Urban Temperature Studies, Bulletin of the American Meteorological Society, volume 93, 12, pp.1879-1900, 2012.
- Line 182, you can recall what code 9 means.
- For the different statistical methods used (§3.5), you can add the corresponding references
- To help visualize your research work, you should zoom in on a part of the study area in Figure 3, so that details can be seen.
- Title 4.3, modify the "cross-validation" because this title is misleading. You have not carried out a cross-validation as such in statistics.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper presents,using a fusion method based on agreement analysis and fuzzy-set theory,the fusion of multi-source land use/land cover classification data ,in order to analyze the accuracy of the fusion results. The empirical study was performed along costal areas of the Maritime Silk Road.

Recommendations :

1.The authors demonstrate as data-set features values have been increased, the overall accuracy results have also been observed better and, this shows that multi-source data integration significantly improves the analysis and classification of land use/land cover types.The authors argue that the employed classification framework improves classification accuracy of the fusion data,which means it can further generate better results for land use/land cover classification.If true,more discussions is needed on how this multi source data fusion used in land use/land cover can be used for decision-making, future prediction, and quick and accurate analysis of land use and land cover, when employing sophisticated rules on multi-source data-sets.

2.Limitation of the study is necessary,while using other performance evaluating factors,different types of data-set etc, the analysis and classification of land use/land cover types may conduct to different results.

3.In addition,the author should outline their contribution and show the novelty of their research in the context of previous works and results obtained. This is important because thus authors could demonstrate what added value bring their research to the area of knowledge.

4.The source of data and software use should be attached to the figures and tables.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

accept