Applicability Analysis of GF-2PMS and PLANETSCOPE Data for Ground Object Recognition in Karst Region
Round 1
Reviewer 1 Report
High temporal resolution data can provide continuous changes of surface elements in time and space and play an irreplaceable role in regional ecological environment monitoring. However, due to the limitations of satellite launch cost, technical conditions and satellite revisit cycle, the remote sensing image of a single satellite has the problem of mutual restriction between spatial resolution and temporal resolution. As a result, the accuracy of land use classification is not high, which limits the practical application of remote sensing data. Based on GF-2 and PlanetSope data, this paper compares and analyzes the applicability of FSDAF, STDFA, and Fit_FC model in different terrain conditions in karst area. The results are promising and can be applied in land use classification in karst areas. There are still several aspects the authors can improve the paper:
1) Figire 1, the authors should put the study region on a China map, so readers can understand where the study area located.
2) in line 307, the authors wrote : As can be seen from Table 2, it can be seen that, the sentence should be made more concise.
3) the subtitles in the results section, like the area of land-water boundary, the area of mountains, these two subtitles should be the land-water boundary, or the accurancy of land-water boundary, the mountain areas, or the accrancy of land classification in mountain areas. Current subtiltes can be mis-understood.
4) For consistency, the location of the area in figure 2, figure 3 and fiture 4 should be noted in Figure 1, so it can be easily understood.
5) in the discussion part, currently each section talks about different methods, it should have one section to discuss how can use these methods to acquire a full land use classification map for a region in karst areas, if differnt methods were used to acquire different objects or land use, how can these methods combined to work in a region?
Author Response
请参阅附件。
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The main idea of spatial temporal fusion presented in this manuscript is interesting. However, there are various weaknesses of this paper; therefore, the overall rating for this manuscript is 70 points out of 100.
The paper is written in grammatic and clear English.
The title is appropriate to the content. However, does the word "KARST" in the title refer to the geological structure of the region or the name of the region? This word causes confusion in the title. Why is the proper name used in the title? Does the status of this word have a relationship with fusion studies in the literature? If it is not a proper name, the importance of the geological situation should be better explained in the introduction.
The abstract describes the content of the paper. But there are some typos like "ap-proach" in the summary section.
In the introduction, some data models are discussed, and the importance of the subject is highlighted. An additional information, as a sub-title, is included for examining the former data models and it is written in detail.
The used models are discussed in qualitative and quantitative manner including image quality metrics and image quality assessment.
The fusion models were described in the manuscript, specifically the relationship between multiscale decomposition and the other tools.
The results and conclusion are written in a scientific and satisfactory manner.
I like this study. I congratulate the authors of the manuscript for their scientific work.
My decision regarding this study is a major revision. In this context, the corrections I suggest are given below.
Line 20
“the FSDAF model has the best fusion effect”
In what respect is the FSDAF model the best?
Lines 26, 27
“but the ac-tual spatial resolution of the fused image is relatively poor.”
the resolutions of the images are all the same (1m). Is it bad in terms of spatial quality?
Line 29
“STDFA model have higher image prediction accuracy,”
In which way does the STDFA model have higher image prediction accuracy?
Lines 33,34
Are high spatial and temporal fusion images or models used in this article? which of these?
Line 39
Is this rate (12%) global or regional? Which region or country is this rate for?
Line 91
Why were only two of the algorithms (FSDAF, STDFA) in the literature mentioned in the previous paragraph chosen? In the previous paragraph, reference was made to these two methods. but there is no literature reference from the Fit_FC method. Why?
Line 92
Why was the GF-2 and PlanetScope satellite data chosen?
Line 113
The display of the study area in Fig 2 should be made more understandable. for example, three different fields used in the analysis should be marked. A cartographic map can be attached to it.
Line 116
Are there other satellites within the satellite? This complex situation must be explained.
Line 118
What does spatial resolution 3-5m mean? This situation should be written more clearly.
Line 141
Why is the NIR band information repeated in table 1?
Line 143
Why were these three techniques chosen according to the literature in the introduction? Information should be given in the text.
Line 162
Spelling errors should be corrected in formula notations such as Tp and tp in the text.
Line 170
“cover type a of the high-spatial-resolution”
The typo here needs to be fixed
Line 304
Why are the original images different in Figures 2,3,4? What date are the fusion images? Which images were fusion processed? what is the difference between the two images? similar information like this should be stated in clearer text.
Line 326
tables 2,3,4 can be made smaller and more legible.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper compared and analyzed the applicability of FSDAF, STDFA, and Fit_FC in different terrain conditions in karst area. This manuscript used GF-2 data and PlanetScope data as experimental data to obtain spatiotemporal fusion results. Overall, this paper seems to be well organized, but there is lack of major innovations in the content. There are some questions the authors may answer or consider before this manuscript goes for further consideration.
1. Please elaborate on the contributions and innovations of this manuscript. The manuscript seems to simply apply the previous proposed method to the karst area.
2. Please complete the abbreviations (e.g. satellite name, method name, etc.) when they first appear in the text of the manuscript.
3. In line 125, line 126, page 3, please check the date of the experimental data. Why GF-2 data from July 13 was used as verification image instead of July 10 ?
4. In line 128, page 3, “low frequency” or “low temporal”, please unify.
5. Fig. 1, page 3, please mark the experimental area of land–water boundary, mountains, and Urban in Fig. 1.
6. There are some problems with Table 1 and Eq. (16), please check.
7. Line 279, page 7, please give the values of C1 and c2.
8. Fig .2 to Fig .4, please add the verification image and the date of fusion images.
9. Table 2 to Table 4, please increase the average of each evaluation index for the 4 bands of different methods please
10. Line 338, page 8, “similar tones” is proposed to be changed to “similar spectral”.
11. Line 351 to Line 353, page 9, these two sentences are repeated.
12. Line 386 to Line 388, page 10, from Fig. 4(c), the Fit_FC method does not achieve excellent result.
13. The discussion section is not convincing without results and tables to support it. It is suggested to discuss from the application of fusion results. The authors are recommended to apply the fusion result to land-cover classification, building extraction, and so on. The accuracy of land cover classification and building extraction results from the different fusion results are used to verify the applicable fields of different fusion methods.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The corrections I have suggested for the article titled “Applicability analysis of GF-2PMS and PLANETSCOPE data for ground object recognition in Karst region” have been sufficiently made by the authors. I believe that the manuscript in its current form is academically and scientifically sufficient for publication in the journal "LAND".
I congratulate the authors for their scientific work.
Kind regards,
Reviewer 3 Report
The authors have answered all of my concerns.
