Spatio-Temporal Knowledge Graph-Based Research on Agro-Meteorological Disaster Monitoring
Round 1
Reviewer 1 Report
The authors have done a great job and presented interesting and relevant material in the article. This paper presents a groundbreaking approach to tackling the challenges of agro-meteorological disaster monitoring. The proposed method, centered around a spatio-temporal knowledge graph and semantic ontology framework, offers a promising solution. The article is generally well-prepared and structured, but some revisions should be made for clearer presentation of the material:
1. I would not recommend repeating the article title in the keywords. It would be beneficial to modify them to encompass all aspects of the work.
2. There's a lack of references in the text describing the problem in the introduction.
3. The data source information should be placed before the results section. For example, in section 3.2, the provided text isn't related to results and should be relocated to the appropriate section.
4. The study area description should also be moved higher, preceding the methodology explanation.
5. Figure 10 contains excessive irrelevant textual information. The link in the figure doesn't offer any significant information. I recommend using a concise description and providing an explanation below the figure. Additionally, it's advisable to review the images in this section and consider the possibility of shortening report titles to prevent overloading the text scheme.
6. If possible, include comparisons with the results of other similar studies in the discussion.
Thanks to the authors for the interesting work.
Author Response
Thank you very much for your review. Your suggestions strongly help us to improve our work. We have corrected the problems that you pointed out in the manuscript, and we have carefully optimized the article organization. We describe the revisions point by point for your comments in this response document.
Author Response File: Author Response.docx
Reviewer 2 Report
Overview:
This paper gives a technical framework how Spatio-Temporal Knowledge Graph can be used to give farmers in China early warning and recommended actions when Agro-Meteorological Disasters are forecasted to take place. The problem is sufficient to warrant study and the methodology is sound. I recommend it be accepted with minor revisions.
1. Introduction Comments:
Please define RVI and ARVI here: Li et al. [6] found that NDVI and RVI are more sensitive than ARVI in detecting the severity of hot-dry wind disasters, making them suitable for large-scale monitoring of such disasters
2. Spatio-Temporal Knowledge Graph Construction Comments:
2.2 Define how abnormal meteorological data is identified
3. Experiment and Results Comments
Delete the following text. “This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.”
4. Discussion Comments
You do not mention the current lagtime between receiving the metrological data and informing the farmers. An estimate of the current technology vs and ideal situation/ amount of warning farmers need to implement recommendations is required.
Figure Comments:
Define GeoSPARQL and SWRL in the caption of Figure 1.
Define ogc in Figure 4.
Table Comments
Tables 5 & 6: I do not understand what the numbers mean for the column headings below. Why are there 2 categories listed?
1 mild, 1 moderate | 3 moderate, 4 moderate | 1 moderate, 1 severe | 2 moderate, 1 severe | 3 moderate, 1 severe | 1 moderate, 2 severe
Tables 7 & 8: I do not understand what the numbers mean for the column headings. Why are there 2 categories listed?
Author Response
Thank you very much for your review. Your suggestions strongly help us to improve our work. We have corrected the problems that you pointed out in the manuscript, and we have carefully optimized the article organization. We describe the revisions point by point for your comments in this response document.
Author Response File: Author Response.docx
Reviewer 3 Report
Dear Authors,
The spatio-temporal knowledge inference framework is essential in the digital era as we gather multi-scale datasets from various sources. Overall, the manuscript covered all the essential parameters and was well explained. A few minor queries need to be addressed.
1. Figure 9 may need minor modification. As I understand, NDVI values are used before disaster for monitoring and early warning systems as an input. However, Fig 9 illustrates remote sensing data in the after-disaster section. Pl rearrange the flow of information.
2. Now we have a lot of high-resolution images available at free of cost. Since the scale of operation is individual farms, why Coarse-resolution MODIS is considered in this framework?
3. The study uses the knowledge graph approach to obtain information from multiple reliable sources. However, the decision is majorly controlled by weather inputs (Fig 8 and 9). What is the sensitivity of other inputs in this framework?
Author Response
Thank you very much for your review. Your suggestions strongly help us to improve our work. We have corrected the problems that you pointed out in the manuscript, and we have carefully optimized the article organization. We describe the revisions point by point for your comments in this response document.
Author Response File: Author Response.docx
Reviewer 4 Report
What was the reason for using MODIS and not Sentinel-2? Or even Planet images or even use HSL images
The disaster to be evaluated must cover an area greater than 1 km2 to be detected by satellite images.
Because only NDVI could have used other vegetation indices that make better use of the wavelengths caused by a disaster.
Author Response
Thank you very much for your review. Your suggestions strongly help us to improve our work. We have corrected the problems that you pointed out in the manuscript, and we have carefully optimized the article organization. We describe the revisions point by point for your comments in this response document.
Author Response File: Author Response.docx