Next Article in Journal
An Assessment of the Influences of Clouds on the Solar Photovoltaic Potential over China
Next Article in Special Issue
Quick Report on the ML = 3.3 on 1 January 2023 Guidonia (Rome, Italy) Earthquake: Evidence of a Seismic Acceleration
Previous Article in Journal
Hyperspectral Image Classification Based on a 3D Octave Convolution and 3D Multiscale Spatial Attention Network
Previous Article in Special Issue
Recent Advances and Challenges in the Seismo-Electromagnetic Study: A Brief Review
 
 
Article
Peer-Review Record

Pre-Seismic Anomaly Detection from Multichannel Infrared Images of FY-4A Satellite

Remote Sens. 2023, 15(1), 259; https://doi.org/10.3390/rs15010259
by Yingbo Yue 1,2, Fuchun Chen 1,* and Guilin Chen 1
Reviewer 1:
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Remote Sens. 2023, 15(1), 259; https://doi.org/10.3390/rs15010259
Submission received: 10 November 2022 / Revised: 26 December 2022 / Accepted: 28 December 2022 / Published: 1 January 2023

Round 1

Reviewer 1 Report

 

In this manuscript (ms), the authors present a preseismic anomalies detection algorithm by using multichannel infrared (IR) images from the FY-4A satellite. They use nighttime data of the seventh to the fourteenth channel of the advanced geostationary radiation imager (AGRI) that operates onboard of FY-4A satellite which cover the wavelength range from 3.5-4.0 μm to 13.2-13.8 μm.  They employ long short-term memory (LSTM) to predict the multichannel infrared data together with the density-based spatial clustering of application with noise (DBSCAN) in order to identify the noise which is later (Section 2.3) used to define anomalies on the basis of Eqs.(17) to (21). The results which are interesting are focused in two strong earthquakes namely the M6.4 Yangbi (or Yunnan [46]) and the M7.4 Madoi (or Qinghai [46]) earthquakes that occurred on 21 May 2021 (UTC), see, e.g., https://earthquake.usgs.gov/earthquakes/eventpage/us7000e54r/executive

and  https://earthquake.usgs.gov/earthquakes/eventpage/us7000e532/executive

The authors conclude a correlation rate of 64.29% and a hit rate of 68.75% with the prediction area radius 600 km and the prediction time window 14 days.

                The presented approach is interesting and possibly useful for the identification of thermal anomalies. The presentation, however, needs improvement in various points, which are mentioned below, in order to convince the reader about the merits of this new method to analyze IR images for the detection of preseismic anomalies. Hence, I suggest that the author resubmit their manuscript after considering the following points:

1)The references used in the Introduction are not appropriate for giving to the reader the wide aspect of the research on the subject of earthquake precursors. For example, the following 4 references should be added in the list of references mentioned in line 6:

Seiya Uyeda, Toshiyasu Nagao, Masashi Kamogawa, Short-term earthquake prediction: Current status of seismo-electromagnetics, Tectonophysics, 470, Issues 3–4, 2009, Pages 205-213, https://doi.org/10.1016/j.tecto.2008.07.019

Huang Q. “Seismicity Pattern Changes Prior to the 2008 Ms7.3 Yutian Earthquake.” Entropy. 2019; 21(2):118. https://doi.org/10.3390/e21020118

Varotsos, P.A.; Sarlis, N.V.; Skordas, E.S. Self-organized criticality and earthquake predictability: A long-standing question in the light of natural time analysis. EPL (Europhysics Letters) 2020, 132, 29001. https://doi.org/10.1209/0295-5075/132/29001

Moreover, the connection of the present work with those in IR preseismic anomalies by other authors like

Jiao, Z.; Shan, X. Pre-Seismic Temporal Integrated Anomalies from Multiparametric Remote Sensing Data. Remote Sens. 2022, 14, 2343. https://doi.org/10.3390/rs14102343

Filizzola, C.; Corrado, A.; Genzano, N.; Lisi, M.; Pergola, N.; Colonna, R.; Tramutoli, V. RST Analysis of Anomalous TIR Sequences in Relation with Earthquakes Occurred in Turkey in the Period 2004–2015. Remote Sens. 2022, 14, 381. https://doi.org/10.3390/rs14020381

should be further discussed in the Discussion and Conclusion Section (Sec. 4).  

2) It is very important for the integrity of the present study that the authors explain in both the abstract (lines 14-15) and in the second paragraph of Section 2.1 (lines 72 to 83) which is the magnitude threshold they use for the earthquakes to be predicted (target earthquakes). They should also provide a list of the earthquakes used (line 73) probably in the form of an electronic supplement. This is essential because a related search in the United States Geological Survey https://earthquake.usgs.gov/earthquakes/map/?currentFeatureId=us7000e54r&extent=-33.57801,22.67578&extent=70.72898,215.85938&range=search&sort=largest&baseLayer=terrain&timeZone=utc&search=%7B%22name%22:%22Search%20Results%22,%22params%22:%7B%22starttime%22:%222021-01-01%2000:00:00%22,%22endtime%22:%222022-01-01%2023:59:59%22,%22maxlatitude%22:60,%22minlatitude%22:0,%22maxlongitude%22:140,%22minlongitude%22:70,%22minmagnitude%22:5,%22orderby%22:%22time%22%7D%7D

results in 287 earthquakes with magnitude greater than or equal to 5.0 in the concerned region from 70 to 140 degrees East  and  0 to 60 degrees North during 2021.

3) As concerns the Figures the following comments are in order: In order to improve the clarity of the presentation units are needed in Figs. 3 and 4. In Figs. 5 and 6, topographic information concerning the height above sea level should be added so that the reader can evaluate the statement the authors make in lines 153 to 156. What is the purpose to the straight lines present in Fig. 12? The authors should explain, or they may use the most recent method of Receiver Operating Characteristics, see, e.g., 

Sarlis, N.V.; Skordas, E.S.; Christopoulos, S.-R.G.; Varotsos, P.A. Natural Time Analysis: The Area under the Receiver Operating Characteristic Curve of the Order Parameter Fluctuations Minima Preceding Major Earthquakes. Entropy 2020, 22, 583. https://doi.org/10.3390/e22050583

and references therein. In Figs. 14 and 15, the captions should describe the figures and the axis should be visible.  “This is a figure. Schemes follow the same formatting” is incomprehensible.

4)The authors should improve the readability of the ms. This can be done by: (i)Setting the References just after mentioning the authors, i.e., “Akhoondzadeh et al. [27] compared” in line 40, instead of mentioning the Reference at the end of the sentence, see line 41. Other similar case are “Zhai et al. [32] used” (l.44), “Jing et al. [33] found” (ll.45-46), etc. (ii)Eliminating the typing mistakes like that in line 90 throughout the ms. (iii)Avoiding using past tense in Sections 3.1, 3.2 and 3.3. Simple present is ok, e.g., “anomaly images are” at line 194, “Figures 10 and 11 show” at line 202, “The anomaly time and the distance between the anomaly and the epicenter are shown” etc.

 

In summary, this ms is potentially a very useful one in the determination of thermal IR anomalies before strong earthquakes. The presentation, however, needs improvement along the line mentioned above before I will be able to suggest its publication in remote sensing.

 

 

 

 

Author Response

Thank you for your review and suggestions. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

In the present manuscript (ms), the authors used a multichannel infrared (IR) remote images from the Fengyun-4 satellite for the detection of pre-earthquake anomalies. Their algorithm combines the long short-term memory with the density-based spatial clustering of application with noise in order to identify anomalies related to two strong earthquakes, i.e., the Madoi and Yangbi earthquakes in May 2021.

The ms falls well in the scope of the journal of Remote Sensing and it is interesting. Before its acceptance, however, the authors should consider the following points.

In lines 52 and 53 the reader comes along with the abbreviations LSTM and DBSAN. In this part of the ms, where abbreviations are used for the first time, the authors should explain what is the meaning of these abbreviations, e.g., long short-term memory (LSTM), and then use the abbreviation in the rest of the ms.

In lines 73-74 the authors say “There are 106 earthquakes in China and its surrounding area (from 70 oE to 140 oE, from 0 oN to 60 oN) in 2021.”. The authors should mention the magnitude threshold they used in order to get these 106 earthquakes from the catalog.

In lines 78-80 the sentence should be rephrased. Aftershocks comes usually in sequencies.

In line 82 I read “… hours before the Madoi earthquake”. How many hours before?

In line 90 reads “[Error! Reference source not found.]”. Please, provide the reference.

Units should be added for the y-axis in figures 3 and 4.

In line 121 “Euclide” should be replaced by “Euclidean”

In line 123 the authors should say that M_0 is an integer.

In line 126, replace “is any corn” to “is no core”

In line 147, replace “four-dimension” with “four-dimensional”

In line 148, replace “was shown in Figure 3” with “are shown in Figure 3”

In line 154, replace “was related” with “are related”

In line 168, replace “were shown” with “are shown”

In line 194, replace “were shown” with “are shown”

At the end of line 205 there is a symbol “R” which should be removed.

In line 218 which is the magnitude of the earthquakes that are excluded?

In lines 232 and 233 “were” should be replaced by “are”.

In my opinion, this ms is interesting but its presentation needs improvement along the points mentioned above before its acceptance for publication.

Author Response

Thank you for your review and suggestions. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript focuses on the novel procedure to detect pre-seismic anomalies from multi-channel infrared images of FY-4A satellite.

Results are important for contributing in understanding the delicate approach to detect anomalies from satellite observations of pending earthquakes. The ms is written with clear language, with a robust mathematic treatment but some remarks should be addressed before publication.

In the following, please find a detailed list of comments/suggestions.

Line 45: [32] schould be moved at the end of the previous sentence, --> ...two earthquakes [32]. Jing et al...

Line 52 and 53: acronyms LSTM and DBSAN should be written extensively the first time in which they are used

Line 75: are 106 earthquakes reporter in fig.1 or only those with magnitude >5?

Line 80: replace the word range with area

Line 82: is this the Yangbi earhquake? 

Line 83: what the red box indicates in the figure? please add a specification in the caption 

Line 90: reference not found

Eq. 14 and 15: put a full stop at the end.

Line 121: ...distance D between...

Line 132: ...statistically. The pre-seismic ... --> statistically. We assume that the ...

Line 142: ...NA is the number 

Line 150: replace through with because of 

Line 152: ...around zero. As exsample, the image of...[HAS THE CHOSEN DATE A SPECIAL MEANING? PLEASE SPECIFY]

Line 154: how the reader can take into consideration the topography? The colour bar in Fig. 5 is relative to the temperature, so what about the elevation?

Line 161: vertical axes in plots of figure 3 do not show units [°K]. Add units in the plots or specify units in the caption

Line 163: what indicates red coloured curve for 2021? Please, specify in the caption together with units

Line 179: parameters are given  or chosen? Authors should motivate this aspect

Line 186 to 195: for the reader's convenience I suggest to add a table which lists the whole sequence of earhquakes with associated parameters [date  epicenters - hypocenters - magnitudes...]

Line 194: the anomaly image... [recall here what the authors consider anomaly and expand such concept in a way that even from the figure can be clearly understood and recognized]

Line 201: in the caption, please explicit what red stars are, together with the meaning of denser clouds of points-lines 

Line 204: ...the anomaly duration (indicated in days). The coordinate... [ THIS SHOULD BE INCLUDED IN THE COLOUR BARS OF FIG. 10 AND 11]

Line 205: ...region [R, special character ?]

Line 206: ...to Equation (24) [45] ->> to the so-called Dobrovolsky area, defined as [45]...

Line 207: Where M is ...--> where M is

Fig.10 and 11: put legends in boxes and place them in white space of the plots avoiding interference with other parts of the plots

Line 223: the Molchan diagram --> please, add reference supporting the kind of diagram used

Line 236: The considered earthquakes with...

Caption of figure 13: please, add more informative descriptions for the reader. What each cell represent, how the time distance from the earhquake work, going from right to left or the contrary or both?

Caption of figure 14: as for the caption of fig 13, the authors should add few informative tools for letting the reader getting easy interpretation.

Line 257 to 262: expand the discussion about the positive and adverse aspects which come from the application of the methods proposed by the authors

 

 

 

Author Response

Thank you for your review and suggestions. Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

Suggestions and corrections have been all addressed by the authors and I have no other issues to highlight.

Back to TopTop