A Geo-Hazard Risk Assessment Technique for Analyzing Impacts of Surface Subsidence within Onyeama Mine, South East Nigeria

Round 1

Reviewer 1 Report

The topic of the research is interesting, but improvements are needed in some parts of this paper. Specific comments are: 

The Literature Review part could be improved by adding other cases where the topic has already been treated and applied in an international context.

For ease of reading indicate the research sites within the study area in Chapter 2.

In general, the figures have poor reading. See for example Figure 7. The quality of the figures needs to be improved.

The results chapter must be rewritten. It's confusing: indicates methodology. For example: “The results of horizontal and vertical subsidence was obtained for every InSAR com-bination after application of (Block F of Figure 4). Next, we proceeded to masking out area of low coherence and finally geo-coding the results to have absolute geographical coordi-nate”.

In figures 9 and 11 it is referred “Result of yearly absolute vertical subsidence (mm) ….. using Holts-Winters forecasting model.” The bibliographic reference is missing. Holt's is referred to in the figure and Holts is mentioned in the figure legend. Which one is correct? I think it's Holt-Winters.

Discussion needs to be revised. The discussion chapter must be supported by the results. Discussion of the results should be expanded substantially.

The conclusions needs improvement. It should objectively indicate the contribution of the work to the field and the progress of the research.

Author Response

RESPONSE TO FIRST REVIEWER COMMENTS

Point 1: The Literature Review part could be improved by adding other cases where the topic has already been treated and applied in an international context.

Response 1: The observation is acknowledged and the literature review has been updated accordingly in the revised manuscript.

Point 2: For ease of reading, indicate the research sites within the study area in Chapter 2.

Response 2: The research site is the whole study area indicated in figure 1.

Point 3: In general, the figures have poor reading. See for example Figure 7. The quality of the figures needs to be improved.

Response 3: The observation is acknowledged and the figures have been improved for proper reading.

Point 4: The results chapter must be rewritten. It's confusing: indicates methodology. For example: “The results of horizontal and vertical subsidence was obtained for every InSAR com-bination after application of (Block F of Figure 4). Next, we proceeded to masking out area of low coherence and finally geo-coding the results to have absolute geographical coordinate”

Response 4: The observation has been noted and corrected accordingly in the revised manuscript.

Point 5: In figures 9 and 11 it is referred “Result of yearly absolute vertical subsidence (mm) ….. using Holts-Winters forecasting model.” The bibliographic reference is missing. Holt's is referred to in the figure and Holts is mentioned in the figure legend. Which one is correct? I think it's Holt-Winters.

Response 5: The observation has been noted and corrected accordingly in the revised manuscript. Holt-Winters is correct. Please see the revised manuscript.

Point 6: Discussion needs to be revised. The discussion chapter must be supported by the results. Discussion of the results should be expanded substantially.

Response 6: The observation was noted. The discussion of results has been revised and expanded to support the results. Please see the revised manuscript.

Point 7: The conclusions needs improvement. It should objectively indicate the contribution of the work to the field and the progress of the research.

Response 7: The observation has been noted and corrected accordingly. The conclusion has been revised to indicate the contribution of the work to the field and the progress of the research.

Author Response File: Author Response.docx

Reviewer 2 Report

REVISION MANUSCRIPT land-2228520: “A GEOHAZARD RISK ASSESSMENT TECHNIQUE FOR ANAYLYZING IMPACTS OF SURFACE SUBSIDENCE WITHIN ONYEAMA MINE, SOUTH EAST NIGERIA”

General comments:

The authors proposed a geo-hazard risk assessment technique to analyze the impacts of surface subsidence monitored within a major coalmine in Nigeria. In order to accomplish that, the authors also proposed an approach that combines the spatial relationship between vulnerability assessment, risk assessment, and elements at risk, to highlight the grave consequences of potential disasters. Therefore, I found the research very interesting. However, major revisions need to be applied to consider it for publication.

Comment 1: Resolution of ALL figures MUST be improved.

Comment 2: Boundaries (latitude and longitude) should appear in Figure 1 (right hand-side). Please remove the text: “red boundary”.

Comment 3: In Section 3.1, the authors mentioned: “We also acquired a secondary dataset for validation purposes from two different sources; (1) GPS (X, Y, Z) field data of fourteen investigation locations…” Could you please explain in detail the characteristics of these locations (Reference System, how they were determined/method? their standard deviation? GPS ambiguity method/technique?

Comment 4: In Section 3.2, the authors described the SBAS-DInSAR processing technique. Could you please add more details about it? For instance, the path direction of the satellite, the threshold for the spatial and temporal baseline, and the software.

Comment 5: In Section 4.1, the authors mentioned: “Next, we proceeded to masking out area of low coherence…” Which was the threshold value?

Comment 6: How does the lack of information produced by the low coherence affect the risk evaluation? How was it solved?

Comment 7: Were the InSAR time series pre-processed before calculating the average value? How was the process?

Comment 8: Was the average subsidence for each year calculated concerning the beginning of each year or the start of 2016? How was it calculated?

Comment 9: Considering that the displacements were calculated from the start date of the stack of images. Why were the cumulative displacements implemented instead of the velocity (mm/year)? In this manner, the cumulative displacements might show lower risk because the time series are limited from 2016 to 2020.

Comment 10: In Section 4.1, the authors claim the year with the lowest level of risk is 2018 with -28.008 mm, while the year with the highest level of risk is 2016 with -49.312 mm. Could you please explain this in more detail?

Author Response

RESPONSE TO SECOND REVIEWER COMMENTS

Point 1: Resolution of ALL figures MUST be improved.

Response 1: The observation is acknowledged and the figures have been improved for proper reading.

Point 2: Boundaries (latitude and longitude) should appear in Figure 1 (right hand-side). Please remove the text: “red boundary”.

Response 2: The observation has been noted and corrected accordingly in the revised manuscript.

Point 3: In Section 3.1, the authors mentioned: “We also acquired a secondary dataset for validation purposes from two different sources; (1) GPS (X, Y, Z) field data of fourteen investigation locations…” Could you please explain in detail the characteristics of these locations (Reference System, how they were determined/method? their standard deviation? GPS ambiguity method/technique?.

Response 3: The validation of SBAS-DInSAR deformation was done using the Global Navigation Satellite System (GNSS) measurements observed at two different epoch within the study area (Ullo et al., 2019). GNSS measurement techniques offer information about surface movement in three dimensional displacement vector (X,Y,Z), while DInSAR gives displacement variations along the satellite line of site (LOS) (Armaş et al., 2016). Given the interest in displacements (surface subsidence) caused by excessive exploitation of ground water, geological influences due to past coal mining activities, and the heterogeneous surface of the study area, there was a need for a network of GNSS receivers on the ground (Cigna et al., 2021). The GNSS control points were randomly observed between 2010 and 2011, at few accessible locations within the study area. The observation and measuring points were selected at few accessible known locations (major settlements) within the study area with visible evidence of subsidence over the years. The few observation points in the middle is because those areas are highly inaccessible and risky. Hence, we have more observation points around the edges (major settlements). For the validation process, it was necessary to get both dataset into the same coordinate geometry (Armaş et al., 2016). However, because we only used data from the ascending satellite track for DInSAR measurement, we instead converted the GNSS data into vectors along the satellite LOS (Armaş et al., 2016). To achieve this, first we extracted the deformation values from the DInSAR processed time series map along the years using the GNSS X, Y, Z points. The LOS direction of the SBAS-DInSAR measurement is known and is expressed in angles referred to the reference geometric system. Secondly, the conversion of the GNSS three-dimensional coordinates on the LOS direction was made by projecting the orientation angle of incident signal (incident angle) on the vertical direction  and calculating the angles made by the LOS direction (satellite heading angle) for each moment in time (Armaş et al., 2016; Mikhailov et al., 2014). Finally, the complementary data provided by DInSAR and GNSS in both LOS directions was geometrically compared by matching both results through statistical correlation analysis of the displacement (Armaş et al., 2016).

Point 4: In Section 3.2, the authors described the SBAS-DInSAR processing technique. Could you please add more details about it? For instance, the path direction of the satellite, the threshold for the spatial and temporal baseline, and the software?

Response 4: The software is used for applying the SBAS-DInSAR processing chain, including Sentinel Toolbox (SNAP—Open Source), ArcGIS (Licensed), R for Spatial Statistics (Open Source), and Virtual Machine Player (VMware—Open Source). According to (Casu et al., 2014), some general notes are in order for the SBAS-DInSAR procedure. First, the processing steps from block A to D (Figure 4) are performed at full spatial resolution, whereas the subsequent steps work on multi-looked data (block E to F of Figure 4). This first step was performed using the SNAP software. Second, a common storage is assumed available to all the processing phases, i.e., each step gains access to the same common storage for reading inputs and writing outputs. This second phase was performed using SNAP and VMware software. ArcGIS and R were both used for visualization and statistical analysis, respectively. The full sequential steps of the SBAS-DInSAR processing chain through widely used metrics (such as speedup, efficiency, and load balance) are shown in Figure 4. The SBAS-DInSAR processing was performed stand alone (Figure 4), while statistical analysis and prediction was performed using Holt–Winter (Figure 2). The ascending satellite track was used for the SBAS-DInSAR measurement.

Point 5: In Section 4.1, the authors mentioned: “Next, we proceeded to masking out area of low coherence…” Which was the threshold value?

Response 5: The accuracy of monitored ground subsidence values is directly related to the coherence of the subsidence zones (Chen et al., 2019). Hence, the coherence between the reference and the secondary image is estimated as an indicator of the quality of the phase information. If the images have strong similarities, they are therefore usable for interferometric processing. We averaged the coherence coefficient of each map (based on level of risk) to determine the spatial distribution and variations in the subsidence values monitored (Chen et al., 2019). The average coherence level for both horizontal and vertical deformation ranges between 0.45 and 0.47 across the time series image. We proceeded to mask out areas of low coherence using band maths and some logical expressions. For each yearly coherence image, we subtracted the minimum value from the maximum value and divided the outcome by two (Chen et al., 2019).

Point 6: How does the lack of information produced by the low coherence affect the risk evaluation? How was it solved?

Response 6: The area of low coherence was masked out and was not used for the risk evaluation. The risk evaluation was based on the impact on exposure over the investigation locations based on the level of human development index (HDI). Please see section 3.3, 3.3.1,3.3.2,3.3.3 and 4.4 for detailed procedure of risk evaluation.

Point 7: Were the InSAR time series pre-processed before calculating the average value? How was the process?

Response 7: The geometrical impacts created by surface subsidence hazards within Onyeama mine and environment from 2016 to 2020 was measured using SBAS-DInSAR technique (Figure 4). The absolute deformation results were calculated using a quantitative comparative analysis (Nduji et al., 2022). The outcome of block E to F (of Figure 4) is an unwrapped phase, which is a continuous raster, not yet in metric measure but radian units. To convert the unwrapped phase in radian units to absolute displacements, the Phase to Displacement operator in SNAP software is applied. It translates the phase into surface changes along the line-of-sight (LOS) in meters. The LOS is the line between the sensor and a pixel. Accordingly, positive values mean uplift and negative values mean subsidence of the surface (Tables 1 and 2). The Phase to Displacement operator has no parameters and produces an output, which looks similar to the unwrapped phase, but now each pixel has a metric value indicating its displacement. The absolute deformation values in millimeters are further computed using logical expressions of band math’s tool in SNAP software. The wavelength of TOPSAR Sentinel-1 SLC SAR image is 5.6 cm or 56 mm. Therefore, applying the Expression: Horizontal_Displacement = (unwrapped phase * wavelength)/(−4 * PI), we obtain the absolute horizontal displacement. Similarly, if we apply the Expression: Vertical_Displacement = (unwrapped phase * wavelength)/(−4 * pi x cos (rad (incident angle))), we obtain the absolute vertical displacement (Nduji et al., 2022).

Point 8: Was the average subsidence for each year calculated concerning the beginning of each year or the start of 2016? How was it calculated?

Response 8: Please see Response 7. Furthermore, detailed procedure can be found in (Nduji et al., 2022).

Point 9: Considering that the displacements were calculated from the start date of the stack of images. Why were the cumulative displacements implemented instead of the velocity (mm/year)? In this manner, the cumulative displacements might show lower risk because the time series are limited from 2016 to 2020.

Response 9: The results of geometrical subsidence within the study area measured using SBAS-DInSAR technique calculated land deformation velocity in mm/year (Figure 7A and 7B). The cumulative displacement was only computed when calculating the vulnerability assessment (Section 4.4). Here, vulnerability assessment was performed using the vulnerability matrix of equation 3.3. We estimate the potential damage using weighting criteria from the forecasted horizontal and vertical subsidence across the fourteen investigation locations. The weighting criteria was computed by dividing the forecasted subsidence values for each investigation location by the sum of subsidence values for each particular year and multiplying by 100%. This weighting criteria is what gives the cumulative displacement. This was only done for the forecasted deformation values (Figure 9 and 11).

Point 10: In Section 4.1, the authors claim the year with the lowest level of risk is 2018 with -28.008 mm, while the year with the highest level of risk is 2016 with -49.312 mm. Could you please explain this in more detail?

Response 10: To convert the unwrapped phase in radian units to absolute displacements, the Phase to Displacement Operator in SNAP software is applied. It translates the phase into surface changes along the line-of-sight (LOS). The LOS is the line between the sensor and a pixel. Furthermore, the Colour Manipulation Slider Tool in SNAP is applied to highlight areas of low, medium and high risk respectively. Accordingly, positive values mean uplift and negative values mean subsidence of the surface (Tables 1 and 2). The risks are divided according to the range of time series deformation values for each year. Depending on the range of deformation values, it may differ year to year; hence the difference in 2016 to 2020.

Author Response File: Author Response.docx

Reviewer 3 Report

Dear authors.

I went through your paper and put quite a number of comments. You can see them in the attached doc. There are more comments in the first half and less in the second half as many of my concerns repeat and I did not point them out again.
From my perspective the paper has a major flaw. I am not convinced that the observed ground movements (a few cm per year) can cause severe impacts for humans or their assets. You artificially set categories of low, medium, high risk without evidence of severe impacts. I am not an expert in SAR but I presume that the displayed displacement rates are more or less realistic. 
You connect the observed ground movements to causes like coal or ground water extraction. I do not see evidence for your claims. It could be correct but there is a need for evidence to support your theory.
There are numerous smaller issues with the paper as you can see in my comments. With best regards

Comments for author File: Comments.docx

Author Response

RESPONSE TO FIRST REVIEWER COMMENTS

Point 1: I went through your paper and put quite a number of comments. You can see them in the attached doc. There are more comments in the first half and less in the second half as many of my concerns repeat and I did not point them out again. From my perspective the paper has a major flaw. I am not convinced that the observed ground movements (a few cm per year) can cause severe impacts for humans or their assets. You artificially set categories of low, medium, high risk without evidence of severe impacts. I am not an expert in SAR but I presume that the displayed displacement rates are more or less realistic. You connect the observed ground movements to causes like coal or ground water extraction. I do not see evidence for your claims. It could be correct but there is a need for evidence to support your theory. There are numerous smaller issues with the paper as you can see in my comments. With best regards.

Response 1: Please understand that the aim of this paper is to underscore the importance of integrating local level inputs in analyzing risk factors and vulnerability indicators for geo-hazard assessment. The proposed approach combines the spatial relationship between vulnerability assessment and elements at risk, to highlight grave consequences of a potential disaster, which may likely occur due to the rising impacts of surface subsidence in the study location. In developing countries like Nigeria, the local level inputs for appropriate assessment of disasters are hugely underscored and in most cases ignored. Unfortunately, these overlooked historical disaster information collected at local and small-scale regional units are effective statistical indicators for complete disaster risk assessment schemes. Ignoring them renders disaster assessment response efforts inadequate. Also, some disasters are rapid, while others are quite subtle. Disasters like hurricanes, earthquakes and floods, are instant with wide spread losses, while others such as desertization, drought, surface subsidence, sea level rise, soil erosion and glacial retreat are quite slow, but may cause larger impacts over time. The study area has experienced sever negative impacts of subsidence over the years. The ranking of categories on of low, medium, high risk is based on the approach or methodology used for computing the risk assessment matrix and vulnerability assessment. Please see subsection 3.3, 3.3.1, 3.3.2, and 4.4 for clarification.

All other concerns raised by the reviewer has been addressed in the revised manuscript.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

---

Author Response

Response 1: There is no comment or suggestion for the authors. The reviewer accepts the publication of the article in its current form.

Author Response File: Author Response.docx

Reviewer 2 Report

Manuscript land-2228520: "A GEOHAZARD RISK ASSESSMENT TECHNIQUE FOR ANAYLYZING IMPACTS OF SURFACE SUBSIDENCE WITHIN ONYEAMA MINE, SOUTH EAST NIGERIA ". I consider that the authors attended ALL the observations and/or corrections suggested to the article. Therefore, I consider that the manuscript after having attended the corrections in the previous stage, it should be ACCEPTED for publication.

Author Response

Response 1: The reviewer accepts the manuscript in its current form for publication.

Author Response File: Author Response.docx

Reviewer 3 Report

I read the comments of the authors and I also noted some changes in the draft paper, however, my main concern is not sufficiently addressed. I do not see the evidence that the observed annual cm of ground movement are a significant contribution to hazards having the potential to result in a disaster. We have to keep in mind that disasters are big, serious disruptions of the wellbeing of a community and there needs to be a strong correlation between what the authors see as a cause for this disruption and the actual
event. A good way of documenting this would be data about past
disasters where ground subsidence was a determining factor, but I do
not see this. I don't dispute that there were floods or other extreme
events but I do not see a significant contribution of ground subsidence to such events. With this I do not agree to the classification of subsidence into low, medium, high risk categories. This seems to be determined without evidence.

Author Response

Response 1: Your views on disasters been big to seriously disrupt the wellbeing of a community are factual. However, you do not consider that some disasters like hurricanes, earthquakes and floods, are instant with wide spread losses, while others such as desertization, drought, surface subsidence, sea level rise, soil erosion and glacial retreat are quite slow, but may cause larger impacts over time. Our study focuses on these quite slow but dangerous hazards like surface subsidence, which over time have capacity to turn into a disaster with great consequences if not addressed.

KEY NOTE: I mention in the literature review that in the South Eastern states of Nigeria, hazards like land subsidence and soil erosion have increasingly become major environmental issues [8]. It is estimated that approximately 500 tons/km2 of soil per year are washed away due to soil erosion [6,8]. Land subsidence has been witnessed due to several human induced and natural causes, such as: earth motion, excessive exploitation of ground water and indiscriminate exploration of minerals like coal, oil and gas [9]. Soil erosion on the other hand is as a result of rain on slopes, floods and poor storm water drainage networks [5,8]. The continuous ephemeral flows along steep slopes during intense rainfall usually causes the soil erosion to advance into gully erosion [10]. The expansion of these gullies has degraded many areas of land and damaged building infrastructure. Often times, this may lead to splitting of communities and destruction of pathways [5,10]. With a booming population, these combined details put significant pressure on land resources and the safety of the population living around the gullies [5,8]. This problem is more prominent as the country is prone to flooding disaster. In the year 2022 alone, more than 600 lives have been lost and hundreds of thousands of people have been displaced because of flooding (BBC News, 2022). This has become a perennial occurrence with long-term impacts on the economy, destruction of major infrastructure such as transportation routes, power transmission lines and water supply. In addition, top soils are washed off from farmlands, thus leading to poor food production [8,10,11]. THIS IS OUR JUSTIFICATIOB FOR THIS STUDY AND AS SUCH, SHOULD NOT BE TRIVIALIZED.

Our study have measured this subsidence between 2016-2020 and made a forecast from 2021-2024. Using both measured and forecasted deformation values, we are able to propose/develop a Geohazard assessment technique by taking into cognizance the usually overlooked local historical disaster information collected at local and small-scale regional units, which are effective statistical indicators for complete disaster risk assessment schemes. Thus, we underscore the importance of integrating local level inputs in analyzing elements at risk and vulnerability indicators for hazard assessment. The geo-hazard risk assessment for horizontal and vertical deformation was analyzed using the risk assessment matrix of equation 3.2. The risk assessment matrix is based on a scale of likelihood and a scale of severity. By matching the scale of severity against scale of likelihood, the level of risk was summarized as low, medium and high. Figure 8 shows the horizontal deformation risk assessment matrix for investigation location Abor. Based on the key assessment factor, the perceived risk of horizontal deformation between 2021 – 2024 ranges from low to medium. This implies that the likelihood of the horizontal deformation turning into a disaster at this locations is possible. However, this potential disaster based on the scale of severity would still be tolerable. Similarly, Figure 10 shows the vertical subsidence risk assessment matrix for investigation location Okwe. Based on the key assessment factor, the perceived risk of vertical deformation between 2021 – 2024 ranges from medium to high. This implies that the likelihood of the vertical subsidence turning into a disaster at this location is likely. The potential disaster based on the scale of severity is unacceptable. Tables 3 and 4 summarizes the horizontal deformation and vertical subsidence risk assessment analysis showing elements at risk, vulnerability parameters and consequences in the event of likely disaster. From both tables, the likelihood of the horizontal deformation and vertical subsidence turning into a disaster at some investigation locations is further justified. Furthermore, Figure 9 and 11 shows the results of forecast for horizontal deformation and vertical subsidence made using Holts-Winters model across the fourteen (14) investigation locations within the study area. We made use of data on yearly rate of horizontal deformation from January 2016 – December 2020. The prediction are for January 2021 to December 2024 (48 months). On the average, there is a gradual increasing trend of potential hazard across all investigation locations over the years. This increase in the horizontal dimension may worsen due to adverse effects of global warming, climate change, increased poverty rate, ever-growing population and fast-growing urbanization with a disregard for sustainability [9].

Author Response File: Author Response.docx