year 12, Issue 1 (Spring 2022)
E.E.R. 2022, 12(1): 183-206 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Hejazi S A, Rezaee Moghadam M H, Valizadeh Kamran K, Mosavi N. Detection and Determination of Subsidence Using Radar Interferometer (D-InSAR) Method (Harris Plain). E.E.R. 2022; 12 (1) :183-206
URL: http://magazine.hormozgan.ac.ir/article-1-645-en.html
URL: http://magazine.hormozgan.ac.ir/article-1-645-en.html
Detection and Determination of Subsidence Using Radar Interferometer (D-InSAR) Method (Harris Plain)
Department of Geomorphology, Faculty of Planning and Environmental Sciences, University of Tabriz, Iran , hejazi@tabriz.ac.ir
Abstract: (2350 Views)
1-Introduction
Subsidence is a morphological phenomenon that occurs under the influence of land subsidence motion. The cause of this phenomenon may be due to human and natural factors. The phenomenon of subsidence in recent decades has created many problems for agricultural lands, residential areas, roads and water supply canals in some parts of the country. In recent years, the decrease in rainfall and the increase in uncontrolled groundwater harvesting by exploitation wells have caused a large drop in groundwater level, which has resulted in subsidence and cracks and fissures in parts of the Harris plain. In this research, in order to determine the affected area and finally estimate the amount of subsidence, the radar interferometer method has been used. The advantages of this method include very high accuracy, wide coverage, high spatial resolution and no need for field work, cost-effectiveness and the possibility of accessing information in any weather conditions. ENVISAT and Sentinel radar satellite images were used for this purpose. Also, two Goldstein and Adaptive filters were used to evaluate the obtained results. Also, to validate the research, the relationship between subsidence and the depth of wells in the region was studied, which showed a positive correlation of 87%; this indicates that the depth of wells is also deeper in the areas where subsidence has occurred.
2-Method
Software of SARscape5. 2 and SNAP6. 0 used in this research are among the powerful tools for monitoring subsidence. The radar interference method by comparing the phases of two radar images taken from an area in two different times is able to determine changes in the earth's surface in that time period.
The phase obtained from a complication on the ground is proportional to its distance from the radar sensor. Therefore, changing this distance affects the measured phase. An image called an interferogram is created using a radar interferometry technique. To eliminate the topographic effect, the digital elevation model SRTM has been used with a spatial resolution of 90 meters. Orbital errors were modeled by a procedure that have no displacement. Atmospheric error can be adjusted with the help of atmospheric information and atmospheric model. Interferometry is produced by the complex multiplication of one SAR image in the mixed conjugate of the second image. The resulting differential interferometer contains some noise. The cause of these noises can be different, but there are two main factors influencing their occurrence. The first factor is related to the time difference between the two main and dependent images. The second main factor influencing the occurrence of noise is the spatial baseline where the amount of noise in the images is directly related to the spatial baseline. Goldstein and Adaptive Window Filters have been used to remove and reduce noise.
As mentioned earlier, Goldstein and Adaptive filters were used in this study, and you will see the main results of these filters below. In principle, the differences between the results of these two widely used filters in the field of radar are discussed. The images have been used in the new 2017 to 2019 returns.
Coherent image is an image that is resulted from the power correlation of two coordinated images. This image shows the correlation index of signal strength values in two images taken at two different times. The value of correlation varies from 0 to 1, which affects the quality of the interferometry process. After interfrogram processing, to determine the average subsidence rate in the time period of the images, the time series analysis method was used with the least squares method. Based on this, the average subsidence rate is equal to 1. 2 cm. Due to subsidence in this area, it is possible to use groundwater resources indiscriminately and reduce rainfall, which has led to a drop in groundwater levels. Of course, the type of cultivation in the region and the weight pressure caused by human structures in this area can be other causes of subsidence. The obtained results showed that the Goldstein filter has subsidence values up to 9 cm in certain ranges and the uplift values up to about 8. 5 cm and the Adaptive filter subsidence has values up to 9. 5 cm in the range and the values of the uplift up to about 8 cm. The reason for the difference in values is the results of these two filters. The Goldstein filter, by manipulating the phases, increases the coherence and the brightness of different parts and is higher and brighter, so the situation is better in this filter. In Adaptive filter, phases are not manipulated, coherence zones remain and the amount of blur is greater in different parts of the image. However, in each picture, the amount of subsidence is significantly observed in the east of the area, which is due to the greater concentration of agricultural activity and the use of groundwater.
3-Results:
There is a close relationship between geological formation and subsidence: the weaker the formation, the more noticeable the subsidence in the area. As can be seen, the subsidence in the alluvial formation is more than the volcanic conglomerate formation and the reason is the hardness of the conglomerate stone. As a result, alluvial rocks, andesite and marl have the most subsidence and latite (volcanic) have the least subsidence. Given that the hypothesis of this research was based on the principle that increasing the amount of groundwater extraction causes subsidence in the region, the relationship between the depth of wells in the area and the amount of subsidence in that area can be a good indicator to assess the accuracy of operations. Therefore, the statistics of piezometric wells in the region have been received from the Regional Water Organization of East Azerbaijan Province. After converting it to Arch GJS software database format, the relationship between subsidence values and piezometric well depths was established using zonal spatial analysis functions. Regression correlation analysis between these two factors showed a positive correlation of 87% between them and the implication is that any place that has a lot of subsidence in it is deeper piezometric wells. This finding confirms the hypothesis of this research that there is a direct and strong relationship between subsidence and groundwater abstraction.
4-Discussion & conclusions
The use of radar interference method in this study introduces a good capacity for its capabilities in determining the amount of subsidence in the study area. Findings from the use of this method showed a relatively high rate of subsidence for about a year. According to the maps, the maximum and minimum subsidence in the mentioned periods are 9.8, 0.6 cm, respectively. After entrophagram processing, time series analysis method with least squares method was used to determine the average
Subsidence is a morphological phenomenon that occurs under the influence of land subsidence motion. The cause of this phenomenon may be due to human and natural factors. The phenomenon of subsidence in recent decades has created many problems for agricultural lands, residential areas, roads and water supply canals in some parts of the country. In recent years, the decrease in rainfall and the increase in uncontrolled groundwater harvesting by exploitation wells have caused a large drop in groundwater level, which has resulted in subsidence and cracks and fissures in parts of the Harris plain. In this research, in order to determine the affected area and finally estimate the amount of subsidence, the radar interferometer method has been used. The advantages of this method include very high accuracy, wide coverage, high spatial resolution and no need for field work, cost-effectiveness and the possibility of accessing information in any weather conditions. ENVISAT and Sentinel radar satellite images were used for this purpose. Also, two Goldstein and Adaptive filters were used to evaluate the obtained results. Also, to validate the research, the relationship between subsidence and the depth of wells in the region was studied, which showed a positive correlation of 87%; this indicates that the depth of wells is also deeper in the areas where subsidence has occurred.
2-Method
Software of SARscape5. 2 and SNAP6. 0 used in this research are among the powerful tools for monitoring subsidence. The radar interference method by comparing the phases of two radar images taken from an area in two different times is able to determine changes in the earth's surface in that time period.
The phase obtained from a complication on the ground is proportional to its distance from the radar sensor. Therefore, changing this distance affects the measured phase. An image called an interferogram is created using a radar interferometry technique. To eliminate the topographic effect, the digital elevation model SRTM has been used with a spatial resolution of 90 meters. Orbital errors were modeled by a procedure that have no displacement. Atmospheric error can be adjusted with the help of atmospheric information and atmospheric model. Interferometry is produced by the complex multiplication of one SAR image in the mixed conjugate of the second image. The resulting differential interferometer contains some noise. The cause of these noises can be different, but there are two main factors influencing their occurrence. The first factor is related to the time difference between the two main and dependent images. The second main factor influencing the occurrence of noise is the spatial baseline where the amount of noise in the images is directly related to the spatial baseline. Goldstein and Adaptive Window Filters have been used to remove and reduce noise.
As mentioned earlier, Goldstein and Adaptive filters were used in this study, and you will see the main results of these filters below. In principle, the differences between the results of these two widely used filters in the field of radar are discussed. The images have been used in the new 2017 to 2019 returns.
Coherent image is an image that is resulted from the power correlation of two coordinated images. This image shows the correlation index of signal strength values in two images taken at two different times. The value of correlation varies from 0 to 1, which affects the quality of the interferometry process. After interfrogram processing, to determine the average subsidence rate in the time period of the images, the time series analysis method was used with the least squares method. Based on this, the average subsidence rate is equal to 1. 2 cm. Due to subsidence in this area, it is possible to use groundwater resources indiscriminately and reduce rainfall, which has led to a drop in groundwater levels. Of course, the type of cultivation in the region and the weight pressure caused by human structures in this area can be other causes of subsidence. The obtained results showed that the Goldstein filter has subsidence values up to 9 cm in certain ranges and the uplift values up to about 8. 5 cm and the Adaptive filter subsidence has values up to 9. 5 cm in the range and the values of the uplift up to about 8 cm. The reason for the difference in values is the results of these two filters. The Goldstein filter, by manipulating the phases, increases the coherence and the brightness of different parts and is higher and brighter, so the situation is better in this filter. In Adaptive filter, phases are not manipulated, coherence zones remain and the amount of blur is greater in different parts of the image. However, in each picture, the amount of subsidence is significantly observed in the east of the area, which is due to the greater concentration of agricultural activity and the use of groundwater.
3-Results:
There is a close relationship between geological formation and subsidence: the weaker the formation, the more noticeable the subsidence in the area. As can be seen, the subsidence in the alluvial formation is more than the volcanic conglomerate formation and the reason is the hardness of the conglomerate stone. As a result, alluvial rocks, andesite and marl have the most subsidence and latite (volcanic) have the least subsidence. Given that the hypothesis of this research was based on the principle that increasing the amount of groundwater extraction causes subsidence in the region, the relationship between the depth of wells in the area and the amount of subsidence in that area can be a good indicator to assess the accuracy of operations. Therefore, the statistics of piezometric wells in the region have been received from the Regional Water Organization of East Azerbaijan Province. After converting it to Arch GJS software database format, the relationship between subsidence values and piezometric well depths was established using zonal spatial analysis functions. Regression correlation analysis between these two factors showed a positive correlation of 87% between them and the implication is that any place that has a lot of subsidence in it is deeper piezometric wells. This finding confirms the hypothesis of this research that there is a direct and strong relationship between subsidence and groundwater abstraction.
4-Discussion & conclusions
The use of radar interference method in this study introduces a good capacity for its capabilities in determining the amount of subsidence in the study area. Findings from the use of this method showed a relatively high rate of subsidence for about a year. According to the maps, the maximum and minimum subsidence in the mentioned periods are 9.8, 0.6 cm, respectively. After entrophagram processing, time series analysis method with least squares method was used to determine the average
subsidence rate in the time period of the images. Based on this analysis, the average subsidence rate is equal to 1.2 cm. One of the causes of subsidence in this area is the indiscriminate use of groundwater resources and reduced rainfall, which has led to a drop in groundwater levels. In the present study, after initial processing on Sentinel-1 satellite data in remote sensing software and GIS, the amount of subsidence in this plain was estimated. Also in this study, both Goldstein and Adaptive filters were used for conducting further investigation. The obtained results indicated that the Goldstein filter has subsidence values up to 9 cm in certain ranges and the uplift values up to about 8.5 cm and the Adaptive filter has the subsidence values up to 9.5 cm in some ranges and uplift values are also shown up to about 8 cm. The reason for the difference in values in the results of these two filters is that the Goldstein filter, by manipulating the phases, increases the coherence rate and the brightness of different parts and the image is brighter, so the situation in this filter improves. But this is not the case with the Adaptive filter, and the phases are not manipulated, and in some areas the coherence remains, and the amount of blur is greater in different parts of the image. However, in both images, the amount of subsidence can be seen significantly to the east of the area, which is due to the greater concentration of agricultural activity and the use of groundwater.
Type of Study: Research |
Received: 2021/06/21 | Published: 2022/03/12
Received: 2021/06/21 | Published: 2022/03/12
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
