Spatial Analysis of Jiroft Plain Subsidence Using the Coherence Pixel Technique (CPT)

AbstractThe phenomenon of land subsidence is one of the most important environmental hazards that have affected many plains of the country today. Jiroft Plain located in Kerman Province is also one of the areas where subsidence effects are evident. In this research, besides analyzing the spatial subsidence of Jiroft Plain and determining the extent and trend of its spread over a period of time, the effective factors in this phenomenon were investigated. For this purpose, the Sentinel-1 radar images related to the years of 2014-2022 were used. The Coherence Pixel Technique (CPT) was utilized to map the affected areas and determine the subsidence rate. The results of this method showed that the subsidence rate in Jiroft Plain had increased from 11 cm in 2014 to 13 cm in 2022. In addition, its area had increased during this period and the expansion trend had moved towards the northern areas of the plain. To analyze the causative factors of this phenomenon, in addition to studying the changes in the groundwater level of the plain and its relationship with subsidence, the roles of faults and soil thickness in creating or intensifying this phenomenon were investigated. The results showed that in addition to the uncontrolled abstraction from the aquifer, the subsidence of Jiroft Plain was affected by Sabzevaran Fault, while subsidence intensity was higher in the areas with higher soil thickness.Keywords: subsidence, groundwater, fault, spatial relationship, Jiroft Plain IntroductionThe phenomenon of subsidence is one of the growing and fundamental problems in most human societies, which often occurs as a result of human activities. Improper use of water in agricultural and industrial sectors due to the increasing population growth has led to adverse quantitative and qualitative effects on water resources. Understanding the spatial extent and measuring the amount of subsidence as accurately as possible can be considered as the first step in studying this phenomenon. Therefore, by recognizing the spatial characteristics and temporal behavior of this phenomenon, it is possible to present and develop a regional model of it as well as practical and basic solutions to reduce the damage associated with it and prevent its future trends. The use of radar interferometry method in recent years as an efficient tool for monitoring displacements caused by various phenomena, such as volcanoes, subsidence, earthquakes, and landslides, etc., has been considered by earth scientists. The advantages of this method compared to the previous one include the possibility of calculating displacements with centimeter-level accuracy, providing continuous and extensive spatial coverage, and having the ability to operate it in any weather conditions. So far, various studies have qualitatively identified the relationship between groundwater level declines and occurrence of subsidence. However, few studies have tested this relationship quantitatively. Jiroft Plain has been facing a serious drought crisis and declining groundwater levels in recent years. In fact, changes in the agricultural pattern, reduced rainfall, and occurrence of continuous droughts have led to unplanned and unprincipled use of groundwater resources and decline of groundwater levels in the catchment area of Jiroft Plain and have provided conditions for the occurrence and expansion of land subsidence. Therefore, subsidence monitoring has been proposed as an efficient method for identifying and displaying the regional situations in terms of the risk of land subsidence by planners and managers and made it possible to plan and implement appropriate prevention programs. Therefore, the purpose of this paper was spatial analysis of Jiroft Plain subsidence and evaluation of the effects of uncontrolled groundwater abstraction on land subsidence and fault development. To this end, 73 Sentinel-1 images related to the period of 2014-2021 were processed by using the CPT technique. The darkest areas were related to agricultural areas, which had the least amounts of coherence since their vegetations had not remained constant over time and caused a temporal correlation in the interfering phase. MethodologyThe first step in the CPT processing was production of differential interferometers. Initially, the images were referenced in pairs and the interferometers were selected from the items that had spatial and temporal baselines of less than 100 and 365 m, respectively. Based on this, 72 mapping overlays were generated. Along with the various interferences, the related coherence maps were generated. Coherence is a good estimator of phase quality and is used in the pixel selection phase. Coherence values range from 0 that shows a completely uncorrelated phase or pure noise to 1 that indicates a coherent or noise-free phase. In the second step, differential interferometers were processed to obtain the deformation time series, which included linear and nonlinear components and DEM error. Not all image pixels are suitable for processing due to lacking correlation. Among the various pixel selection criteria, a coherence-based criterion was used. Therefore, all the pixels that had a mean coherence value of less than 0.6 were discarded. Figure 3 shows the coherence map created by the CPT for Jiroft Plain where the brightest areas showed the most coherent areas, which corresponded to barren lands, mountainous lands and highlands, and residential areas because they showed very little change over time. DiscussionFigure 4 shows the magnitude of the shifts that occurred along the satellite's line of sight as a result of Sentinel-1 data processing from April 12, 2014 to September 21, 2021. The total number of pixels calculated from the Sentinel-1 data set by the CPT technique reached 2571. The calculations revealed a high deformation rate with a maximum speed of up to -13 cm per year for satellite visibility in the central and southern parts of Jiroft Plain from 2014 to 2021. The positive values indicated that the surface was rising. Most of these values were located in the mountains around the plain. Motion may be related to tectonic factors and the isostatic process. In contrast, the negative values indicated subsidence, which was mainly concentrated in the central and southern parts of the plain. As can be seen in Figure 4, the displacements in the direction of satellite view of the points varied from +3 to -13 mm. In addition, the results showed the increasing trend of subsidence over time in Jiroft Plain. From 2014 to 2022, the area affected by subsidence had increased from 530 km2 in 2014 to 580 km2 in 2022, showing an expansion from the southern to the northern areas of the plain. In addition to the extent of the subsidence rate, it had increased from 11 cm in 2014 to 13 cm in 2022. ConclusionIn this study, the subsidence phenomenon of Jiroft Plain was investigated and its relationship with various factors was analyzed. The displacements that had occurred and were then obtained from the CPT technique indicated that the study area had undergone progressive subsidence. The subsidence rate in the southern and southeastern parts of the plain had increased from 11 ear to 13 cm per year over an 8-year period. In addition, during this period, the area affected by this phenomenon had increased from 530 to 580 km2 and had been drawn to the northern parts of the plain over time. Investigation of groundwater level changes in Jiroft Plain and its compliance with the subsidence areas showed that improper abstraction from the aquifer had been an important and key factor in creating this phenomenon. In addition, it is worth noting that the areas with the highest subsidence rate corresponded to the areas with the highest soil thickness. Thus, their impacts on the subsidence were investigated due to the enclosure of Jiroft Plain by faults. The results of this study showed that Sabzevaran Fault had controlled subsidence and affected this phenomenon. Finally, the subsidence of Jiroft Plain could be considered as a result of two factors, one was the uncontrolled abstraction of groundwater and the other one was the activity of faults, which could affect each other. This issue had intensified the subsidence phenomenon.