Document Type : Research Article
Authors
Department of Geomorphology, Faculty of Planning and environmental sciences, University of Tabriz, Tabriz, Iran
Abstract
Land subsidence represents one of the most critical geomorphological hazards across global alluvial plains, primarily driven by excessive groundwater exploitation. This study evaluates land subsidence in the Miandoab Plain (northwestern Iran) and investigates its relationship with groundwater level decline and environmental covariates. A total of 38 Sentinel-1 Synthetic Aperture Radar (SAR) acquisitions from January 2018 to January 2024 were processed using the Small Baseline Subset time-series interferometric SAR (SBAS-InSAR) algorithm within GMTSAR software to quantify surface deformation. Additionally, historical records from 12 piezometric monitoring wells were analyzed to evaluate groundwater depletion dynamics. A subsidence susceptibility map was generated using a hybrid model integrating fuzzy logic and the Relative Weighting Ratio (RWR) coupled with key environmental parameters. The results revealed that cumulative subsidence ranged from 2.9 to 36.7 cm over the 6-year period, corresponding to an average annual deformation rate of approximately 6.0 cm/yr (60 mm/yr). The maximum subsidence rates were concentrated in the southern sectors of the plain and across the urban extent of Miandoab. Piezometric analysis indicated a mean annual groundwater table decline of 81.6 cm. Pixel-based spatial correlation analysis confirmed a significant positive correlation (r=0.66) between groundwater decline and ground subsidence. Furthermore, susceptibility zonation demonstrated that 17.8% of the study area is classified as high to very high subsidence hazard. These findings underscore that groundwater depletion is the primary controlling driver of land subsidence in the Miandoab Plain, highlighting the urgent need for sustainable aquifer management to mitigate ongoing ground deformation.
Extended Abstract
Introduction
Land subsidence is one of the major geomorphological and environmental hazards affecting rapidly developing regions worldwide. It is manifested as gradual or localized downward displacement of the ground surface, primarily resulting from the compaction of unconsolidated and compressible subsurface sediments. Among the various controlling factors, excessive groundwater abstraction is considered a dominant anthropogenic driver. Continued groundwater withdrawal reduces pore-water pressure and increases effective stress within aquifer systems, particularly in fine-grained clayey and silty deposits, resulting in irreversible compaction and progressive surface deformation. Subsidence can consequently cause ground cracks, damage to buildings and infrastructure, disruption of transportation and energy networks, reduction in aquifer storage capacity, alteration of surface drainage, and increased flood susceptibility.
Iran is particularly vulnerable to land subsidence because of increasing pressure on groundwater resources, climatic aridity, and extensive dependence on groundwater for agricultural activities. The Miandoab Plain in northwestern Iran is an important agricultural and settlement area characterized by extensive alluvial deposits and intensive groundwater use. Increasing agricultural development, population growth, and reliance on groundwater have contributed to a persistent decline in groundwater levels, raising concerns regarding the development and spatial expansion of subsidence. The presence of fine-grained alluvial sediments and concentrated agricultural activities further increases the susceptibility of the plain to ground deformation. Despite the significance of these processes, detailed spatial information on the temporal evolution and controlling factors of subsidence in the Miandoab Plain remains limited.
Previous studies in different regions, including Mexico City, Shanghai, Peshawar, and several Iranian plains, have demonstrated the effectiveness of Interferometric Synthetic Aperture Radar (InSAR) for detecting and quantifying surface deformation and have emphasized the relationship between groundwater depletion and subsidence. However, the spatial interaction between groundwater decline and other environmental controls requires further investigation. Therefore, this study aims to analyze the spatial and temporal characteristics of land subsidence in the Miandoab Plain using Sentinel-1 InSAR observations and to investigate its relationship with groundwater-level decline. In addition, environmental and hydrogeological factors are integrated to identify areas with greater susceptibility to future subsidence, providing a spatial basis for groundwater management and land-use planning.
Material and Methods
This study employed an integrated approach combining satellite radar interferometry, hydrogeological observations, and GIS-based spatial analysis to investigate land subsidence and its controlling factors in the Miandoab Plain, northwestern Iran. The analysis was conducted using 38 Sentinel-1 Synthetic Aperture Radar (SAR) images acquired in Interferometric Wide (IW) mode with VV polarization and descending orbit geometry during the period from 11 January 2018 to 4 January 2024. The radar data were processed using GMTSAR and GMT under the Linux operating system. Orbit correction, topographic phase removal using a digital elevation model (DEM), interferogram generation, and filtering were performed during preprocessing. Interferometric pairs were selected based on temporal and perpendicular spatial baselines, with a spatial baseline threshold of less than 30 m and a coherence threshold of 0.2 to reduce geometric and decorrelation-related errors. A total of 37 suitable interferometric pairs were subsequently processed using the Small Baseline Subset (SBAS) approach to estimate the spatial and temporal characteristics of surface deformation and generate the subsidence-rate map.
Groundwater conditions were evaluated using long-term observations from 12 piezometric wells distributed across the study area. Groundwater-level data covering the period 2001–2024 were analyzed to determine the spatial pattern and annual rate of groundwater decline. The Inverse Distance Weighting (IDW) interpolation method was applied in ArcGIS to generate a continuous groundwater-decline surface. The spatial relationship between groundwater decline and InSAR-derived subsidence was then assessed using the Pearson correlation coefficient.
To identify areas susceptible to land subsidence, six environmental and hydrogeological parameters were considered: groundwater-level decline, distance from the river, elevation, slope, land use, and lithology. The spatial layers were prepared in GIS and standardized using fuzzy membership functions, with values approaching one representing greater susceptibility. The relative importance of the parameters was determined using the Relative Weighting Ratio (RWR) method. In this approach, an initial weight of one was assigned to each parameter and 0.1 was added for each pairwise preference, after which the resulting values were normalized. The standardized layers and their relative weights were integrated using a fuzzy gamma operator. Different gamma values were evaluated, and the value providing the greatest spatial agreement with the observed InSAR-derived subsidence pattern was selected. A gamma value of 0.9 was ultimately adopted for the final susceptibility model.
Finally, the resulting susceptibility map was compared with the observed subsidence distribution to evaluate model performance and identify areas where groundwater depletion and environmental characteristics jointly contribute to higher subsidence potential.
Results and Discussion
The Sentinel-1 time-series analysis revealed a substantial spatial and temporal pattern of land subsidence across the Miandoab Plain during 11 January 2018–4 January 2024. The cumulative subsidence ranged from 2.9 to 36.7 cm, with an average annual subsidence rate of approximately 6 cm. The most pronounced deformation occurred in the southern parts of the plain and extended into substantial portions of Miandoab urban area, whereas the lowest deformation was generally observed in the surrounding upland areas. The concentration of severe deformation in the southern plain corresponds spatially with intensive agricultural activities, high groundwater demand, and the distribution of compressible alluvial sediments.
Analysis of groundwater observations from the 12 piezometric wells indicated a persistent decline in groundwater levels throughout the study area. The mean annual groundwater decline was approximately 81.6 cm, with values ranging from 59 to 100 cm per year. The highest decline, approximately 100 cm/year, was recorded at well 9 near the agricultural station, while the lowest value, approximately 59 cm/year, occurred at well 8 near Lilan. The IDW interpolation showed that groundwater depletion was particularly pronounced in the southern part of the plain. This spatial pattern is consistent with the concentration of groundwater extraction wells and water-intensive agricultural activities. Groundwater decline generally decreased toward the northern and peripheral areas of the plain.
The spatial comparison between groundwater decline and InSAR-derived subsidence demonstrated a significant positive relationship, with a Pearson correlation coefficient of 0.66. Areas experiencing greater groundwater-level decline generally corresponded to areas with greater subsidence, supporting the interpretation that groundwater depletion is a major controlling factor in the deformation of the Miandoab Plain. This relationship is consistent with the established mechanism whereby groundwater withdrawal reduces pore-water pressure, increases effective stress, and promotes compaction of compressible fine-grained sediments.
The fuzzy-RWR model further demonstrated the spatial significance of groundwater decline in controlling subsidence susceptibility. Among the six investigated parameters, groundwater decline received the highest relative weight of 0.2. The final fuzzy combination using a gamma value of 0.9 identified the southern plain as the principal high-susceptibility zone. Approximately 98.7 km², equivalent to 17.8% of the study area, was classified as having very high subsidence susceptibility, while 121.9 km², approximately 22% of the area, was categorized as highly susceptible. These zones are characterized by substantial groundwater depletion, low elevation and slope, intensive agricultural land use, and the presence of relatively compressible alluvial materials.
The general agreement between the modeled susceptibility pattern and the observed InSAR deformation indicates that integrating hydrogeological and environmental parameters can effectively identify areas vulnerable to land subsidence. Nevertheless, uncertainties remain due to atmospheric effects in C-band radar observations, temporal and spatial decorrelation associated with agricultural land-cover changes, the limited number and uneven distribution of piezometric wells, differences between the groundwater and radar observation periods, and the absence of some potentially influential parameters such as aquifer thickness and groundwater-well density.
Conclusion
The results demonstrate that land subsidence has become a significant and spatially extensive environmental hazard in the Miandoab Plain. Sentinel-1 SBAS analysis identified cumulative subsidence of 2.9–36.7 cm during 2018–2024, with the highest deformation concentrated in the southern plain and parts of Miandoab urban area. Groundwater observations showed an average annual decline of 81.6 cm, and the significant positive correlation between groundwater decline and subsidence (r = 0.66) confirms the major role of groundwater depletion in controlling surface deformation. The fuzzy-RWR model, using six environmental and hydrogeological parameters and a gamma value of 0.9, identified 17.8% of the plain as having very high and approximately 22% as having high subsidence susceptibility. These findings highlight the necessity of continuous InSAR monitoring, improved groundwater-level observation, control of groundwater abstraction, adoption of water-efficient irrigation practices, artificial aquifer recharge, and integration of subsidence susceptibility into land-use and water-resource management.
Keywords
Subjects
