DOI QR코드

DOI QR Code

Monitoring Land Surface Deformation Using the PS-InSAR Technique: A Case Study of Kakinada City, India

  • Received : 2025.03.25
  • Accepted : 2025.06.26
  • Published : 2025.06.30

Abstract

In recent years, land subsidence has markedly increased in urban areas. This phenomenon poses a significant threat, especially in coastal areas at risk of rising sea levels, emphasizing the need for continuous monitoring of land subsidence and investigation of the underlying causes. This research focuses on the spatiotemporal analysis of land subsidence in Kakinada, India, from 2020 to 2024. Utilising 58 Sentinel-I images from the descending track, this research implemented the Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) methodology to process the data. The findings revealed a deformation rate within the study area ranging from -26 mm/year to +14 mm/year. The highest subsidence was observed in densely populated urban areas and regions with extensive aquaculture and saltpans. Croplands and low-density urban areas exhibited minimal deformation. Interestingly, some peripheral urban regions show signs of uplift. The region's main drivers of land subsidence are compaction of unconsolidated sediments due to overlying loads in densely built-up areas and groundwater pumping. This research provides valuable information for urban planners, policy developers, and environmental managers to develop sustainable urban development, groundwater management, and coastal protection strategies.

Keywords

1. Introduction

The increasing density of coastal populations is causing land strain and land subsidence, a significant global threat marked by the gradual compaction or sudden sinking of the Earth’s surface (Takagi et al., 2017). Subsidence poses a substantial risk to populated coastal cities, where subsidence rates often surpass the global sea-level rise (Erkens et al., 2015; Hallegatte et al., 2013; Ingebritsen and Galloway, 2014). Consequently, these areas experience increased vulnerability to flooding and substantial infrastructure loss.

Anthropogenic actions contribute to approximately 80% of the subsidence cases worldwide, with groundwater extraction being the primary cause (Bagheri-Gavkosh et al., 2021). Additionally, the load of new construction and engineering worksin the delta areas can lead to shallow subsurface consolidation (Yan et al., 2020). Due to its slow rate, subsidence often goes unnoticed for extended periods(Gambolati et al., 2005). Therefore, continuous monitoring of the spatiotemporal behaviour of land deformation is crucial (Garg et al., 2022).

The past few decades have witnessed significant advancements in satellite-based synthetic aperture radar interferometry, enabling highly accurate and detailed monitoring of ground deformation. Over the past decade, the improvements in both the quality and volume of satellite imagery has made it possible to measure ground motion with sub-millimetre precision per year across the globe (Biggs and Wright, 2020). Interferometric Synthetic Aperture Radar(InSAR)techniques have been effectively implemented in various applications, including the study of surface displacement caused by earthquakes (Kandregula et al., 2022; Lasserre et al., 2005), landslides(Bénédicte et al., 1997;Colesanti and Wasowski, 2006; Tofani et al., 2013), volcanic activity (Champenois et al., 2014; Pavez et al., 2006), mining (Baek et al., 2008; Carnec and Delacourt, 2000; Gupta et al., 2019), hydrocarbon extraction (Fielding et al., 1998; Moghaddam et al., 2013). These advancements have significantly improved our capability of understanding and monitoring surface deformation phenomena.

The Persistent Scatterer Interferometry technique introduced by (Ferretti et al., 2000, 2001) works by exploiting image pixels called permanent scatterers (PS), distinguished by consistent radar reflections through time, and are often found on artificial structures or natural objects with negligible deformation. It effectively overcomes the limitations of traditional Differential Interferometric Synthetic Aperture Radar (DInSAR) methods, such as coherence loss due to temporal and geometric factors, and phase disturbances caused by atmospheric conditions (Crosetto et al., 2016). It has been successfully used for land deformation studies in various cities worldwide, including Jakarta (Hakim et al., 2020), Beijing (Cao et al., 2021), Lahore (Ahmad et al., 2022), Delhi (Agarwal et al., 2021), and Varanasi (Tripathi et al., 2022).

The objectives of the study were to i) estimate the line-of-sight (LOS) deformation of the study area between 2020 and 2024 using the Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) technique, and ii) study various natural and human-induced driving factors that may have influenced the detected ground deformation.

Kakinada was chosen for this study due to its unique combination of characteristics that contribute to subsidence phenomena, including intensive oil and gas extraction activities in the Krishna-Godavari Basin, rapid urbanization, industrial development, and highly compressible quaternary sediments with a high clay content, which are particularly susceptible to consolidation. As a major port city, the area’s importance hasled to extensive infrastructure development and land reclamation projects, creating substantial loading on compressible sediments. Previous InSAR-based studies have documented significant subsidence rates, with Kumar et al.(2017) reporting approximately 14 mm/yr from 2007 to 2011 and Reshma et al. (2023) finding rates of approximately 10 mm/yr between 2017 and 2019, primarily attributed to rapid urbanization. However, these studies covered Kakinada within broader Godavari Delta analyses, leaving a gap for detailed city-scale investigations. The region’s vulnerability is further increased by its low elevation, exposure to coastal erosion, rising sea levels, and the loss of natural coastal protection through mangrove destruction. This combination of geological susceptibility, anthropogenic activities, and documented subsidence history makes Kakinada a case study for understanding the complex spatiotemporal patterns and driving mechanisms of coastal land subsidence.

2. Study Area

Kakinada is a coastal city in Andhra Pradesh, India, located between 16°51′–17°04′N latitude and 82°06′–82°17′E longitude, covering approximately 250 km² and comprising 43 administrative units, including urban, semi-urban, and rural settlements. The region experiences a tropical wet-and-dry climate with an annual rainfall of 1,110 mm, predominantly delivered by southwest and northeast monsoons. The study area lies within the onshore Krishna–Godavari Basin, a passive rift-margin basin with sedimentary fill exceeding 5,000 m thickness from Late Carboniferous to Holocene age (Rao, 2001). Quaternary deposits dominate the uppermost 300 m, comprising fluvial formations (Ramachandrapuram, Yeleru) of clay-rich silts and sands, and coastal formations (Kaikaluru, Polatitippa) of fine sands and black silty clays in paleo-beach and tidal-flat facies.

Over 80% of Kakinada lies below 10 m above mean sea level, rendering it highly susceptible to storm surges and sea-level rise. The city frequently faces cyclones in the Bay of Bengal, where the storm surges inundate low-lying coastal areas. South of the city, the Coringa mangrove forest, the second largest in India, provides essential coastal protection, although its extent has recently declined.

Following the bifurcation of Andhra Pradesh, Kakinada has been marked for rapid industrial development, emerging as one of the key deep-water ports on India’s east coast. The city’s proximity to the Krishna-Godavari (KG) Basin for oil and gas exploration and establishing a 7,440-acre Special Economic Zone (SEZ) hosting petrochemical, gas-based, and pharmaceutical industries has caused substantial economic growth.

However,this development, including extensive mudflat reclamation for industrial parks and port expansions, has significantly altered natural coastal morphologies, increasing land subsidence risks and stress on the natural environment and the densely populated surrounding areas. The growing industrial activity contributes to habitat loss, shoreline instability, and heightened vulnerability to climate-induced hazards. Additionally, Uppada, located near Kakinada, is recognized as one of the most erosion-prone coastal stretches in the region, further highlighting its fragility. The location of the study area is depicted in Fig. 1.

OGCSBN_2025_v41n3_605_3_f0001.png 이미지

Fig. 1. Geographical location and boundary of the study area.

3. Materials and Methods

3.1. Data

In this study, Sentinel-1 Synthetic Aperture Radar (SAR) data were utilized to investigate land deformation in the study area. A total of 58 Single Look Complex (SLC) datasets were acquired in Interferometric Wide (IW)swath mode of the descending orbit, covering the period from 20April 2020 to 30 March 2024. These data were obtained from the Copernicus Open Access Hub and were acquired in VV polarization. The IW mode provides a spatial resolution of 5 m by 20 m (range by azimuth). The mean radar incidence angle over the study area is approximately 39.3°, with values ranging from around 33° near the near range to approximately 46° at the far range of the swath. The Shuttle Radar Topography Mission (SRTM) 30m Digital Elevation Model (DEM) was used for topographic phase removal during interferometric processing. A flowchart summarizing the detailed methodology employed in this study is provided in Fig. 2.

OGCSBN_2025_v41n3_605_4_f0001.png 이미지

Fig. 2. Methodology flowchart.

3.2. Data Processing and Analysis

The Sentinel Application Platform(SNAP), an open-source software developed by ESA, was used to preprocess the Sentinel-1 SAR imagery data, which was processed using a combination of SNAP and Stanford Method for Persistent Scatterers(StaMPS), a MATLAB-based processing tool used for PS-InSAR first introduced by (Hooper et al., 2007). It exploits stable radar reflectors on Earth’s surface to detect millimeter-level ground deformation. Its key features include phase-based selection of persistent scatterers via spatial phase correlation, integrating multi-temporal and small-baseline interferometry for atmospheric correction, and enhanced stability.

To reduce processing time, the Terrain Observation with Progressive Scans(TOPS)split tool was used to extract the study area in the IW2 sub-swath within bursts 2, 3, and 4. The orbit files were updated for each image to improve geometric accuracy. A batch processing tool was used to automate the above process for all 58 datasets. The InSAR Stack overview tool uses the acquisition date and perpendicular and temporal baseline information to select a reference image. The reference image was chosen to achieve the highest possible coherence within the interferometric stack and a low perpendicular baseline dispersion. 14 August 2022 was selected as a reference image forthe analysis (Fig. 3).

OGCSBN_2025_v41n3_605_4_f0002.png 이미지

Fig. 3. Plot showing perpendicular and temporal baselines selected for PS-InSAR processing.

Co-registration is performed in the next step to align the pixels of the secondary images to the reference images with sub-pixel accuracy. Interferograms were generated from the co-registered stack, and both the interferogram and co-registered stack were then debursted and exported to StaMPS format for PS-InSAR processing. The interferometric phase of two SAR acquisitions is defined by Eq. (1).

ø = øflat + øtopo + ødisp + øatm + øorbit + ønoise + øflat + øLAE       (1)

where øflat represents the flat earth phase, øtopo is the phase component related to topography (surface elevation), both of which are removed in the preprocessing stage. ødisp represents the phase variation resulting from the displacement of the pixel along the satellite’s LOS direction, øatm denotes the atmospheric influence on the phase signal, øorbit refers to errors related to satellite orbital inaccuracies, and ønoise is noise due to sensor specifications like variability inscattering and thermal disturbances, øLAE represents errors caused by error in the look angle.

The first step in the process involves converting the exported data into formats required for PS processing. The amplitude-based method was applied to detect the persistent scatterer based on an evaluation of pixel amplitude stability across the time series. Scatterers with an amplitude dispersion of less than 0.4 were chosen as PS points. The equation for amplitude dispersion is given in Eq. (2).

\(\begin{align}\text {Amplitude Dispersion}=\frac{\sigma_{A}}{\mu_{A}}\end{align}\)       (2)

where σa is the amplitude’s standard deviation and μa is the amplitude’s mean. The phase noise of each candidate pixel is then estimated for each interferogram. Eligible candidate pixels were identified based on the noise characteristics. The next step involves a weeding process, which is performed to eliminate noisy candidates that neighbouring elements or scatterers may influence. The spatially uncorrelated look-angle error is then quantified and removed from thewrapped phase. This error may arise from topography variations or incidence angle differences. Phase unwrapping, an essential step in this process, is performed to remove phase ambiguities and obtain continuous phase values. A spatially correlated look angle error was subsequently estimated and corrected using the 30-meter resolution SRTM DEM, which provided detailed topographic information.

The Toolbox for reducing InSAR noise (TRAIN) package was employed to rectify the data for the tropospheric delay. Linear phase-based correction methods provided by the toolbox were applied to account for tropospheric delays. Once the corrections were applied, the output data were exported as vector shapefiles compatible with Geographic Information Systems (GIS) for visualisation and further analysis. These shapefiles contain the coordinates of each persistent scatterer(PS) point, along with an attribute table that provides the average velocity derived from least-squares regression analysis, as well as a time series of line-of-sight displacements (Mancini et al., 2021).

The PS points only show the displacement in the line-of-sight direction. Due to the unavailability of ascending track images of the study area, the vertical displacement, which shows uplift or subsidence, was obtained using the following Eq. (3):

\(\begin{align}d_{v e r}=\frac{d_{l o s}}{\cos \theta}\end{align}\)       (3)

where θ is the incidence angle and dver and dlos are the displacements in the vertical and line-of-sight directions, respectively

3.3. Ancillary Data

Ancillary data are important in analysing land subsidence results and understanding the contributing factors. High-resolution satellite imagery of Kakinada was downloaded from Planet Labs, and a detailed land cover map was prepared using object-based image analysis in the eCognition software. Groundwater levels were acquired from the India WRIS portal. Geological and lithological maps were sourced from the Geological Survey of India’s Bhukosh portal to understand the surface and subsurface conditions and their influence on subsidence. All thematic layers were integrated with the ArcGIS Pro environment for further analysis, and Google Earth historical images were used to observe changes in the region.

4. Results and Discussion

4.1. Rates of Deformation

This study employed the SNAP-StaMPS-based Persistent Scatterer InSAR methodology to process 58 descending track images captured by the Sentinel-1A satellite between 2020 and 2024. A total of 39,958 PS pixels were generated with a density of 162 scatterers per km². This corresponds to an average density of approximately 162 PS/km². However, the density is not uniform across the region. Higher PS concentrations were observed in the urban core due to the abundance of stable built-up structures, while lower densities occurred in surrounding semi-rural and rural areas in the north and south of the city. The deformation analysis revealed that the central region of the city experienced notable subsidence, with a transition toward uplift in the northwestern part of the study area. The vertical deformation rate ranged from –26 to +14 mm/yr, with negative values representing subsidence and positive values representing uplift. These values were derived by converting LOS displacements to vertical deformation using the incidence angle. The mean deformation rate across all PS points was approximately –2.8 mm/yr. The main subsidence areas were found to be the city center, Indrapalem, Chollangi, Gurajanapalli, and the Industrial area near the coast, where the average subsidence was approximately 12 mm/yr. Uplift was observed in the northern and northwestern settlements of Unduru, Panasapadu, Panduru, and Nemam at a rate of 4 mm/year.

OGCSBN_2025_v41n3_605_6_f0001.png 이미지

Fig. 4. Mean vertical deformation velocity map between April 2020 and March 2024.

4.2.Relationship between Land Use and Subsidence Rates

Land use plays a significant role in influencing land subsidence because various human activities and land management practices can alter the natural balance of the subsurface environment. For instance, urbanisation often involves extensive excavation, groundwater extraction, and soil compaction due to heavy infrastructure, which can lead to increased subsidence rates, particularly in deltas with unconsolidated sediments (Haq and Milliman, 2023). Agricultural practices, such as irrigation and drainage of wetlands, can also contribute to land subsidence by altering water tables(Chung, 2019).To understand the subsidence in Kakinada, a high-resolution land cover map was prepared using data from Planet Labs at a resolution of 3 m. The Object Based Image Analysis (OBIA) method was employed for the preparation of the land cover map using eCognition software. Multi-resolution segmentation was performed with optimized parameters (scale: 50, shape: 0.2, compactness: 0.5) determined through multiple trial iterations. The Near-Infrared (NIR) band is given a weight of 2 to enhance vegetation discrimination. Following segmentation, ten land cover classes were defined considering the features of the study area. Training samples for each class were selected from the segmented image objects by visual inspection of high-resolution Google Earth imagery and knowledge gained from field visits to the study area. The Nearest Neighbour classification algorithm was applied using spectral, spatial, and contextual properties as input variables. Post-classification improvement included manual correction of misclassified segments, merging objects belonging to the same category, and exporting. The final classified map attained an overall accuracy of 91% and a Kappa coefficient of 0.88.

Analysis of the classified map shows the distribution of various land-use categories within the study region (Fig. 5). Croplands represent regions primarily utilised for agricultural activities, encompassing an extensive area of 108 square kilometres and signifying a notable agriculturalfootprint.Plantationswere classified as areas with cultivated trees, while Natural Vegetation, spanning 13 square kilometres, comprises trees, shrubs, and mangroves, which form sensitive ecosystems in the region. High-density builtup areas, covering 55 square kilometres, are characterised by a high concentration of buildings and infrastructure with minimal green cover, exacerbating subsidence by reducing groundwater recharge and depleting aquifers over time (Shastri et al., 2023). Low-density built-up areas, accounting for 21 square kilometres, have a lower density of buildings. In contrast, Built-up with Tree Cover, extending over 8 square kilometres, includes regions where significant tree cover within urban spaces mitigates subsidence risks by promoting rainwater infiltration and aiding groundwater recharge.

OGCSBN_2025_v41n3_605_7_f0001.png 이미지

Fig. 5. Land use map of the study area.

Other identified classes include aquaculture, which spans 15 square kilometres and highlights the areas used for fish and shrimp farming, and saltpans, which are primarily used for salt production, a major occupation in some coastal areas. Bare Ground, with little to no vegetation cover, and Waterbodies, such as rivers, lakes, and other water sources, each account for 8 square kilometres. Together, these land use classes reflect a heterogeneous landscape, with a dominant agricultural presence alongside significant urban development and smaller specialised land uses contributing to the region’s complex land use mosaic.

To analyse the relationship between ground deformation and land cover type, PS-InSAR point data were spatially overlaid onto the land cover map in ArcGIS, and the deformation points for each land use class were extracted. Subsequently, box plots were generated for each class, visualising the distribution of deformation values (e.g., subsidence rates) within each land cover type (Fig. 6). Box plots effectively visualise the distribution of subsidence values within each land cover class, allowing for easy comparison of deformation patterns across different land use types. Boxplots revealed potential correlations between specific land cover types and varying degrees of subsidence susceptibility by comparing medians, ranges, and identifying outliers.

OGCSBN_2025_v41n3_605_7_f0002.png 이미지

Fig. 6. Boxplot showing the distribution of deformation rates across different land use classes.

The deformation patterns across land use classes exhibited distinct behaviours in terms of subsidence, uplift, and variability. The high-density built-up areas experienced high deformation rates, as shown in the plots. In addition, the deformation rates vary over the study area, likely owing to different densities of buildings, loads, and other localised factors.

Among the more subsidence-prone classes, saltpans experienced significant subsidence, likely because of soil compaction and water extraction for salt production (Higgins et al., 2013).Around 1,000 acresin the peripherals ofKakinada, namely Gurajanapalli, Coringa, Chollangi, and nearby villages, where salt production occurs between December and July. Aquaculture areas also exhibited high subsidence, likely due to heavy groundwater extraction, and low-density built-up areas, croplands, and plantations exhibited high variability in deformation. In contrast, built-up areas with tree cover and aquaculture areas were more uniform. The most intense subsidence was observed in highdensity built-up and saltpan and aquaculture areas, while vegetation and plantation classes were stable and showed some uplift. Soil compaction due to building loads and groundwater extraction was a key driver of subsidence in the area, although the spatial distribution varied.

4.3.Impact of Groundwater Level Changes

Groundwater extraction has been recognized as the principal cause of land subsidence in the majority of cities worldwide (Bakr, 2015). As the population increases in cities, the demand for water increases, causing excessive pumping of groundwater and leading to compaction of the underlying sediments. The unconsolidated sediments on which the study area is situated make it particularly vulnerable. Even though more than 10 groundwater observation points were present across the study area, data from only four locations were available forthe study period; therefore, the groundwaterlevels at these locations between 2020 and 2024 were collected, and changes in water levels were compared to deformation in the surrounding area. The drop in water levels coincided with subsidence in the location, indicating a positive relationship between deformation and change in groundwater levels.

Table 1. Groundwater levelchanges with respect to LOS deformation rates in the study area

OGCSBN_2025_v41n3_605_8_t0001.png 이미지

5. Main Subsidence Centers

5.1. City Center

Fig. 7 shows time series deformation plots of selected locations in the urban center of Kakinada, a region exhibiting significant deformation with an annual average of up to 12 mm. The area also has a very high population density of around 9,000/km2 , contributing to high groundwater extraction. The soil in the area is brown fine sand, which has a loose granular structure and is highly susceptible to compaction under loads. Owing to the very limited permeable surfaces, the infiltration of rainwater is also very low. Plots (b) and (c) show significant deformation in the Suryanarayanapuram area of Kakinada, one of the region’s oldest settlements and a major residential and commercial hub. The highest subsidence of approximately 95 mm was found in Dummulapeta, from which plot (d) is highlighted. Most of the buildings in this region were built on land reclaimed from the adjoining mangroves and mudflats, which resulted in subsidence. Plot (e) exhibited deformation of around 75 mm, where the mechanical vibrations caused by the railway over the softsoil may have contributed to the deformation. Plot(f)shows deformation over a multi-storied apartment complex constructed over the mangroves. The findings highlight the intense urbanization and land reclamation as driving factors of the observed patterns.

OGCSBN_2025_v41n3_605_9_f0001.png 이미지

Fig. 7. Deformation map of the central urban area and time series velocities of selected locations.

5.2.Industrial Area near the Port

Fig. 8 shows the Kakinada anchorage port area. Most warehouses and structures to the north of the port were constructed on land acquired by clearing mangroves in the area, which are characterised by black clays that are highly compressible. The deformation data indicate a stark contrast in the subsidence behaviour between the port structures and warehouses farther from the port. Plots (b), (e), and (f) show deformation patterns over warehouses in the area exhibiting deformation rates from 8 to 15mm/yr. Plots (c) and (d), which belong to the main port, showed relative stability, which can be attributed to their construction on deep piles, mitigating the impact of underlying soil conditions. Due to their proximity to the coast and low elevation, many structures in the region are vulnerable to storm surges and rising sea levels, which is further amplified by the observed deformation.

OGCSBN_2025_v41n3_605_10_f0001.png 이미지

Fig. 8. Deformation map of the port area and time series velocities at selected locations.

6. Conclusions

Our research examined land subsidence in Kakinada between March 2020 and March 2024 utilising PS-InSAR methodology. The findings revealed a cumulative deformation ranging from –90 mm to +30 mm over the four years. The deformation map highlighted significant surface deformation acrossthe study region, with the city centre exhibiting more pronounced deformation than its northern and southern regions. Areas with high-density construction displayed severe deformation, suggesting that building loadwas a contributing factor. Furthermore, subsidence was noted in regions previously reclaimed from mangrove forests. Groundwater with drawal was also identified as a key driver of subsidence in Kakinada. Because satellite images from the ascending trackwere unavailable for our study area, we were limited to using only descending track images. As a result, our analysis could only measure ground movements along the satellite’s line-of-sight direction, rather than being able to separate vertical and horizontal components of displacement. Nevertheless, this investigation represents a crucial initial step towards understanding and addressing the effects of land subsidence in Kakinada.

Acknowledgments

The authors would like to express their sincere gratitude to the European Space Agency (ESA) Copernicus Open Access Hub for providing freely accessible Sentinel-1 Synthetic Aperture Radar(SAR) data. We also acknowledge the India-WRIS (Water Resources Information System) portal, developed by the Central Water Commission(CWC) and Indian Space Research Organisation (ISRO), for providing groundwater data essential to this study. Additionally, we extend our thanks to Planet Labs for granting access to high-resolution satellite imagery.

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

References

  1. Agarwal, V., Kumar, A., Gee, D., Grebby, S., Gomes, R. L., and Marsh,S.,2021.Comparative study of groundwater-induced subsidence for London and Delhi using PSInSAR. Remote Sensing, 13(23), 4741. https://doi.org/10.3390/rs13234741
  2. Ahmad, A., Wang, C., Tang, Y., Sultan, M., Falak, A., Duan, W., and Wang, J., 2022. SAR-based subsidence monitoring and assessment of the factors involved in the occurrence of subsidence, Lahore City. Journal of Resources and Ecology, 13(5), 826–841. https://doi.org/10.5814/j.issn.1674-764x.2022.05.007
  3. Baek, J., Kim, S. W., Park, H. J., Jung, H. S., Kim, K. D., and Kim, J.W., 2008. Analysis of ground subsidence in coal mining area using SAR interferometry. Geosciences Journal, 12(3), 277–284. https://doi.org/10.1007/s12303-008-0028-3
  4. Bagheri-Gavkosh, M., Hosseini, S. M.,Ataie-Ashtiani, B., Sohani, Y., Ebrahimian, H., Morovat, F., and Ashrafi, S., 2021. Land subsidence:Aglobal challenge. Science of The Total Environment, 778, 146193. https://doi.org/10.1016/j.scitotenv.2021.146193
  5. Bakr, M., 2015. Influence of groundwater management on land subsidence in deltas. Water Resour Manage, 29(5), 1541–1555. https://doi.org/10.1007/s11269-014-0893-7
  6. Biggs, J., and Wright, T. J., 2020. How satellite InSAR has grown from opportunistic science to routine monitoring over the last decade. Nature Communications, 11, 3863. https://doi.org/10.1038/s41467-020-17587-6
  7. Cao, J., Gong, H., Chen, B., Shi, M., Zhou, C., Lei, K., Yu, H., and Sun, Y., 2021. Land subsidence in Beijing's sub-administrative center and its relationship with urban expansion inferred from Sentinel-1/2 observations.Canadian Journal of Remote Sensing, 47(6), 802–817. https://doi.org/10.1080/07038992.2021.1964944
  8. Carnec, C., and Delacourt, C., 2000. Three years of mining subsidence monitored by SAR interferometry, near Gardanne, France. Journal of Applied Geophysics, 43(1), 43–54. https://doi.org/10.1016/S0926-9851(99)00032-4
  9. Champenois, J., Pinel, V., Baize, S.,Audin, L.,Jomard, H., Hooper, A., Alvarado, A., and Yepes, H., 2014. Large-scale inflation of Tungurahua volcano (Ecuador) revealed by Persistent Scatterers SAR interferometry. Geophysical Research Letters, 41(16), 5821–5828. https://doi.org/10.1002/2014GL060956
  10. Chung,C.-K., 2019. The exploration of relationship between land subsidence and landscape transformation. Natural Hazards, 97(3), 1051–1068. https://doi.org/10.1007/s11069-019-03685-2
  11. Colesanti, C., and Wasowski, J., 2006. Investigating landslides with space-borne Synthetic Aperture Radar(SAR) interferometry. Engineering Geology, 88(3–4), 173–199. https://doi.org/10.1016/j.enggeo.2006.09.013
  12. Crosetto, M., Monserrat, O., Cuevas-González, M., Devanthéry, N., and Crippa, B., 2016. Persistent Scatterer Interferometry: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 115, 78–89. https://doi.org/10.1016/j.isprsjprs.2015.10.011
  13. Erkens, G., Bucx, T., Dam, R., De Lange, G., and Lambert, J., 2015. Sinking coastal cities. Proceedings of IAHS, 372, 189–198. https://doi.org/10.5194/piahs-372-189-2015
  14. Ferretti, A., Prati, C., and Rocca, F., 2000. Nonlinear subsidence rate estimation using permanent scatterers in differential SAR interferometry. IEEE Transactions on Geoscience and Remote Sensing, 38(5), 2202–2212. https://doi.org/10.1109/36.868878
  15. Ferretti, A., Prati, C., and Rocca, F., 2001. Permanent scatterers in SAR interferometry. IEEE Transactions on Geoscience and Remote Sensing, 39(1), 8–20. https://doi.org/10.1109/36.898661
  16. Fielding, E. J., Blom, R. G., and Goldstein, R. M., 1998. Rapid subsidence over oil fields measured by SAR interferometry. Geophysical Research Letters, 25(17), 3215–3218. https://doi.org/10.1029/98GL52260
  17. Fruneau, B., Achache, J., and Delacourt, C., 1996. Observation and modelling of the Saint-Étienne-de-Tinée landslide using SAR interferometry. Tectonophysics, 265(3–4), 181–190. https://doi.org/10.1016/s0040-1951(96)00047-9
  18. Gambolati, G., Teatini, P., and Ferronato, M., 2005. 158 Anthropogenic land subsidence. Encyclopedia of Hydrological Sciences, 17. https://doi.org/10.1002/0470848944.hsa164b
  19. Garg, S., Motagh, M., Indu, J., and Karanam, V., 2022. Tracking hidden crisisin India's capital from space:Implications of unsustainable groundwater use. Scientific Reports, 12(1), 1–17. https://doi.org/10.1038/s41598-021-04193-9
  20. Gupta, A., Asopa, U., and Bhattacharjee, R., 2019. Land subsidence monitoring in Jagadhri City using Sentinel-1 data and DInSAR processing. Proceedings, 24(1), 25. https://doi.org/10.3390/iecg2019-06230
  21. Hakim, W.L., Achmad, A.R., and Lee, C.W., 2020. Land subsidence susceptibility mapping in Jakarta using functional and meta‐ensemble machine learning algorithms based on time‐series InSAR data. Remote Sensing, 12(21), 3627. https://doi.org/10.3390/rs12213627
  22. Hallegatte, S., Green, C., Nicholls, R. J., and Corfee-Morlot, J., 2013. Future flood losses in major coastal cities. Nature Climate Change, 3(9), 802–806. https://doi.org/10.1038/nclimate1979
  23. Haq, B., and Milliman, J., 2023. Perilous future for river Deltas. GSA Today, 33(10), 4–12. https://doi.org/10.1130/GSATG566A.1
  24. Higgins, S., Overeem, I., Tanaka, A., and Syvitski, J. P. M., 2013. Land subsidence at aquaculture facilities in the Yellow River delta, China. Geophysical Research Letters, 40(15), 3898–3902. https://doi.org/10.1002/grl.50758
  25. Hooper, A., Segall, P., and Zebker, H., 2007. Persistent scatterer interferometric synthetic aperture radar for crustal deformation analysis, with application to Volcán Alcedo, Galápagos. Journal of Geophysical Research: Solid Earth, 112(B7). https://doi.org/10.1029/2006JB004763
  26. Ingebritsen, S. E., and Galloway, D. L., 2014. Coastal subsidence and relative sea levelrise. Environmental Research Letters, 9(9), 091002. https://doi.org/10.1088/1748-9326/9/9/091002
  27. Kandregula, R. S., Kothyari, G. C., Swamy, K. V., Kumar Taloor, A., Lakhote, A., Chauhan, G., et al., 2022. Estimation of regional surface deformation post the 2001 Bhuj earthquake in the Kachchh region, Western India using RADAR interferometry. Geocarto International, 37(18), 5249–5277. https://doi.org/10.1080/10106049.2021.1899299
  28. Kumar, L. R., Vaka, D. S., and Rao, Y. S., 2017. Mapping land subsidence of Krishna–Godavari Basin using persistent scatterer interferometry technique. In Proceedings of the 38th Asian Conference on Remote Sensing (ACRS 2017), New Delhi, India, Oct. 23–27, vol. 4, p. 2490. https://acrsaars.org/proceeding/ACRS2017/ID_753_1652/840.pdf
  29. Lasserre, C., Peltzer, G., Crampé, F., Klinger, Y., Van der Woerd, J., and Tapponnier, P., 2005.Coseismic deformation of the 2001 Mw = 7.8 Kokoxili earthquake in Tibet, measured by synthetic aperture radar interferometry. Journal of Geophysical Research: Solid Earth, 110(B12). https://doi.org/10.1029/2004JB003500
  30. Mancini, F., Grassi, F., and Cenni, N., 2021. A workflow based on snap–stamps open‐source tools and GNSS data for PSI-based ground deformation using dual‐orbit Sentinel-1 data: Accuracy assessment with error propagation analysis. Remote Sensing, 13(4), 753. https://doi.org/10.3390/rs13040753
  31. Moghaddam, N. F., Sahebi, M. R., Matkan, A. A., and Roostaei, M., 2013. Subsidence rate monitoring ofAghajari oil field based on Differential SAR Interferometry. Advances in Space Research, 51(12), 2285–2296. https://doi.org/10.1016/j.asr.2013.01.023
  32. Pavez, A., Remy, D., Bonvalot, S., Diament, M., Gabalda, G., Froger, J. L., Julien, P., Legrand, D., and Moisset, D., 2006. Insight into ground deformations at Lascar volcano (Chile) from SAR interferometry, photogrammetry and GPS data: Implications on volcano dynamics and future space monitoring. Remote Sensing of Environment, 100(3), 307–320. https://doi.org/10.1016/j.rse.2005.10.013
  33. Rao, G. N., 2001. Sedimentation, stratigraphy, and petroleum potential of Krishna-Godavari Basin, East Coast of India. AAPG Bulletin, 85(9), 1623–1643. https://doi.org/10.1306/8626CCDF-173B-11D7-8645000102C1865D
  34. Reshma, K. N., Agrawal, R., Ramakrishnan, R., Sreejith, K. M., and Rajawat, A. S., 2023. Land subsidence studies in the Godavari Delta regions of the East coast of India using ALOS and Sentinel 1 data. Ecological Informatics, 78, 102373. https://doi.org/https://doi.org/10.1016/j.ecoinf.2023.102373
  35. Shastri, A., Sreejith, K. M., Rose, M. S., Agrawal, R., Sunil, P. S., Sunda, S., and Chaudhary, B. S., 2023. Two decades of land subsidence in Kolkata, India revealed by InSAR and GPS measurements: implications for groundwater management and seismic hazard assessment. Natural Hazards, 118(3), 2593–2607. https://doi.org/10.1007/s11069-023-06107-6
  36. Takagi, H., Fujii, D., Esteban, M., and Yi, X., 2017. Effectiveness and limitation of coastal dykes in Jakarta: The need for prioritizing actions against land subsidence. Sustainability, 9(4), 619. https://doi.org/10.3390/su9040619
  37. Tofani, V., Raspini, F.,Catani, F., andCasagli, N., 2013. Persistent Scatterer Interferometry (PSI) technique for landslide characterization and monitoring. Remote Sensing, 5(3), 1045–1065. https://doi.org/10.3390/rs5031045
  38. Tripathi, A., Reshi, A. R., Moniruzzaman, M., Rahaman, K. R., Tiwari, R. K., and Malik, K., 2022. Interoperability of CBand Sentinel-1 SAR and GRACE satellite sensors on PSInSAR-based urban surface subsidence mapping of Varanasi, India. IEEE Sensors Journal, 22(21), 21071–21081. https://doi.org/10.1109/JSEN.2022.3208117
  39. Yan, X., Yang, T., Xu, Y., Tosi, L., Stouthamer, E., Andreas, H., et al., 2020. Advances and practices on the research, prevention and control of land subsidence in coastal cities. Acta Geologica Sinica‐English Edition, 94(1), 162–175. https://doi.org/10.1111/1755-6724.14403