Search

Introduction: Surface deformation in the Three Gorges Reservoir area poses significant threats to infrastructure and safety due to complex geological and hydrological factors. Despite existing studies, systematic exploration of long-term deformation characteristics and their driving mechanisms remains limited. This study combines SBAS-InSAR technology and machine learning to analyze and predict surface deformation in Fengjie County, Chongqing, China, between 2020 and 2022, focusing on riverside urban ground, riverside road slopes, and ancient landslides in the reservoir area. Methods: SBAS-InSAR technology was applied to 36 Sentinel-1A images to monitor surface deformation, complemented by hydrological and meteorological data. Machine learning models—Random Forest (RF), Extremely Randomized Trees (ERT), Gradient Boosting Decision Tree (GBDT), Support Vector Regression (SVR), and Long Short-Term Memory (LSTM)—were evaluated using six metrics, including RMSE, R2, and SMAPE, to assess their predictive performance across diverse geological settings. Results: Deformation rates for riverside urban ground, road slopes, and ancient landslides were −3.48 ± 2.91 mm/yr, −5.19 ± 3.62 mm/yr, and −6.02 ± 4.55 mm/yr, respectively, with ancient landslides exhibiting the most pronounced deformation. A negative correlation was observed between reservoir water level decline and subsidence, highlighting the influence of seasonal hydrological adjustments. Urbanization and infrastructure development further exacerbated deformation processes. Among the models, LSTM demonstrated superior predictive accuracy but showed overestimation trends in ancient landslide areas. Discussion: Reservoir water level adjustments emerged as a critical driver of subsidence, with rapid water level declines leading to increased pore pressure and soil compression. Seasonal effects were particularly evident, with higher subsidence rates during and after the rainy season. Human activities, including urbanization and road construction, significantly intensified deformation, disrupting natural geological conditions. Progressive slope failure linked to road expansion underscored the long-term impacts of engineering activities. For ancient landslides, accelerated deformation patterns were linked to prolonged drought and reservoir-induced hydrological changes. While LSTM models showed high accuracy, their limitations in complex geological settings highlight the need for hybrid approaches combining machine learning with physical models. Future research should emphasize developing integrated frameworks for long-term risk assessment and mitigation strategies in reservoir environments. Conclusions: This study provides new insights into the complex surface dynamics in the Three Gorges Reservoir area, emphasizing the interplay of hydrological, geological, and anthropogenic factors. The findings highlight the need for adaptive management strategies and improved predictive models to mitigate subsidence risks. [ABSTRACT FROM AUTHOR]

The long-term and extensive exploration of energy sources, such as oil and gas, has led to the large-scale surface deformation of the Songnen Plain in China. To study the impact of oil and gas exploitation on surface deformation in the Songnen Plain, we selected a typical oil exploitation area (OEA) to conduct an analysis and prediction using multimodal data, Interferometry Synthetic Aperture Radar (InSAR) technology, and Autoregressive Integrated Moving Average (ARIMA) – Back Propagation (BP) neural network model. The study revealed a subsidence trend in the OEA using Small Baseline Subset InSAR (SBAS-InSAR), indicating a deformation velocity of −2.4 mm/year throughout the monitoring period. Additionally, we analysed the factors impacting surface deformation by considering six factors: oil and gas exploitation, stratum structure, groundwater, surface water, meteorological conditions, and seismic activity. We found that oil and gas exploitation production had a significant negative correlation with surface deformation, with a distance effect of about 800 m. The stratum structure causing surface subsidence was mainly located in the silty clay layers. Groundwater, surface water, and cumulative precipitation were all negatively correlated with surface deformation, while temperature and seismic activity showed no clear relationship. Furthermore, we proposed a novel ARIMA-BP model to predict the deformation trend of OEA and found that the prediction accuracy of this model is higher than the single models. This study can provide decision support for analysing surface deformation and implementing disaster prevention measures in the OEA, thereby contributing significantly to the sustainable development of society. [ABSTRACT FROM AUTHOR]

Interferometric Synthetic Aperture Radar (InSAR) technology has emerged as a vital tool for monitoring surface deformation due to its high accuracy and spatial resolution. With the rapid economic development of Nanchang, extensive infrastructure development and construction activities have significantly altered the urban landscape. Underground excavation and groundwater extraction in the region are potential contributors to surface deformation. This study utilized Sentinel-1 satellite data, acquired between September 2018 and May 2023, and applied the Permanent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) technique to monitor surface deformation in Nanchang's urban area. The findings revealed that surface deformation rates in the study area range from −10 mm/a to 6 mm/a, with the majority of regions remaining relatively stable. Approximately 99.9% of the monitored points exhibited deformation rates within −5 mm/a to 5 mm/a. However, four significant subsidence zones were identified along the Gan River and its downstream regions, with a maximum subsidence rate reaching 9.7 mm/a. Historical satellite imagery comparisons indicated that certain subsidence areas are potentially associated with construction activities. Further analysis integrating subsidence data, monthly precipitation, and groundwater depth revealed a negative correlation between surface deformation in Region A and rainfall, with subsidence trends aligning with groundwater level fluctuations. However, such a correlation was not evident in the other three regions. Additionally, water level data from the Xingzi Station of Poyang Lake showed that only Region A's subsidence trend closely corresponds with water level variations. We conducted a detailed analysis of the spatial distribution of soil types in Nanchang and found that the soil types in areas of surface deformation are primarily Semi-hydromorphic Soils and Anthropogenic Soils. These soils exhibit high compressibility, making them prone to compaction and significantly influencing surface deformation. This study concludes that localized surface deformation in Nanchang is primarily driven by urban construction activities and the compaction of artificial fill soils, while precipitation also has an impact in certain areas. [ABSTRACT FROM AUTHOR]

IntroductionSurface deformation in the Three Gorges Reservoir area poses significant threats to infrastructure and safety due to complex geological and hydrological factors. Despite existing studies, systematic exploration of long-term deformation characteristics and their driving mechanisms remains limited. This study combines SBAS-InSAR technology and machine learning to analyze and predict surface deformation in Fengjie County, Chongqing, China, between 2020 and 2022, focusing on riverside urban ground, riverside road slopes, and ancient landslides in the reservoir area.MethodsSBAS-InSAR technology was applied to 36 Sentinel-1A images to monitor surface deformation, complemented by hydrological and meteorological data. Machine learning models—Random Forest (RF), Extremely Randomized Trees (ERT), Gradient Boosting Decision Tree (GBDT), Support Vector Regression (SVR), and Long Short-Term Memory (LSTM)—were evaluated using six metrics, including RMSE, R2, and SMAPE, to assess their predictive performance across diverse geological settings.ResultsDeformation rates for riverside urban ground, road slopes, and ancient landslides were −3.48 ± 2.91 mm/yr, −5.19 ± 3.62 mm/yr, and −6.02 ± 4.55 mm/yr, respectively, with ancient landslides exhibiting the most pronounced deformation. A negative correlation was observed between reservoir water level decline and subsidence, highlighting the influence of seasonal hydrological adjustments. Urbanization and infrastructure development further exacerbated deformation processes. Among the models, LSTM demonstrated superior predictive accuracy but showed overestimation trends in ancient landslide areas.DiscussionReservoir water level adjustments emerged as a critical driver of subsidence, with rapid water level declines leading to increased pore pressure and soil compression. Seasonal effects were particularly evident, with higher subsidence rates during and after the rainy season. Human activities, including urbanization and road construction, significantly intensified deformation, disrupting natural geological conditions. Progressive slope failure linked to road expansion underscored the long-term impacts of engineering activities. For ancient landslides, accelerated deformation patterns were linked to prolonged drought and reservoir-induced hydrological changes. While LSTM models showed high accuracy, their limitations in complex geological settings highlight the need for hybrid approaches combining machine learning with physical models. Future research should emphasize developing integrated frameworks for long-term risk assessment and mitigation strategies in reservoir environments.ConclusionsThis study provides new insights into the complex surface dynamics in the Three Gorges Reservoir area, emphasizing the interplay of hydrological, geological, and anthropogenic factors. The findings highlight the need for adaptive management strategies and improved predictive models to mitigate subsidence risks.

The Changbaishan volcano is recognized as one of China's most potentially dangerous active volcanoes. In response to the unrest event of 2002, GPS technology was utilized to monitor the surface deformation it induced. However, the understanding of the volcano's dynamics and the intricacies of GPS data at the time were limited, which affected the quality of the conclusions. For example, the commencement of GPS observations followed the escalation of seismic activities, leading to the loss of some deformation information. Moreover, observational noise in the GPS coordinate sequences introduced oscillations in the evolution of deformation characteristics. To overcome these limitations, we have developed an advanced GPS data processing methodology. This includes the establishment of a meticulous three-tier control network, the employment of high-precision geophysical models in GAMIT/GLOBK software, the creation of a volcanic regional reference frame, and the formulation of a sophisticated motion model for monitoring stations. With these approaches, we have captured the maximum surface deformation caused by the unrest and have re-evaluated the volume change (25.95× 106 m3/year) of the magma chamber based on Mogi model, yielding results that significantly surpass the mean of previous estimates of 8.58 × 106 m3/year and enhancing our understanding of the magma chamber's dimensions. Additionally, the surface deformation following the unrest displayed a pattern of continuous decay, which is in contrast to the seismic activity that initially rose and then declined, peaking notably after the surface deformation's peak. Considering the geological context of the volcano's formation, we have also provided an extensive dataset of GPS velocity fields. We have preliminarily discussed the possible relationship between the subduction of the Pacific Plate and the unrest in 2002, as well as the recent low-level unrest in 2021, acknowledging that this hypothesis requires further confirmation through stress modeling related to the disturbances. The deformation data resulting from the unrest, as well as the background deformation caused by plate subduction presented in this study, provide essential data constraints for the construction of subsequent stress models. • Capture the 2002 unrest's maximum deformation to better understand the magma chamber's dimensions • Deformation shows steady decay and seismic activity surges post-deformation peak • Pacific Plate subduction possibly linked to 2002 unrest and 2021 low-level activity [ABSTRACT FROM AUTHOR]