Back to search

University of West Attica

Monitoring topographic subsidence with the implementation of InSAR satellite imagery, and GNSS vector analysis techniques

Abstract

dc:description

Land subsidence is a progressive geohazard with profound implications for infrastructure safety, environmental stability, and land-use planning. This doctoral research addresses the challenge of accurately detecting, monitoring, and predicting land subsidence over extended spatial and temporal scales using remote sensing and machine learning techniques. By integrating multitemporal InSAR datasets with GNSS observations and geospatial factors, the study presents a scalable and transferable framework for land subsidence susceptibility (LSS) mapping, validated over the tectonically and hydrologically complex region of Attica, Greece. The thesis is structured around three major scientific approaches. The first approach focuses on enhancing the monitoring of ground displacement across multiple temporal scales by combining Persistent Scatterer InSAR (PS-InSAR), Small Baseline Subset (SBAS) InSAR, and GNSS data using Kalman filtering. Five case studies—Anavyssos, Corinth, Metamorfosi, Penteli, and Eleonas—were examined, covering both six-year (2015–2021) and one-year (2021) observation periods. The fusion methodology provided smoother, higher-confidence displacement time series, eliminating noise while preserving significant deformation signals. The GNSS Kalman-filtered fusion yielded displacement time series consistent with those from PS-InSAR and SBAS techniques, thereby validating the robustness of the integrated approach. Feature importance and Partial Dependence Plot (PDP) analyses further revealed that proximity to faults and streams were the most dominant factors driving long-term ground deformation, followed by lithology and land use, particularly in fault-affected alluvial plains. These causal relationships were consistently observed across case studies and temporal scales. The second approach introduces an LSS model by integrating SBAS InSAR data with key geospatial factors using machine learning approaches. The study focuses on the Attica prefecture, Greece, and utilizes SBAS InSAR data from 2015 to 2021 to extract ground deformation velocities by classifying them into four susceptibility levels: stable, low, moderate, and high. The susceptibility results indicate that stable zones constitute 58.2% of the study area, followed by low (27.2%), moderate (11.2%), and high susceptibility zones (3.4%), predominantly concentrated in areas undergoing hydrological stress and urbanization. Random Forest (RF) and XGBoost (XGB) models incorporate a comprehensive set of causal factors, including slope, aspect, land use, groundwater level, geology, and rainfall. The evaluation of the models includes accuracy metrics and confusion matrices. The XGB model achieved the highest performance, recording an accuracy of 94%, with well-balanced predictions across all susceptibility classes. Addressing class imbalance during model training improved the recall of minority classes, though with slight tradeoffs in precision. Feature importance analysis identifies proximity to streams, land use, aspect, rainfall, and groundwater extraction as the most influential factors driving subsidence susceptibility. This methodology demonstrates high reliability and robustness in predicting LSS, providing critical insights for land-use planning and mitigation strategies. These findings establish a scalable framework for regional and global applications, contributing to sustainable land management and risk reduction efforts. The third approach explored the spatial transferability of susceptibility models through a crossregional learning framework. The Attica AOI was divided into two subregions—Dataframe 2 (DF2) for model training and testing, and Dataframe 1 (DF1) as an independent region for prediction. The models trained on DF2 were evaluated for their predictive performance in DF1 using both spatial overlays and a 100-meter nearest-neighbor matching algorithm. Among the evaluated models, RF_ADASYN and XGB_ADASYN demonstrated the highest alignment with the full AOI susceptibility reference, particularly in Moderate and High classes—zones of critical geohazard concern. These results affirm the operational readiness of ensemble models combined with adaptive sampling for reliable susceptibility forecasting in unseen, geologically similar areas. This research makes significant contributions to geohazard monitoring by establishing an integrated, multi-sensor displacement analysis pipeline and a transferable machine learning framework for subsidence susceptibility assessment. By linking observed deformation patterns with geomorphological, lithological, and anthropogenic variables, the thesis offers a reproducible and scalable methodology for ground motion mapping in tectonically active and urbanizing regions. The outcomes are particularly valuable for risk-informed urban planning, infrastructure resilience strategies, and early warning systems.

Degree

thesis:*
Grantor dc:publisher
University of West Attica
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Yaragunda, Vishnuvardhan Reddy
  • Γιαραγούντα, Βισνουβαρντάν-Ρέντι

Subjects

dc:subject × 23

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/60294

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
related terms
citation

Yaragunda, Vishnuvardhan Reddy; Γιαραγούντα, Βισνουβαρντάν-Ρέντι. Monitoring topographic subsidence with the implementation of InSAR satellite imagery, and GNSS vector analysis techniques. University of West Attica, 2025. http://hdl.handle.net/10442/hedi/60294