Contribution of InSAR coherence images to evaluate surface elevation change detected from multi-temporal photogrammetric imagery - Archive ouverte HAL Accéder directement au contenu
Poster De Conférence Année : 2024

Contribution of InSAR coherence images to evaluate surface elevation change detected from multi-temporal photogrammetric imagery

Résumé

The landscape pattern and topography are affected by natural events and anthropic processes. Detecting and estimating these topographic changes is crucial for understanding the underlying processes (i.e. erosion, morphological changes, vegetation growth, anthropization …). Efficient change detection requires recurrent surveys of the area of interest, effective accuracy to identify and estimate the rate of change, and a reference frame for accuracy assessment (Cook, 2017). The combination of various datasets obtained from different remote sensing approaches, each with specific resolution, enables the 3D monitoring of surface elevation change. The aim of this work is to explore the potential of InSAR data to validate the detected changes in the 3D Earth surface, extracted from the photogrammetric approach. The study area is a watershed, known as Lebna, located in the coastal plain in northeastern Tunisia, in the Cap Bon peninsula. This watershed is prone to soil erosion mostly enhanced by agricultural practices and Mediterranean climate (Desprats et al., 2013; Mekki et al., 2018). Additionally to a long-term agro-hydrological observatory implemented in the Lebna watershed, providing in-situ data (Molenat et al., 2018), the area of interest is also covered by tri-stereo Pléiades images acquired in 2015, 2018, and 2023, and Sentinel-1 SAR interferometric data from 2017 to 2022. The tri-stereo Pléiades images were processed using the Structure from Motion coupled with the Multi View Stereo (SfM-MVS) algorithms for 3D reconstruction performed with Agisoft Metashape Pro v. 1.6.3. The different processing steps follow the Time-SIFT approach introduced by Feurer & Vinatier (2018). The DEM of Differences (DoD), calculated by subtracting the early epoch DEM from the following period DEM, allows detecting geomorphological changes between multi-epoch surveys. The InSAR imagery products were used to compare/validate the results derived from the photogrammetric approach. Interferometric coherence images enable the detection of changes of the surface’s scattering properties. The coherence value approaches zero in areas where ground variation occurs between two acquisitions and modifies the structure of scattering targets, and one where the area is stable (Zebker and Villasenor, 1992). Although erosion is triggered by rainfall events, it is cumulative for long periods. In coherence images, it is expected to be defined as features continuously losing coherence with an expanded temporal baseline. For three coherence images spanning the period from 27/10/2018 to 02/12/2018 the coherence values, for an identified eroded area on the DoD, are gradually decreasing. For newly constructed buildings identified on the DoD, the coherence value abruptly decreases and the decoherence features show structured spatial patterns. However, the interpretation of the coherence variation should be approached carefully. Several components can contribute to its value, such as the viewing geometry, vegetation cover and soil moisture. Further investigation on the decomposition of the coherence according to the above mentioned decorrelation factors is required to explore the changes related to environmental processes such as erosion, by refining the extraction of the erosion information from other signals. To complement erosion detection from coherence images, integrating precipitation dataset helps in understanding the environmental conditions that contribute to soil erosion.
Fichier principal
Vignette du fichier
Poster_NTrICs_2024.pdf (2.25 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04410453 , version 1 (22-01-2024)

Identifiants

  • HAL Id : hal-04410453 , version 1

Citer

Imen Brini, Denis Feurer, Riadh Tebourbi, Damien Raclot, Fabrice Vinatier, et al.. Contribution of InSAR coherence images to evaluate surface elevation change detected from multi-temporal photogrammetric imagery. New trends in Communications, Signal and Image Processing, Jan 2024, Tunis (Cité des Sciences), Tunisia. ⟨hal-04410453⟩
52 Consultations
10 Téléchargements

Partager

Gmail Facebook X LinkedIn More