Skip to Main content Skip to Navigation

Synthetic Aperture Radar : Algorithms and Applications in Forests and Urban Areas

Abstract : The objective of this thesis is to exploit Multi-baseline Synthetic Aperture Radar (SAR) for studying the remote sensing of natural scenarios, such as forest structure characterization and land subsidence monitoring. In the case of forested areas, tropical forest structure parameters are derived by Tomography SAR (TomoSAR) technique. For urban areas, Land subsidence is investigated through Interferometry SAR (InSAR) techniques. TomoSAR and InSAR will be treated by using Multi-baseline SAR images over different sites. Prior to tomographic analysis, a phase calibration algorithm is needed to compensate for phase residuals that corrupt the data and influence the focusing of Multi-baseline data. First, a tomographic study has carried out in tropical forest, where the forest characterization was assessed by using SAR tomography at L and P-band. Second, different InSAR techniques have been compared with respect to their performance in monitoring earth’s surface deformation, taking Lebanon as a case study.The first part of the thesis presents the TomoSAR analysis in the tropical forest. A review of phase calibration techniques employed on TomoSAR data is shown. The problem formulation starts with the phase calibration of the data stack that is considered as the main gate to begin with SAR processing algorithms. Thus, the main phase calibration algorithms proposed in the literature are discussed. Two of the most important phase calibration approaches are then described and discussed in detail. The potential of L-band TomoSAR data to characterize tropical forest structure is evaluated. The challenge here is the short wavelength of L-band data, and whether can penetrate tropical forests down to the ground. Tomographic analysis is carried out using L-band UAVSAR data from the AfriSAR campaign conducted over Gabon Lopé Park in February 2016. It was found that L-band TomoSAR was able to penetrate into and through the canopy down to the ground, and thus the canopy and ground layers were detected correctly. Then, monitoring tropical forest structure using SAR tomography at L- and P-band are treated. For this, a comparison of the P- and L-band TomoSAR profiles, Land Vegetation and Ice Sensor (LVIS), and discrete return LiDAR is provided in order to assess the ability for TomoSAR to monitoring and estimating the tropical forest structure parameters for enhanced forest management and to support biomass missions. The L- and P-band's performances for canopy penetration are assessed to determine the underlying ground locations. Additionally, the 3D records for each configuration are compared regarding their ability to derive forest vertical structure.The second part of the thesis tackle the utilization of InSAR techniques in land subsidence monitoring. The idea is to split the estimation of earth's surface deformations into two steps. The first step is to use Maximum Likelihood technique to jointly process Permanent scaterrers and Distributed scaterrers in order to yield the best estimates of interferometric phases. Then, the second step is to separate the contributions to the interferometric phases due to the scene topography and deformation field from those caused by decorrelation noise and atmospheric disturbances. As a case study, an extensive InSAR analysis of Lebanon site is shown, relying on a data-set of 117 Sentinel-1 satellite data acquired over Lebanon between 2015 and 2019, with high temporal resolution (i.e. 6 days).
Document type :
Complete list of metadatas

Cited literature [207 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Thursday, May 14, 2020 - 7:27:07 PM
Last modification on : Friday, October 23, 2020 - 5:02:32 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02579180, version 1



Ibrahim El Moussawi. Synthetic Aperture Radar : Algorithms and Applications in Forests and Urban Areas. Earth Sciences. Université Montpellier; Lebanese international university, 2019. English. ⟨NNT : 2019MONTG078⟩. ⟨tel-02579180⟩



Record views


Files downloads