The Impact of Temporal Decorrelation over Forest Terrain in Polarimetric SAR Interferometry (original) (raw)
Related papers
Quantification and compensation of temporal decorrelation effects in polarimetric SAR interferometry
2012 IEEE International Geoscience and Remote Sensing Symposium, 2012
Temporal decorrelation is a critical issue for a successful Pol-InSAR inversion in case of repeat-pass SAR data, as provided by conventional satellite or airborne SAR systems. This paper proposes estimation and compensation of temporal decorrelation effects by using multi-baseline Pol-InSAR data. A new approach to quantify different temporal decorrelation levels (one for volume and the other for the ground layer) is performed without resort to the special case of zero spatial baseline interferograms. Both temporal decorrelation coefficients were separately estimated at temporal baselines ranging from 1 to 15 days and compared to height inversion errors caused by them.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2013
Temporal decorrelation is the most critical issue for the successful inversion of polarimetric SAR interferometry (Pol-InSAR) data acquired in an interferometric repeat-pass mode, typical for satellite or lower frequency airborne SAR systems. This paper provides a quantitative estimation of temporal decorrelation effects at L-band for a wide range of temporal baselines based on a unique set of multibaseline Pol-InSAR data. A new methodology that allows to quantify individual temporal decorrelation components has been developed and applied. Temporal decorrelation coefficients are estimated for temporal baselines ranging from 10 min to 54 days and converted to height inversion errors caused by them.
Multibaseline Polarimetric Sar Interferometry Forest Height Inversion Approaches
2011
Polarimetric SAR interferometry (Pol-InSAR) is a radar remote sensing technique that is sensitive to the vertical distribution of scattering processes in volumes. The Random Volume over Ground (RVoG) model is a powerful tool used to invert forest height from Pol-InSAR data. But Pol-InSAR inversion performance depends critically on uncompensated decorrelation contributions (i.e. temporal decorrelation in repeat pass system) and the height sensitivity of the effective baseline, represented by the vertical wavenumber Z .
Forest Height Estimation by means of Pol-InSAR Limitations posed by Temporal Decorrelation
2009
Polarimetric Synthetic Aperture Radar (SAR) Interferometry (Pol-InSAR) is a radar remote sensing technique, based on the coherent combination of radar polarimetry (Pol-SAR) and SAR interferometry (InSAR) which is substantially more sensitive to structural parameters of forest volume scatterers (e.g. forest) than conventional interferometry or polarimetry alone. However, temporal decorrelation is probably the most critical factor towards a successful implementation of Pol-InSAR parameter inversion techniques in terms of repeat-pass InSAR scenarios. This report focuses on the quantification of the effect of temporal decorrelation at L-band as a function of temporal baseline based on multi-temporal airborne experimental data acquired in the frame of dedicated air-borne experiments. Conclusions on the suitability of ALOS/PalSAR for Pol-InSAR applications are drawn and recommendations for mission characteristics of a potential follow on mission are addressed.
2011
World forests contain a significant amount of carbon that is, in consequence of natural and human induced deforestation and regrowth processes, affected by rapid changes and therefore difficult to quantify in terms of biomass/carbon storage. This uncertainty remains because of the lack of reliable and frequent information of biomass levels on a global scale. Spaceborne Synthetic Aperture Radar (SAR) could provide the required global and temporal coverage of the world's forested areas. Innovative new inversion approaches allow today accurate meassurement of vegetation structure parameters such as forest height and biomass [1][2] . The coherent combination of polarimetric and interferometric SAR at lower frequencies by means of Pol-InSAR is sensitive to the vertical distribution of scattering processes within a resolution cell and can be used for model-based inversion of forest height and structural parameters.
Temporal decorrelation in polarimetric differential interferometry using a ground-based SAR sensor
Proceedings. 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. IGARSS '05., 2005
In this paper, the first results of the Polarimetric Ground-Based SAR system developed at the Technical University of Catalonia (UPC) are presented. A heterogeneous scenario containing different kinds of targets such as low vegetation, forest and urban areas has been chosen for performing a measurement campaign. The monitoring activity has dealt with the observation of the test-site at X-Band for a whole day with a revisiting time of approximately one hour: a temporal sequence of PolDInSAR dataset has been gathered. A first study of temporal decorrelation effects on multi-polarization differential coherence concerning different kinds of targets is here shown. Preliminary results are commented and future activities are proposed.
Comparison of Methods to Derive Forest Height from Polarimetric SAR Interferometry
2008
Six forest tree height inversion methods of polarimetric SAR interferometry were validated using repeat pass E-SAR datasets and the corresponding ground measured forest stand height. We show that SINC, three stage full model and three stage SINC inversions have good performance, and the SINC inversion with the first optimum coherence (OPT1) produces the best result among the eleven coherence types validated. Taking SINC inversion as one example, two inversion correction methods were investigated, RMSE can be reduced below 4.0m after correction. The phase and coherence inversion method achieved the second lowest RMSE among the six inversion methods validated, but the estimated tree height of it has very low correlation with ground truth, and this drawback was found caused by the poor performance of RVoG phase inversion method involved.
2003
Polarimetric Interferometric Synthetic Aperture Radar (SAR) is a recent area of research that has had significant attention from the mid-1990s. This area of research has combined the utility of two SAR technologies: Polarimetric SAR (PolSAR) and Interferometric SAR (InSAR). Polarimetric SAR provides four channels which can be used to determine the polarimetric ellipse, and hence, structural information of the scatterer. Therefore PolSAR is suitable for target recognition and detection applications. InSAR data combines two SAR image data sets acquired from nearly the same perspective. The phase difference between these images provides information about the topography, or changes in the topography between the two image dates. InSAR methods have been used to map terrains, detect environmental changes and determine velocities of moving targets. By combining both technologies, polarimetric InSAR (Pol InSAR) permits distinction between different distributed targets at different elevations. In particular, most current research is investigating the use of this technology for measuring the height of forest, and to help estimate its biomass. Other applications under research include terrain moisture estimation, terrain roughness estimation, and (of more interest in mapping applications) vertical obstruction detection.
Multi-baseline Pol-InSAR Forest Height Estimation in the presence of temporal decorrleation
2010
This paper addresses the effect of temporal decorrelation on the inversion of forest parameters using Pol-InSAR techniques. The modeling of temporal decorrelation and the inversion of single-baseline Pol-InSAR data in the presence of temporal decorrelation is discussed. Model based simulations and experimental multi-temporal fully polarimetric and repeat pass interferometric data from the SIR-C Space shuttle mission are used for the performance analysis of the proposed approach.