Assessing Surface Textural Variations on the Piton de la Fournaise Volcano Using L-Band Insar and LIDAR Fusion Study

Interferometric Synthetic Aperture Radar (InSAR) is a powerful tool to monitor deformation in active volcanoes, such as the Piton de La Fournaise (Reunion Island, France). However vegetation and pyrochlast covers could constrain the L-band InSAR coherence and then the reliability for measurement of pre-eruptive surface displacements. To correct this deficiency, we combine normalized airborne LiDAR intensity data with spaceborne InSAR coherence images from ALOS PALSAR L-band acquired over the Piton de la Fournaise in 2008 and 2009, just after the 2007 major eruption. The fusion of the two data sets improves the calculation of coherence and the textural classification of different pyrochlastic and lava flows. For the DESDynI (Deformation, Ecosystem Structure and Dynamics of Ice) mission, such data fusion studies can provide a better analysis of the spatiotemporal variations in InSAR coherence in order to enhance the monitoring of pre-eruptive ground displacements. The LiDAR intensity data are used to improve the accuracy of InSAR-derived estimates of surface elevation and roughness, especially in vegetated areas. The airborne campaigns carried out on overlapping areas of the Piton de la Fournaise cover different types of vegetation and terrain roughness on the central and western part of the volcano. The LiDAR data are first processed to generate an accurate and high-resolution digital terrain model of the volcanic edifice so as to characterize its surface features and to analyze morphological variations during this two-year period. Then normalized LiDAR intensity images are computed and compared to coherence L-band InSAR images for different zones of the volcano to assess the Lidar-inSAR statistical behavior of different lava flows, pyrochlastics, and vegetated surfaces. Empirical models that use normalized LiDAR intensity to correct L-band polarimetric coherence for vegetated surfaces are finally tested. Preliminary results suggest that different A'A and Pahoehoe lava flows have a unique LiDAR-InSAR intensity-coherence function, which remains unchanged even in the case of moderate vegetation cover. Moreover ashes, cinders and spatter cones show a well-distinguished intensity-coherence function for vegetated and bare terrains.