Lava flow mapping and area estimation using radar coherence images
Résumé
Interferometric synthetic aperture radar (InSAR), more specifically calculation of coherence images, can be used to infer changes in the ground surface’s geometry. If these changes come from the emplacement of a lava flow, coherence images can then be used to map the flow. For this purpose, we developed an algorithm which separates the lava flow pixels from the others depending on their median shade. After processing the picture, we get a map and a surface for a given date. By doing this for several dates, we thus can follow the time and space evolution of the lava flow. For the August – October 2015 eruption of Piton de la Fournaise, available InSAR data allowed us to map the lava flow at nine different dates between 08/29/15 and 11/01/15. We estimated the flow’s area for each date; at the end of the eruption, the total surface estimation is 4.251 10⁶ ± 4.5 10⁴ m². This work could be pursued to map active parts of the flow and to estimate the volume of emitted products.