Change Detection for Mapping Volcanic Ash Fall Using Sentinel-1 Data

CAMILO NARANJO; PABLO EUILLADES; GUILLERMO TOYOS; LEONARDO EUILLADES; GUSTAVO VILLAROSA

Leeds

Workshop; Fringe 2023; 2023

European Space Agency

Inthe face of a volcanic eruption, it is important to know the affected regionsto manage an emergency. Volcanic ashfall affects buildings, vegetation andharvest, water bodies, livestock and population. Aiming to identify and map theregions affected by volcanic ashfall, we adapted and applied a TemporalDecorrelation Model (TDM) [1] using Synthetic Aperture Radar (SAR) C band dataacquired by the Sentinel-1 constellation.Coherencemaps derived from SAR interferometry (InSAR) are useful for change detectionbut these maps alone can be limited for distinguishing the changes resultingfrom the presence of volcanic ash on the ground from changes generated by othercauses such as rain, snow, wind or the seasons. To this end, the implementationof TDM aims to improve our capabilities to discriminate among both kinds ofchanges. Theeruption of Taal volcano (Philippines) on January 12, 2020 was selected as acase study. We prepared a dataset of 93 SAR images acquired between January 02,2017 and February 16, 2020 in interferometric wide (IW) beam mode and Level 1 Single Look Complex(SLC), polarization VV and ascending orbit. We arranged these scenes into twosubsets, one that contained 89 images acquired before the eruption and anotherwith 4 images acquired after the eruption. Subsequently, we generated 3916pre-eruption interferometric pairs (reference pairs) and 362 post-eruptionpairs. First,the total coherence is estimated for each pixel with the TDM by using as inputthe pre-eruption coherence maps. Temporal decorrelation changes are related tochanges in the dielectric and physical structure of the scatterers and in thelocation where these changes occur (i.e. volume and ground). The estimationtakes into account three aspects: (1) the random motion of the vegetationstructure, (2) the temporally correlated dielectric changes (seasonal andnormal decreasing of the coherence over time) and (3) the temporallyuncorrelated dielectric changes (random). Considering these aspects, the totaltemporal coherence depends on three Pair-Invariant variables: (1) the Ground toVolume Ratio, (2) the Characteristic Time for Ground and (3) the CharacteristicTime for Volume. The total temporal coherence for each pixel is also dependenton two Pair-Variant random variables: (1) the Random Dielectric Changes ofGround, (2) the Random Dielectric Changes of Volume. Thus, the number of modelparameters becomes twice the total number of interferometric pairs (N) plusthree Pair-Invariant variables (2N + 3); in this sense, coherence depending onthe land cover, the properties of the vegetation and the dominant scattering.[1]Second,the extracted model parameters are used to estimate the affected area. ACumulative Distribution Function (CDF) is constructed by using the randomdielectric changes of ground and volume variables for each pixel. This CDF isthe core for the estimation because it contains statistical information on hownatural occurrence or randomness affect the coherence of each pixel [1]. Third,the whole procedure explained above is implemented again but with thepost-eruption subset in order to generate a new post-eruption CDF. Finally,both CDFs are compared in order to assess their similarity and subsequentlyestimate the probability that each pixel has been affected by ashfall. Thus, aprobability map is generated, where pixels with probabilities closer to oneindicate the presence of volcanic ash on the ground. Ourimplementation of the TDM resulted in a probability map that shows highervalues that are in agreement with the distribution of volcanic ash observed onthe ground by the volcano observatory of the Philippine Institute ofVolcanology and Seismology (https://www.phivolcs.dost.gov.ph/) and with thedispersion of volcanic ash in the atmosphere observed from Himawari-8 satellitedata. However, areas where coupled effects occur, like crustal deformation andashfall, are more complex to interpret. Subsequent work will aim at correlatingthe probabilities with the presence and thickness of ashfall deposits and atimplementing the TDM with both C and L Band SAR data.