Using TanDEM-X Satellite Interferometry for Measuring Pyroclastic Flow Processes: Bulking and Run-out During the 2018 Eruption of Fuego Volcano, Guatemala

Abstract:

During an explosive eruption, the evaluation of the area impacted and the flow volumes from field observations are challenging due to terrain access and poor visibility, but this information is crucial for mitigating the impact of volcanic hazards. Fuego eruption on 3^rd June 2018 was one of the deadliest eruptions of the 21^st century so far. A series of pyroclastic density currents (PDCs) travelled more than 12 km from the summit down to Barranca Las Lajas, which caused the evacuation of 12,000 people, the destruction of infrastructures and the death of hundreds of inhabitants. Here, we use a combination of satellite data to constrain the location, thicknesses and volume of Fuego PDCs. From Sentinel-2 post-eruptive images, we map the five different units of the PDCs: collapse, transitional, channel, overbank and ash cloud surge. Using TanDEM-X SAR data, we produce a series of 10m-resolution Digital Elevation Models (DEM) and retrieve PDC thicknesses by differencing pre-eruptive and post-eruptive DEMs. By comparing TanDEM-X and Sentinel-2 results, we show that topographic changes are spatially correlated with: i) large negative changes (30-60m) close to the summit due to collapse of materials, ii) negative changes (up to 25 m) for the overbanks and the ash cloud surge, which correspond to the destruction of the vegetation and iii) positive changes (up to 25 m) in the main channel related to deposition. The net volume of deposition is 15.1 ± 0.42 x10⁶ m³, and the analysis of the volume budget indicates that a large proportion of the materials accumulated in the main channel originates from the collapse at the summit, which suggests a bulking process. Bulking was an important mechanism during the 2018 Fuego eruption and impacted the run-out distance of PDCs by increasing the mobility of the flow. Therefore, an effort should be made to incorporate bulking volumes in the next-generation numerical models of PDC to improve the forecasting of inundated areas and to support the mitigation of associated hazards. Finally, our approach is not limited to Fuego, but could be applied for the near real-time monitoring of the mass changes at other stratovolcanoes worldwide, which will provide more insights about the frequency and the volume of such collapse events.