A comparison of volcanic ash source term characteristics estimated by source inversion and plume rise modelling methods: Raikoke 2019

Predictions of volcanic ash location and concentration following an eruption rely heavily on estimates of source term characteristics including mass eruption rate, vertical distribution of ash and particle size distribution. These characteristics can be provided by several methods including (i) preset values based on historical data, (ii) near-source plume rise model simulations, (iii) a combination of satellite retrievals and long-range dispersion model simulations (known as source inversion). For the first time, this study presents a comparison of source term characteristics from these different methods. The study focuses on the 2019 Raikoke eruption and analysis of the volcanic ash cloud 150 km downwind from the volcano vent, representing an effective source term for the dispersion of ash in the distal volcanic cloud. Results indicate good agreement in the vertical distribution of ash between the plume rise and source inversion methods but large differences in estimates of the horizontal mass flux at this distance. The plume rise model demonstrates the rapid sedimentation and deposition of coarse ($>100\mu m$ diameter) ash particles close to the volcano vent resulting in a particle size distribution comparable to the preset distribution used operationally by the London VAAC at this range. These results suggest that source inversion can provide a computationally cheaper alternative to the 3D plume rise method for estimating the vertical distribution of ash, and that the assumption of near-source fallout of coarse particles in the preset particle size distribution holds fairly well. Further investigations are recommended including particle aggregation effects to understand differences in estimates of the effective mass eruption rate.