Improving Volcanic SO2 Cloud Modeling Through Data Fusion and Trajectory Analysis: A Case Study of the 2022 Hunga Tonga Eruption

Abstract

The January 2022 eruption of the Hunga Tonga–Hunga Ha'apai volcano in the South Pacific emitted significant sulfur dioxide $\left({SO}_2\right)$ into the atmosphere, forming a large stratospheric cloud. This study employs the HYSPLIT model, a Lagrangian atmospheric transport and dispersion model, along with satellite retrievals of ${\text{SO}}_2$ cloud properties to model the long range transport of the cloud. To reduce the uncertainty and complexity of modeling the near-source behavior of the umbrella cloud, we utilize a data insertion technique that initializes the model at a downwind plume location. Satellite retrievals provide estimates of column mass loading and plume top height, though the plume top height may be uncertain above the tropopause. Additionally, the vertical mass distribution must be estimated by making assumptions about the cloud thickness. We use a back trajectory analysis to provide better estimations of plume top height and thickness. Our findings reveal that trajectory-derived cloud top heights substantially exceeded satellite estimates, with 60% ranging between 20 and 40 km, compared to most satellite-derived estimates being around 15 km. Long range 5-day forecasts produced with data insertion using the revised cloud top heights and estimated thickness are compared with forecasts using retrieved cloud top heights and an assumed simple thickness of 1 km, and a control run initiated from the vent at the eruption start time. A qualitative comparison with satellite retrievals and data from ground based lidar stationed at Réunion Island shows the use of the back trajectory analysis significantly improves the forecast.