Distinguishing Inflation Drivers at Shallow Magmatic Systems using Ensemble-Based Data Assimilation

Abstract

In this study, synthetic numerical experiments are conducted to investigate how well the Ensemble Kalman Filter (EnKF) data assimilation approach distinguishes between two potential drivers of ground deformation at volcanic systems: pressurization and lateral reservoir expansion. Numerical models indicate that pressure-driven inflation creates larger radial displacements relative to inflation driven by lateral expansion. However, the introduction of noise can obscure these differences in simulated geodetic data. Although the EnKF does not fully reproduce the original synthetic models, it remains sensitive to changes in the magma reservoir's aspect ratio and is able to distinguish between the two inflation mechanisms. Ultimately, there remains significant non-uniqueness in how changes in reservoir pressure and size are reflected in surface deformation for any given aspect ratio, but future innovations may continue to improve filter performance.