Yan Zhan


  • M.S. Structure Geology, Peking University, China
  • B.S. Geochemsitry, Peking University, China

Forecasting volcanic unrest of Atka volcanic center, Alaska, using the Ensemble Kalman Filter


The proposed magma plumbing model for the Atka volcanic center.

Forecasting volcanic unrest for vulnerable populations requires understanding the dynamics of the underlying magma reservoirs by linking volcanic unrest observations (e.g., surface deformation, seismicity, and gas emissions) to eruption potential. The application of a model-data fusion framework based on the Ensemble Kalman Filter (EnKF) has been shown to be a powerful method for providing updates of the volcanic storage models through time (Gregg and Pettijohn, 2016; Zhan and Gregg, 2017Zhan et al., 2017). However, this approach is still in its initial stages of development. To take full advantage of observational data (e.g., InSAR) and to test more sophisticated finite element models (FEMs), I propose to develop new high-performance computing (HPC) techniques to conduct the EnKF data assimilation. Then, a more 3D sophisticated magma reservoir model can be tested under this new framework to explain the triggering mechanism of the 2006 eruption of the Korovin Volcano in the Atka Caldera, whose precursory deformation and seismicity is offset ~5 km away from the eruption center.