RUSSIAN JOURNAL OF EARTH SCIENCES, VOL. 17, ES5003, doi:10.2205/2017ES000612, 2017
Sergey I. Malkovsky, Aleksei A. Sorokin, Sergey P. Korolev
The ash clouds and plumes produced during explosive volcano eruptions are extremely hazardous for the population and economic activities and, for this reason, their detection and control over their movements are considered an urgent and vital problem. The solution of the problem is primarily linked to the satellite data analysis and modeling volcanic ash cloud movement trajectories. One of the most accepted models widely used in these investigations is the PUFF model and its Puff-UAF software implementation. The experience of applying the model in different regions of the world for more than a decade to investigate volcano explosive eruptions has shown a sufficient correlation established between the forecasted and observed volcanic ash cloud movement trajectories obtained using remote sensing methods. The paper describes the existing approaches and program solutions applied to meteorological dataset preprocessing for numerical simulation of the ash cloud movement and dispersion from explosive volcano eruptions using the Puff-UAF software suite. The detected problems of wgrib2-generated NetCDF file usage for the puff program operation, as well as their solutions, are discussed. The test numerical simulation results derived using different Puff-UAF versions, the original Puff-UAF 2.2.2 and the Puff-UAF with the authors' puff source code modifications are presented. Their comparison with remote sensing data obtained for the Shiveluch volcano eruption is made and the correctness of the puff source code modifications is concluded.
Received 11 December 2017; accepted 11 December 2017; published 30 December 2017.
Citation: Malkovsky Sergey I., Aleksei A. Sorokin, Sergey P. Korolev (2017), Improving the system of numerical simulation of volcanic ash propagation using the PUFF model, Russ. J. Earth Sci., 17, ES5003, doi:10.2205/2017ES000612.
Copyright 2017 by the Geophysical Center RAS.