About

Three eruptions (Kilauea 2018, Iceland SW rift 2021, la Palma 2021) highlight the needs for computationally efficient codes to model lava flow inundation.  In this project we are exploring the use of the MOLASSES code to model lava flow inundation.

MOLASSES and its predecessors have been developed since 2012 (Connor et al., 2012), going through a series of iterations designed to make the code computationally efficient.

The code is currently available on Github, and can be run online via gscommunitycodes.

https://gscommunitycodes.usf.edu/geoscicommunitycodes/public/molasses/molasses.php

MOLASSES stands for MOdular LAva Simulation Software for the Earth Sciences.  The code relies on a cellular automata algorithm to estimate the area inundated by lava flows.  MOLASSES does not model the rate of lava emplacement or the physical dynamics of lava flows. A digital elevation model (DEM) of the volcanic region is pre-loaded on the server; lava flows are simulated using this DEM. Rheology is accounted for by specifying a modal lava thickness. The thickness of lava flows, especially their margins, is a function of the lava flow yield strength. Yield strength (Pa) is a more useful rheological parameter than viscosity (Pa s), since the code does not account for dynamics.

The goals of this VICTOR project are to:

1. Broaden use of lava flow inundation models, like MOLASSES, by simplifying the use of alternative digital elevation models (DEMs)

2. Use MOLASSES as an example of how to construct workflows for probabilistic lava flow hazard assessment

3. Explore implementation of alternative constitutive equations in MOLASSES, which may broaden use.

4. Use MOLASSES in benchmarking activities related to code verification and validation.

 

Team