Project Description: The project focuses on a common but poorly understood behavior during effusive eruptions: erupting vents and lava ponds act as sources that feed flows into topographically confined basins. As those basins fill, lava eventually spills over their rims, generating new flow branches that can themselves fill and spill into adjacent low-lying areas. The result is a cascading, branching network of interconnected lava flows—a system that can be elegantly described using graph theory, in which source vents and ponds are represented as nodes, and individual flow segments between them are represented as edges. The intern will help build and apply this graph-based framework to real volcanic systems on Earth and Mars. Using ArcMAP and Python, flow behavior along each edge will be modeled using PyFLOWGO, an open-source Python platform that simulates the thermo-rheological evolution of a lava control volume as it cools and cools and crystallizes while moving down a channel.
NASA Relevance: This project directly supports NASA's Science Mission Directorate (SMD) Planetary Science Division by advancing quantitative models of volcanic emplacement processes that are fundamental to interpreting the geologic history of Mars—a core objective of NASA's Mars Exploration Program. The graph-theoretic and MCMC-based modeling framework developed here has broad applicability to future mission planning, including the identification and characterization of volcanically resurfaced terrains relevant to habitability and geologic evolution across the Solar System.
Work Description:
1. Theory — Graph-Theoretic Framework Development: Develop a mathematical framework representing lava flow networks as directed graphs, with vents and ponds as source nodes, flow segments as weighted edges, and topographic thresholds governing spill-over connections.
2. Validation — Benchmarking: Test the framework against well-documented terrestrial eruptions by comparing model outputs—flow lengths, cooling rates, and branching behavior—against field measurements and published parameters.
3. Parameter Estimation via Markov Chain Monte Carlo (MCMC) — Apply MCMC methods to quantify uncertainty in eruption rate, lava rheology, and vent geometry, using Python-based probabilistic sampling to identify conditions consistent with observed flow-field geometries.
4. Application to Terrestrial and Martian Lava Flow Fields — Extract and process digital elevation models (DEMs) from terrestrial and Martian orbital datasets to map basin geometries and flow pathways to initialize PyFLOWGO simulations.
Open or Reserved Project: Reserved/Open, 1 position reserved for student but mentor may be willing to work with another student if reserved student isn't selected.