Simulating Random Natural Variability in Time-Varying Atmospheric Concentrations of Toxic Gas from Pipeline Ruptures

February 4, 2004 - 12:00pm
Bldg. 90
Seminar Host/Point of Contact: 

Random time series are found everywhere in nature.  The Brownian motion of small particles; the price of assets (stocks) in financial markets; the diffusion of individual molecules through a membrane; the ballistic deposition of nano-particles onto a lattice substrate; and the time-varying concentration fluctuations at a point downwind from a pollution source all have a common dynamic description.  All are stochastic processes where the local rate of change of the variable has a natural drift back to some equilibrium state, combined with a random fluctuating component.  We will explore how a stochastic model for concentration fluctuations can be constructed for a fixed receptor (your nose), and used to generate a large ensemble of simulated concentration time series.  As a concrete example, these concentration time series will be used to estimate the event-to-event variability in how far downwind you must stand to avoid being killed by exposure to "identical" ruptures of a sour gas pipeline transporting a mixture of natural gas and hydrogen sulfide.  Finally, we will demonstrate the advantages of using direct event simulation (rather than probability distributions) when non-linear processes such as toxicology are involved.  An essential element of this demonstration will require breaking up a large sheet of peanut-brittle with a hammer, and having the audience digest the experimental data.

Add event to Google Calendar

Announcement List

For notification about future seminars, contact J.M. Lambert.

Site Access

Schedule subject to change without notice. If you are coming from off-site, please call first to verify. UC staff and guests are welcome. LBNL shuttle buses stop every few minutes at marked sidewalk locations along Bancroft and Hearst Avenues and Rockridge.