Coupled Weather and Wildfire Behavior Modeling at Los Alamos: An Overview

Conference · Wed Dec 31 04:00:00 EST 1997

OSTI ID:635205

Bossert, James E; Harlow, Francis H; Linn, Rodman R; Reisner, Jon M; White, Andrew B; Winterkamp, Judith L

Over the past two years, researchers at Los Alamos National Laboratory (LANL) have been engaged in coupled weather/wildfire modeling as part of a broader initiative to predict the unfolding of crisis events. Wildfire prediction was chosen for the following reasons: (1) few physics-based wild-fire prediction models presently exist; (2) LANL has expertise in the fields required to develop such a capability; and (3) the development of this predictive capability would be enhanced by LANL`s strength in high performance computing. Wildfire behavior models have historically been used to predict fire spread and heat release for a prescribed set of fuel, slope, and wind conditions (Andrews 1986). In the vicinity of a fire, however, atmospheric conditions are constantly changing due to non-local weather influences and the intense heat of the fire itself. This non- linear process underscores the need for physics-based models that treat the atmosphere-fire feedback. Actual wildfire prediction with full-physics models is both time-critical and computationally demanding, since it must include regional- to local-scale weather forecasting together with the capability to accurately simulate both intense gradients across a fireline, and atmosphere/fire/fuel interactions. Los Alamos has recently (January 1997) acquired a number of SGI/Cray Origin 2000 machines, each presently having 32 to 64 processors. These high performance computing systems are part of the Department of Energy`s Accelerated Strategic Computing Initiative (ASCI). While offering impressive performance now, upgrades to the system promise to deliver over 1 Teraflop (10(12) floating point operations per second) at peak performance before the turn of the century.

• Research Organization: Los Alamos National Lab., NM (United States)
• Sponsoring Organization: USDOE Assistant Secretary for Management and Administration, Washington, DC (United States)
• DOE Contract Number: W-7405-ENG-36
• OSTI ID: 635205
• Report Number(s): LA-UR--97-4073; CONF-980121--; ON: DE98002669
• Country of Publication: United States
• Language: English

Related Subjects

54 ENVIRONMENTAL SCIENCES
ATMOSPHERES
COMPUTERIZED SIMULATION
FEEDBACK
FIRES
FUELS
HEAT
INTERACTIONS
NONLINEAR PROBLEMS
SIMULATION
VARIATIONS
WEATHER
WIND