Results from Daysmoke for weak smoke plumes

This paper describes Daysmoke, an empirical/stochastic smoke plume model for wildland fires and its use for modeling weak smoke plumes from three prescribed burns conducted at Fort Benning, GA, during April 2008. The burns, conducted on lands with low fuel loadings, produced smoke plumes with low heat content and weak vertical velocity – weak plumes. The weak updrafts were intercepted by strong convective mixing in the ambient air. These circulations broke the plume into deep turrets or pillars of smoke some of which extended through the depth of the mixing layer. The vertical plume structures simulated by Daysmoke were confirmed with lidar observations taken approximately 2 mi (3.6 km) downwind from the burns.

Daysmoke also generated spatial patterns of ground-level PM2.5 smoke concentrations. Time series of 5-min average patterns of PM2.5 concentration show centers (or patches) of high concentration surrounded by areas of low concentration transported with the wind. Peaks and spikes of time series of PM2.5 observed at three trucks positioned strategically beneath the plume from 1 - 4 miles (1.6 – 6.4 km) from the burn site confirm the Daysmoke results.

Smoke plumes from wildland fires are characterized by numerous updraft “cores” that merge convectively into a single core or a multiple-core plume by the time the plume has ascended to an equilibrium level. The plumes from the Fort Benning prescribed fires were described by 6 – 8 updraft cores. The multiple-core updraft structure of the plumes was confirmed by photo-images.

Daysmoke is planned as a smoke-injector for CMAQ. The accuracy of the model in simulating details of weak smoke plumes adds to our confidence in the accuracy of Daysmoke to place smoke in the atmosphere.