Assessment of the plume theory predictions of crown scorch using transport models

We studied numerically crown scorch as the effects of a fire line spreading through surface fuel under a tree canopy. The objective was to assess the usual assumptions made when one uses the Van Wagner criterion or similar models for crown scorch height. For this purpose the FIRESTAR 2D and FIRETEC wildfire behavior models were used. We simulated the fire line by a heat source at ground level and mainly investigated the temperature field.

As a first step, we ran computations of thermal plumes with no wind and with no canopy, for first comparison to plume theory. The influence of crown existence on the temperature field above the heat source, as well as on crown scorch, was then investigated. It was found that the presence of the tree crowns caused an increase of scorch height. This effect was attributed to the reduction of fresh air entrainment due to additional drag of the crown layer.

As a second step, the effect of a wind to the plume, as well as to crown scorch, was studied with FIRETEC, for the no-canopy and canopy cases. Four values of wind speed (1, 2, 5 and 10 m/s at 40 m height) and four fire line intensities (250, 500, 1000, 2000 kW/m) were tested.

Even a weak wind (1 m/s) strongly modified the plume structure with respect to the no-wind (idealized) situation and scorch heights were decreased. As one can expect, increasing the fire line intensity increased the scorch height. In contrast, increasing the wind speed tended to increase very slightly the scorch height. Also we observed that the tree canopy presence (LAI=4) led to an increase of scorch height with respect to the no-canopy simulation. We also studied the influence of canopy bulk density (LAI =2 or 6).

The numerical results were compared with experimental observations of scorch height (mainly field experiments), which were usually obtained in the presence of a weak wind (prescribed burning experiments under tree canopies). Experimental data show a large variability from one study to the other. FIRETEC predictions of scorch height as a function of fire line intensity were in the lower range of experimental data.