Using Background-Oriented Schlieren to Visualize Convection in a Propagating Wildland Fire

ABSTRACT

Heat and mass transfer are important processes associated with wildland fire. Both radiant and convective heat transfer are important processes with convection often being the dominant mechanism. Unlike radiation, there is no direct method of measuring convection. Since convective heat transfer is governed by the fluid flow, understanding the fluid flow provides good understanding on the convective heat transfer. In fluid mechanics, flow visualization is a common methodology used to understand flow characteristics. Schlieren imagery is a common flow visualization technique which captures changes in fluid density such as the ones occur around a fire. Background-Oriented Schlieren (BOS) is a flow visualization technique that uses a background image with various patterns to visualize the density gradient caused by density fluctuations in a fluid. We applied BOS to measure the flow associated with laboratory-scale line fires. The reproducible fires were spreading in pine needle fuel beds in a wind tunnel with and without imposed wind. This initial application of BOS in a fire environment successfully visualized the flow around the flame. The visualized flow underwent a secondary process to produce the velocity field of the flow. Results indicate that even in conditions where the fire is known to be dominated by radiation, wind carried the thermal plume ahead of the flame front and expanded the thermal plume. In contrast, in the no wind condition, the thermal plume remained vertical above the fire. Using the BOS imagery, a new model for estimation of convective heat transfer was introduced. In addition to estimation of the convective heat transfer ahead of the fire, this new model enables visualization of convective motion.

KEYWORDS:

• Fire spread
• background-oriented schlieren
• flow visualization
• convective heat transfer

Acknowledgments

The authors want to thank Ms. Gloria Burke, Ms. Bonni Corcoran. and Mr. Joey Chong for their invaluable assistance during the experimental work. This research was supported by funding from the DOD/DOE/EPA Strategic Environmental Research and Development Program project RC-2640 administered through agreement 16JV11272167026 between the USDA Forest Service PSW Research Station and the University of California, Riverside.

All the optical flow algorithms used in this paper are available in the Open-CV library under 3-clause BSD License.

Notes

¹ The use of trade or firm names in this publication is for reader information and does not imply endorsement by the U.S. Department of Agriculture of any product or service.

Additional information

Funding

This research was supported by funding from the DOD/DOE/EPA Strategic Environmental Research and Development Program project RC-2640 administered through agreement 16JV11272167026 between the USDA Forest Service PSW Research Station and the University of California, Riverside.