Inferring energy incident on sensors in low-intensity surface fires from remotely-sensed radiation and using it to predict tree stem injury

Abstract:

Remotely-sensed radiation measurements, attractive for their spatial coverage, offer a means of inferring energy deposition in fires (e.g., on soils, fuel particles, and tree stems) but coordinated remote and in-situ (in flame) measurements are lacking. We relate remotely-sensed measurements of fire radiative energy density (FRED) from nadir (overhead) radiometers on towers and the Wildfire Airborne Sensor Program (WASP) infrared camera on a piloted, fixed-wing aircraft to energy incident on in situ, horizontally-oriented, wide-angle total-flux sensors positioned at ~0.5 m above ground level. Measurements were obtained in non-forested herbaceous and shrub dominated sites and in (forested) longleaf pine (Pinus palustris Miller) savanna. Incident energy was strongly related to nadir radiometer FRED (R² = 0.79) and moderately related to WASP FRED (R² = 0.50). As a demonstration of how this approach could be applied to ecological effects, we predict stem injury for turkey oak (Quercus laevis Walter), a common tree species at our study site, using incident energy inferred from remotely-sensed FRED. Larger diameter stems were expected to be killed in the forested than in the non-forested sites. Measurement challenges remain for remote and in-situ measurements including further consideration of error associated with inferring energy incident on tree stems.