The effect of sampling rate on interpretation of the temporal characteristics of radiative and convective heating in wildland flames

Time-resolved radiative and convective heating measurements were collected on a prescribed burn in coniferous fuels at a sampling frequency of 500 Hz. Evaluation of the data in the time and frequency domain indicate that this sampling rate was sufficient to capture the temporal fluctuations of radiative and convective heating. The convective heating signal contained significantly larger fluctuations in magnitude and frequency than did the radiative heating signal. The data were artificially down-sampled to 100, 50, 10, 5 and 1 Hz to explore the effect of sampling rate on peak heat fluxes, time-averaged heating and integrated heating. Results show that for sampling rates less than 5 Hz the difference between measured and actual peak radiative heating rates can be as great as 24%, and is on the order of 80% for 1-Hz sampling rates. Convective heating showed degradation in the signal for sampling rates less than 100 Hz. Heating rates averaged over a 2-s moving window, as well as integrated radiative and convective heating were insensitive to sampling rate across all ranges explored. The data suggest that peak radiative and convective heating magnitudes cannot be fully temporally resolved for sampling frequencies lower than 20 and 200 Hz.