Deconvolution Technique to Separate Signal from Noise in Gravel Bedload Velocity Data

摘要: A deconvolution procedure is presented to estimate the probability density function of bedload transport velocity from noisy stationary data collected using bottom tracking feature acoustic Doppler current profilers (aDcps). The involves optimization a computational summation random variables for instrument noise (assumed be Gaussian with zero mean) and spatially averaged within insonified area each beam (V) . was tested on two aDcp time series, measured in different gravel-bed rivers (Fraser River Norrish Creek). Models generated either semitheoretical compound Poisson-gamma distribution or an empirical gamma V were similar did not differ significantly original data. Optimized distributions highly positively skewed. comparable water measurements, i.e., order magnitude greater t...