Flux footprint model evaluations and multifractal analysis of plume concentration fluctuations

Two lines of research were conducted. The first of these was an evaluation of surface layer flux footprint models. The footprint is the contribution, per unit surface flux, of each unit element of the upwind surface area to a measured vertical flux. Knowledge of the footprint is necessary for accurately quantifying fluxes from the surface. Two analytical models and one Lagrangian stochastic model were evaluated. Within the limits of the uncertainties of the experimental measurements, all three were found to give predictions of the footprint in agreement with fluxes measured during a footprint tracer gas field study for moderately stable to moderately unstable conditions ({dollar}-{dollar}0.01 {dollar}<{dollar} {dollar}zsb{lcub}rm o{rcub}/L < 0.005).{dollar} Uncertainties in model use are related to discrepancies between the magnitudes of the measured and model-predicted fluxes (up to 25%). Large discrepancies exist between model and measurement for large {dollar}vert zsb{lcub}m{rcub}/Lvert.{dollar}; An extension to the footprint research was pursued with the goal of determining the evolution of the footprint in essentially real time. Individual realizations of the footprint can vary significantly from the mean, quasi-stationary representations given by the analytical and Lagrangian footprint models. A semi-empirical, near real-time model was developed for defining the footprint using actual turbulence measurements of wind and temperature and elements of Lagrangian and surface layer similarity theory to calculate back-trajectories of air parcels to their estimated point of last surface contact.; The second research topic was the multifractal analysis of plume concentration fluctuations for the purpose of characterizing extreme departures from the mean concentration. The plume measures of peak-to-mean concentration ratio, intermittency, and concentration fluctuation intensity were found to be well described by a multifractal characterization. An extremal Levy stochastic model generated time series which gave a good representation of the measured plumes within the levels of uncertainty associated with the model predictions and experimental measurements. The uncertainties associated with both model and measurement were sometimes very large reflecting the large variability inherent in turbulent transport and mixing.