The mode of vertical velocity in convective boundary layers is usually negative and the probability distribution function (PDF) of w, P(sub w), is rarely symmetric. Consequently, vertical diffusion from elevated sources is usually asymmetric and exhibits a descending mode of concentration, causing higher peak surface concentrations than predicted by Gaussian models. The main concentration (X) effects, the authors argue, can be modeled using simple PDF diffusion models, with tracers responding to P(sub w) at the source height with straight line trajectories and simple reflection at the surface and z(sub i), the mixing depth. The critical element is the choice of P(sub w). The PDF models are compared with diverse selection of atmospheric turbulence measurements. Recent lidar measurements of oil fog plumes are presented that show a large variability.