Author: Owen Neiman

Publisher:

Published: 1979

Total Pages: 0

ISBN-13:

A wind tunnel model was used to study stack plume dispersion at a surface-mining site. The interaction between plumes and the wake flows behind the stacks and large terrain obstacles (dikes) was investigated. Mean flow Reynolds number was about 25,000 times smaller in the wind tunnel than in the full scale atmospheric flow, affecting the dynamic similarity of the model. Stack wake flows were found to model downwash effects properly despite Reynolds number mismatch, as indicated by measured values of base pressure coefficient, -Cp^ - 0.86, for cylinders at model conditions, which were close to values from Roshko (1960) for cylinder flows at full scale Reynolds numbers. However, model dikes were found to greatly exaggerate the size of recirculation zones and influence of the dike wake flows on downwind velocity and turbulence levels. This problem was corrected using a deflector vane mounted on the model dike crests. A combined-rise formulation was developed which predicts both the momentum and buoyancy effect on plume rise. Momentum rise entrainment constants, predicted from a simple model from Wilson (unpublished) were found to yield accurate momentum rise predictions using the combined-rise formulation. Buoyancy rise entrainment constants $2 were found to vary realistically with changing flow conditions. The momentum flux ratio d).. was found to be well correlated with stack downwash effects, providing a means of prediction for this phenomenon. Dike wakes were found to increase measured ground level concentrations (GLC) from plumes, with a non-linear degree of severity with dike height. The Gaussian dispersion model was found to be adequate in predicting GLC. A correction model was proposed to account for dike effects on GLC through increased vertical plume spread o in the Gaussian model.