Studies on mathematical models for characterizing plume and drift behavior from cooling towers. Volume 4. Mathematical model for multiple-source (multiple-tower) cooling tower plume dispersion

This volume presents a generalization of our single source model (presented earlier in Volume 2) to multiple sources. The generalized model can treat vapor plume dispersion from any number of cooling towers in any geometrical configuration in any orientation to the direction of the wind. Important characteristics of the model include: (1) methodology of plume merging which accounts for differing plume entrainment rates during merging depending on wind direction; and (2) treatment of the effects of tower downwash from multiple towers; namely, additional entrainment and an additional pressure drag force acting vertically. Limited calibration of the model to laboratory data was undertaken to determine two downwash coefficients. Verification of the model by comparing model predictions to new lab and field data revealed a superior performance of our model as compared to the models commonly used in environmental impact evaluation. The ANL multiple-source model predicts visible plume height within a factor of 2 and/or visible plume length within a factor of 2 1/2 in 80% of our field data test cases. For comparison, the Orville and Slawson-Wigley Models satisfy this criterion for only 67% and 49% of the time, respectively. Additional ANL Model improvement is possible mainly through further development of the plume merging criterion.