Citation:

Carissimo, B., S. Hanna, S. Trini Castelli, G. Tinarelli, M. Pirhalla, S. Perry, D. Heist, AND J. Chang. A comparison of building effects simulations in different atmospheric dispersion models with detailed sonic measurements from the Jack Rabbit II field experiment. 37th International Technical Meeting on Air Pollution Modelling and its Application, HamburgG, September 23 - 27, 2019.

Impact/Purpose:

Dispersion models are frequently used in complex urban areas where building effects are predominant. Urban structures considerably modify the local flow, impacting concentrations from hazardous releases. Simulations of building effects must be properly validated with detailed datasets in real atmospheric conditions to gauge model performance and suggest improvements. This presentation uses sonic anemometer wind flow data from a mock urban field study to select near-neutral atmospheric conditions for comparison with detailed CFD simulations. The wake size and turbulent structure in lee of building obstacles in these simulations are analyzed and presented, which can be used to better inform urban dispersion after harmful releases.

Description:

Atmospheric dispersion models have frequently been used around complex industrial sites or in urban areas where building effects are predominant. Urban structures tend to considerably modify the local flow, potentially impacting concentrations from hazardous releases. These dispersion models range from simple empirical formulas to very detailed CFD simulations. However, simulations of building effects must be properly validated with detailed datasets in real atmospheric conditions to gauge model performance and suggest improvements. One such dataset is the Jack Rabbit II Special Sonic Anemometer Study, which took place in November 2015 and March 2016 at Dugway Proving Ground, Utah, USA. These experiments were carried out by keeping 14 CONEX shipping container obstacles from the original Jack Rabbit II chlorine release. The measurements involved 34 sonic anemometers placed within the CONEX array to document the flow field around single CONEX containers and a 2 by 3 CONEX tower. Upstream of the study area, the approach flow was captured by a 32 m meteorological tower instrumented with sonic anemometers at five levels. The array of buildings were aligned in the meteorological direction 165° (axis -15° from the true north), and we then selected several wind situations between 150-180° with various wind speeds for the comparison with the model simulations. The selected periods were chosen during March 2016 when more sonics were available. The current results of this ongoing project will be presented. They involve at least one full CFD model, a fast momentum solver and a diagnostic mass-consistent code. Detailed comparisons were carried out for the wind speed, direction and turbulence around the obstacles (with more instruments in the wake) and at different levels (up to 4 levels around the CONEX tower). Some statistical measures of the differences will also be discussed. In the comparison of the simulations, we will also use virtual tracer releases (no measurements) to investigate how the observed wake differences affect the dispersion in the different modeling approaches, with Eulerian CFD and Lagrangian stochastic particle models.

Record Details: