Toward understanding the sensitivity of a fast response dispersion modeling system to real input data

QUIC (Quick Urban and Industrial Complex) is a fast response urban wind and particle dispersion modeling system. QUIC can be used in real-time or in training scenarios such as evacuation and clean up as a result of intentional or accidental chemical or biological releases. QUIC consists of a wind flow model (QUIC-URB), a particle dispersion model (QUIC-PLUME) and graphic user interface (GUI-GUI). QUIC-URB computes a 3-dimensional flow field in an urban area by enforcing mass conservation on an empirically based wind field. QUIC-PLUME is a Lagrangian dispersion model that solves the Langevin random walk equations. The objective of this study is to validate QUIC using data obtained from the JU2003 (Joint Urban 2003) field experiment conducted in Oklahoma City. This experiment consisted of tracer gas releases from varying locations in the CBD (central building district) of Oklahoma City with a dense array of samplers located in the downtown area as well as downstream of the CBD at varying distances up to 4km.

The primary objective of this work is to validate the QUIC model. To do so in real urban environments involves utilizing field measurements to construct an appropriate inflow boundary condition that allows for the most reasonable evaluation of the model. The work presented here describes the procedure involved in selecting an appropriate boundary condition. The sensitivity of the QUIC generated solution to the specified inflow wind profile is discussed. This analysis will show quantitative point-to-point comparisons of measured quantities between the QUIC solutions and the JU2003 data as well as qualitative behavior. The focus of the comparisons will be on Intensive Observation Period number 3 (IOP3) which was a daytime release on 7 July 2003. The results consist of a comparison of concentrations and winds obtained from many QUIC simulations with a wide range of varying inflow parameters. Each of the inflow parameters used were obtained from the JU2003 experiment from many wind profilers and sonic anemometers with a wide range of sampling frequencies. The aim is to find the most appropriate way to assimilate inflow field data into the simulation given the error and disparity associated with the measurements.