Impact of meteorological inflow uncertainty on tracer transport and source estimation in urban atmospheres

Abstract

A computational Bayesian inverse technique is used to quantify the effects of meteorological inflow uncertainty on tracer transport and source estimation in a complex urban environment. We estimate a probability distribution of meteorological inflow by comparing wind observations to Monte Carlo simulations from the Aeolus model. Aeolus is a computational fluid dynamics model that simulates atmospheric and tracer flow around buildings and structures at meter-scale resolution. Uncertainty in the inflow is propagated through forward and backward Lagrangian dispersion calculations to determine the impact on tracer transport and the ability to estimate the release location of an unknown source. Our uncertainty methods are compared against measurements from an intensive observation period during the Joint Urban 2003 tracer release experiment conducted in Oklahoma City. The best estimate of the inflow at 50 m above ground for the selected period has a wind speed and direction of 4.6−2.5+2.0 m s−1 and 158.0−23+16, where the uncertainty is a 95% confidence range. The wind speed values prescribed in previous studies differ from our best estimate by two or more standard deviations. Inflow probabilities are also used to weight backward dispersion plumes and produce a spatial map of likely tracer release locations. For the Oklahoma City case, this map pinpoints the location of the known release to within 20 m. By evaluating the dispersion patterns associated with other likely release locations, we further show that inflow uncertainty can explain the differences between simulated and measured tracer concentrations.