Study of atmospheric dispersion under low wind conditions in an urban environment

Impact studies of industrial sites for air pollutant emissions must consider all representative meteorological conditions. For low wind conditions, the impact evaluations present large uncertainties. Dispersion mechanisms and turbulence properties in these situations are modified and favor the stagnation of emitted pollutants in the atmosphere. The improvement of the understanding and modeling of these situations is hampered by a lack of data at the international level, particularly for the built environment.

The objectives of this study are to present and analyze experimental data of dispersion of a tracer gas (Helium) under low winds in urban environment, as well as analyze the meteorological conditions corresponding to these situations to characterize the processes dominating the dispersion (especially meandering).

Two measurement campaigns were realized in 2020 and 2022 respectively on the SIRTA (Site Instrumental de Recherche par Télédétection Atmosphérique). This built area is located near Paris in a peri-urban environment. The height of the buildings varies between 5 and 30 m. Helium concentrations are measured using air samples and mass spectrometers in real time, in the near field of the emission point (<300m). Wind and turbulence conditions are measured by ultrasonic anemometers positioned at different heights in the surface layer and in the canopy layer.  Atmospheric Transfer Coefficients (ATC) are determined to quantify plume dispersion.

Data processing consists in establishing atmospheric turbulence parameters of each experiment (friction velocity, heat flux, Monin-Obukhov Length), in analyzing the spatial distribution of Helium concentrations in the built environment, especially searching correlation between the variations of the concentrations and of the wind directions and making spectral and autocorrelation analysis of the wind speed components to characterize the flow meandering.

The meandering is characterized by an oscillating behavior of the Eulerian autocorrelation function with the presence of a negative loop for the horizontal components u and v of the wind speed. The autocorrelation function of the vertical component w presents a classical exponential curve. The periods of this process vary between 20 and 60 min in the first experimental campaign. The meandering disperses the plume over a wide angular range. The analysis of the time series shows the oscillation of the wind direction in low wind.

The meandering modifies the low frequencies part of the spectrum of the wind speed horizontal components. The impact of the urban environment on this phenomenon is shown using comparisons between spectrum and autocorrelation functions measured at 3 m (inside the buildings canopy) and 30 m heights (above the buildings). The next step is to link the dispersion and meandering processes by performing  cospectrum of ATC and wind direction, in order to have information on the frequency dependence of the covariance of these two variables.