Dispersion of small inertial particles in characteristic atmospheric boundary layer flows
Dispersion of small inertial particles in characteristic atmospheric boundary layer flows
The aim of this work is to improve the accuracy of atmospheric dispersion modelling for small inertial particles. Using data available in the literature it could be shown that small particles in the order of 1 µm are expected to exhibit dispersion behaviour similar to a gas, whereas particles of diameters ∼ 10 µm are expected to be dominated by inertia in windy situations and dominated by gravity during calm weather. Direct numerical simulations of particles in homogeneous isotropic turbulence confirm this observation. Estimates of the near-wall behaviour of atmospheric aerosols suggest that both at the ground and at building walls particle clustering can be expected to be a relevant process that potentially has a strong influence on the results in urban dispersion modelling. A universal scaling relation for the concentration-based D measure for particle clustering in homogeneous isotropic turbulence has been established, which suggests that simulations at moderately high Reynolds numbers can be sufficient to understand the small-scale behaviour of particles. This needs to be confirmed for more complex flows. In order to obtain realistic simulation results, real validation data is needed. So far it has been considered difficult to use anything other than wind-tunnel data for the validation of time-resolved atmospheric dispersion simulations. A new method that is robust enough to evaluate noisy time series has been developed in this work. One of its distinguishing features is the fact that it gives stronger weight to good-quality releases than to bad quality (i.e. low dosage) releases. These filtering properties allow for the evaluation of field data without the need to introduce empirical thresholds. Therefore the method is objective and reliable in situations where established methods have not been successful. The DNS code PANDORA 2.0 has been extended to allow for any form of homogeneous turbulence that can be described by a deformation tensor for the mean flow. First tests of shear flow simulations show typical characteristics of homogeneous shear flows in terms of shear stresses, turbulent kinetic energy and the development of elongated structures. However, further verification and validation is needed.
University of Southampton
Wittemeier, Thorsten
7389419e-de70-4f43-84f6-823d5827bc72
Wittemeier, Thorsten
7389419e-de70-4f43-84f6-823d5827bc72
Shrimpton, John
9cf82d2e-2f00-4ddf-bd19-9aff443784af
Wittemeier, Thorsten
(2021)
Dispersion of small inertial particles in characteristic atmospheric boundary layer flows.
University of Southampton, Doctoral Thesis, 175pp.
Record type:
Thesis
(Doctoral)
Abstract
The aim of this work is to improve the accuracy of atmospheric dispersion modelling for small inertial particles. Using data available in the literature it could be shown that small particles in the order of 1 µm are expected to exhibit dispersion behaviour similar to a gas, whereas particles of diameters ∼ 10 µm are expected to be dominated by inertia in windy situations and dominated by gravity during calm weather. Direct numerical simulations of particles in homogeneous isotropic turbulence confirm this observation. Estimates of the near-wall behaviour of atmospheric aerosols suggest that both at the ground and at building walls particle clustering can be expected to be a relevant process that potentially has a strong influence on the results in urban dispersion modelling. A universal scaling relation for the concentration-based D measure for particle clustering in homogeneous isotropic turbulence has been established, which suggests that simulations at moderately high Reynolds numbers can be sufficient to understand the small-scale behaviour of particles. This needs to be confirmed for more complex flows. In order to obtain realistic simulation results, real validation data is needed. So far it has been considered difficult to use anything other than wind-tunnel data for the validation of time-resolved atmospheric dispersion simulations. A new method that is robust enough to evaluate noisy time series has been developed in this work. One of its distinguishing features is the fact that it gives stronger weight to good-quality releases than to bad quality (i.e. low dosage) releases. These filtering properties allow for the evaluation of field data without the need to introduce empirical thresholds. Therefore the method is objective and reliable in situations where established methods have not been successful. The DNS code PANDORA 2.0 has been extended to allow for any form of homogeneous turbulence that can be described by a deformation tensor for the mean flow. First tests of shear flow simulations show typical characteristics of homogeneous shear flows in terms of shear stresses, turbulent kinetic energy and the development of elongated structures. However, further verification and validation is needed.
Text
PhD thesis_Thorsten_Wittemeier
- Version of Record
Text
PTD_Thesis_Wittemeier-SIGNED
Restricted to Repository staff only
More information
Submitted date: 18 October 2021
Identifiers
Local EPrints ID: 467298
URI: http://eprints.soton.ac.uk/id/eprint/467298
PURE UUID: 4ad22251-04c6-425a-bb2c-67a07f62ca26
Catalogue record
Date deposited: 05 Jul 2022 16:49
Last modified: 17 Mar 2024 07:19
Export record
Contributors
Author:
Thorsten Wittemeier
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics