Numerical Modelling of Wind Borne Pollution Dispersion from Open Windrow Compost Sites

Open environmental flow is a multivariable, non-linear, unsteady flow system which presents a challenge to those who engage in understanding, monitoring and predicting it. This thesis sets the foundation for future work and suggests best practice techniques for the numerical simulation of such flow taking into account important phenomena and assumptions. These factors include boundary conditions, location and size of the control volume and level of resolution. The use of more comprehensive techniques such as Computational Fluid Dynamics proposed herein, compared to those currently employed by regulating agencies, may take into account local meteorology and topography in order to provide a more accurate estimation of pollutant dispersion by site specific investigation. Wind flow and effluent pollutant dispersion was examined for idealised flat and hilly terrains where topography was seen to largely influence wind flow phenomena and later pollution dispersion. These simulations assumed an incompressible, steady state air, flowing isothermally over vertical and horizontal line sources positioned near ground and upstream one or two sinusoidal axisymmetric hills in a single row. Results suggest that the wind speed increases on the upslopes of hills. Then the flow recirculates strongly on the downslope of a hill or the lee of any bluff body such as a house or trees. Furthermore, a case study has been carried out for two locations in the UK, employing techniques and conclusions from idealised simulations. A proposed compost site, located in South Yorkshire was examined, where the local topography is characterised by sloping terrain and large woods; such factors were seen to influence wind flow and dispersion of particles and concentration. Results from the compost site were compared to results from a contrasting flat-land green waste processing plant in West Midlands. All simulations assumed an incompressible, steady state wind over three presumed compost piles. For all cases studied herein, Lagrangian particle tracking was employed to show particle dispersion from the alleged compost piles and calculate average travel distances for trapped particles and entrapment probability. Particle tracking results show that particulate pollutants can get trapped at recirculation areas, in-between trees and village houses. Mass-less particles travel further before hitting an obstacle when released in low winds and their orbit is largely affected by topography when released from near ground sources. In addition, Species Transport was enabled to examine the effect of topographical and meteorological conditions to pollutant concentrations. Species transport results for the examined composting sites show an increased number of particles trapped in-between rows of compost piles and a decrease of emissions to near background values within 200m from the source. However, it was evident that flat land favours the spread of species which can reach nearby houses at larger percentages.