Predicting Winds Inside Deep Urban Canopies Using Vortex Dynamics
DescriptionKnowledge of urban winds is crucial to applications in air quality and pollutant dispersion. Yet despite advances in our ability to forecast large-scale winds, wind profiles cannot be determined quickly and accurately for arbitrary urban geometries. Existing theoretical predictions do not apply to the important limit of tall, closely packed buildings or generalise to realistic geometries. Downscaling of large-scale winds is inaccurate. Building-resolving numerical simulations are computationally expensive.An alternative strategy is to develop a simplified but accurate model of urban winds by exploiting the close connection between the velocity and the vorticity or local rotation. Recent results demonstrate that the flow regime, or qualitative structure of the velocity field, can be predicted from the vorticity at the roof and pedestrian levels of a street canyon. This proposal hypothesizes that the mean winds inside deep urban canopies are also controlled by thin and intense layers of vorticity that form near solid boundaries. Strong spatial localisation of the vorticity yields a reduced mathematical description and fast numerical solutions.The aim of this proposal is to develop a new method for estimating wind profiles inside deep urban canopies. This will be achieved by applying techniques for vorticity dominated flows and extending recent work on the prediction of street-canyon flow regimes. (i) Reduced models of the vorticity field inside idealised and realistic street canyons will be derived from large-eddy simulation data. (ii) The velocity field implied by each vortex sheet or layer will be obtained from the numerical solution of a boundary-value problem and urban wind data provided by the Hong Kong Observatory. (iii) A velocity profile valid for the entire canopy will be constructed by matching.This proposal addresses an important unsolved problem. It will deepen understanding of winds in urban areas by highlighting the role of vorticity layers and boundaries. The underlying method offers enormous computational savings over computational fluid dynamics and will improve the accuracy of operational air quality and dispersion models.
|Effective start/end date||1/01/18 → …|