Understanding and modelling the dynamics of the atmosphere over urban terrains still represent a scientific challenge with high stakes for our ability to handle the ever growing population in urban areas and the associated issues of air quality and urban management. The project URBANTURB aims at elucidating the interaction mechanisms between the lower-atmosphere and the urban-canopy flow and developing new modelling strategies to account for the interaction between these two regions of the flow.
More than 50% of the world population currently lives in urban areas and this proportion is expected to keep increasing during the next decades due to the fact that most of the population natural growth will be absorbed in cities (World Urbanization Prospects, The 2011 Revision Highlights, United Nations, NY – http://esa.un.org/unup/Documentation/highlights.htm). In order to ensure factors important for living conditions in these areas, such as air quality, inhabitants comfort or sustainable energy management, it is of crucial importance to be able to apply well-designed strategies of urban design and management. In the meantime, urban areas are highly complex environments because of their spatial heterogeneity and the numerous and complex physical processes that take place. Due to these complexities, understanding urban micro-climate and developing forecasting tools to help urban management, while considering the global evolution of the climate, require the development of models and methodologies specifically designed to take into account the multi-scales and multi-processes character of the urban regions. Being able to reproduce the time-evolving atmospheric flow over a wide range of spatial scales, modelling strategies based on the concept of large-eddy simulation (LES) have proved to be very promising. However, LES still relies on the use of models to account for the non-resolved turbulence. Moreover, the spatial complexity and heterogeneity of the urban canopy render accurate simulations of the canopy flow out of reach with the current computational capabilities. This type of simulations also requires the use of models to account for the presence of the urban environment. Understanding the key physical processes that drive the exchanges of mass, momentum, heat, moisture and energy between the urban canopy and the lower-atmosphere is therefore essential to improve and develop new models.
In order to overcome the scientific challenges arising from the multi-scale character and the high complexity of both the urban terrain and the atmospheric flows, the URBANTURB project is based on the realization of well-designed experiments simulating atmospheric flows over idealized urban canopies at smaller scale (1:200 scale) representative of the urban environment but also on numerical simulations of similar configurations to complement the experimental database. Wind-tunnel experiments have been chosen as they provide well-controlled and repeatable experimental conditions, allowing for very accurate measurements. State-of-the-art measurement techniques mainly based on flow imaging methods (such as stereoscopic particle image velocimetry) via the use of high-end cameras, high-energy lasers and advanced image or signals processing methods are employed. The main advantage of these techniques is that it provides detailed spatial information about the flow structure with very-fine resolution. Flow-image based methods will be combined and complemented with more classical point-wise techniques (thermal anemometry) for validation and to obtain a detailed description of the flows. To overcome the limitations of experimental techniques which can only provide data in a slice of the flow, numerical simulations will be also performed on high performance super-computers to obtain a time-evolving three-dimensional view of the flow within and over the same canopy configurations as in the wind-tunnel experiments.
Research conducted in the framework of URBANTURB is an occasion to strengthen a collaboration with the University of Western Ontario, Canada, focusing on the specific configuration of street canyons. It also falls within scope of an ongoing collaboration with the National Center for Atmospheric Research, Colorado, USA, on the interaction between the lower atmosphere and dense canopies such as forests. In direct link with the experimental research conducted within URBANTURB, the 2017 edition of the International Workshop on Physical Modelling of Flow and Dispersion Phenomena (PHYSMOD) will be held in Nantes, France, and organized by the LHEEA lab (L Perret), the project leader.
The expected outcomes of URBANTURB are an improved understanding of the physical processes that drive the exchanges between the urban-canopy and the lower-atmosphere, the derivation of accurate predictive models of the near-surface turbulence driven by the large eddies of the flow, which will help to develop new modelling strategies to be eventually included in operational numerical simulation codes.
Since the beginning of the project (January 2015), our work has led to the participation to 7 international conferences, 7 French conferences and to the publication of four articles in international peer-reviewed journals.
Understanding and modelling the dynamics of the atmospheric flow over urban terrains still represent an important scientific challenge with high stakes for our ability to handle the ever growing population in urban areas and the associated issues of air quality and urban management. The encountered difficulties arise from the high geometrical complexity of built areas, the existence of numerous interacting thermodynamics processes of both natural and anthropogenic origin that take place in the urban-canopy, and the mutual influence of the atmospheric processes. In particular, the wind field and turbulence play a crucial role in driving the instantaneous exchanges of various quantities such as momentum and heat or particles. From a purely aerodynamic point of view, the atmospheric flow over the urban-canopy can be considered as a high Reynolds number boundary layer flow developing over a heterogeneous and multi-scale surface. This flow has therefore an important multi-scale character in both the spatial and temporal domain with strong and complex inter-scale interactions. The resulting high complexity still limits our ability to understand the dynamics of urban flows but also to model them due to the prohibitive computational cost of performing obstacle resolving simulations at the district- or city-scale that would be necessary to take into account the whole range of phenomena at play.
The proposed research project URBANTURB aims at elucidating the interaction mechanisms between the lower-atmosphere and the urban-canopy flow and developing new modelling strategies to account for the interactions between these two regions of the flow. The focus will be on neutrally-stratified configurations. Recent results obtained for high-Reynolds number boundary-layer flows over smooth-walls have shed light on the nature of the coupling between the near-wall turbulence and the large-scales of the flow developing above and enabled the development of simple predictive models. Building upon these findings, the proposed project will consist in three main phases. The goal of the first one is to identify and characterize, both in the canopy region and in the boundary-layer developing above, the turbulent coherent structures that play a major role in the interaction between theses flow regions and in the transfer processes in order to extract scaling laws. This analysis will be based on the realisation of well-controlled wind-tunnel simulations of high-Reynolds number flow over rough wall or dense urban-like canopies but also on large-eddy simulation of similar configurations. The second main step will concern the investigation and the modelling of the interaction mechanisms between the large-scales present in the boundary-layer and the canopy turbulence. The aim here is to develop predictive models that accurately account for the identified interactions. In the last phase of the project, the developed models will be implemented in operational numerical codes to serve as wall- or canopy-model. The performance of the proposed modelling strategies will be assessed and their potential to improve the prediction of the aeolian resources in urban environment tested.
The expected outcomes of the URBANTURB project are an improved understanding of the physical processes that drive the exchanges between the urban-canopy and the lower-atmosphere, the derivation of accurate predictive models of the near-surface turbulence driven by the large-scales of the flow, which will help to develop new modelling strategies to be eventually included in operational numerical simulation codes.
Monsieur Laurent Perret (Laboratoire en Hydrodynamique Energétique et Environnement Atmosphérique)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LHEEA Laboratoire en Hydrodynamique Energétique et Environnement Atmosphérique
IMFT Institut de Mécanique des Fluides de Toulouse
Help of the ANR 398,840 euros
Beginning and duration of the scientific project: September 2014 - 48 Months