Computer models used to predict the weather (known as numerical weather prediction or NWP models) and climate (known as climate models) do not simulate the layer of the atmosphere nearest to the ground, known as the atmospheric boundary layer (ABL), accurately by default, because their computational grids are too coarse to represent the turbulence that transports most flow properties in that layer of the atmosphere. The ABL regulates the fluxes of momentum, heat and pollutants that determine the lower boundary conditions in NWP and climate models, and therefore the resolved profiles of the mean wind, temperature and pollutant concentrations, and their variability, near the ground. The surface layer (SL), which corresponds to the lowermost 10% of the ABL, is described by the classical Monin-Obukhov Similarity Theory (MOST), which is applicable in a comprehensive set of meteorological conditions over flat terrain. The physics of the SL in NWP or climate models must be approximated using so-called parametrizations and, given its success, these parametrizations are almost universally based on MOST. However, this approach, which assumes statistically horizontally homogeneous flow, has been shown to produce large errors over mountainous terrain, and more generally over any horizontally heterogenous terrain, for example characterized by sharp transitions in land use.
The ultimate aim of this PhD is to develop a new paradigm to represent the SL, to be able to apply an appropriate lower boundary condition in NWP and climate models over mountainous terrain. Some progress has been made in this direction by Stiperski et al. (2019), who showed that MOST can be adapted in such a way that it becomes more accurate over mountainous terrain, if the turbulence is sorted by different types of anisotropy (i.e. depending on which components of the turbulent velocity fluctuations are dominant). However, a connection between anisotropy and flow or terrain characteristics that can be used as input parameters to a parametrization based on this new theory is currently lacking. The main aim of this PhD project is to discover such a link. This will be done under the umbrella of the TEAMx international partnership (http://www.teamx-programme.org/), whose objectives align with the topic of this PhD project, and which congregates the University of Reading (the host institution), the Met Office and the University of Innsbruck, all of them involved in the supervision of this PhD project. This will be done using a combination of theory, observations and numerical simulations, as follows.
A large database of field measurements within the SL over mountainous terrain, collected over the last 7 years in 6 mountain terrain locations by the partner institution University of Innsbruck, available to the lead supervisor through the TEAMx partnership and an ongoing Royal Society International Exchanges Project, will be processed to discover relationships as general as possible between ratios of meteorological quantities characterizing the SL, to build understanding of the processes currently neglected in MOST. Additional relevant measurements of unprecedented quality and detail will be made during the TEAMx observational campaign, to take place between August 2024 and September 2025. A theoretical method known as Rapid Distortion Theory, alongside the same kind of scaling arguments that originated MOST and its extension by Stiperski et al. (2019), will be used to obtain insights into SL turbulence over mountainous terrain, and inspire new ideas to parametrize it. This aspect will build on the lead supervisor’s experience of applying RDT to the upper ocean affected by waves (Teixeira, 2018), as the waviness of mountainous terrain has somewhat similar effects on the turbulence, through vorticity distortion by streamline curvature. From a numerical modelling perspective, DNS data obtained from simulations carried out in the context of an ongoing ERC grant held by the co-supervisor at the partner institution University of Innsbruck will be used to understand turbulence processes by simulating them explicitly (i.e. without any turbulence closures), and inspire further the formulation of parametrization ideas. A parametrization supported by scaling, RDT and DNS results will then be formulated and implemented in a NWP model, which will be run in km-scale simulations over mountainous terrain, to test how well these ideas work in practice.
This project will potentially lead to improvements in the accuracy of models used to predict the weather and climate over mountainous regions, where many communities reside, benefitting both the academic community and society as a whole.
Eligibility requirements - This project is suitable for students with a good (1st class or upper 2nd class) degree in physics, mathematics or a closely related environmental or physical science. Good computational skills are essential. Experience in numerical modelling, applied mathematics and data processing would be preferred but are not essential. The student should be enthusiastic, eager to learn, and have a keen interest in physical and mathematical aspects of atmospheric dynamics.
Continue with Facebook
