Understanding and Modelling Natural and Forced Variability of Near-Surface Winds

点击翻译按钮获取中文摘要

Near-surface winds control the exchange of mass, energy, and momentum between the atmosphere and the underlying surface; they influence surface temperature and air quality and represent an energy resource of growing importance; and their extremes represent hazards to the natural and built environments. The proposed research is comprised of four complimentary wind-related research projects addressing urgent questions in the field. Future climate changes will be associated with changes in wind and solar energy (as well as of their co-variability which can mitigate the intermittency of renewables). In the first project, we will use ensembles of climate model simulations to quantify uncertainties related to North American renewable energy in the 21st century. We will also relate interannual and interdecadal variability in power resources to dominant modes of climate variability such as El Nino. Near-surface wind speed is determined by processes across the Planetary Boundary Layer (PBL). The nocturnal PBL displays two regimes, with very different temperature and wind profiles. These structures are poorly represented in weather and climate models, resulting in temperature and wind biases. The second project will apply novel mathematical tools to idealized physical models to investigate these regimes. We will then consider the role of irregular weather variations in driving regime transitions. Extremes of sea surface wind components are substantially affected by asymmetries in their fluctuations around the mean. I have developed a set of idealized models to account for this behaviour. A novel perspective on the physical origin of these asymmetries is provided by the bispectrum, which partitions this asymmetry into different frequency contributions. In the third project, we will use bispectra to further investigate the physics of wind component fluctuation asymmetry in observations and in an idealized model. Air-sea exchanges in weather and climate models depend on wind speed. Small-scale winds not resolved in these models contribute to flux enhancements that the models must parameterize. While parameterizations normally assign a unique subgrid-scale flux to each unique state of the resolved flow, we have shown that in fact these corrections are better represented by probability distributions. The fourth project will extend our earlier analysis to longer records and larger domains, with the goal of implementing such a parameterization in an operational weather model.