abstract.tex

\begin{abstract}

The major zonally asymmetric features of the Southern Hemisphere (SH) extratropical circulation are the zonal wavenumber one (ZW1), zonal wavenumber three (ZW3) and the Pacific-South American (PSA) pattern. These tropospheric waveforms play a critical role in the meridional transport of heat and moisture and in the development of blocked flow, causing the regional surface climate to vary strongly depending on the strength, frequency and phase of their activity. The PSA pattern is widely regarded as the primary mechanism by which the El Ni\~{n}o-Southern Oscillation (ENSO) influences the high southern latitudes, and in recent years it has been suggested as a mechanism by which longer-term tropical sea surface temperature trends have influenced the Antarctic climate.

This thesis presents novel approaches to identifying both the zonal waves and PSA pattern in reanalysis and model output. In comparison to existing wave identification methods, the approaches more fully exploit the information available from Fourier analysis. For the zonal wave analysis, this was achieved by adapting the wave envelope construct recently used in the identification of synoptic-scale Rossby wave packets. In order to apply similar methods to the non-zonal PSA pattern, a grid rotation method traditionally used in ocean modelling was used to orient the equator along the approximate great circle path of the pattern. These new wave identification methods were applied to ERA-Interim reanalysis data in order to analyse the climatological characteristics of the waveforms and their influence on regional climate variability. The results reveal that both the zonal waves and PSA pattern are important drivers of temperature, precipitation and sea ice variability in the mid-to-high southern latitudes. While ZW1 and ZW3 are both prominent features of the climatological circulation, the defining feature of highly meridional hemispheric states is an enhancement of the ZW3 component. Identified seasonal trends towards the negative phase of the PSA pattern were largely inconsistent with recent high latitude temperature and sea ice trends. Only a weak relationship was identified between the PSA pattern and ENSO, suggesting that the pattern might be better conceptualised as preferred regional atmospheric response to various external (and internal) forcings.

The analysis of large datasets such as ERA-Interim typically requires extensive use of various software tools and packages, to the point where coding/programming is a major component of the research methodology. Despite this strong reliance on computation, traditional academic publishing formats and conventions do not allow for the documentation of computer software and code, which means it is impossible to replicate and verify much of today's academic literature. In an attempt to provide a practical solution to this so-called reproducibility crisis, the zonal wave and PSA pattern results have been presented in a reproducible manner. The procedure used to document the computational aspects of the research was developed to be consistent with recommended best practices in scientific computing and seeks to minimise the time burden on authors. It should provide a starting point for weather and climate scientists looking to publish reproducible research, and it is proposed that relevant academic journals could adopt the procedure as a formal minimum standard. 


\end{abstract}

\clearpage