Bibliography: leaves 93-101. / Relationships between the El Niño-Southern Oscillation phenomenon (ENSO), seasonal rainfall, and atmospheric circulation patterns at the 1000 hPa and 500 hPa levels are investigated. Firstly, correlations between early-summer (October to December), late-summer (January to March) rainfall over South Africa, and sea-surface temperatures in the NINO3 region in the equatorial Pacific Ocean were investigated, where the correlations in the case for late-summer showed much better spatial coherence than in the case for early-summer. Consequently, the study further concentrated only on late summer. The influence of the quasi 18-20 year oscillation of summer rainfall on the effect of El Niño and La Niña events was also investigated, and it was found that during an epoch of above-normal/below-normal rainfall a moderating effect is evident on the severity of El Niño/La Niña events so that on average even above-normal/below-normal rainfall is experienced during such events. Canonical correlation analysis (CCA) was applied to different combinations of years, to find associations between equatorial Pacific sea-surface temperatures and deviations in circulation patterns during certain years. Some at the above results were then verified by model runs, to further prove the relationships not to be coincidental, and to add more degrees of freedom. The results of CCA were then separately interpreted for each combination of seasons (e. g. El Niño during the above-normal phase at the 500 hPa level) with the aid of average circulation maps for different combinations of years. Above - or below-normal rainfall during such years could then be explained in terms of deviations of general synoptic features at the surface and 500 hPa levels.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

As the largest mode of coupled climate variability, the El Niño Southern Oscillation (ENSO) carries consequences for weather patterns worldwide. In turn, communities that live in areas affected by ENSO variability would benefit from reliable and timely information on the occurrence of such events. To address this need, there has been an on-going effort within the scientific community to investigate and characterize the mechanisms that give rise to ENSO events. One of the greatest impediments to this effort, however, is that the ENSO system can behave both as a self-sustained, deterministic oscillation, and as a response to stochastic forcing. In this dissertation, we uncover a key determinant of these two types of ENSO behavior – namely North Pacific Oscillation (NPO)-induced variations in the northeasterly trade winds – and analyze how the variations in these trade winds influence ENSO variability historically and into the future. The first Chapter of this dissertation provides a thorough review of previous efforts to understand the initiation, onset, and evolution of ENSO events with a particular focus on the relationship between ENSO events and two NPO-related precursors, namely the Trade Wind Charging and the Northern Pacific Meridional Mode (TWC/NPMM). In Chapter 2 (Pivotti and Anderson 2021), we study the TWC/NPMM-ENSO relation over 140 years and uncover that there has been a multi-decadal modulation in the strength of TWC/NPMM’s influence upon ENSO events. Further, as a consequence of this modulation, ENSO behavior shifted between a stochastically forced interannual mode of variability to a multi-annual, quasi-regular one with a self-sustained oscillation and back again over the course of the last 140 years. In Chapter 3, we assess how the TWC/NPMM-ENSO relationship is reconstructed in state of the art coupled climate models. We find that across the ensemble the TWC/NPMM is reconstructed by all models as the leading ENSO precursor. Further, a subset of better-performing models confirms that during those periods when the TWC/NPMM-ENSO coupling is weak, ENSO oscillates regularly with constant periodicity, whereas when the coupling is strong, ENSO shows a more stochastic behavior. In Chapter 4, we introduce experiments with increasing anthropogenic radiative forcings from the same ensemble of coupled climate models evaluated in Chapter 3. We find that ENSO events initiated by TWC/NPMM are consistently stronger than standard ENSO events, independent of the presence/absence of increasing external forcings, however neither TWC/NPMM-induced ENSO events nor standard events show any systematic change in intensity in the presence of increasing atmospheric forcings. Further, there is no systematic change in TWC/NPMM-ENSO coupling in the presence of increasing external forcing and hence no systematic change in the oscillatory (or stochastic) behavior of the ENSO system. Finally, Chapter 5 of this dissertation includes a concluding summary and suggestions for future work. In summary, this dissertation shows for the first time that the ENSO system can behave both as a self-sustained oscillation and as a response to stochastic forcing, that the modulation of this behavior is linked to the strength of TWC/NPMM-ENSO coupling, and that the strength of this coupling is the result of multi-decadal internal climate variability and not human-induced climate change.