| dc.description.abstract | Tropical precipitation naturally exhibits considerable spatial and interannual variability. The dominant mode of temporal variability is the El Niño-Southern Oscillation (ENSO), where sea surface temperature (SST) fluctuations in the eastern equatorial Pacific cause global weather effects. Meanwhile, changes to tropical precipitation are beginning to emerge from natural variability, caused by anthropogenic global warming. Future projections using state-of-the-art coupled climate models indicate substantial mean state and variability changes in response to increasing greenhouse gas emissions over the twenty first century. A large scale wet-get-wetter and dry-get-drier response is expected, mainly due to increased moisture in a warmer atmosphere. However, at policy-relevant regional scales this paradigm is inadequate, and the pattern of tropical precipitation changes remains uncertain, despite decades of modelling developments. Hence, the aim of this thesis is to improve our understanding of the mechanisms which control tropical precipitation, its variability, and future changes. Three major results are presented in this thesis. Firstly, a common mechanism is found which bridges present day tropical precipitation interannual variability with simulated future mean state changes. Generally, the heaviest precipitation remains situated above the relatively warmest and most humid regions of the tropics, even as these regions shift geographically. Secondly, in an ensemble of HadGEM2-A simulations, there is evidence of a strong link between tropical land minus ocean temperature and relative humidity contrasts and the overlying atmospheric circulation, in response to idealised global warming perturbations. Since circulation and precipitation changes are strongly related in the tropics, this implies changes over tropical land are likely to be linked to regional patterns of future precipitation change. This extends results in the literature describing the important role that SST pattern and circulation changes play in tropical precipitation changes. Finally, significant changes in ENSO anomalous precipitation are simulated in the atmospheric response to either uniform or patterned SST warming, or increased CO2 forcing. Hence, in coupled projections and reality, changes to ENSO-related precipitation are likely to result from a careful balance of several interacting process. | en_GB |
