Abstract/Summary

Societies in the Maritime Continent depend on monsoon rainfall for their water supply. Large spatio-temporal variability of the rains has a significant socio-economic impact and affects the global circulation. However, the Maritime Continent remains a major modelling challenge. In this thesis, the fidelity of the fifth Coupled Model Intercomparison Project (CMIP5) models at simulating mean climate and its variability over the Maritime Continent is assessed. We find that model horizontal resolution is not a good indicator of performance in atmosphereonly models. Instead, a given model’s local Maritime Continent biases are somewhat related to biases in the local Hadley circulation and global monsoon. Cluster analysis on Maritime Continent annual cycle precipitation results in two distinct clusters: Cluster I models are able to capture both the winter monsoon and summer monsoon shift, whereas Cluster II models simulate weaker seasonal migration than observed. A model’s Maritime Continent climatological mean-state precipitation is shown to be negatively correlated with sea-surface temperature (SST) biases in central tropical Pacific Ocean (CTPO) and western tropical Indian Ocean (WTIO) regions in coupled CMIP5 models. On interannual timescales, the El Nino-Southern Oscillation (ENSO) and Indian Ocean ˜ Dipole (IOD) teleconnections to Maritime Continent precipitation are well simulated by both uncoupled and coupled CMIP5 models. However, the spatial pattern of these teleconnections is not well captured, especially in coupled models. Idealized Maritime Continent SST-perturbation experiments are performed using the HadGEM3-GA6 atmospheric model in the CPTO and WTIO regions. These result in remote responses of Maritime Continent mean precipitation, somewhat comparable with the signals related to biases found in the CMIP5 coupled models. This suggests that remote Indo-Pacific SST biases in the CMIP5 coupled models could plausibly cause the precipitation biases over the Maritime Continent, and thus highlights the importance of reducing these biases to improve Maritime Continent mean state and climate variability.