Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.744512
Title: Characterisations of different El Nino types, their physical causes and predictions
Author: Lai, Wang Chun
ISNI:       0000 0004 7226 6756
Awarding Body: University of Cambridge
Current Institution: University of Cambridge
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
El Niño Southern Oscillation (ENSO) is the most important interannual mode of climate variability in the tropical Pacific affecting the globe through teleconnections. The evolution of ENSO is studied with focus on individual El Nino (EN) events; factors and processes explaining the behaviours of different EN flavours are identified. The comparison to model simulations reveals a number of biases that explain differences in model behaviour. Based on reanalysis data, ENs are divided into Central Pacific (CPEN), Eastern Pacific (EPEN), and Hybrid (HBEN). ENs are found to form a continuous spectrum of events with CPEN and EPEN as its end members depending on: (1) the Western Pacific subsurface potential temperature anomaly (PTA) about 1 year before the EN peak, and (2) the Western to Central Pacific cumulative zonal wind anomaly (ZWA) between the onset and peak of the EN. Using these two parameters, about 70% of the total variance of the maximum EN SSTA can be explained up to 6 months in advance. ZWA describes the potential for triggering Kelvin waves for a given initial West Pacific recharge state as captured by PTA. A cross-validated statistical model is developed to hindcast the 1980-2016 Nov-Dec-Jan (NDJ) mean Niño3.4 SSTA based on the two parameters. The model is comparable to, or even outperforms, many NOAA Climate Prediction Centre's statistical models during the boreal spring predictability barrier. The explained variance between observed and predicted NDJ Niño3.4 SSTA at a lead-time of 8 months is 57% using five years for cross-validation. Predictive skills are lower after 2000 when the mean climate state is more La Niña-like due to stronger equatorial easterly ZWA caused by an intensification of both, Walker and Hadley cell. The ability of climate models to simulate and predict EN is assessed with data from the Climate Model Inter-comparison Project 5 (CMIP5). Most models are able to capture the main features of different EN types. But models struggle to reproduce large intensity ENs as found in observations. This issue can be traced back to a failure to realistically simulate the oceanic recharged state and the subsequent Kelvin waves for intense EN. Causes of EN involve Kelvin waves that are triggered by westerly wind bursts (WWB). From higher temporal resolution of reanalysis data, WWBs above a certain threshold are required to trigger a Kelvin wave. Kelvin waves are triggered in locations of positive Ocean Heat Content (OHC) anomalies. Intensity, longitudinal coverage and duration of a WWB, the strength of the OHC anomaly and gradient influence the amplitude of Kelvin waves as they propagate. Synoptic pattern analysis suggests that most WWBs are caused by cyclones with the combination of an active Madden-Julian Oscillation. The NorESM is able to reproduce many characteristics of observed WWBs, OHC anomalies and their relation to Kelvin waves. However, differences are noticeable for the distribution of synoptic patterns causing WWBs in the model. In future work, climate models can be used to disentangle causes and effects of EN for correlations identified here with the ultimate goal to advance our understanding of ENSO, its variability and future changes.
Supervisor: Herzog, Michael Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.744512  DOI:
Keywords: El Nino ; Climate variability ; ENSO
Share: