This work package aims to further our understanding of global climate dynamics with the overall aim of improving regional climate predictions.
Many dynamical patterns of variability - such as the El Niño Southern Oscillation (ENSO), the North Atlantic Oscillation (NAO), the Pacific Decadal Oscillation (PDO) -have near hemispheric or global impacts on regional climate. This work package aims to critically examine the performance of the latest climate models in reproducing these dynamical patterns and the fidelity of their remote teleconnections to surface climate climate and extreme events which impact society. With a focus on East Asia and Europe, we aim to develop dynamical analysis tools and diagnostics, to evaluate how well the latest generation of climate models simulate observed climate variability. In parallel with this, we will also assess the fidelity and predictability of these patterns in initialised monthly to decadal climate predictions. The ultimate aim of this work package is to improve climate predictions of extreme events in Europe and Asia.
In association with Chinese partners (CMA and IAP) we are jointly analysing our respective climate models, in free-running mode, to assess the fidelity of the large scale patterns of climate variability. In addition to the simulation of the patterns themselves, we are also evaluating the associated teleconnections to surface climate impacts, with a particular focus on East Asia and Europe. In Year 1 we are particularly focussing on identifying key common model errors in patterns of variability and surface climate teleconnections.
In Year 2 we will be examining mechanisms in the ocean and atmosphere that lead to these key model errors and how model resolution impacts these errors. One of the key tools to achieve this is the Met Office (and CMA) initialised seasonal and decadal predictions. By examining these initialised forecasts it will be possible to determine the impact of model biases on simulated patterns of variability and their remote teleconnections. We will also examine how quickly these model errors develop and whether there is any overshoot or damped oscillatory behaviour.
Running parallel to the analysis of the fidelity of model patterns of variability and teleconnections, we are also examining their predictability using initialised hindcasts from the Met Office seasonal and decadal systems. In this work we are also interested in the optimal design of forecast systems, such as ensemble size and the inherent signal-to-noise ratio in the forecast systems versus reality. For example, an exploratory piece of work into the predictability of East Asian summer rainfall has been carried out in Year 1.
Starting in Year 2, the Met Office will also be exploring how unusual dynamical situations could give rise to extreme events that we are yet to observe. Each ensemble member, from each start date (in both our seasonal and decadal prediction systems), is a potential realization of reality (limited of course by the fidelity of the model simulation). We therefore have a very large dataset to mine for extreme regional climate impacts. If unprecedented events (those beyond the observed range of variability) are found, then we can examine the dynamical situation that generated them and hence learn what is possible in a 'perfect storm' situation.
WP2.1 - Regional climate and modes of variability: El Niño Southern Oscillation (ENSO)
WP2.2 - Teleconnections: TEACliM - Teleconnections over EurAsia in Climate models
WP2.3 - Predictability of regional climate and modes of variability: subseasonal to decadal predictability in East Asia
WP2.4 Dynamics of regional extremes
Significant levels of skill (r=0.76) over the Yangtze River catchment is present in the GloSea5 hindcasts. This high skill, compared to previous forecast systems, appears to be due to improved modelling of the SST variability over the tropical western Pacific and its associated teleconnection to East Asian climate. A paper is in final stages of preparation for submission (Li et al., 2015).
Last updated: 27 February 2015