WP3: Natural Climate Variability
We aim to establish a theory that connects changes in network properties to specific physical processes of climate variability on interannual-to-interdecadal time scales. An important issue in this part of LINC will therefore be to connect network characteristics and measures (e.g. closeness centrality, betweenness centrality) to specific physical processes, which provides coupling between WP3 and WP1). As a first step we will construct networks from results of minimal models for important phenomena of climate variability (e.g. ENSO, AMO) and study their properties while varying model parameters. Based on these results, as a second step we will aim to develop new mechanistic indicators based on network analysis. These indicators are crucially important in the interpretation and evaluation of GCM results and observations. As a third step we will aim to characterize and understand the interaction between patterns of climate variability (e.g. ENSO-PDO) using network measures. To do so, we will use long (1000+ years) control (constant CO2) simulations from the IPCC-AR413 and the future CMIP5 database in which the variability patterns as well as their global connections are well represented (which provides coupling between WP3 and WP4). Focus will be on how to characterize the global coupling of the patterns in terms of network properties and how to connect these to the physics of how information (e.g. heat and mass) is transferred in the climate system.
A strong reduction of the Atlantic MOC also has a global impact on the patterns of interannual-to-multidecadal variability in GCMs. In LINC we also want to study the changes in network properties of SST anomalies and MOC resulting from a strong reduction in the MOC. We will use two types of simulations with GCMs: one in which the reduction is associated with a crossing of a tipping point and one in which it is not (which provides coupling between WP3 and WP5). The aim is to develop new network-based measures to distinguish both types of behaviour of the MOC.
In summary, the specific objectives of WP3 are: 1) improving the physical interpretation of changes in network properties for phenomena of low-frequency climate variability; 2) develop test quantities (mechanistic indicators) based on network analysis for specific physical mechanisms of natural variability and apply them to GCM results and observations; 3) characterize the global interaction between low-frequency patterns of variability using networks; and 4) study changes in network properties associated with Atlantic MOC variability with the aim to distinguish behaviour associated with the presence of a tipping point.