This proposed research activity builds on the progress of the MJO sub-project and teleconnections sub-projects in Phase I. During Phase I much of the MJO research associated with the S2S sub-projects focused on the predictive skill for the MJO (e.g. Vitart, 2017), including: the relationship between predictive skill and model biases (e.g. Kim, 2017; Lim et al., 2017); and changes in basic state associated with the MJO (e.g. Marshall et al., 2016). Whilst it is expected that much of this research activity will continue, within Phase II wepropose two new foci for activity to be promoted within the S2S project, and in collaboration with the WGNE MJO Task Force, described below.

A number of studies have identified the relationship between the MJO and high-impact weather throughout the tropics (e.g. Jones et al., 2004; Chang et al., 2005; Xavier et al., 2014; Sossa et al., 2017), however little research has been carried out to evaluate the ability of operational S2S models to capture these relationships between the MJO and high-impact weather, or the implications for sub-seasonal predictive skill for these events. The S2S project will promote research to address the following related questions:

This activity will benefit from links developed within the international community through e.g. the S2S Africa sub-project; the Years of Maritime Continent; and the WWRP High Impact Weather Project.

The statistical relationship between the MJO and extratropical circulation patterns is well established (e.g. Matthews et al., 2004; Cassou et al., 2008; Mori and Watanabe, 2008). This relationship provides enormous potential for sub-seasonal predictive skill for the extratropical circulation and associated weather. Realizing this potential predictability requires a faithful simulation of these teleconnections by operational prediction models. For example, Vitart (2014) showed that prior to 2007 the ECMWF forecast monthly forecast system had lower predictive skill for the NAO when the MJO was active in the initial conditions than without an active MJO, likely because of the model’s loss of MJO amplitude. Following improvements in the model the ECMWF forecast skill has improved skill of the NAO when the MJO is active, realizing some of that potential predictability. However, S2S models still underestimate the strength of the extratropical response to the MJO (Vitart et al., 2017) which can impact on their skill for the weather variables (e.g. temperature and precipitation). Furthermore, a number of studies have identified the dependence of the MJO-extratropical teleconnections on both the propagation characteristics of individual MJO events (e.g. Yadav and Strauss, 2017); and the slowly varying basic state (e.g. Roundy et al., 2010). This research activity will address the following research questions which relate both to the observed MJO teleconnections to the extratropics and the predictive skill of operational S2S models.

This activity will benefit from links developed with the Year of Tropical –Mid-latitude Interactions, Teleconnections sub-project, and the Working Group on Subseasonal to Interdecadal Prediction (WGSIP) Initiative Interaction/Teleconnection between tropics and extratropics.