Impact of Pacific sea surface temperature anomalies on the tropical Atlantic and role of the
background mean state

Project facts

Project promoter
Complutense University of Madrid
Project Number:
Target groups
Researchers or scientists
Initial project cost:
Final project cost:
From EEA Grants:
€ 5,854
The project is carried out in:


El Niño is the most prominent mode of variability at interannual time scales. There is a debate regarding the relationship between this mode and sea surface temperature (SST) variability in the Tropical Atlantic. This project aims to understand the impact of the Pacific El Niño on the Tropical Atlantic region. With this, we expect to gain insight on the connection of the Pacific and Atlantic basins and the dynamical mechanisms at play and to test whether this impact can change depending on the mean background SST. The project approach is based on dedicated sensitivity experiments with two state of the art general circulation models. A deeper knowledge of El Niño impacts on the Tropical Atlantic can help prediction in this area, which still remains quite uncertain as regards seasonal predictions and long-term projections. The project promoter, Complutense University of Madrid, has long experience in Tropical Climate variability and can provide knowledge of the tele-connections mechanisms affecting the Tropical Atlantic variability and predictability. The donor partner, the University of Bergen, has strong competence in decadal climate predictability.

Summary of project results

After an initial assessment of the situation at the beginning of the stay (August 2015), partners decided to collaborate in the investigation of the role of the background mean state (decadal variability modes) for prediction purposes in the Tropical Atlantic but with a different approach: we focused on the decadal time scale and on the Sahel region. Though this was not the initially intended focus, the collaboration fostered by the NILS grant has shown to be very fruitful both, for the understanding of decadal prediction skill and to promote further collaboration. During the stay partners have analysed the decadal prediction runs performed with the MPI-ESMLR model (Muller et al. 2014) in the 1901-2010 period to estimate the skill in predicting Sahel rainfall at decadal time scales and the sources of such skill. Several experiments have been performed with interesting results. Partners are currently working on the draft of a paper with the main results of the work carried out. They hope it will be submitted before the end of the year. In parallel, they established other collaborations and hope the work will be communicated to the Scientific Community in the form of publications. Decadal prediction is just starting. The first experiments began in the late 2000s. In this framework, project results will improve the understanding of the sources of decadal predictability, particularly for Sahel rainfall. They also provide a framework to better understand the skill in decadal predictions and what could be feasible with such predictions. Partners hope that, in the long term, the results will also contribute to enhance the skill in the decadal predictions performed with dynamical models.

Summary of bilateral results

1. In the short term, partners plan to work on a manuscript that summarizes the results of the analysis on the sources of skill (or lack of) in the decadal prediction of Sahel rainfall. 2. Partners are currently collaborating to set up and coordinate the anomaly coupling experiments that, unfortunately, were not ready at the time of the stay. 3. Partners also plan to collaborate in researching the causes for the strong trends shown by the decadal predictions in the tropics, especially over the Pacific. For such research, they are currently collaborating with Wolfgang Muller (see above). Main hypothesis is that such trends could be due to the use of the 20CR reanalysis product for the initialization. 4. Another common line of research is variability at intra-seasonal timescales. In particular, partners have agreed to analyse the outputs from a model with a particularly well simulated MJO (Tseng et al. 2015) to evaluate the impact on West Africa under global warming conditions (Chang et al. 2015). 5. Partners will also address the current scientific debate on the causes for the Sahel rainfall recovery (Dong and Sutton 2015) with the use of experiments performed in the framework of the GREENICE project.