Description:
<jats:title>Abstract</jats:title><jats:p>Sea surface temperature (SST) variability in the tropical Pacific (TP) has world‐wide climate and economic influences; hence, improving its prediction is of great interest. Recent modeling studies suggested that north tropical Atlantic variability can influence TP SST variability and prediction. However, models are subject to biases in both the tropical Atlantic and Pacific, and it remains an open question whether these biases affect the trans‐basin linkages and the prediction of TP SST. To investigate this issue, we apply linear inverse modeling to observations and climate models. We find that removing north tropical Atlantic forcing indeed strongly lowers TP SST multiyear prediction skills in observations and models. Models show a large spread of the Atlantic trans‐basin effect, which leads to markedly different TP SST multiyear prediction skills. Our study highlights the necessity to narrow the inter‐model spread for more reliable TP SST multiyear prediction.</jats:p>