The forecast skill of a coupled multi-model ensemble (MME) in predicting two main types of the El Niño-Southern Oscillation (ENSO), (namely the canonical or “cold tongue” and the “warm pool” or El Niño Modoki), and their regional climate impacts, is assessed for the boreal winter. Based on a hindcast set of boreal winter predictions for the period 1972–1988 and 1989–2005, we show that the MME is able to predict and discern important differences between the canonical and warm pool ENSOs in the patterns of tropical Pacific sea surface temperature anomalies, 1 and 4 month ahead, for the two periods, respectively. The coupled MME predicts decadal changes of the distinct impacts of the canonical ENSO reasonably well, including the intensified winter monsoon rainfall over East Asia, the weakened below-normal rainfall over Australia, the strengthened anomalously wet conditions across the southern USA, and the anomalously dry conditions over South America. However, there are some limitations in the MME’s ability to capture rainfall anomalies associated with the different types of ENSO for the early period, at a lead time of 1 and 4 months, particularly over the Indian Ocean, equatorial western and central Pacific. Nevertheless, the forecast skills for rainfall over East Asia during the different types of ENSO are comparable to, or slightly higher than, those during canonical ENSO events.

1. INTRODUCTION

1.1 Climate regime shift

1.2 Two types of ENSO and their impacts

1.3 Objectives of this study

2. METHODOLOGY

2.1 Data used

2.2 Evaluation methods

3. RESULTS

3.1 Changes in mean and variance

3.2 Changes in the major mode

3.3 Regional impact by major mode changes

4. CONCLUSION