The North American Multimodel Ensemble prediction experiment is described, and forecast quality and methods for accessing digital and graphical data from the model are discussed.

Real-time model predictions of ENSO conditions during the 2002–11 period are evaluated and compared to skill levels documented in studies of the 1990s. ENSO conditions are represented by the Niño- 3.4 SST index in the east-central tropical Pacific. The skills of 20 prediction models (12 dynamical, 8 statistical) are examined. Results indicate skills somewhat lower than those found for the less advanced models of the 1980s and 1990s. Using hindcasts spanning 1981–2011, this finding is explained by the relatively greater predictive challenge posed by the 2002–11 period and suggests that decadal variations in the character of ENSO variability are a greater skill-determining factor than the steady but gradual trend toward improved ENSO prediction science and models. After adjusting for the varying difficulty level, the skills of 2002–11 are slightly higher than those of earlier decades. Unlike earlier results, the average skill of dynamical models slightly, but statistically significantly, exceeds that of statistical models for start times just before the middle of the year when prediction has proven most difficult. The greater skill of dynamical models is largely attributable to the subset of dynamical models with the most advanced, highresolution, fully coupled ocean–atmosphere prediction systems using sophisticated data assimilation systems and large ensembles. This finding suggests that additional advances in skill remain likely, with the expected implementation of better physics, numeric and assimilation schemes, finer resolution, and larger ensemble sizes.

The successive failure of the East African short rains (typically October‐December) and subsequent long rains (March‐May) in 2010–11 plunged much of the region into severe drought, impacting millions of people and triggering a humanitarian crisis. While poor short rains in 2010 were generally anticipated given linkages with La Niña, the subsequent long rains do not exhibit similar predictability. Here we show the long rains failure in boreal spring of 2011 is consistent with a recurrent large‐scale precipitation pattern that followed their abrupt decline around 1999. Using observations and climate model simulations, we show the abrupt decline in long rains precipitation is linked to similarly abrupt changes in sea surface temperatures, predominately in the tropical Pacific basin.

An ensemble of twenty four coupled oceanatmosphere models has been compared with respect to their performance in the tropical Paci®c. The coupled models span a large portion of the parameter space and dier in many respects. The intercomparison includes TOGA (Tropical Ocean Global Atmosphere)-type models consisting of high-resolution tropical ocean models and coarse-resolution global atmosphere models, coarse-resolution global coupled models, and a few global coupled models with high resolution in the equatorial region in their ocean components. The performance of the annual mean state, the seasonal cycle and the interannual variability are investigated. The primary quantity analysed is sea surface temperature (SST). Additionally, the evolution of interannual heat content variations in the tropical Paci®c and the relationship between the interannual SST variations in the equatorial Paci®c to¯uctuations in the strength of the Indian summer monsoon are investigated. The results can be summarised as follows: almost all models (even those employing¯ux corrections) still have problems in simulating the SST climatology, although some

Table A1. Basic information and references for the 20 ENSO prediction models whose real-time forecasts, and some of their longer-term hindcasts, are evaluated in this paper. Fully Coupled or Anomaly Coupled Dynamical ModelsThe European Centre for Medium-Range Weather Forecasts (ECMWF) fully coupled ocean-land-atmosphere model has had two versions over the forecast period. System 2 (S2; Anderson et al. 2003; Balmaseda et al. 2004; Alves et al. 2004) was operational from the beginning of the period through System 3 (S3) was the operational model (Anderson et al. 2007; Balmaseda et al. 2008; Stockdale et al. 2011). The atmospheric model in S3 is cycle 31r1 of the ECMWF Integrated Forecast System (IFS) with a spectral truncation at T159 with 62 vertical levels, while that for S2 was cycle 23r4 with a spectral truncation of T95 and 40 vertical levels. Both S2 and S3 used the same oceanic model, the Hamburg Ocean Primitive Equation Model (HOPE), version 2 (Latif et al. 1994; Wolff et al. 1997), with a horizontal resolution of 1° zonally and a telescoping resolution in the meridional direction of 1° in midlatitudes and 0.33° near the equator. The ocean model has 29 vertical layers with 10-m resolution over the upper 100 m. The Met Office (UKMO) uses the Global Seasonal (GloSea) coupled model. GloSea has had four versions over the period of this study. GloSea1 became operational in March 2003. It was based on the third climate configuration of the Met Office Unified Model (HadCM3) coupled climate model (Gordon et al. 2000). The AGCM component model had a horizontal resolution of 3.75° in the zonal direction and 2.5° in the meridional direction with 19 levels in the vertical. The OGCM component model had a horizontal resolution of 1.25° in the zonal direction and a telescoping grid in the meridional direction with 0.28° resolution near the equator and stretching to 1.25° in the middle and high latitudes. GloSea2 became operational in October 2004 and used essentially the same AGCM and OGCM as version 1; in particular, the component models had the same resolution. GloSea3 became operational in September 2005. The AGCM component model had the same resolution as previous versions and the OGCM horizontal resolution was also the same. GloSea4 (Arribas et al. 2011) became operational in October 2009.The AGCM resolution is 1.25° in latitude and 1.875° in longitude with 38 vertical levels. The OGCM has a horizontal resolution of 1.0° in the zonal direction; in the meridional direction the resolution telescopes from 0.3° near the equator to 1.0° in the middle and higher latitudes with 42 vertical levels.