A coupled atmosphere–ocean model is developed to study the El Niño–Southern Oscillation (ENSO) phenomenon. The model, with no external forcing, reproduces key features of ENSO, including the recurrence of warm events at irregular intervals with a preference for 3–4 years. The model shows that the mean sea surface temperature, wind, and ocean current fields determine the spatial structure of ENSO anomalies. The phase-locking of anomalies is explained by variations in coupling strength associated with the annual cycle in the mean fields. Sensitivity studies reveal that the amplitude and time scale of the oscillation are sensitive to parameters affecting atmosphere–ocean coupling. Stronger coupling leads to larger oscillations with a longer time scale. A critical element of the model is the variability in the equatorial heat content of the upper ocean, which increases prior to warm events and decreases sharply during them. A theory for this variability and the transitions between non-El Niño and El Niño states is presented. The model results suggest that the oscillation is driven by the interaction between the tropical ocean and atmosphere, with the annual cycle playing a key role in determining the preferred period of 3–4 years and the aperiodicity of ENSO. The model shows that the annual cycle influences the development of anomalies, with the summer period being most favorable for rapid growth of both positive and negative anomalies. The results indicate that the annual cycle modulates the coupling strength, which is essential for the phase-locking of anomalies to the annual cycle. The model also shows that the annual cycle is not essential for the system's tendency for interannual oscillation, though it contributes to the aperiodicity of the full model. The model's oscillation is driven by the interaction between the tropical ocean and atmosphere, with the annual cycle playing a key role in determining the preferred period of 3–4 years and the aperiodicity of ENSO. The model's results suggest that the annual cycle modulates the coupling strength, which is essential for the phase-locking of anomalies to the annual cycle. The model also shows that the annual cycle is not essential for the system's tendency for interannual oscillation, though it contributes to the aperiodicity of the full model.A coupled atmosphere–ocean model is developed to study the El Niño–Southern Oscillation (ENSO) phenomenon. The model, with no external forcing, reproduces key features of ENSO, including the recurrence of warm events at irregular intervals with a preference for 3–4 years. The model shows that the mean sea surface temperature, wind, and ocean current fields determine the spatial structure of ENSO anomalies. The phase-locking of anomalies is explained by variations in coupling strength associated with the annual cycle in the mean fields. Sensitivity studies reveal that the amplitude and time scale of the oscillation are sensitive to parameters affecting atmosphere–ocean coupling. Stronger coupling leads to larger oscillations with a longer time scale. A critical element of the model is the variability in the equatorial heat content of the upper ocean, which increases prior to warm events and decreases sharply during them. A theory for this variability and the transitions between non-El Niño and El Niño states is presented. The model results suggest that the oscillation is driven by the interaction between the tropical ocean and atmosphere, with the annual cycle playing a key role in determining the preferred period of 3–4 years and the aperiodicity of ENSO. The model shows that the annual cycle influences the development of anomalies, with the summer period being most favorable for rapid growth of both positive and negative anomalies. The results indicate that the annual cycle modulates the coupling strength, which is essential for the phase-locking of anomalies to the annual cycle. The model also shows that the annual cycle is not essential for the system's tendency for interannual oscillation, though it contributes to the aperiodicity of the full model. The model's oscillation is driven by the interaction between the tropical ocean and atmosphere, with the annual cycle playing a key role in determining the preferred period of 3–4 years and the aperiodicity of ENSO. The model's results suggest that the annual cycle modulates the coupling strength, which is essential for the phase-locking of anomalies to the annual cycle. The model also shows that the annual cycle is not essential for the system's tendency for interannual oscillation, though it contributes to the aperiodicity of the full model.