Abstract
Abstract. This study investigates parameter estimation (PE) to enhance climate forecasts of a coupled general circulation model by adjusting the background vertical diffusivity coefficients in its ocean component. These parameters were initially identified through sensitivity experiments and subsequently estimated by assimilating the sea surface temperature and temperature–salinity profiles. This study expands the coupled data assimilation system of the Community Earth System Model (CESM) and the ensemble adjustment Kalman filter (EAKF) to enable parameter estimation. PE experiments were performed to establish balanced initial states and adjusted parameters for forecasting the El Niño–Southern Oscillation (ENSO). Comparing the model states between the PE experiment and a state estimation (SE) experiment revealed that PE can significantly reduce the uncertainty of these parameters and improve the quality of analysis. The forecasts obtained from PE and SE experiments further validate that PE has the potential to improve the forecast skill for ENSO.