The Initial Condition Errors Occurring in the Indian Ocean Temperature That Cause “Spring Predictability Barrier” for El Niño in the Pacific Ocean

The Community Earth System Model is used to explore the influences of the initial condition errors in sea temperatures in the Indian Ocean (related initial condition errors) on the uncertainties in the Pacific El Niño predictions. Two categories of Indian Ocean‐related initial condition errors are shown to frequently induce the “spring predictability barrier” for El Niño events. The category‐1 error consists of a positive Indian Ocean Dipole‐like sea temperature pattern over the Indian Ocean while the category‐2 initial condition error has a negative Indian Ocean Dipole‐like sea temperature structure and is nearly opposite to that of the category‐1. Both categories of errors underestimate the Pacific El Niño in terms of its amplitude, although they present different spatial structures and undergo different mechanisms. Specifically, the category‐1 error induces negative errors in El Niño predictions in terms of its amplitude by tropical oceanic channel Indonesian Throughflow while the category‐2 error exerts its impact on El Niño prediction and makes it underestimate through the atmospheric bridge. Both categories of initial condition errors emphasize the sensitivity of El Niño predictions to the initial uncertainties in sea temperatures in the Indian Ocean. They may therefore provide useful information on how initialization in the Indian Ocean can be utilized to improve El Niño predictions.