Skillful Prediction of Length of Day One Year Ahead and Its Relationship with El-Nino Southern Oscillation and Quasi-Biennial Oscillation

Length of day (LOD), or the rate of the Earth’s rotation, is an important variable in geodesy. For instance, despite a small amplitude, it determines the accuracy of GPS time calculations. The LOD is not stationary in time but varies from one year to another due to changes in atmospheric angular momentum (AAM) partly driven by El-Nino Southern Oscillation (ENSO) and Quasi-Biennial Oscillation (QBO). A recent study reported that LOD, AAM, and AAM-related extratropical climate signals can be predicted one year in advance by decadal prediction system. In this study, we extend the previous study by examining nine decadal prediction systems archived for the Coupled Model Intercomparison Project phase 6 (CMIP6) Decadal Climate Prediction Project (DCPP). A persistence and rebound in LOD prediction skill at a lead time of one year or longer is found not only in multi-model mean but also in individual models, confirming the recent study. A poleward propagation of AAM anomaly via wave-mean flow interaction is also qualitatively well reproduced. This long-lead prediction skill is attributed to reliable predictions of ENSO and QBO over a year, being LOD prediction skill more systematically related with ENSO prediction skill in the model. This result indicates that improved ENSO prediction may lead to better prediction of LOD, AAM and the related extratropical climate in the decadal prediction models.