Abrupt breakdown of the predictability of early season typhoon frequency at the beginning of the twenty-first century

Maintenance of the anomalous anticyclone (cyclone) over the Philippine Sea following strong peak El Nino (La Nina) leads to a negative correlation between the pre-winter (October-December) El Nino-Southern Oscillation (ENSO) index and the tropical cyclone (TC) genesis frequency in the ensuing early TC season (January-June). This negative lead-lag correlation has established the predictability of the western North Pacific (WNP) TC genesis frequency in the early season. However, little is known about whether this relationship remains steady to date, especially in the context of evolving features of ENSO in recent decades. Here we show that the negative relationship experiences an abrupt breakdown around the early 2000s (1999-2006), which primarily occurs in the southeastern quadrant of the WNP. During the breakdown period, the anomalous anti-cyclone shifts westward to the South China Sea and an anomalous cyclonic circulation occupies the southeastern quadrant of the WNP, causing enhanced TC formation. We further reveal that the westward displacement of the WNP anticyclonic anomaly during 1999-2006 is induced by the westward shift of sea surface temperature (SST) and associated precipitation anomalies during the decaying phase of ENSO, which increases cyclonic circulation and ascending motion while reduces vertical wind shear, and thereby enhances TC formation over the southeastern WNP. The results suggest that the frequent occurrence of decaying ENSO events with SST anomaly center over the central Pacific (CP) accounts for the abrupt breakdown of predictability, and the impacts of ENSO on TC formation in the early TC season are sensitive to the location of the maximum SST anomaly. Distinguishing eastern Pacific (EP) and CP types of ENSO and/or strong and weak ENSO events may help to improve the early season TC genesis forecast over the WNP. The results also have important implications for understanding the future projection of TC changes under global warming scenarios.