A simulation study on pre-landfall erratic track of typhoon Haitang (2005)

Summary

A new typhoon model named as GRAPES_TCM is applied to study the pre-landfall erratic track of Typhoon Haitang (2005), which hit China twice in mid-July by making landfall in Taiwan and Fujian provinces consecutively. The model is based on the Chinese Global/Regional Assimilation and PrEdiction System (GRAPES) and put into real-time operational test since 2004. It predicts almost correctly the pre-landfall loop and sharp turn of Haitang and its asymmetric rainfall distribution.

Haitang’s erratic track is well explained by the potential vorticity (PV) theory on tropical cyclone motion, with the typhoon center moving toward the area of maximum wave-number 1 (WN1) PV tendency most of the time. Among the terms contributing to the WN1 PV tendency, the horizontal advection of PV is dominant with the diabatic heating and residual terms also being not negligible.

A sensitivity experiment is carried out with removal of the Taiwan terrain to determine its importance in the erratic track of Haitang and it is found that the basic erratic feature of Haitang’s track remians unchanged although it tends to have a larger loop and a weaker northward turn, which suggests that Taiwan terrain may not be a key factor here.

The sudden change of Haitang’s moving direction is always accompanied by a newly-generated or re-intensified WN1 PV center in the southern semicircle, which circles around the TC center cyclonically afterwards and weakens in the north or northwestern part. A phase-lock WN1 PV forcing related to diabatic heating is proposed to be the major contributor, the importance of which is magnified as it is in phase with the WN1 horizontal advection of PV. The intrusion of mid-level warm and dry air, as well as the existence of a low-level southwesterly jet, is considered to be the main reasons for such a phase-lock of the diabatic heating forcing on the PV tendency field that finally results in the erratic track of Haitang.