| dc.description.abstract | This research includes two parts using an idealized WRF model. The first part simulates the track evolution of a tropical cyclone moving westward at different departure positions toward an elongated mesoscale mountain (mimicking Taiwan) at different orientation angles. The second part examines the impact of dynamic vortex initialization on typhoon track and intensity forecasts.
The first part involves simulations of cyclones departing from different initial positions toward the terrain at different angles to cover a range of typhoon impacts on Taiwan. The track deflection is primarily controlled by the meridional departure position as well as the ratio of the vortex size and effective terrain length (the aspect ratio, or the nondimensional vortex size). When the cyclone is still somewhat distant away from the terrain, the outer circulation of the cyclone is influenced by the terrain, causing a slightly southward deviation in the track (compared to the experiments without the terrain). When the cyclone approaches the terrain, a “cyclonic deflection” of track is observed where the track deviates northward upstream of the terrain and then southward after passing through the center of the terrain. The wavenumber-1 vortex flow and potential vorticity (PV) budget analyses have helped explain the track deflection that is dominated by horizontal PV advection in comparison to both vertical PV advection and diabatic heating that somewhat modulate the tracks in the vicinity of the mountain. The circulation has already bypassed the topography and recirculated back to the cyclone. The horizontal circulation of the cyclone′s center becomes asymmetric. The vertical profile also reveals significant differences between the eastern and western sides, with stronger vertical motions and wind speeds on the eastern side. The adjustment of the cyclone′s asymmetric structure is related to the horizontal PV advection. Subtracting the wavenumber-1 asymmetric wind fields with and without topography results in a pair of gyres. The gyres move counterclockwise with time and the cyclone center, providing the components of cyclone′s direction of its movement. Consequently, the cyclone moves northward initially, then westward, and southward finally, thus presenting a cyclonic deflection.
The second part will focus on the dynamical vortex initialization (DVI). A modified vortex is inserted into a basic uniform flow as a nature run (representing the best track of a real typhoon, denoted as NR). The vortex after DVI is weakened to produce a control run (representing the initial field for forecasting, denoted as CTL). The CTL vortex then can be strengthened using the DVI method to match the minimum sea-level pressure or maximum wind speed of the NR vortex. The results show that most of the tracks for the experiments are similar. CTL gradually intensifies, with the wind speed catching up with other experiments at 36 simulation hours and the pressure catching up at 60 simulation hours. The experiments using the DVI can well maintain or slightly enhance the vortex intensity at early forecast times. The vortex in all the DVI experiments, however, becomes weakened after 24 hours. The azimuthal-mean transverse circulation across the typhoon center for the DVI experiments reveals significant differences are primarily produced in the inner core region. The impact of DVI is also investigated for the real case of Typhoon Meranti (2016) using the WRF model. The track forecast can be significantly improved when the DVI is conducted with the matching of maximum wind speed. Similar to the idealized experiments, the DVI for this real case can effectively intensify the initial typhoon vortex and thus results in better intensity forecasts up to 60 hours. | en_US |