Typhoon WRF (TWRF) based on the Advanced Research Weather Research and Forecasting Model (ARW WRF) was operational at the Central Weather Bureau (CWB) for tropical cyclone (TC) predictions since 2010 (named TWRF V1). CWB has committed to improve this regional model, aiming to increase the model predictability toward typhoons over East Asia. In 2016, an upgraded version designed to replace TWRF V1 became operational (named TWRF V2). Compared with V1, which has triple-nested meshes with coarser resolution (45/15/5 km), V2 increased the model resolution to 15/3 km. Since V1 and V2 were maintained in parallel from 2016 to 2018, this study utilized the real-time forecasts to investigate the impact of model resolution on TC prediction. Statistical measures pointed out the superiority of the high-resolution model on TC prediction. The forecast performance was also found competitive with that of two leading global models. The case study further pointed out, with the higher resolution, the model not only advanced the prediction on the TC track and inner core structure but also improved the representativeness of the complex terrain. Overall, the high-resolution model can better handle the so-called terrain phase-lock effect and, therefore, improve the TC quantitative precipitation forecast over the complex Taiwanese terrain.

Tropical Cyclone Atsani occurred in late October 2020 and moved westward offshore south of Taiwan. During its offshore passage, the cyclone deflects northward as it closes to the southern end of Taiwan. A global model MPAS at a multi-resolution of 60-15-3-km is applied to explore the track responses of Atsani and identify the topographic effects of the Central Mountain Range (CMR) on the cyclone circulation and the associated track deflection. With a 3-km resolution targeted at the Taiwan area, the cyclone track deflection can be reasonably simulated, with more sensitivity to physics schemes and dynamic vortex initialization and less sensitivity to initial environmental perturbations. When the Taiwan terrain is removed, the cyclone indeed deflects more northward earlier, in particular for simulations with a stronger cyclone that tends to generate stronger east-west wind asymmetry in the absence of the terrain. Idealized simulations with a regional model WRF at 3-km resolution are also utilized to contrast the track deflection of different departing cyclones, similar to the real case. It was found that northward deflection will be induced near south of the CMR-like terrain for both stronger and weaker westbound cyclones departing at different latitudes south of the terrain. We have explained why a further northward track at earlier stages is induced in the absence of the terrain effects in regard to model initial states. In both real and idealized cases, the track deflection of the cyclone moving around the terrain is dominated by the wavenumber-one horizontal potential vorticity (PV) advection that is somewhat offset by both vertical PV advection and differential diabatic heating.

With violent wind and severe rainfall, the tropical cyclone is one of the most disastrous weather systems over ocean and the coastal area. To provide accurate tropical cyclone (TC) track and intensity forecasts is one of the most important tasks of the national weather service of countries affected. Taiwan is one of the areas frequently influenced by tropical cyclones. Improving the tropical cyclone forecast is the highest priority task of Taiwan’s Central Weather Bureau (CWB).Recent improvement of the TC forecast is due to the improvement of the numerical weather prediction. A version of the Advanced Research Weather Research and Forecasting Model (WRF), named TWRF (Typhoon WRF), was developed and implemented in CWB for operational TC forecasting from 2011. During the years, partial update cycling, cyclone bogus scheme, relocation scheme, 3DVAR with outer loop, analysis blending scheme, new trigger Kain–Fritsch cumulus scheme, and so on have been studied and applied in TWRF (Hsiao et al. 2010, 2012, 2015) to improve the model. We also improved the model by changing the TWRF configuration from a triple nested to a double nested grid and increasing the model resolution from 45/15/5 km 45-levels to 15/3 km 52-levels from 2016. Results showed increasing the model resolution improving the track, intensity and rainfall forecast. However, The averaged 24/48/72 hours TC track forecast errors of TWRF are 91/147/223, 84/133/197, 74/127/215, 64/122/202, 70/120/194 and 70/122/180 km in year 2014, 2015, 2016, 2017, 2018 and 2019 respectively.In this study, WRF Four-dimensional data assimilation (FDDA) is adopted to assimilate the temperature, pressure, water vapor content which processed from the FORMOSAT-7 constellation, high-temporal frequency atmospheric motion vector (AMV) retrieved from Himawari-8 satellite images and radar data to generate a model balanced TC structure and thermodynamic state at the initial time. The specific goal is to improve the track, structure and intensity prediction of TCs and their associated rainfall distribution in Taiwan. The detail will be presented in the conference.Keywords: tropical cyclone, Himawari-8 AMV, Four-dimensional data assimilation, FORMOSAT-7, radar data.Corresponding author address:Der-Song Chen,  song@cwb.gov.twCentral Weather Bureau, 64 Gongyuan Rd., Taipei, Taiwan, R.O.C., 10048.