In this paper, the impact of outer loop and partial cycling with the Weather Research and Forecasting Model’s (WRF) three-dimensional variational data assimilation system (3DVAR) is evaluated by analyzing 78 forecasts for three typhoons during 2008 for which Taiwan’s Central Weather Bureau (CWB) issued typhoon warnings, including Sinlaku, Hagupit, and Jangmi. The use of both the outer loop and the partial cycling approaches in WRF 3DVAR are found to reduce typhoon track forecast errors by more than 30%, averaged over a 72-h period. The improvement due to the outer loop approach, which can be more than 42%, was particularly significant in the early phase of the forecast. The use of the outer loop allows more observations to be assimilated and produces more accurate analyses. The assimilation of additional nonlinear GPS radio occultation (RO) observations over the western North Pacific Ocean, where traditional observational data are lacking, is particularly useful. With the lack of observations over the tropical and subtropical oceans, the error in the first-guess field (which is based on a 6-h forecast of the previous cycle) will continue to grow in a full-cycling limited-area data assimilation system. Even though the use of partial cycling only shows a slight improvement in typhoon track forecast after 12 h, it has the benefit of suppressing the growth of the systematic model error. A typhoon prediction model using the Advanced Research core of the WRF (WRF-ARW) and the WRF 3DVAR system with outer loop and partial cycling substantially improves the typhoon track forecast. This system, known as Typhoon WRF (TWRF), has been in use by CWB since 2010 for operational typhoon predictions.

This paper introduces a relocation scheme for tropical cyclone (TC) initialization in the Advanced Research Weather Research and Forecasting (ARW-WRF) model and demonstrates its application to 70 forecasts of Typhoons Sinlaku (2008), Jangmi (2008, and Linfa (2009) for which Taiwan's Central Weather Bureau (CWB) issued typhoon warnings. An efficient and dynamically consistent TC vortex relocation scheme for the WRF terrainfollowing mass coordinate has been developed to improve the first guess of the TC analysis, and hence improves the tropical cyclone initialization. The vortex relocation scheme separates the first-guess atmospheric flow into a TC circulation and environmental flow, relocates the TC circulation to its observed location, and adds the relocated TC circulation back to the environmental flow to obtain the updated first guess with a correct TC position. Analysis of these typhoon cases indicates that the relocation procedure moves the typhoon circulation to the observed typhoon position without generating discontinuities or sharp gradients in the first guess.Numerical experiments with and without the vortex relocation procedure for Typhoons Sinlaku, Jangmi, and Linfa forecasts show that about 67% of the first-guess fields need a vortex relocation to correct typhoon position errors while eliminates the topographical effect. As the vortex relocation effectively removes the typhoon position errors in the analysis, the simulated typhoon tracks are considerably improved for all forecast times, especially in the early periods as large adjustments appeared without the vortex relocation. Comparison of the horizontal and vertical vortex structures shows that large errors in the first-guess fields due to an incorrect typhoon position are eliminated by the vortex relocation scheme and that the analyzed typhoon circulation is stronger and more symmetric without distortions, and better agrees with observations. The result suggests that the main difficulty of objective analysis methods [e.g., three-dimensional variational data assimilation (3DVAR)], in TC analysis comes from poor first-guess fields with incorrect TC positions rather than not enough model resolution or observations. In addition, by computing the eccentricity and correlation of the axes of the initial typhoon circulation, the distorted typhoon circulation caused by the position error without the vortex relocation scheme is demonstrated to be responsible for larger track errors. Therefore, by eliminating the typhoon position error in the first guess that avoids a distorted initial typhoon circulation, the vortex relocation scheme is able to improve the ARW-WRF typhoon initialization and forecasts particularly when using data assimilation update cycling.

A mesoscale mountain such as the Central Mountain Range (CMR) of Taiwan can considerably affect the motion and the structure of an approaching tropical cyclone in very complicated ways. Many studies have presented and explained the mechanism for the change of moving speed, track deflection, or track discontinuity. In the past, the tracks of Typhoon Mary (1965), Tropical Cyclone Polly (1993), and Typhoon Haitang (2005) have a looping motion prior to making landfall on Taiwan. When the looping occurs, the timing and the intensity of the typhoon to affect Taiwan can be very different from the one moving with smooth track. It is very important to have a correct forecast for disaster mitigation operations. However, the mechanism causing such looping has not been fully understood. In this study, the case of Typhoon Haitang is examined. The radar reflectivity diagrams are used to illustrate looping in its track. By applying piecewise potential vorticity inversion analysis and numerical model simulations, it is shown that the looping motion of Typhoon Haitang near Taiwan resulted from the CMR terrain effect. When Typhoon Haitang approaches Taiwan and its circulation encounters the CMR, the terrain induces a cyclonic vortex in the lee side of the mountain over southeast of Taiwan. This terrain-induced vortex interacts with Typhoon Haitang in a way similar to the binary vortex interaction. The two vortex centers rotate around their system center, which causes looping of Typhoon Haitang.

A blending method to merge the NCEP global analysis with the regional analysis from the WRF variational data assimilation system is implemented using a spatial filter for the purpose of initializing the Typhoon WRF (TWRF) Model, which has been in operation at Taiwan's Central Weather Bureau (CWB) since 2010. The blended analysis is weighted toward the NCEP global analysis for scales greater than the cutoff length of 1200 km, and is weighted toward the WRF regional analysis for length below that. TWRF forecast experiments on 19 typhoons from July to October 2013 over the western North Pacific Ocean show that the largescale analysis from NCEP GFS is superior to that of the regional analysis, which significantly improves the typhoon track forecasts. On the other hand, the regional WRF analysis provides a well-developed typhoon structure and more accurately captures the influence of the Taiwan topography on the typhoon circulation. As a result, the blended analysis takes advantage of the large-scale analysis from the NCEP global analysis and the detailed mesoscale analysis from the regional WRF analysis. In additional to the improved track forecast, the blended analysis also provides more accurate rainfall forecasts for typhoons affecting Taiwan. Because of the improved performance, the blending method has been implemented in the CWB operational TWRF typhoon prediction system.

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.