Advanced Search
Article Contents

Remote Impact of Blocking Highs on the Sudden Track Reversal of Tropical Cyclones


doi: 10.1007/s00376-015-4284-7

  • Previous work showed that some tropical cyclones (TCs) in the western Pacific Ocean undergo sudden track reversal, and the onset, maintenance and decay of blocking highs (BHs) coexisted with 19 of the studied TCs with sudden track reversal. In these cases, the phase relations between the BH, the continental high (CH), the subtropical high (SH) and the suddenly reversed TCs could be classified into types A, B, C and D. Types C and D were the focal point of this follow-up study, in which Typhoon Pabuk (2007) and Lupit (2009) were employed to conduct numerical simulations. The results showed that the reversed tracks of Pabuk (2007) and Lupit (2009) could have been affected by the BH, particularly in terms of the turning location and the trend of movement after turning. Specifically, the two main features for Pabuk (2007) in the BH perturbations were the deflection of its turning point and a distinct anticlockwise rotation. Lupit (2009) deviated to the southwest and finally made landfall in the Philippines, or experienced further eastward movement, in the perturbed BH. The impact mechanisms can be attributed to the change in the vorticity field transported from the BH, leading to an intensity variation of midlatitude systems. BHs may have a positive feedback effect on the strength of the westerly trough (TR), as indicated by a weakened and strengthened TR corresponding to negative and positive BH perturbations, respectively.
  • 加载中
  • Armi L., D. Hebert, N. Oakey, J. Price, P. L. Richardson, T. Rossby, and B. Ruddick, 1988: The history and decay of a Mediterranean salt lens. Nature, 333, 649- 651.
    Carr III, L. E., R. L. Elsberry, 1995: Monsoonal interactions leading to sudden tropical cyclone track changes. Mon. Wea. Rev., 123, 265- 290.
    Carr III, L. E., R. L. Elsberry, 2000a: Dynamical tropical cyclone track forecast errors. Part I: Tropical region error sources. Wea. Forecasting, 15, 641- 661.
    Carr III, L. E., R. L. Elsberry, 2000b: Dynamical tropical cyclone track forecast errors. Part II: Midlatitude circulation influences. Wea. Forecasting, 15, 662- 678.
    Carr III, L. E., A. B. Mark, R. L. Elsberry, 1997: Observational evidence for alternate modes of track-altering binary tropical cyclone scenarios. Mon. Wea. Rev., 125, 2094- 2111.
    Davis C. A., K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev., 119, 1929- 1953.
    Doyle J., C. Reynolds, C. Amerault, J. Goerss, J. Mclay, and R. M. Hodur, 2010: Tropical cyclone track and intensity predictability. 2010 DoD High Performance Computing Modernization Program Users Group Conference (HPCMP-UGC), Schaumburg,IL, IEEE, 281- 287.
    Dudhia J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 3077- 3107.
    Elsberry R. L., 1990: International experiments to study tropical cyclones in the western North Pacific. Bull. Amer. Meteor. Soc., 71, 1305- 1316.
    Evans J. L., G. J. Holland , and R. L. Elsberry, 1991: Interactions between a barotropic vortex and an idealized subtropical ridge. Part I: Vortex motion. J. Atmos. Sci., 48, 301- 314.
    George J. E., W. M. Gray, 1997: Tropical cyclone recurvature and nonrecurvature as related to surrounding wind-height fields. J. Appl. Meteor., 16, 34- 42.
    Guard C. P., 1983: Monsoonal influences on unusual development and movement of tropical cyclones. Proc. Tropical Cyclone Conference, Baguio, Republic of the Philippines, Environmental Group Pacific Command, 157- 180.
    Gu, X. Z. and P. G. Pan, 2010: Diagnostic analysis of typhoon heavy rain of Pabuk and Sepat during August 2007. Torrential Rain and Disasters, 29( 2), 129- 134. (in Chinese)
    Holland, G. J., Y. Q. Wang, 1995: Baroclinic dynamics of simulated tropical cyclone recurvature. J. Atmos. Sci., 52, 410- 426.
    Hong S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 2318- 2341.
    Hoskins B. J., M. E. Mclntyre, and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111, 877- 946.
    JMA, 2009: Annual report on the activities of the RSMC Tokyo-Typhoon Center 2009. Japan Meteorological Agency.
    Kain J. S., 2004: The Kain-Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170- 181.
    Kieu C. Q., N. M. Truong, H. T. Mai, and T. Ngo-Duc, 2012: Sensitivity of the track and intensity forecasts of typhoon Megi (2010) to satellite-derived atmospheric motion vectors with the ensemble Kalman filter. J. Atmos. Oceanic Technol., 29, 1794- 1810.
    Kurihara Y., M. A. Bender, R. E. Tuleya, and R. J. Ross, 1993: An initialization scheme of hurricane models by vortex specification. Mon. Wea. Rev., 121, 2030- 2045.
    Kurihara Y., M. A. Bender, R. E. Tuleya, and R. J. Ross, 1995: Improvements in the GFDL hurricane prediction system. Mon. Wea. Rev., 123, 2791- 2801.
    Land er, M. A., 1995: Specific tropical track types and unusual tropical cyclone motions associated with a reverse-oriented monsoon trough in the western North Pacific. Wea.Forecasting, 11, 170- 186.
    Li W. J., 1999: General atmospheric circulation anomaly in 1998 and their impact on climate anomaly in China. Meteorological Monthly, 25, 20- 25. (in Chinese).
    Luo X., J. F. Fei, X. G. Huang, X. P. Cheng, J. L. Ding, and Y. Q. He, 2015: Statistical characteristics and mechanistic analysis of suddenly reversed tropical cyclones over the western North Pacific Ocean. Adv. Atmos. Sci.,32(4), 565-576, doi: 10.1007/s00376-014-4064-9.
    Ma X. Q., Y. H. Ding, H. M. Xu, and J. H. He, 2008: The relation between strong cold waves and low-frequency waves during the winter of 2004/2005. Chinese J. Atmos. Sci., 32( 2), 380- 394. (in Chinese)
    Mlawer E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmospheres: RRTM, a validated correlated-k model for the long wave. J. Geophys. Res., 102, 16 663- 16 682.
    Namais J., 1947: Characteristics of the general circulation over the Northern Hemisphere during the abnormal winter 1946-47. Mon. Wea. Rev., 75, 145- 152.
    Pike A. C., 1985: Geopotential heights and thickness as predictors of Atlantic tropical cyclone motion and intensity. Mon. Wea. Rev., 113, 931- 940.
    Riehl H., R. J. Shafer, 1944: The recurvature of tropical storms. J. Meteor., 1, 42- 54.
    Shi W. L., J. F. Fei, X. G. Huang, X. P. Cheng, J. L. Ding, and Y. Q. He, 2014: A numerical study on the combined effect of midlatitude and low-latitude systems on the abrupt track deflection of typhoon Megi (2010). Mon. Wea. Rev., 142, 2483- 2501.
    Tao S. Y., Y. J. Zhao, and X. M. Chen, 1958: The relationship between May-y\"u in far east and the behavior of circulation over Asia. Acta Meterologica Sinica, 29( 2), 119- 134. (in Chinese)
    Wang Y. Q., G. J. Holland, 1996: Tropical cyclone motion and evolution in vertical shear. J. Atmos. Sci., 53, 3313- 3332.
    Willoughby H. E., F. D. Marks Jr., and R. J. Feinberg, 1984: Stationary and moving convective bands in hurricanes. J. Atmos. Sci., 41, 3189- 3211.
    Wu L. G., B. Wang, 2000: A potential vorticity tendency diagnostic approach for tropical cyclone motion. Mon. Wea. Rev., 128, 1899- 1911.
    Wu L. G., Z. P. Ni, J. J. Duan, and H. J. Zong, 2013: Sudden tropical cyclone track changes over the western North Pacific: A composite study. Mon. Wea. Rev., 141, 2597- 2610.
    Wu C.-C., T.-S. Huang, W. P. Huang, and K.-H. Chou, 2003: A new look at the binary interaction: Potential vorticity diagnosis of the unusual southward movement of tropical storm Bopha (2000) and its interaction with supertyphoon Saomai (2000). Mon. Wea. Rev., 131, 1289- 1300.
    Wu C.-C., T.-S. Huang, and K.-H. Chou, 2004: Potential vorticity diagnosis of the key factors affecting the motion of typhoon Sinlaku (2002). Mon. Wea. Rev., 132, 2084- 2093.
    Xu X. D., Y. Y. Xie, and P. Y. Li, 1994: Numerical experiment on effect of the blocking high structure on unusual track of typhoon in coastal waters. Quart. J. Appl. Mereor., 5( 1), 104- 108. (in Chinese)
    Yeh T.C., 1949: On energy dispersion in the atmosphere. J. Meteor., 6, 1- 16.
    Zhang W., Y. Leung, and J. C. L. Chan, 2013: The analysis of tropical cyclone tracks in Western North Pacific through data mining, Part I: Tropical cyclone recurvature. Journal of Applied Meteorology and Climatology, 52, 1394- 1416.
  • [1] Lu Riyu, Huang Ronghui, 1996: Numerical Simulation of the Effect of the SST Anomalies in the Tropical Western Pacific on the Blocking Highs over the Northeastern Asia, ADVANCES IN ATMOSPHERIC SCIENCES, 13, 411-424.  doi: 10.1007/BF03342034
    [2] Chang-Hoi HO, Joo-Hong KIM, Hyeong-Seog KIM, Woosuk CHOI, Min-Hee LEE, Hee-Dong YOO, Tae-Ryong KIM, Sangwook PARK, 2013: Technical Note on a Track-pattern-based Model for Predicting Seasonal Tropical Cyclone Activity over the Western North Pacific, ADVANCES IN ATMOSPHERIC SCIENCES, 30, 1260-1274.  doi: 10.1007/s00376-013-2237-6
    [3] GAO Feng*, Peter P. CHILDS, Xiang-Yu HUANG, Neil A. JACOBS, and Jinzhong MIN, 2014: A Relocation-based Initialization Scheme to Improve Track-forecasting of Tropical Cyclones, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 27-36.  doi: 10.1007/s00376-013-2254-5
    [4] ZHAO Haikun, WU Liguang*, and WANG Ruifang, 2014: Decadal Variations of Intense Tropical Cyclones over the Western North Pacific during 19482010, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 57-65.  doi: 10.1007/s00376-013-3011-5
    [5] Tian Yongxiang, Luo Zhexian, 1994: Vertical Structure of Beta Gyres and Its Effect on Tropical Cyclone Motion, ADVANCES IN ATMOSPHERIC SCIENCES, 11, 43-50.  doi: 10.1007/BF02656992
    [6] QIN Xiaohao, MU Mu, 2014: Can Adaptive Observations Improve Tropical Cyclone Intensity Forecasts?, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 252-262.  doi: 10.1007/s00376-013-3008-0
    [7] HUANG Hong, JIANG Yongqiang, CHEN Zhongyi, LUO Jian, WANG Xuezhong, 2014: Effect of Tropical Cyclone Intensity and Instability on the Evolution of Spiral Bands, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 1090-1100.  doi: 10.1007/s00376-014-3108-5
    [8] Pan SONG, Jiang ZHU, Zhong ZHONG, Linlin QI, Xiaodan WANG, 2016: Impact of Atmospheric and Oceanic Conditions on the Frequency and Genesis Location of Tropical Cyclones over the Western North Pacific in 2004 and 2010, ADVANCES IN ATMOSPHERIC SCIENCES, 33, 599-613.  doi: 10.1007/s00376-015-5046-2
    [9] Yihui DING, Yunyun LIU, Zeng-Zhen HU, 2021: The Record-breaking Mei-yu in 2020 and Associated Atmospheric Circulation and Tropical SST Anomalies, ADVANCES IN ATMOSPHERIC SCIENCES, 38, 1980-1993.  doi: 10.1007/s00376-021-0361-2
    [10] MA Zhanhong, FEI Jianfang, HUANG Xiaogang, CHENG Xiaoping, 2014: Impacts of the Lowest Model Level Height on Tropical Cyclone Intensity and Structure, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 421-434.  doi: 10.1007/s00376-013-3044-9
    [11] Yan ZHENG, Liguang WU, Haikun ZHAO, Xingyang ZHOU, Qingyuan LIU, 2020: Simulation of Extreme Updrafts in the Tropical Cyclone Eyewall, ADVANCES IN ATMOSPHERIC SCIENCES, 37, 781-792.  doi: 10.1007/s00376-020-9197-4
    [12] Kelvin T. F. CHAN, Johnny C. L. CHAN, 2016: Sensitivity of the Simulation of Tropical Cyclone Size to Microphysics Schemes, ADVANCES IN ATMOSPHERIC SCIENCES, 33, 1024-1035.  doi: 10.1007/s00376-016-5183-2
    [13] ZENG Zhihua, DUAN Yihong, LIANG Xudong, MA Leiming, Johnny Chung-leung CHAN, 2005: The Effect of Three-Dimensional Variational Data Assimilation of QuikSCAT Data on the Numerical Simulation of Typhoon Track and Intensity, ADVANCES IN ATMOSPHERIC SCIENCES, 22, 534-544.  doi: 10.1007/BF02918486
    [14] Zhenhua HUO, Wansuo DUAN, Feifan ZHOU, 2019: Ensemble Forecasts of Tropical Cyclone Track with Orthogonal Conditional Nonlinear Optimal Perturbations, ADVANCES IN ATMOSPHERIC SCIENCES, 36, 231-247.  doi: 10.1007/s00376-018-8001-1
    [15] Tong XIE, Liguang WU, Yecheng FENG, Jinghua YU, 2024: Alignment of Track Oscillations during Tropical Cyclone Rapid Intensification, ADVANCES IN ATMOSPHERIC SCIENCES, 41, 655-670.  doi: 10.1007/s00376-023-3073-y
    [16] Chen Yuejuan, Zheng Bin, Zhang Hong, 2002: The Features of Ozone Quasi-Biennial Oscillation in Tropical Stratosphere and Its Numerical Simulation, ADVANCES IN ATMOSPHERIC SCIENCES, 19, 777-793.  doi: 10.1007/s00376-002-0044-6
    [17] YANG Jing, BAO Qing, JI Duoying, GONG Daoyi, MAO Rui, ZHANG Ziyin, Seong-Joong KIM, 2014: Simulation and Causes of Eastern Antarctica Surface Cooling Related to Ozone Depletion during Austral Summer in FGOALS-s2, ADVANCES IN ATMOSPHERIC SCIENCES, 31, 1147-1156.  doi: 10.1007/s00376-014-3144-1
    [18] LUO Xia, FEI Jianfang, HUANG Xiaogang, CHENG Xiaoping, DING Juli, HE Yiqiang, 2015: Statistical Characteristics and Mechanistic Analysis of Suddenly Reversed Tropical Cyclones over the Western North Pacific Ocean, ADVANCES IN ATMOSPHERIC SCIENCES, 32, 565-576.  doi: 10.1007/s00376-014-4064-9
    [19] LI Wei-Wei, WANG Chunzai, WANG Dongxiao, YANG Lei, DENG Yi, 2012: Modulation of Low-Latitude West Wind on Abnormal Track and Intensity of Tropical Cyclone Nargis (2008) in the Bay of Bengal, ADVANCES IN ATMOSPHERIC SCIENCES, 29, 407-421.  doi: 10.1007/s00376-011-0229-y
    [20] YU Xing, DAI Jin, LEI Hengchi, FAN Peng, 2005: Comparison Between Computer Simulation of Transport and Diffusion of Cloud Seeding Material Within Stratiform Cloud and the NOAA-14 Satellite Cloud Track, ADVANCES IN ATMOSPHERIC SCIENCES, 22, 133-141.  doi: 10.1007/BF02930877

Get Citation+

Export:  

Share Article

Manuscript History

Manuscript received: 28 December 2014
Manuscript revised: 08 April 2015
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Remote Impact of Blocking Highs on the Sudden Track Reversal of Tropical Cyclones

    Corresponding author: HUANG Xiaogang, huang.x.g@163.com
  • 1. College of Meteorology and Oceanography, PLA University of Science and Technology, Nanjing 211101

Abstract: Previous work showed that some tropical cyclones (TCs) in the western Pacific Ocean undergo sudden track reversal, and the onset, maintenance and decay of blocking highs (BHs) coexisted with 19 of the studied TCs with sudden track reversal. In these cases, the phase relations between the BH, the continental high (CH), the subtropical high (SH) and the suddenly reversed TCs could be classified into types A, B, C and D. Types C and D were the focal point of this follow-up study, in which Typhoon Pabuk (2007) and Lupit (2009) were employed to conduct numerical simulations. The results showed that the reversed tracks of Pabuk (2007) and Lupit (2009) could have been affected by the BH, particularly in terms of the turning location and the trend of movement after turning. Specifically, the two main features for Pabuk (2007) in the BH perturbations were the deflection of its turning point and a distinct anticlockwise rotation. Lupit (2009) deviated to the southwest and finally made landfall in the Philippines, or experienced further eastward movement, in the perturbed BH. The impact mechanisms can be attributed to the change in the vorticity field transported from the BH, leading to an intensity variation of midlatitude systems. BHs may have a positive feedback effect on the strength of the westerly trough (TR), as indicated by a weakened and strengthened TR corresponding to negative and positive BH perturbations, respectively.

1. Introduction
  • Over the last few years, the prediction of tropical cyclone (TC) tracks has become more accurate and better developed. However, dynamical models still possess large errors, such as 300 n mile at 72 h, 400 n mile at 96 h and 500 n mile at 120 h (Carr and Elsberry, 2000a, 2000b), which are usually associated with sudden track changes (Wu et al., 2013). Sudden track changes of TCs, including their movement direction and transition variations, occur nearly every year and sometimes can result in huge disasters, particularly in offshore areas. The prominent factor determining TC motion is the environmental steering flow supplied by surrounding synoptic systems, such as subtropical ridges, midlatitude troughs, monsoon systems, and other vortexes (Holland and Wang, 1995). Thus, improper observations and misrepresented features of the above surrounding systems might contribute greatly to false representations of the physical mechanisms controlling the TC's motion (Carr and Elsberry, 2000a, 2000b; Kieu et al., 2012).

    For example, the sudden turning of a TC is often due to the eastward retreat or the double ridge lines' dissipation of the subtropical high (Riehl and Shafer, 1944; Elsberry, 1990; Evans et al., 1991; George and Gray, 1997; Zhang et al., 2013). A sudden northward turning is always accompanied by a neutral point (or "break") in the midlatitude high-pressure zone or the asymmetry saddle field. Additionally, monsoon surges and reverse-oriented monsoon troughs have been proven to be dominant factors in some TCs' strong poleward acceleration (Guard, 1983; Lander, 1995), sometimes assisted by an anticyclone located to the southeast of the TC due to Rossby dispersion (Carr and Elsberry, 1995; Wu et al., 2013). Sudden track changes or deflections can also be observed when two TCs or multiple TCs coexist in a certain region, referred to as binary cyclone interaction. Although conceptual models of binary cyclone interaction (direct, semi-direct, and indirect) have been presented in detail (Carr et al., 1997), the development and dissipation processes vary from case to case and are always responsible for some larger forecast errors.

    When the steering flow is relatively weak, neglecting the nonlinear interaction between mesoscale and small-scale eddies accounts for a large proportion of inaccurate forecasts. The perturbations of nearby mesoscale convective systems and oceanic eddies also play a role in a TC's meandering motion (Willoughby et al., 1984; Armi et al., 1988). The internal nonlinear interaction also includes the impact of baroclinic processes, like the vertical structure of the TC and the vertical shear in the steering flow (Wang and Holland, 1996). For instance, most TCs tend to drift to the left of the vertical shear, which is accompanied by positive potential vorticity (PV) advection (Wang and Holland, 1996; Wu and Wang, 2000). The investigations of Wu et al. (2003, 2004) have demonstrated great advantages in the use of PV diagnostics for identifying the influence of each PV perturbation on TC motion.

    As discussed above, the sudden track change of a TC has emerged as an important issue. But how do we define sudden track change? A unified answer has yet to be reached (Wu et al., 2013), largely because it depends heavily on the purpose of the particular piece of research. However, a common characteristic is that marked changes in the TC's track take place during a very short period. In the companion paper (Luo et al., 2015), some special cases (referred to as suddenly reversed TCs, or Suddenly Reversed Tropical Cyclones) in the western Pacific Ocean were documented as having experienced sudden reversed turning, from northeast to southwest (sometimes approaching 180°), within a short timeframe. It is these cases that form the key focus of the present, follow-up paper.

    Atmospheric circulation can be regarded as a systematic entity. This means that the adjustment of systems at high latitudes, such as blocking highs (BHs), as a relatively stable flow pattern, is always accompanied by midlatitude circulation evolution. The initial interest in BH activities originated mainly from their evident seasonal and geographic distribution characteristics (Namais, 1947; Yeh, 1949). Since these formative studies, much research has focused on the connection between BHs and the East Asian climate, especially with respect to cold-air outbreaks and continuous droughts/floods in China (Tao et al., 1958; Ma et al., 2008). These abnormal weather and climate phenomena mainly result from a large circumfluent adjustment due to the onset and decay of a BH that, as its name suggests, blocks the intensive meridional momentum and heat exchange. In summer, the subtropical high (SH) can combine with the BH via the East Asia-Pacific teleconnection, and the location of the SH will deviate to the south if the BH persists for a long time, or vice versa (Li, 1999). (Xu et al., 1994) conducted numerical experiments to investigate the effect of varied BH intensity on TC Rita's (1972) trajectory and found Rita's track in strengthened BH experiments showed a parabolic characteristic, just like its actual track, but differed in terms of the track's curvature. Interestingly, BHs can also form in the environmental field of some SRTCs and can persist after their turning. Is this coincidental or mechanistically determined? Is there correlation between BHs and SRTC tracks or do BHs help TCs form their unique trajectories? Statistical analyses and numerical simulations were used in the present study to examine the underlying connections between BHs and SRTCs. Among the large numbers of SRTCs, Typhoon Pabuk (2007) and Lupit (2009) were selected as typical cases to implement the numerical simulations.

    The data and methods used in this study are described in section 2, including a brief description of the model's configuration. The definition of an SRTC and the related statistical results are provided in section 3. The synoptic analyses of Typhoon Lupit (2009) and Pabuk (2007) are presented in section 4. The effects of the BH in each of the two cases are described in section 5. The results and mechanistic analysis of the comparison experiments are presented in section 6. And finally, a discussion and summary follow in section 7.

2. Data and methods
  • Best-track data of western North Pacific TCs during 1949-2011 were derived from the Japan Meteorological Agency. These data cover the TC's center position (latitude and longitude) and intensity (central pressure and maximum sustained wind speed) at 6 h intervals. The black body temperature (TBB) data, with a horizontal resolution of 0.05°× 0.05°, were obtained from Japan's Geostationary Meteorological Satellite at 1 h intervals. The National Centers for Environmental Prediction (NCEP) Final (FNL) Operational Global Analysis data, on a 1.0°× 1.0° grid at every 6 h, were used for the synoptic environment features.

  • 2.2.1 PV diagnostics

    In isentropic coordinates, Ertel's PV can be defined as \[ q=-g\dfrac?θ? P(ζ1.1. θ+f) , \] where ζθ is the vertical component of the relative vorticity evaluated on an isentropic surface, θ is the potential temperature, g is the gravitational acceleration, P indicates the air pressure, f is the Coriolis parameter, and ?θ/? P is the hydrostatic relation. Isentropic PV is obtained through interpolating the isobaric data to the isentropic surface. Here, a total of 15 layers were used, and the lowest layer was set to 300 K at 5 K intervals in the θ coordinate.

    Besides isentropic PV, another distinct characteristic of PV is its conserved quality, which can be used to measure synoptic features by tracing abnormal PV areas via the method of PV inversion (Hoskins et al., 1985; Davis and Emanuel, 1991). Currently, the most widely applied PV inversion method is derived from [Davis and Emanuel, 1991], including piecewise PV inversion. In view of the large number of related studies, the specific formula is not introduced in detail here; however, interested readers can find further information in [Wu et al., 2004]. In this study, the synoptic analysis of Lupit (2009) mainly relied upon PV inversion to assess the impact of the flow associated with specific PV perturbations.

    2.2.2 System perturbation and separation

    In order to examine the specific impacts brought by the surrounding synoptic systems and the BH on the TC's reversed track, a filtering method was employed in the sensitivity experiments. Initially, this method was utilized in the TC's initialization scheme [Kurihara et al., 1993], aimed at resolving the false spin-up problem. Subsequently, the initialization scheme was improved due to the lack of accuracy in weak TCs' simulations [Kurihara et al., 1995]. Specifically, this method contains three key steps: (1) The initial field h is split into the basic field h B and the disturbance field h D, h=h B+h D. By using a three-point smoothing operator [Kurihara et al., 1993], the waves are smoothed in longitude-latitude orientation:

    \begin{eqnarray*} \overline{h}_{\lambda,\varphi}&=&h_{\lambda,\varphi}+K(h_{\lambda-1,\varphi}+h_{\lambda+1,\varphi}-2h_{\lambda,\varphi}) ,\\ \overline{h}_B&=&\overline{h}_{\lambda,\varphi}+K(\overline{h}_{\lambda,\varphi-1}+\overline{h}_{\lambda,\varphi+1}-2\overline{h}_{\lambda,\varphi}) , \end{eqnarray*}

    where Λ and φ representing latitude and longitude, respectively. The parameter K is defined as $K=[1-cos(2π/m)/2, m=2,3,4,5,6,7,2,8,9,2. For the background with 1° resolution, the components with less than 9° wavelength will be removed and the amplitudes of those with 15°, 20° and 30° wavelength will be reduced by 82%, 60% and 32% under ideal conditions. (2) Determine the filtered area, based on the longitude-latitude distribution and the center of the system. (3) Reconstruct the analysis field. The new analysis field h_ new is defined as the original field h plus the disturbed field h_ D multiplied by the coefficient Si, which determines the magnitude of systems' intensity changes h_ new=h+S_ ih_ D. The detail of this method can be found in [Shi et al., 2014].

    2.2.3 Model configuration

    The numerical model employed in this study was the Advanced Research Weather Research and Forecasting (WRF ARW) model, version 3.4. The initial and lateral boundary conditions for ARW were acquired from the NCEP FNL analysis and forecast fields. ARW contains a single, fixed grid with horizontal resolution of 30 km and 35 vertical levels (1000-50 hPa). The domain is centered at 35°N, 135°E with 225 (lat) × 254 (lon) grid points. Apart from the microphysics scheme (Ferrier, used for Lupit (2009), and Kessler, used for Pabuk(2007)), all schemes were the same because simulated tracks are very sensitive to the combination of different parameterization schemes, and the one that can obtain the best tracks is expected in control experiments. The Kain-Fritsch scheme [Kain, 2004] was applied as the cumulus parameterization. The Yonsei University scheme [Hong et al., 2006] was chosen for the planetary boundary layer parameterization. The model also includes the Rapid Radiative Transfer Model longwave radiation scheme [Mlawer et al., 1997] and the Dudhia shortwave radiation scheme [Dudhia, 1989].

3. Statistical results
  • The SRTC standards are defined in the companion paper [Luo et al., 2015] mainly dependent on the TC turning angle variations. SRTCs' turning angle could meet any of the following characters: (1) the track-direction change exceeds 90° during the 6 h period; (2) the turning angle is ≤90° but >45° (90°) during the 6 (12) h period; (3) the turning angle is ≤90° but >45° (90°) during the 6 and 12 h (24 h) period. Further findings showed that the BH always develops before the SRTC's turning and persist for a certain period after their turning, referred to as a TC coexisting with a BH. Based on these TCs' environmental field characteristics, the phase relations between the BH, the continental high (CH), the SH and the SRTC could be divided into four types (A, B, C and D). The phase properties of these four types are as follows (Fig. 1):

    Figure 1.  Sketches of phase relations between the BH, CH, SH and SRTCs: the solid and dotted lines respectively represent the initial and ultimate flow pattern; the flag marks the TC’s possible position (the rectangle represents two initial positions).

    Figure 2.  (a) The best track of Pabuk (2007) from 1200 UTC 7 August to 1200 UTC 12 August at 6 h intervals; (b) the best track of Lupit (2009) from 0000 UTC 19 October to 0000 UTC 26 October at 6 h intervals.

    In type A, the SRTC is initially situated to the southwest of the SH, the midlatitude westerlies are relatively narrow, and the SH remains stable during the whole evolution process. In type B, the SRTC is situated to the east of the CH. At the beginning, the continental ridge develops into a long-wave ridge of great amplitude because of the BH's eastward movement. Then, the axis of the CH presents a clockwise rotation toward the northeast with the decay of the BH. In type C, the SRTC is situated to the west or south of the SH. Note: the SH is zonally distributed in the initial phase. The overlapping of the BH and the SH is accompanied by shrinkage of the SH due to the low pressure's eastward movement. Gradually, the SH ridge line turns from northeast to southwest in meridional distribution. In type D, the SRTC is mainly situated to the southwest of the SH. In comparison with types A and C, the westerlies are broader, in excess of 10° of latitude. It should be stressed that the SH remains steady without the flow pattern transition.

    The serial numbers of the SRTCs, the moments of onset of the BHs, the moments of turning of the SRTCs, and the phase relations of the BHs and SRTCs, are listed in Table 1. Because of the high occurrence frequency of types C and D, Typhoon Pabuk (2007) and Lupit (2009) were adopted as typical cases.

4. Diagnostic analysis
  • Typhoon Pabuk (2007) formed at around (18.4 °N, 137.5°E) and developed quickly into a tropical storm at 0600 UTC 5 August (Fig. 2a). The ambiguous low-level circulation merged with another depression (Wupit) in the southeast until Wupit dissipated at 1800 UTC 8 August. Pabuk (2007) meandered near the coastline of Guangdong Province (China) prior to making a sudden northeastward turn at 1200 UTC 9 August, as a result of large prediction errors approaching Shanwei (22.4°N, 115.4°E). Pabuk (2007) continued to move slowly to the northeast with its maximum wind speed and dissipated after several hours.

    Typhoon Lupit (2009) (Fig. 2b) was upgraded to a typhoon at 1800 UTC 16 October and reached its peak intensity at 1800 UTC 18 October. After 1800 UTC 20 October, Lupit (2009) experienced a dramatic directional shift from the southwest to the northeast with a low translation speed, which brought heavy rain to the northeast of Taiwan. By 1800 UTC 25 October, Lupit (2009) lay over the north coast of Taiwan and weakened to a tropical storm. A series of numerical weather prediction models (Coupled Ocean-Atmosphere Mesoscale Prediction System, Navy Operational Global Atmospheric Prediction System, National Oceanic and Atmospheric Administration Global Forecast System and European Centre fir Medium-Range Weather Forecasts) forecasted that Lupit (2009) would keep moving westward and cause tremendous damage in Luzon [Doyle et al., 2010], implying that the rapid reversal of TCs' tracks is often too difficult for numerical models to predict. In fact, Lupit (2009) gave rise to the largest track errors in 2009, according to the verification data of operational forecasts from the Annual Report on the Activities of the Regional Specialized Meteorological Centre Tokyo-Typhoon Center 2009 [JMA, 2009].

    Figure 3.  The 500 hPa geopotential height (the areas with values larger than 5880 gpm are shaded) at (a) 1200 UTC 7, (b) 1200 UTC 8, (c) 1200 UTC 9, and (d) 12 UTC 10. The 850 hPa averaged water vapor flux (units: g cm-1 hPa-1 s-1) and steering flow on (e) 8 August and (f) 9 August. The TBB data and 500 hPa wind field (one full barb represents 10 m s-1) at (g) 0000 UTC 8 and (h) 0000 UTC 9.

    Figure 4.  The 500 hPa geopotential height (left) and the 350 K isentropic potential with the corresponding wind field (one full barb represents 10 m s-1; 1 PVU = 10-6 m2 s-1 K kg-1) at (a) 0000 UTC 20, (b) 0000 UTC 21, and (c) 0000 UTC 22. (d) Time-series of Lupit's (2009) motion [V (obs)], the actual steering flow from NCEP data [V (NCEP)], the steering flow associated with the CH [V (ch)], SH [V (sh)], TR [V (tr)] and HZ [V (hz)], and the sum of the flow components of the CH, SH and TR.

    Figure 5.  The simulated tracks for Pabuk (2007) in the control (solid circles) at 6 h intervals and the best track (typhoon symbol): (a) from 1200 UTC 7 to 1200 UTC 11 August; (b) from 1800 UTC 8 to 0600 UTC 10 August. The simulated tracks for Lupit (2009) in the control at 6 h intervals and the best track from 0000 UTC 20 to 25 October (c) and the corresponding movement speed variation (d).

    Figure 6.  Pabuk (2007) simulated tracks (a, b) in the perturbed BH from 1200 UTC 7 to 11 and (c, d) from 0600 UTC 8 to 11 August; Lupit (2009) simulated tracks (e, f) in the perturbed BH from 0000 UTC 20 to 25 and (g, h) from 1800 UTC 20 to 24 October.

    Figure 7.  The perturbed simulated tracks for Pabuk (2007) TR (a, b), SH (c) and combined experiments (both TR and SH are perturbed) at 6 h intervals from 1200 UTC 7 to 11 August.

  • As shown in Fig. 3a, at 1200 UTC 7 August, the ridge line of the SH was located at around 32°N, much farther north compared to normal (28°N) and this would have facilitated the interaction between Pabuk (2007) and the ITCZ because both Pabuk (2007) and Wupit were constrained within the ITCZ during this stage for a limited distance (less than 1000 km) [Gu and Pan, 2010]. Note that the BH developed in the north of the SH at 1200 UTC 8 August (Fig. 3b). At almost 1200 UTC 9 August (Fig. 3c), the SH started to retreat due to the eastward movement of the upstream westerly trough (TR). The northeastward movement of Pabuk (2007) was accelerated, while the BH and SH were superimposed in the same phase (Fig. 3d). The water vapor transmission (Figs. 3e and f) was mainly supported by the southwesterly and its northward deviation was accompanied by Pabuk (2007) strengthening (Fig. 3h). Pabuk (2007) was expected to weaken at 0000 UTC 9 August due to the less warm moist flow and downgraded spin-interaction accompanying Wupit's dissipation (Fig. 3g). Under the situation of the BH, the sudden track change of Pabuk (2007) could be largely ascribed to the enhanced southwesterly and accelerated eastward shift of the SH.

  • As illustrated in Fig. 4, the synoptic systems around Lupit (2009) were the TR, SH and CH. Then, a warm high ridge (45°-65°N, 140°-170°E) developed into a closed high pressure center (the BH). Meanwhile, an easterly wave developed to the south of the SH and strengthened the pressure gradient around 160°E. As the TR with high PV migrated to the south (Fig. 4b), the CH was affected and became weaker. Interestingly, a northeast to southwest oriented high pressure zone (HZ) formed after the SH (Fig. 4c), with continuous movement to the south and finally merged into the CH. A similar system was also found in Megi's (2010) environment field and has been proven to have provided the southwesterly steering flow when Megi underwent the northward turn [Shi et al., 2014]. The question naturally arises: did the HZ play a similar role in Lupit's (2009) movement? The piecewise PV inversion was taken to further analyze Lupit's (2009) synoptic environment.

    The whole PV perturbation field was divided into four parts, each respectively representing PV perturbations associated the SH (10°-30°N, 145°-170°E), CH (5°-20°N, 100°-120°E), TR (25°-35°N, 100°-130°E) and HZ (5°-20°N, 130°-140°E). The balanced wind field corresponding to each PV perturbation was ultimately attained by the individual PV inversion from 0000 UTC 19 to 0000 UTC 24 October (Fig. 4e). V (obs), V (sh), V (ch), V (tr) and V (hz) respectively represent the actual translation speed of Lupit (2009) and the balanced flow of the SH, CH, TR and HZ. In order to examine the role of HZ, V (com) was defined as the combined steering flow of the SH, CH and TR without the HZ. The environmental steering flow [V (NCEP)] was averaged within a 5°× 5° grid with the TC as the level between 850 hPa and 400 hPa [Pike, 1985] and was basically consistent with V (obs). The steady westerly was supplied by the TR, which made the greatest contribution to the track change, and the persistent southeasterly and northerly were respectively provided by the SH and CH. Note that the HZ played a supporting role in facilitating the turning occurrence with an enhancing southwesterly to southerly, proving the role of this kind of system forming near TCs.

    The common characteristic of the two cases was that the BH could develop just before the turning occurrence. But are there some special effects of the BH on the SRTC's unique motion? The results of the numerical simulation can be utilized to further explain this issue.

5.BH effect
  • The method of system perturbation and separation presented in section 2 was applied using ARW, aimed at changing the target system strength. Besides the BH, the SH and TR were also involved in target systems in order to compare the simulated tracks. The coefficient Si determines the magnitude of systems' intensity changes indicates the strength factor controlling the extent of systems' intensity changes. A positive and negative Si respectively represent the enhancement and reduction of the system's intensity. The details of the experiments are provided in Table 2, and the combined experiments implied that both the TR and SH were perturbed. The perturbations (including the BH, TR and SH) of Pabuk (2007) and Lupit (2009) were respectively initialized at 1200 UTC 7 August and 0000 UTC 20 October, and it is noteworthy that the BH experiments for the two cases were also started at different moments in order to stress the track change characteristics in the perturbed BH. This section concentrates mainly on reporting the results of the BH experiment. Comparisons with other systems' sensitivity experiments are discussed in section 6.

  • Pabuk's (2007) track in the control experiment (ctrl) was compared with its best-track positions (Fig. 5a), including the sudden reversal period from 1800 UTC 8 to 0600 UTC 10 August. Except for the large bias in the northeastward motion, the westward movement and the sudden reversal periods agreed well with the observed track, particularly around the time of turning (1200 UTC 9), so the simulated track of Pabuk (2007) was completely satisfied in the present investigation. After examining the 500 hPa geopotential height field, the discrepancy of the simulated track could be attributed to the slightly weaker SH in ctrl.

    The reversed track of Lupit (2009) was successfully reproduced in ctrl (Fig. 5c), accompanied by the transition speed variation (Fig. 5d). Although the simulated track deflected to the southwest slightly around the turning location, the result can still be regarded as a good reference.

  • For Pabuk (2007), the BH experiments were respectively started at 1200 UTC 7 (Figs. 6a and b) and 0600 UTC 8 August (Figs. 6c and d), both including Si values as 0.5, 1.0, -0.5 and -1.0. Note that the tracks in the ctrl experiments set at different initial moments had slight differences, but this did not affect the simulated results in the BH perturbations because they were only compared with their corresponding ctrl experiments. Apparently, when the initial moments were respectively set at 1200 UTC 7 and 0600 UTC 8 August, the track deviation trends were consistent in the perturbed BH experiments. This means that, after the BH intensity was artificially altered, the track changes were not accidental phenomena. Specifically, in the weakened BH experiments (BH-), Pabuk (2007) tended to move to the southwest around the turning moment and, because of this, its later movement deflected to the west by about 2°-3° of longitude. In the strengthened BH (BH+), its track showed a distinct anticlockwise rotation, a feature distinct from the results in BH-.

  • For Lupit (2009), the initial moments of the BH experiments were respectively at 0000 UTC 20 (Figs. 6e and f) and 1800 UTC 20 October (Figs. 6g and h). Even though these experiments were performed at different initial moments, the whole track biases were in accordance, which proved the effect of the BH. In the BH- experiments, Lupit (2009) kept moving to the southwest at a speed greater than in ctrl at around the turning time, and approached the north island of the Philippines. Although the turning location leaned toward the southwest, both the turning time and the northeastward movement were unanimous with those in ctrl. Meanwhile, in BH+, track changes presented a near opposite trend in that the turning location deviated slightly to the east and its later movement deflected to the southeast.

    According to these results, although the magnitude of the track changes may have differed to some extent, their moving trends were always coincident after the BH's intensity was strengthened or weakened.

6. Comparison experiments
  • 6.1.1. Results of the SH and TR experiments

    Figure 7 presents the Pabuk (2007) track changes caused by the SH and TR perturbations, plus the combined experiments of the SH and TR. The magnitude of the track change in the sensitivity experiments was certain to be influenced by the size of Si. For example, compared with ctrl, the track change in tr-0.5 was not significant. As the intensity of TR continued to be weakened, Pabuk (2007) kept moving to the southwest rather than meandering off the coast, so the northeastward movement after Pabuk (2007) turned was restrained. When Si was set as 0.5 or 1.0, Pabuk (2007) always appeared to rotate near the turning location, as in the BH+ experiments.

    The SH experiments contained two groups, as shown in Fig. 7c. On the one hand, in SH+, the northward movement after reversed turning was remarkably inhibited and the turning location moved much closer to the west, suggesting that the westward flow became stronger in individual SH perturbations. On the other hand, in SH-, Pabuk's (2007) track became more abnormal and also showed a clear anticlockwise rotation.

    In the combined experiments (Figs. 7d and e), the intensities of the TR and SH were altered simultaneously with the same or different Si. It has been shown that the main characteristic associated with the perturbed TR is the anticlockwise rotation, and another, accompanied with the perturbed SH, is the change of the turning location. Clearly, the general trend of movement in the weakened groups only revealed minor change, much weaker than that in SH- and TR-. However, in the strengthened groups, the perturbation of SH played the dominant role in the track change because the two main features were produced and the change in the turning location in tr+1.0 and sh+0.5 was more distinct than the anticlockwise rotation.

    6.1.2. BH effect mechanism

    The simulated tracks obtained from six experiments were compared (Fig. 8a) and the deep-layer mean flow for Pabuk (2007) averaged with a 5°× 5° averaged grid as the level between 850 hPa and 400 hPa [Pike, 1985] was used to help analyze the BH effects. The y-axis in Fig. 8b is the deep-layer mean flow in the sensitivity experiments minus that in ctrl, such as 1.0bh-c, implying the deep-layer mean flow in bh+1.0 minus that in ctrl. Two prominent features were found from the steering flow variation (Fig. 8b). One was the accompaniment of the significantly enhanced north wind, and that was the reason why Pabuk (2007) could not undergo its northeastward motion after 1800 UTC 10 October. Another was that the steering flow variation in 1.0bl-c almost coincided with that in 0.5sh-c after 1800 UTC 9 October when the SH and BH were gradually superimposed and the SH started to show a meridional distribution. But were there any connections between the strengthened BH and the strengthened SH? The area size of 5880 gpm in ctrl was compared to that in bh+1.0 because 5880 gpm could represent the intensity of the SH (figures not shown). It turns out that, after 1800 UTC 9, the north border of the SH in BH+ was always located farther north than that in ctrl. This means that the strengthened BH may have stretched the SH to the north and amplified its meridional distribution. But how was the BH able to exert its influence on the TC? The first possibility relies on its nearby system——the TR.

    From the synoptic analysis for Pabuk (2007), the TR was mainly distributed in (40°-45°N, 100°-120°E). The deep-layer geopotential height within 100°-120°E could be respectively averaged in the 40° and 45°N zone and made the difference with the ctrl. The height difference was utilized to represent the development magnitude of the TR. In this regard, the positive term of the height implied the intensity of the TR was weaker than the ctrl, while a negative value meant a stronger TR. In BH- (Fig. 8c), the difference value ranged from -2 to 5 gpm, and the abrupt negative value at 1200 UTC 11 could be neglected because the focal point was its overall trend. In BH+ (Fig. 8e), the value was close to 4 gpm during the initial period and ultimately stabilized at -5 to 0 gpm. The results obtained from the TR experiments are also shown for comparison. Compared with the results in the BH experiments, the difference values in the TR experiments were more prominent, but the same variation trends were found in the corresponding TR- and TR+ experiments, which proves that the TR in BH- was weakened and in BH+ was reinforced.

    The vorticity advection was also calculated to unveil its physical mechanisms and to complement the above analysis. The 500 hPa vorticity field was respectively averaged over the TR distribution area (40°-45°N) and Pabuk (2007) activity area (22°-27°N), as shown in each sensitivity experiment (Fig. 8). Apparently, the positive vorticity area was mainly distributed between 120°E and 130°E, oriented from northeast to southwest, associated with the TR's eastward movement. Another positive vorticity area was located to the east of 150°E, indicating that the SH had been weakened by a low pressure system. In the BH experiments, it was found that the size and scope of the vorticity field in BH- (Fig. 8d) were smaller than those in BH+ (Fig. 8f), representing a weaker TR in BH-. Meanwhile, among the BH and TR experiments, the vorticity transport in TR+ (Fig. 8j) may have been the strongest, and in TR- (Fig. 8h) the smallest. Consequently, although the intensity change of the TR in the BH perturbations was not as significant as that in TR, the direct impact of the BH on the TR could be validated by the vorticity and height difference variations.

    Figure 8.  (a) The simulated tracks for six perturbed experiments and (b) the corresponding deep-layer mean steering flow difference (one full wind barb represents 1 m s-1). The averaged deep-layer geopotential height difference within 100°-120°E respectively in the 40°N and 45°N zone of (c) bh-1.0, (e) bh+1.0, (g) tr-1.0, (i) tr+1.0 and the longitude-time cross section for the vorticity (10-5 s-1) averaged over 40°-45°N (shading) and the vorticity averaged over 22°-27°N (solid lines) at 500 hPa of (d) bh-1.0, (f) bh+1.0, (h) tr-1.0, (j) tr+1.0 from 1200 UTC 7 to 11 August.

    Figure 9.  The perturbed simulated tracks for Lupit (2009) TR (a, b), SH (c) and combined experiments (both TR and SH are perturbed) at 6 h intervals from 0000 UTC 20 to 25 October.

  • 6.2.1. Results of the SH and TR experiments

    Based on PV inversion, the westerly steering flow was mainly provided by the TR, so the perturbed TR was accorded with the strength of the westerly. That means, when steered by the weaker west wind, Lupit (2009) was supposed to make landfall in the Philippines and linger longer off the coast in the weakened TR (Fig. 9a). On the contrary, the stronger west wind may have driven Lupit (2009) to experience an eastward track deviation accompanied by its translation speed decreasing in advance (Fig. 9b).

    The SH perturbations for Lupit (2009) were also divided into two groups (Fig. 9c). It was also found from the PV inversion that the northwestward movement was mainly controlled by the southeast steering flow supplied by the SH. Thus, the turning location in SH+ lay farther west than that in TR- due to the intensively reinforced southeasterly. Except for the magnitude of eastward deflection, the tracks simulated in the weakened SH were essentially the same as those in TR+, indicating intensification of a TR is regularly accompanied by a weakened SH, and vice versa. The combined experiments could be used to evaluate the relative weight of the TR and SH. For instance, Lupit's (2009) movement trends in different combined experiments (Figs. 9d and e) were always in good agreement with those in the SH experiments. In other words, the effect of the SH on the TC's motion may be more remarkable when the SH and TR are under the same degree of perturbations.

    6.2.2. BH effect mechanism

    Similarly, six simulated tracks (Fig. 10a) were employed to make the comparison. Interestingly, the most obvious trait obtained from the steering flow variation (Fig. 10b) was that the flow variations in BH- and BH+ were respectively well correspondent to those in TR- and TR+. The TR referred to here was in the area (30°-40°N, 110°-130°E). The deep-layer mean geopotential height (between 400 and 850 hPa) was also averaged within 110°-130°N. The difference value was the deep-layer mean geopotential height obtained from sensitivity experiments minus the one obtained from the ctrl. As with the difference value in the 30°N and 40°N zone, shown in Fig. 10c, from 0000 UTC to 1800 UTC 20 October, the minimum value in the 40°N zone reached -6 gpm in BH- (Fig. 10c); later, the value rapidly increased to 2 gpm and remained positive, consistent with the difference variation in the 30°N zone. The difference variation in BH+ (Fig. 10e) was unstable during the initial period, but it gradually moved below the zero line. From the results obtained from the TR experiments (Figs. 10g and i), it can be seen that the positive and negative difference values were respectively shown in the TR- and TR+ experiments, just as expected, which confirms the positive effect mentioned in the Pabuk (2007) results.

    The vorticity advection was also calculated for this case. Considering the only area where the high latitude system could interact with the midlatitude system was between 30°N and 40°N, the 500 hPa vorticity field was respectively averaged over the TR distribution area (30°-40°N) and Lupit (2009) activity area (15°-23°N). From the longitude-time cross variation of the vorticity field (Figs. 10 d-j), it can see that, after 0000 UTC 22, the vorticity transport became more significant and the size of the vorticity field in BH+ (TR+) was bigger than that in BH- (TR+). Consequently, the onset of the BH in some SRTCs' environment fields may help maintain the balance of the vorticity transport and systems' evolution, thereby contributing to the formation of TCs' unique tracks.

7. Discussion and summary
  • According to the observational analysis reported in the companion paper [Luo et al., 2015], a peculiar phenomenon was found in that a BH developed and persisted in the environment field of 19 SRTC cases. Four types of phase relations between the BH, the CH, the SH and the SRTC were proposed and some relevant features examined in detail. Because types C and D accounted for greater occurrence frequency, they were taken for as the focal point of the present study.

    Considering the large number of cases, Pabuk (2007) and Lupit (2009) were respectively adopted as representatives of types C and D. The synoptic analysis for Pabuk (2007) attributed the reversed turning to the suddenly enhanced monsoon surge and accelerated eastward shift of the SH. However, synoptic analysis failed to capture some important characteristics leading to Lupit's (2009) reversed turning, and the piecewise PV inversion method was introduced to obtain the steering flows associated with individual perturbations. As a result, although the strong westerly provided by the TR played a major role in Lupit's (2009) track change, the southwesterly associated with the HZ also had a very beneficial effect on the turning. So, if new systems can form during a TC's movement, multiple-system interaction needs further in-depth study.

    Figure 10.  (a) The simulated tracks for six perturbed experiments and (b) the corresponding deep-layer mean steering flow difference (one full wind barb represents 1 m s-1). The averaged deep-layer geopotential height difference within 110°-130°E respectively in the 30°N and 40°N zones of (c) bh-1.0, (e) bh+1.0, (g) tr-1.0, (i) tr+1.0 and the longitude-time cross section for the vorticity (10-5 s-1) averaged over 30°-40°N (shading) and the vorticity averaged over 15°-23°N (solid lines) at 500 hPa of (d) bh-1.0, (f) bh+1.0, (h) tr-1.0, (j) tr+1.0 from 0000 UTC 20 to 25 October.

    A series of BH experiments for the two cases were conducted using the method of target system perturbations. Pabuk (2007) tended to deflect to the southwest in BH-, including its later movement trend, while in BH+ its track showed a clear anticlockwise rotation without its distinct northward movement. In the BH- experiments, Lupit (2009) was supposed to keep its southwestward movement and approach the north island of the Philippines, while in BH+$, simulated tracks presented an opposite motion with eastward deflections. Here, all the TR, SH and combined experiments were utilized as comparisons and the results obtained from these experiments also confirmed the remote impact of the BH, because the track changes always had common features. The analyses of the deep-layer steering flow variations led us to consider how the BH might have exerted its influence, i.e., the physical mechanisms of this remote impact. After investigating the variations of the geopotential height difference and vorticity field, it can be concluded that the onset and persistence of the BH in some SRTCs' environment fields may help maintain the balance of the vorticity transport and systems' evolution, thereby contributing to the formation of TCs' unique tracks. Further work should be carried out and may be of great importance in improving SRTC forecasting and mechanistic analyses.

Reference

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return