• Berger, H., , R. Langland, , C. S. Velden, , C. A. Reynolds, , and P. M. Pauley, 2011: Impact of enhanced satellite-derived atmospheric motion vector observations on numerical tropical cyclone track forecasts in the western North Pacific during TPARC/TCS-08. J. Appl. Meteor. Climatol., 50, 23092318, doi:10.1175/JAMC-D-11-019.1.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 1990: Observational evidence for predictions of tropical cyclone propagation relative to environmental steering. J. Atmos. Sci., 47, 542548, doi:10.1175/1520-0469(1990)047<0542:OEFPOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 1998: Objective diagnosis of binary tropical cyclone interactions for the western North Pacific basin. Mon. Wea. Rev., 126, 17341740, doi:10.1175/1520-0493(1998)126<1734:ODOBTC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 2000a: Dynamical tropical cyclone track forecast errors. Part I: Tropical region error sources. Wea. Forecasting, 15, 641661, doi:10.1175/1520-0434(2000)015<0641:DTCTFE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 2000b: Dynamical tropical cyclone track forecast errors. Part II: Midlatitude circulation influences. Wea. Forecasting, 15, 662681, doi:10.1175/1520-0434(2000)015<0662:DTCTFE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , M. A. Boothe, , and R. L. Elsberry, 1997: Observational evidence for alternate modes of track-altering binary tropical cyclone scenarios. Mon. Wea. Rev., 125, 20942111, doi:10.1175/1520-0493(1997)125<2094:OEFAMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Chan, J. C., , and W. Gray, 1982: Tropical cyclone movement and surrounding flow relationships. Mon. Wea. Rev., 110, 13541374, doi:10.1175/1520-0493(1982)110<1354:TCMASF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Charney, J. G., 1955: The use of primitive equations of motion in numerical prediction. Tellus, 7, 2226, doi:10.1111/j.2153-3490.1955.tb01138.x.

    • Search Google Scholar
    • Export Citation
  • Davis, C. A., , and K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev.,119, 1925–1953, doi:10.1175/1520-0493(1991)119<1929:PVDOC>2.0.CO;2.

  • Dong, K. Q., , G. J. Holland, , and B. C. Diehl, 1991: Typhoon recurvature and environmental circulation features. WMO Tech. Doc. WMO/TD-472, World Meteorological Organization, Geneva, Switzerland, 16–22.

  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, doi:10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Elsberry, R. L., 1995: Tropical cyclone motion. Global Perspectives on Tropical Cyclones, R. L. Elsberry, Ed., WMO Tech. Doc. WMO/TD-693, World Meteorological Organization, Geneva, Switzerland, 106–197.

  • Galarneau, T. J., , and C. A. Davis, 2013: Diagnosing forecast errors in tropical cyclone motion. Mon. Wea. Rev., 141, 405–430, doi:10.1175/MWR-D-12-00071.1.

    • Search Google Scholar
    • Export Citation
  • Harr, P. A., , and R. L. Elsberry, 1991: Tropical cyclone track characteristics as a function of large-scale circulation anomalies. Mon. Wea. Rev., 119, 14481468, doi:10.1175/1520-0493(1991)119<1448:TCTCAA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Harr, P. A., , and R. L. Elsberry, 1995: Large-scale circulation variability over the tropical western North Pacific. Part I: Spatial patterns and tropical cyclone characteristics. Mon. Wea. Rev., 123, 12251246, doi:10.1175/1520-0493(1995)123<1225:LSCVOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., , Y. Noh, , and J. Dudhia, 2006: A new vertical di¤usion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341, doi:10.1175/MWR3199.1.

    • Search Google Scholar
    • Export Citation
  • Japan Meteorological Agency, 2012: Annual Report on the Activities of the RSMC Tokyo-Typhoon Center 2011. Japan Meteorological Agency, 82 pp. [Available online at http://www.jma.go.jp/jma/jma-eng/jma-center/rsmc-hp-pub-eg/AnnualReport/2011/Text/Text2011.pdf.]

  • Jian, G.-J., , and C.-C. Wu, 2008: A numerical study of the track deflection of Supertyphoon Haitang (2005). Mon. Wea. Rev., 136, 598615, doi:10.1175/2007MWR2134.1.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, doi:10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kehoe, R. M., , M. A. Boothe, , and R. L. Elsberry, 2007: Dynamical tropical cyclone 96- and 120-h track forecast errors in the western North Pacific. Wea. Forecasting, 22, 520538, doi:10.1175/WAF1002.1.

    • Search Google Scholar
    • Export Citation
  • Kieu, C., , T. Nguyen, , M. Hoang, , 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, 17941810, doi:10.1175/JTECH-D-12-00020.1.

    • Search Google Scholar
    • Export Citation
  • Komaromi, W. A., , S. J. Majumdar, , and E. D. Rappin, 2011: Diagnosing initial condition sensitivity of Typhoon Sinlaku (2008) and Hurricane Ike (2008). Mon. Wea. Rev., 139, 32243242, doi:10.1175/MWR-D-10-05018.1.

    • Search Google Scholar
    • Export Citation
  • Kurihara, Y., , M. A. Bender, , R. E. Tuleya, , and R. J. Ross, 1990: Prediction experiments of Hurricane Gloria (1985) using a multiply nested movable mesh model. Mon. Wea. Rev., 118, 21852198, doi:10.1175/1520-0493(1990)118<2185:PEOHGU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kurihara, Y., , M. A. Bender, , and R. J. Ross, 1993: An initialization scheme of hurricane models by vortex specification. Mon. Wea. Rev., 121, 20302045, doi:10.1175/1520-0493(1993)121<2030:AISOHM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lander, M. A., 1996: Specific tropical cyclone track types and unusual tropical cyclone motions associated with a reverse-oriented monsoon trough in the western North Pacific. Wea. Forecasting, 11, 170186, doi:10.1175/1520-0434(1996)011<0170:STCTTA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Li, T., , and B. Fu, 2006: Tropical cyclogenesis associated with Rossby wave energy dispersion of a preexisting typhoon. Part I: Satellite data analyses. J. Atmos. Sci., 63, 13771389, doi:10.1175/JAS3692.1.

    • Search Google Scholar
    • Export Citation
  • Luo, Z. X., , N. E. Davidson, , F. Ping, , and Weican Zhou, 2011: Multiple-scale interactions affecting tropical cyclone track change. Adv. Mech. Eng., 2011, 19, doi:10.1155/2011/782590.

    • Search Google Scholar
    • Export Citation
  • 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 66316 682, doi:10.1029/97JD00237.

    • Search Google Scholar
    • Export Citation
  • Neumann, C. J., 1992: Final report: Joint Typhoon Warning Center (JTWC92) model. SAIC Contract Rep. N00014-90-C-6042 (Part 2), 83 pp.

  • Patla, J. E., , D. Stevens, , and G. M. Barnes, 2009: A conceptual model for the influence of TUTT cells on tropical cyclone motion in the Northwest Pacific Ocean. Wea. Forecasting, 24, 12151235, doi:10.1175/2009WAF2222181.1.

    • Search Google Scholar
    • Export Citation
  • Pike, A. C., 1985: Geopotential heights and thicknesses as predictors of Atlantic tropical cyclone motion and intensity. Mon. Wea. Rev., 113, 931940, doi:10.1175/1520-0493(1985)113<0931:GHATAP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Rappaport, E. N., and Coauthors, 2009: Advances and challenges at the National Hurricane Center. Wea. Forecasting, 24, 395419, doi:10.1175/2008WAF2222128.1.

    • Search Google Scholar
    • Export Citation
  • Riehl, H., , and R. J. Shafer, 1944: The recurvature of tropical storms. J. Meteor., 1, 4254, doi:10.1175/1520-0469(1944)001<0001:TROTS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Thompson, G., , P. R. Field, , R. M. Rasmussen, , and W. D. Hall, 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Mon. Wea. Rev., 136, 50955115, doi:10.1175/2008MWR2387.1.

    • Search Google Scholar
    • Export Citation
  • Torn, R. D., , and C. A. Davis, 2012: The influence of shallow convection on tropical cyclone track forecasts. Mon. Wea. Rev., 140, 21882197, doi:10.1175/MWR-D-11-00246.1.

    • Search Google Scholar
    • Export Citation
  • Velden, C. S., , and L. M. Leslie, 1991: The basic relationship between tropical cyclone intensity and the depth of the environmental steering layer in the Australian region. Wea. Forecasting, 6, 244253, doi:10.1175/1520-0434(1991)006<0244:TBRBTC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-C., , and K. A. Emanuel, 1995a: Potential vorticity diagnostics of hurricane movement. Part I: A case study of Hurricane Bob (1991). Mon. Wea. Rev., 123, 6992, doi:10.1175/1520-0493(1995)123<0069:PVDOHM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-C., , and K. A. Emanuel, 1995b: Potential vorticity diagnostics of hurricane movement. Part II: Tropical Storm Ana (1991) and Hurricane Andrew (1992). Mon. Wea. Rev., 123, 93109, doi:10.1175/1520-0493(1995)123<0093:PVDOHM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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, 12891300, doi:10.1175/1520-0493(2003)131<1289:ANLATB>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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, 20842093, doi:10.1175/1520-0493(2004)132<2084:PVDOTK>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-C., , S. G. Chen, , C. C. Yang, , P. H. Lin, , and S. D. Aberson, 2012: Potential vorticity diagnosis of the factors affecting the track of Typhoon Sinlaku (2008) and the impact from dropwindsonde data during T-PARC. Mon. Wea. Rev., 140, 26702688, doi:10.1175/MWR-D-11-00229.1.

    • Search Google Scholar
    • Export Citation
  • Wu, L., , H. Zong, , and J. Liang, 2011: Observational analysis of sudden tropical cyclone track changes in the vicinity of the East China Sea. J. Atmos. Sci., 68, 30123031, doi:10.1175/2010JAS3559.1.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    The best track of Megi (2010) from 0000 UTC 14 Oct to 0000 UTC 24 Oct 2010 at 6-h intervals (from CMA-STI).

  • View in gallery

    (left) The steering flow of steering layers determined by Megi’s central pressure at (a) 0000 UTC 17 Oct, (b) 1200 UTC 18 Oct, (c) 1500 UTC 19 Oct, and (d) 1500 UTC 20 Oct 2010. (right) The DLM (850–250 hPa) wind (one full wind barb represents 4 m s−1) and geopotential height (the areas with values larger than 5280 gpm are shaded) at the corresponding times.

  • View in gallery

    The inversion area and the potential vorticity perturbations [contour interval of 0.1 potential vorticity units (PVU), the area of PV absolute value more than 0.1 PVU is shaded; 1 PVU = 10−6 m2 s−1 K kg−1] in SH, CH, and TR individually at 500 hPa at 0000 UTC 17 Oct 2010.

  • View in gallery

    (a) Time series of Megi’s motion (), the actual steering flow from the National Centers for Environmental Prediction (NCEP) data [], the steering flow associated with SH [], CH [], TR [], and the sum of these three flow components above []. (b) (left) The negative PV perturbation of SA at 500 hPa and the balanced wind associated with it (vector) on 20 Oct 2010; (right) The steering flow associated with SA [] from 20 to 21 Oct 2010. (c) As in (a), but has been added in and at 0000 UTC 20 and 21 Oct 2010. One full wind barb (flag) represents 1(5) m s−1.

  • View in gallery

    The simulated tracks in the control (solid circles) at 3-h intervals and the best track (typhoon symbol) from 0000 UTC 17 Oct to 22 Oct 2010.

  • View in gallery

    The sketch map of the initial perturbations valid 0000 UTC 17 Oct 2010 in (left) a single system and (right) the perturbed simulated tracks in (a) SCH, (b) SSH, and (c) STR, respectively. Both the Pacific subtropical high (SH) and continental high (CH) are represented by a 5880-gpm geopotential height contour at 500 hPa, while the midlatitude trough (TR) is represented by a 5400-gpm geopotential height contour. The contours of different styles in each sensitivity experiment groups stand for the perturbed simulations with different values of the disturbance coefficient.

  • View in gallery

    The DLM (850–250 hPa) wind, geopotential height [(a),(c) the areas with values larger than 5290 gpm are shaded], and steering flow of corresponding steering layers [(b),(d) the areas with values larger than 5 m s−1 associated with CH are shaded] at (a),(b) 0600 UTC 17 Oct and (c),(d) 0600 UTC 19 Oct 2010 in (left) SCH+1 and (right) SCH−1. One full wind barb represents 10 m s−1.

  • View in gallery

    The synoptic-scale (a) surface wind field at 1800 UTC 19 Oct 2010 and 500-hPa wind field at (b) 0000 UTC 16 Oct, (c) 0000 UTC 17 Oct, and (d) 1800 UTC 19 Oct 2010. The typhoon mark represents the center location of Megi. The vertical–latitudinal cross section of relative vorticity (10−5 s−1) along (e) 12°N and (f) the Rossby wave train axis is labeled by the solid line in (a). The y axis denotes pressure (hPa). The solid circle in (f) represents the position of Typhoon Megi. The letters “A” and “C” represent the centers of the anticyclone and cyclone, respectively.

  • View in gallery

    The sketch map of the initial perturbations valid 0000 UTC 17 Oct 2010 in (left) a single system and (right) the perturbed simulated tracks in SEW. The easterly wave (EW) is represented by a 5870-gpm geopotential contour. The contours of different styles in SEW stand for the perturbed simulations with different values of the disturbance coefficient.

  • View in gallery

    The DLM (850–250 hPa) wind, geopotential height [(a),(c) the areas with values larger than 5290 gpm are shaded], and steering flow of corresponding steering layers [(b),(d) the areas with values larger than 5 m s−1 are shaded] at (a),(b) 0900 UTC 17 Oct and (c),(d) 0000 UTC 20 Oct 2010 in (left) SEW+1 and (right) SEW−1. One full wind barb represents 10 m s−1.

  • View in gallery

    The simulated results of the 10 combined experiments at 6-h intervals.

  • View in gallery

    The simulated results of the control, COMB3, COMB4, SCH+1, and SEW−1 at 6-h intervals.

  • View in gallery

    The motion vectors (km h−1) of Megi before (dashed arrows) and after deflection (solid arrows) in the control, SCH+1, SEW−1, COMB3, and COMB4, with the translation speed labeled at the tips of the arrows. The solid points show the location of deflection.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 12 12 7
PDF Downloads 10 10 8

A Numerical Study on the Combined Effect of Midlatitude and Low-Latitude Systems on the Abrupt Track Deflection of Typhoon Megi (2010)

View More View Less
  • 1 College of Meteorology and Oceanography, PLA University of Science and Technology, Nanjing, China
  • 2 Beijing Aerospace Control Center, Beijing, China
© Get Permissions
Full access

Abstract

In 2010, Supertyphoon Megi experienced an abrupt track deflection in the South China Sea (SCS) after traversing Luzon Island. To reveal the physical processes responsible for the timing and location of the sudden track deflection, the potential vorticity (PV) diagnosis and numerical simulations with initial strength perturbations are applied to investigate the individual and combined effects of environmental systems on Megi’s motion based on the steering flow theory. Results indicate that Megi’s northward track deflection was mainly determined by the effect of the midlatitude circulation, or rather, the break of the high pressure belt joined by the continental high (CH) and the Pacific subtropical high (SH). The retraction of CH played a particularly critical role in the break of the high pressure belt, making it the most important feature of the midlatitude circulation to determine Megi’s deflection. In addition, a small low-latitude anticyclone (SA) southeast of Megi was crucial in affecting the timing and location of the deflection, although the steering effect provided by SA itself was relatively weak. The development of SA was associated with both the tropical cyclone energy dispersion and the activity of an easterly wave. This study suggests that the abrupt track deflection of Megi was attributed to the combined effect of the midlatitude and low-latitude systems, in addition to the combined effects of the large-scale and small-scale systems.

Corresponding author address: Xiaogang Huang, College of Meteorology and Oceanography, PLA University of Science and Technology, No. 60, Shuanglong Rd., Nanjing 211101, China. E-mail: huang.x.g@163.com

Abstract

In 2010, Supertyphoon Megi experienced an abrupt track deflection in the South China Sea (SCS) after traversing Luzon Island. To reveal the physical processes responsible for the timing and location of the sudden track deflection, the potential vorticity (PV) diagnosis and numerical simulations with initial strength perturbations are applied to investigate the individual and combined effects of environmental systems on Megi’s motion based on the steering flow theory. Results indicate that Megi’s northward track deflection was mainly determined by the effect of the midlatitude circulation, or rather, the break of the high pressure belt joined by the continental high (CH) and the Pacific subtropical high (SH). The retraction of CH played a particularly critical role in the break of the high pressure belt, making it the most important feature of the midlatitude circulation to determine Megi’s deflection. In addition, a small low-latitude anticyclone (SA) southeast of Megi was crucial in affecting the timing and location of the deflection, although the steering effect provided by SA itself was relatively weak. The development of SA was associated with both the tropical cyclone energy dispersion and the activity of an easterly wave. This study suggests that the abrupt track deflection of Megi was attributed to the combined effect of the midlatitude and low-latitude systems, in addition to the combined effects of the large-scale and small-scale systems.

Corresponding author address: Xiaogang Huang, College of Meteorology and Oceanography, PLA University of Science and Technology, No. 60, Shuanglong Rd., Nanjing 211101, China. E-mail: huang.x.g@163.com

1. Introduction

Western North Pacific (WNP) tropical cyclone (TC) track forecasts have greatly improved over the past 30 years. According to the verification data of operational forecasts from the Japan Meteorological Agency (2012), by 2011 the short-range forecast (24–48 h) errors were roughly half of those in 1982. However, over the past decade, and particularly since 2005, the pace of improvement has slowed considerably. Furthermore, though track forecast errors are decreasing, there are still large year-to-year fluctuations at lead times of 72 h and beyond, and these mostly occur in the problematic forecast situations associated with unusual TC motions, such as abrupt recurving (Rappaport et al. 2009). As societal expectations increase with the overall improvement in track forecasts, poor forecasts in abnormal cases not only become a scientific challenge but also a social issue (Galarneau and Davis 2013).

Factors influencing TC motion include the environmental steering flow, beta effect, diabatic heating, asymmetric structure, and orographic effects. Among all these factors, the environmental steering flow is responsible for about 50%–80% of TC motion (Elsberry 1995), and in some cases up to 90% (Neumann 1992). Thus, TC motion is mainly controlled by the environmental steering flow (Chan and Gray 1982; Carr and Elsberry 1990; Harr and Elsberry 1995; Berger et al. 2011), which consists generally of a combination of the flows associated with these systems surrounding it, including subtropical ridges, midlatitude troughs, and cyclonic circulations. Because of the complicated interactions between these atmospheric features, the improper estimation of crucial steering factors by forecasters and the misrepresentation of the interaction in numerical models can result in large track forecast errors (Harr and Elsberry 1991, 1995; Carr and Elsberry 2000a,b; Kehoe et al. 2007; Galarneau and Davis 2013; Wu et al.2012). In particular, for a TC with an abnormal track the complicated interaction with the surrounding systems becomes the predominant factor limiting predictability. In this respect, a systematic and in-depth investigation of subtle interactions between these atmospheric features would substantially improve track forecasts.

Various research studies have been conducted to examine the complicated interactions between TCs and environmental features. Several advanced methodologies were applied to examine related factors contributing to unusual tracks. Potential vorticity (PV) diagnosis was successfully used to investigate the factors affecting the track of Hurricane Bob (1991) (Wu and Emanuel 1995a) and Typhoon Sinlaku (2002) (Wu et al. 2004). Numerical simulations using advanced models were then conducted to examine the physical processes responsible for TC motion (Jian and Wu 2008) and the sensitivity of tracks to initial perturbations of certain atmospheric features (Komaromi et al. 2011). In addition, based on the observational analysis, a number of conceptual models were proposed to describe the physical mechanisms responsible for TC motion and forecast challenges, which improved track forecasts in similar cases (Carr et al. 1997; Carr and Elsberry 1998). Furthermore, error mechanisms developed from the observational evidence were described by conceptual models, and these can guide forecasters to analyze complex situations accurately, and to determine whether the processes have been misrepresented in the dynamical models (Carr and Elsberry 2000a,b). However, the mechanisms responsible for certain complex interaction processes or for novel dynamical processes in new TC cases are still inadequately understood, and it is undoubtedly necessary to continuously develop and enrich our understanding of these mechanisms.

In terms of research related to an abrupt northward-turning track, consensus has now been reached on a midlatitude mechanism that the approach of a deep westerly trough and the eastward retreat of the subtropical ridge are found to be favorable to a northward shift (Riehl and Shafer 1944; Dong et al. 1991; Wu et al. 2004). Moreover, interactions between TCs and certain tropical systems, such as the low-latitude monsoon trough (Lander 1996), binary tropical cyclone interaction (Carr et al. 1997; Wu et al. 2003), and the tropical upper-tropospheric trough (TUTT; Patla et al. 2009), cannot be ignored when TCs are situated in the tropics. Misrepresentation of these atmospheric features accounts for a considerable portion of TC track forecast errors (Carr and Elsberry 2000a; Kehoe et al. 2007). However, it is evident that because of the complex physical processes occurring in the tropics, and the limited focus on them, dynamic models describing interactions between TCs and tropical features (in particular certain small, weak features) are far from complete compared with those describing the midlatitude systems.

One typical category of anomalous TC tracks in the WNP presenting a challenge to forecasters is sudden track deflection (Wu et al. 2011). For example, Typhoons Saomai (2000), Maemi (2003), Sinlaku (2008), Jangmi (2008), and several others underwent sudden track changes. Supertyphoon Megi (2010), the focus of this research, experienced a sharp northeast deflection at low latitudes over the South China Sea (SCS). The SCS is a basin with frequent TC activity, and is renowned for the complicated interaction between TCs and various circulation features, including the Pacific subtropical high, South China Sea high, cross-equatorial flow, South Asian monsoon, and easterly waves. However, the steering effect from these systems tends to cancel one another out, which makes it rather difficult to conduct a precise track forecast. The abnormal track of Megi, and its record-breaking intensity, posed a unique scientific and forecasting issue. Results from relevant research show that the midlatitude circulation dominated its track deflection (Kieu et al. 2012). However, taking into account the location of Megi, the contribution of tropical features in the SCS may be significant, and therefore the dynamical mechanism of the abnormal track remains undetermined and inexplicit. The purpose of this study is, therefore, to investigate the dynamical mechanism and address the following issues: 1) whether the low-latitude systems played an important role in Megi’s track deflection, 2) what effect the combination of the midlatitude and low-latitude circulation had on Megi, and 3) whether there was any meaningful physical mechanism that could be identified in this anomalous track case. Finding answers to such issues will not only enrich our understanding of the mechanisms involved in abnormal tracks, but also contribute to the ability to provide accurate predictions for the analogous troublesome TC tracks.

This paper continues as follows. Section 2 describes the data and methodology, including the model configuration. Section 3 provides an overview of Megi. The case synopsis of Megi is presented in section 4. Sections 5 and 6 show the steering effects of individual mechanisms in the midlatitudes and low latitudes, respectively, and their combination is discussed in section 7. Discussion and conclusions are provided in section 8.

2. Data and methodology

a. Data

TC data in this study are obtained from the Chinese Meteorological Administration’s Shanghai Typhoon Institute (CMA-STI) best track datasets, which contain measurements of TC center positions (latitude and longitude) and intensity (the maximum of the 2-min sustained surface wind and central pressure) at 6-h intervals. This study also employs the data on 0.5° × 0.5° grids at every 3 h from the Global Forecast System (GFS) database (http://www.nco.ncep.noaa.gov/pmb/products/gfs). The data are available on the surface, at 26 pressure levels from 1000 to 10 hPa, including surface pressure, sea level pressure, geopotential height, temperature, relative humidity, and wind.

b. Methodology

1) The calculation of steering flow

The vertical heights to which TCs can extend are usually decided by their intensities, leading to steering layers that vary in each case and even within different phases of a particular case (e.g., Velden and Leslie 1991). According to the standard from the Cooperative Institute for Meteorological Satellite Studies (CIMSS; http://tropic.ssec.wisc.edu/real-time/dlmmain.php?&basin=atlantic&sat=wg8&prod=dlm1&zoom=&time; see Table 1), the steering layer is determined by the minimum central pressure (obtained from CMA-STI) of Megi at each point in time.

Table 1.

Central pressures and its corresponding steering layers.

Table 1.

Following the methodology in Pike [1985, Eq. (1)], the deep-layer mean (DLM) flow of the steering layer is calculated as
e1
where is the deep-layer mean field, and is the field F at pressure . Taking a TC of 945 hPa as an example, the levels between 850 and 250 hPa are used to define the steering flow (from Table 1). Equation (1) is used to compute the steering flow using the prescribed weights for each level. All DLM fields referred to in this paper are constructed in the same way.

2) Piecewise potential vorticity inversion

In this study, piecewise PV inversion is selected to conduct a quantitative analysis of the steering effect from individual atmospheric features. PV inversion was developed by Davis and Emanuel (1991) to obtain balanced fields associated with each individual PV perturbation from perturbation PV equations. Wu and Emanuel (1995a,b; Wu et al. 2003, 2004, 2012) then successfully applied this method to analyze the interaction between TCs and environmental features, using various case studies: Typhoon Sinlaku (2002, 2008), Hurricane Bob (1991), Hurricane Ana (1991), and Hurricane Andrew (1992).

In this study, the global zonal average field is defined as the mean field, and the rest as the perturbation field. The mean field is first calculated to obtain the mean streamfunction , so that the associated mean geopotential height and PV fields can be derived from the nonlinear balance equation (Charney 1955):
e2
and from the Ertel PV equation on the coordinate and spherical coordinates:
e3
where q represents PV, represents geopotential height, represents streamfunction, a is the Earth’s radius, f is the Coriolis parameter, is latitude, and is longitude. Through the perturbation form of the above equations, we can take the perturbation field and associated with each PV perturbation . The balanced flow and mass fields can then be calculated. For the detailed algorithm, please see Wu et al. (2004).

3) Systems perturbation based on the filtering method

In this study, a filtering method is applied to split the initial field into basic and disturbance components and to perturb the intensity of specific environmental circulation features. In doing so, we first separate the original scalar field into the basic field and the disturbance field by using a filtering operator (Kurihara et al. 1993):
e4
The basic field represents the large-scale general features in the analysis, while the disturbance field is the deviation from the basic field. The filtering operator is a local three-point smoothing operator (Kurihara et al. 1990), and is obtained as follows, with and representing latitude and longitude, respectively (in degrees; and being longitudes differing from by a distance of 1°):
e5
e6
The coefficient K is the filtering parameter defined by
e7
When using the above filtering operator, all waves with a wavelength of less than 9° are completely removed, and the amplitudes of those with wavelengths of 15°, 20°, and 30° are reduced by 82%, 60%, and 32%, respectively.
Because the filtering technique can separate the disturbance field from the original field, the initial strength of an individual system can be changed by strengthening or weakening its disturbance field in the original field:
e8
and thus, the new initial field with perturbation in a specific system is composed of the original field coupled with the disturbance field , multiplied by a disturbance coefficient , which controls the magnitude of the perturbation. For example, a perturbation with means that the disturbances associated with a specific system are doubled in the original field, while a perturbation with means that the disturbances are completely filtered out from the original field.

4) Model configuration

The regional Advanced Research Weather Research and Forecasting Model (ARW-WRF, version 3.4, hereafter WRF) is employed in this study. WRF uses fully compressible Eulerian nonhydrostatic equations with a mass vertical coordinate. A single, fixed grid with a horizontal resolution of 30 km and 27 vertical levels (1000–50 hPa) is used. The domain is centered at 20°N, 125°E with 241 × 226 grid points that include primary synoptic features.

The model physics include the New Thompson microphysics scheme (Thompson et al. 2008), the Kain–Fritsch cumulus parameterization (Kain 2004) scheme, and the Yonsei University (YSU) boundary layer scheme (Hong et al. 2006). In addition, the Rapid Radiative Transfer Model (RRTM) longwave radiation scheme (Mlawer et al. 1997), and Dudhia shortwave radiation scheme (Dudhia 1989) are incorporated. Note that the Kain–Fritch cumulus scheme is demonstrated as having a problem with the short-term forecast of TC positions in terms of its improper treatment of shallow convection over the western Atlantic basin (Torn and Davis 2012). However, as the physical processes over oceans vary from basin to basin, and an acceptable simulation result was acquired in the control simulation, the Kain–Fritsch scheme is employed in our research.

3. Overview of Supertyphoon Megi

Supertyphoon Megi was the strongest TC to occur in around ocean regions worldwide during 2010. Based on the best track data provided by CMA-STI (Fig. 1), Megi developed from a tropical depression on 13 October 2010 near 11.9°N, 141.4°E, and was then upgraded to a typhoon (33 m s−1) at 0000 UTC 15 October. Megi maintained a steady northwestward motion and reached its peak intensity with an estimated maximum surface wind of 72 m s−1 and a minimum mean sea level pressure (MSLP) of 895 hPa, ranked as a super typhoon at 1200 UTC 17 October.

Fig. 1.
Fig. 1.

The best track of Megi (2010) from 0000 UTC 14 Oct to 0000 UTC 24 Oct 2010 at 6-h intervals (from CMA-STI).

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

After making landfall on the Luzon, it degraded dramatically to a strong typhoon on 18 October due to the strong friction effect of terrain. However, Megi restrengthened when it moved into the SCS on 19 October prior to making a sharp northeastward turn exceeding 90°. At 0000 UTC 20 October, the southwestward motion was replaced by a northeastward motion with a low translation speed. Megi made landfall in the Fujian province of China at 1600 UTC 23 October and then quickly weakened to a tropical depression before dissipating completely several hours later.

Megi is a representative case illustrating the scenario where a number of forecasts from operational centers failed to predict its abrupt deflection [e.g., the CMA mistakenly predicted its landfall in the Hainan province of China]. This study thus focuses on examining the mechanisms responsible for the sudden deflection of Megi.

4. Synopsis

It is an established fact that TC motion is mainly controlled by the large-scale environmental steering flow, and this fact is verified in Megi (see the bottom two rows of Fig. 4a). As the steering flow is consistent with the motion of Megi, its abnormal track is studied in terms of the environmental flow, while other related factors such as the wind shear and the effect of terrain will be discussed in future works.

The steering flow and the DLM geopotential height field are shown in Fig. 2. At 0000 UTC 17 October, Megi was situated between two large deep-layer anticyclones, the continental high (CH; over China) and the Pacific subtropical high (SH; over the WNP), which joined together as a high pressure belt to steer Megi southwestward continuously (Fig. 2a). Simultaneously, an easterly wave (EW) to the south of SH developed around 150°E, creating an inverted V-shaped deformation of SH with a strengthened wind speed in the region. The western part of the deformed anticyclone (SH) extended southward, close to 5°N, between Megi and EW.

Fig. 2.
Fig. 2.

(left) The steering flow of steering layers determined by Megi’s central pressure at (a) 0000 UTC 17 Oct, (b) 1200 UTC 18 Oct, (c) 1500 UTC 19 Oct, and (d) 1500 UTC 20 Oct 2010. (right) The DLM (850–250 hPa) wind (one full wind barb represents 4 m s−1) and geopotential height (the areas with values larger than 5280 gpm are shaded) at the corresponding times.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

As Megi moved southwestward, the high pressure belt gradually weakened (Fig. 2b) and was finally cut off on 19 October (Fig. 2c), providing favorable conditions for the northward track deflection of Megi. About 24 h later, the abrupt track deflection occurred. This is consistent with the results from Kieu et al. (2012) who attributes the track change to the midlatitude circulation. In addition, at 1500 UTC 19 October (Fig. 2c), a small closed anticyclone (SA) formed at low latitudes, taking the place of the southward extending ridge of SH. The growing SA led to the weakening of the steering effect from SH and the strengthening of the southeasterly flow. Several hours later, Megi deflected from a westward motion to a northeastward motion (Fig. 2d). It is evident that the configuration of the flow patterns became complex before the deflection of Megi, including the weakening SH and CH, the passing midlatitude trough (TR), and evolving SA, thus making it difficult to accurately estimate the steering flow, not to mention the storm motion prediction.

To grasp the key factors dominating the track change, we conduct our research by separating the whole complex environmental circulation into two parts: the midlatitude circulation (consisting of CH, SH, and TR), and the low-latitude circulation (including SA and EW).

5. The effect of midlatitude systems on Megi

a. PV diagnosis

Based on the analyses of these environmental systems, piecewise PV inversion is applied to calculate the steering flow associated with different PV features quantitatively, in order to examine how these systems affected the abnormal motion of Megi. First, the whole PV perturbation is divided into three parts, , , and , representing PV perturbations associated with the Pacific SH, CH, and TR, respectively. To highlight these PV features, an example time (0000 UTC 17 October) is chosen to show the distribution of PV perturbations (Fig. 3). The SH and CH are taken as negative PV perturbations within the regions of 5°–40°N, 125°E–180° and 15°–30°N, 80°–125°E, respectively, whereas TR is taken as a positive PV perturbation within 25°–50°N, 100°–140°E. The vertical levels for each system range from 850 to 250 hPa.

Fig. 3.
Fig. 3.

The inversion area and the potential vorticity perturbations [contour interval of 0.1 potential vorticity units (PVU), the area of PV absolute value more than 0.1 PVU is shaded; 1 PVU = 10−6 m2 s−1 K kg−1] in SH, CH, and TR individually at 500 hPa at 0000 UTC 17 Oct 2010.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

The balanced wind field associated with each system is obtained through the PV inversion of individual PV perturbation from 0000 UTC 17 October to 0000 UTC 21 October. The steering flow associated with q′ () is defined as the DLM (850–250 hPa) wind vector averaged over the inner 440 km around Megi’s center. The time series of the DLM steering flow associated with , , and [, , and ] and their summation [] are compared with the translation speed of Megi (; estimated from the 6-h best track positions) and the actual steering flow obtained from the GFS data [] in Fig. 4a.

Fig. 4.
Fig. 4.

(a) Time series of Megi’s motion (), the actual steering flow from the National Centers for Environmental Prediction (NCEP) data [], the steering flow associated with SH [], CH [], TR [], and the sum of these three flow components above []. (b) (left) The negative PV perturbation of SA at 500 hPa and the balanced wind associated with it (vector) on 20 Oct 2010; (right) The steering flow associated with SA [] from 20 to 21 Oct 2010. (c) As in (a), but has been added in and at 0000 UTC 20 and 21 Oct 2010. One full wind barb (flag) represents 1(5) m s−1.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

It can be seen in Fig. 4a that the total steering effect of the three systems [] is in general agreement with the translation speed of Megi [] and the steering flow obtained from the GFS data []. During the time period 17–21 October, the sudden track change of Megi from westward to northward is generally consistent with the clockwise rotation of , and the decreasing translation speed of Megi agrees fairly with the deceleration of the steering flow. TR provides a relatively stable and strong eastward steering wind to hinder Megi’s westward motion. As the retraction of SH and CH after the break of the high pressure belt, the westward steering flow from them (which acts to drive Megi westward) reduces significantly and turns clockwise slightly; thus, contributing to the deceleration of Megi’s westward motion and acceleration of its northward motion.

The PV diagnosis shows that these midlatitude systems play a dominant role in affecting Megi’s movement; in particular the break in the high pressure belt that provided a favorable condition for the northward turn of the westward-moving Megi. It is noted, however, that although the total effect of SH, CH, and TR is in general agreement with the actual steering flow, it fails to capture the northeastward steering of Megi after its deflection on 20 and 21 October. This will be discussed in section 6.

b. Numerical study

1) Experiment design

To further examine the relative steering effect from SH, CH, and TR on Megi’s abrupt track deflection, a series of 120-h integration experiments (with specific atmospheric features disturbed initially) are initialized at 0000 UTC 17 October. Based on the model configuration listed in section 2, the control experiment is designed with the model’s initial conditions taken from GFS 0.5° × 0.5° data at every 3 h, while the sensitivity experiments are a series of perturbed simulations in which the initial conditions are modified by locally balanced perturbations of the individual systems. The filtering technique introduced in section 2 is used to achieve the local initial perturbations, and the extent of the perturbation is controlled by the disturbance coefficient in Eq. (8). Three groups of sensitivity experiments, named SSH, SCH, and STR, are designed by initially perturbing these three individual targets: SH, CH, and the TR. The coefficient for SH, CH, or TR is distinguished using letters S, C, or T, respectively, and is set to 0, +1, +0.5, −0.5, and −1 in each group. According to the value of the coefficient, the experiments are denoted as, for example, SCH+1, SCH−1, SCH+0.5, and SCH−0.5 in the group of SCH (please see the detailed description of the experiment configuration in Table 2).

Table 2.

The description of the control and the sensitivity experiments on midlatitude systems.

Table 2.

2) Results

The simulated track in the control (Fig. 5) generally mimics the best track. The fast westward movement and the sudden northeastward track deflection agree well with the observed best track, in particular the location (about 17°N, 117°E) and the time (0000 UTC 20 October 2010) of the deflection. Although it cannot be denied that there is a sharper northward turn in the control compared with the observed track, the simulated track in the control still provides useful information for examining the physical processes that caused Megi’s track deflection, and serves as a reference for analyzing the results of the sensitivity experiments.

Fig. 5.
Fig. 5.

The simulated tracks in the control (solid circles) at 3-h intervals and the best track (typhoon symbol) from 0000 UTC 17 Oct to 22 Oct 2010.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

The perturbations of local systems the sensitivity experiments are shown in the left panels of Fig. 6, in which notable changes are seen to appear in CH, TR, and the northern part of SH, while a slighter change appears in the southern part of SH. Because of the initial perturbation of atmospheric features, the simulated tracks differ from the control in various degrees (right panels of Fig. 6). Interestingly, it is evident that Megi’s motion is most sensitive to the initial perturbation of CH, but less sensitive to perturbations of SH and TR, which appears inconsistent with the powerful steering effect associated with SH and TR. What causes the simulated results? Discussion related to each group of the sensitivity experiments is conducted as follows.

Fig. 6.
Fig. 6.

The sketch map of the initial perturbations valid 0000 UTC 17 Oct 2010 in (left) a single system and (right) the perturbed simulated tracks in (a) SCH, (b) SSH, and (c) STR, respectively. Both the Pacific subtropical high (SH) and continental high (CH) are represented by a 5880-gpm geopotential height contour at 500 hPa, while the midlatitude trough (TR) is represented by a 5400-gpm geopotential height contour. The contours of different styles in each sensitivity experiment groups stand for the perturbed simulations with different values of the disturbance coefficient.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

The PV diagnosis verifies that the weakening of the westward steering flow and the strengthening of the northward flow produced by the break in the high pressure belt provides the favorable condition for the track deflection. However, despite the strong steering flow provided by SH throughout its entire existence, the track of Megi is less sensitive to its initial perturbation (the second row of Fig. 6), and this can be attributed to the practical stability of SH itself (as evident from the inconspicuous initial disturbance, particularly in its southern part). In other words, the changes in a reasonable degree of SH’s intensity have only a relatively small influence on the strength of the high pressure belt, and it is therefore unable to produce significant changes in Megi’s track. In contrast, although CH supplies a relatively weak steering, a wide spread of the simulated tracks (the first row of Fig. 6) is induced by CH perturbations. To the left of the best track, the strengthening of CH (SCH+1) produces a track that gradually deflects, with a southwestward bias of the deflection position, a smaller turning angle, and a northwestward motion after deflection. The opposite situation appears when CH initially weakens (SCH−1).

Figure 7 illustrates the DLM (850–250 hPa) wind, geopotential height field (Figs. 7a,c), and steering flow in SCH+1 and SCH−1 (Figs. 7b,d; the shaded zones highlight the steering flow associated with CH). As CH is strengthened, the high pressure belt grows stronger and more stable (left panels in Figs. 7a,b), providing a stronger southwestward steering effect upon Megi, thus inducing a quicker southwestward motion of Megi in SCH+1 than in the control. In contrast, the weakened CH acts to decelerate Megi toward the southwest (right panels in Figs. 7a,b). Thus, Megi travels farther southwest in SCH+1 than in the control and SCH−1. The high pressure belt in SCH−1 begins to break at 2100 UTC 18 October, almost 12 h before that in SCH+1 (figures are omitted). As the high pressure belt cutoff at 0600 UTC 19 October (Figs. 7c,d) in SCH−1, Megi deflects northeastward at the point when it loses the westward steering wind. However, the long-lasting high pressure belt in SCH+1 remains, providing a westward steering effect and leading to a gentle shift from the southwestward to the northwestward motion. As a result, CH plays the most important role in determining the timing and location of the breaking of the belt, which is critical to the sudden deflection of Megi.

Fig. 7.
Fig. 7.

The DLM (850–250 hPa) wind, geopotential height [(a),(c) the areas with values larger than 5290 gpm are shaded], and steering flow of corresponding steering layers [(b),(d) the areas with values larger than 5 m s−1 associated with CH are shaded] at (a),(b) 0600 UTC 17 Oct and (c),(d) 0600 UTC 19 Oct 2010 in (left) SCH+1 and (right) SCH−1. One full wind barb represents 10 m s−1.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

The initial perturbations of TR lead only to a slight variation in the simulated tracks (the third row of Fig. 6). The passing TR to the north of the high pressure belt is able to transport a positive relative vorticity to the high pressure belt, thus weakening it, but the changes in TR’s initial intensity do not bring a significant influence on the strength of the high pressure belt; hence, producing slight effect on the simulated track. This indicates that TR may be not the sole factor causing the break in the high pressure belt. It is possible that certain other factors, such as the effect from the strong Megi, may also contribute. Therefore, the effect of TR on Megi’s abnormal track is limited, and it cannot be regarded as a critical factor.

All in all, the results in this section show the crucial influence of the midlatitude circulation on Megi’s motion, which is similar to the results of Kieu et al. (2012). In addition, in comparison with SH and TR, CH is the most important feature on the midlatitude circulation to determine Megi’s deflection. We can deduce that it is unreasonable to attribute the formation of the abnormal track to systems that provide a strong steering effect on typhoon motion, and more cases are necessary to validate the universal and practical significance of this conclusion.

6. The effect of low-latitude systems on Megi

a. PV diagnosis

As shown in Fig. 4a, the summation of , , and is easterly all along, which is different from the actual southwesterly steering flow and Megi’s northwestward motion after its deflection. It is, therefore, considered that an eastward steering from another dynamical system is missing from our calculations. We therefore consider whether a tropical system provides the missing eastward steering, and take the PV perturbation of the small closed anticyclone (SA) on 20 and 21 October (since it formed after 19 October) into consideration alone. The balanced flow associated with the PV perturbation of SA presents an anticyclonic circulation southeast of Megi, thereby providing the southwesterly wind (Fig. 4b). The DLM steering flow over Megi from SA [] are west-southwesterly winds with speeds of around 3 and 3.5 m s−1 on 20 and 21 October, respectively (Fig. 4b).

To highlight the steering effect of SA on the motion of Megi, the time series of together with on 20 and 21 October are shown in Fig. 4c. Compared with the time series of the steering flow in Fig. 4a, in Fig. 4c becomes a northeastward steering flow on 20 and 21 October by encompassing the eastward flow from SA, which is perfectly consistent with Megi’s motion direction. The result strongly suggests that the strengthening of the northeastward steering flow, caused by the development of SA, is important in affecting the motion of Megi after its deflection.

Interestingly, the configuration of these dynamical systems surrounding Megi is analogous to the indirect cyclone interaction model put forward by Carr and Elsberry (1998, known as ITI), and (Carr and Elsberry 2000a, known as ICIW), with SA corresponding to the peripheral anticyclone and EW to the eastern cyclone. Based on an idealized study employing a shallow-water model, Luo et al. (2011) verified that the anticyclone southeast of the TC can lead to a sudden track shift. Both Carr and Elsberry (1998, 2000a) and Luo et al. (2011), in addition to other related studies, proposed that the peripheral anticyclone is the result of TC energy dispersion (TCED), which may lead to a poleward track deflection, and the cyclone to its east can inhibit the development of the peripheral anticyclone.

What is the mechanism responsible for the development of SA? A brief discussion is provided here. Megi’s Rossby wave train is examined based on the synoptic-scale field filtered through the same filtering technique as used in the study of Li and Fu (2006). On the sea surface, an apparent wave train pattern is discerned in the synoptic-scale surface wind field (Fig. 8a) at 1800 UTC 19 October, manifesting the existence of the Rossby wave energy dispersion of Megi. The circulation of Megi penetrates through the entire tropospheric layer and emits energy dispersion on the upper level, inducing a wave train with alternate anticyclone and cyclone at 500 hPa at 1800 UTC 19 October (Fig. 8d). Notice that SA is a tropical system distinguished on the steering flow field (Figs. 2c,d). As shown in the vertical cross section along 12°N of the steering flow relative vorticity (Fig. 8e) at 1800 UTC 19 October, SA and EW are located alternately to the east of Megi and centered between 300 and 500 hPa. Comparing the evolution and structure of the upper-level anticyclone with SA, it is easy to deduce that the upper-level anticyclone is consistent with the circulation of SA. However, different from the equivalent barotropic vertical structure of a usual TCED-induced wave train (see Fig. 7b in Li and Fu 2006), there is an apparent phase difference between the upper-level wave train and the low-level one. From Fig. 8f, the upper-level centers of the anticyclone (labeled as “A”) and the cyclone (labeled as “C”) are seen to have a significant westward bias compared with the lower-level ones. Therefore, we deduce that the distinct vertical train structure, which is closely related to the development of SA, is affected by some other factors, and is not only subjected to TCED.

Fig. 8.
Fig. 8.

The synoptic-scale (a) surface wind field at 1800 UTC 19 Oct 2010 and 500-hPa wind field at (b) 0000 UTC 16 Oct, (c) 0000 UTC 17 Oct, and (d) 1800 UTC 19 Oct 2010. The typhoon mark represents the center location of Megi. The vertical–latitudinal cross section of relative vorticity (10−5 s−1) along (e) 12°N and (f) the Rossby wave train axis is labeled by the solid line in (a). The y axis denotes pressure (hPa). The solid circle in (f) represents the position of Typhoon Megi. The letters “A” and “C” represent the centers of the anticyclone and cyclone, respectively.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

By an analysis of the upper-level field evolution, we attribute the phase difference to the effect of the EW south of SH (located around 20°N, 150°E), a system initially independent to the wave train (Fig. 8b). EW moves westward, extends longitudinally, and deforms the southern part of SH into an inverted V shape, resulting in the anticyclonic circulation extending to the south of 10°N between EW and Megi (Fig. 8c). The TCED-induced anticyclonic circulation combines with the southward extending SH, creating the closed anticyclone, SA (Fig. 8d). As a result, the original upper-level TCED-induced wave train pattern is disturbed by EW and replaced by the wave train compose of SA and EW. Therefore, SA can be regarded as the outcome of the TCED and the interaction between EW and SH. The formation mechanism of SA causes it to be distinguished from the peripheral anticyclone in previous studies. As this study focuses on the contribution of environmental systems, only the effect of EW is considered here.

b. Numerical study

As the critical eastward steering flow associated with SA is demonstrated in the PV diagnosis, we consider what extent the steering effect of SA has on the influence of Megi’s track, and use a group of sensitivity experiments to elucidate this. Considering that SA was formed much later than the initial time of the simulation at 0000 UTC 17 October 2010, the initial perturbation is created in EW to reflect the impact of SA, since EW is closely related to the development of SA (Fig. 9, left), and a detailed description of sensitivity experiments with EW initially perturbed (SEW) is shown in Table 3, where letter “E” refers to the disturbance coefficient for EW.

Fig. 9.
Fig. 9.

The sketch map of the initial perturbations valid 0000 UTC 17 Oct 2010 in (left) a single system and (right) the perturbed simulated tracks in SEW. The easterly wave (EW) is represented by a 5870-gpm geopotential contour. The contours of different styles in SEW stand for the perturbed simulations with different values of the disturbance coefficient.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

Table 3.

The description of the sensitivity experiments on low-latitude systems.

Table 3.

As shown in Fig. 9b, a wide spread of the simulated tracks after deflection results from the perturbations of EW, particularly when it is weakened. To examine whether the track variation is a result of the change of SA, the circulation patterns of SEW+1 and SEW−1 are compared in Figs. 10a,b. The strengthened EW in SEW+1 creates a more serious distortion on the south boundary of SH, where it enhances the geopotential gradient and flow, resulting in a more apparent southward extension of anticyclonic circulation. This favorable initial condition induces a well-developed SA southeast of Megi at 0000 UTC 20 October, and provides stronger southwesterly winds (Figs. 10c,d). Accompanied by the more mature SA, the serious shrinking of the main body of SH results in a significant reduction of the northwestward steering associated with it. In contrast, the weakened EW in SEW−1 creates a less developed SA, leading to a weaker southwesterly wind and a stronger remaining northwestward flow from SH. Therefore, the net effect of the weakened (strengthened) SA is to reduce (enhance) the eastward-steering effect. As demonstrated by Fig. 9b, the weakened SA (determined by the initial weaker EW) produces a westward track bias, and a much smaller direction change of less than 90° compared with the control, while the extent of the track change is inconspicuous between SEW+1 and the control.

Fig. 10.
Fig. 10.

The DLM (850–250 hPa) wind, geopotential height [(a),(c) the areas with values larger than 5290 gpm are shaded], and steering flow of corresponding steering layers [(b),(d) the areas with values larger than 5 m s−1 are shaded] at (a),(b) 0900 UTC 17 Oct and (c),(d) 0000 UTC 20 Oct 2010 in (left) SEW+1 and (right) SEW−1. One full wind barb represents 10 m s−1.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

In addition, the simulated results also help to verify that the development of SA is closely related to the strength of EW via the interaction between SH and EW. In all, the low-latitude mechanism, consisting of the direct effect of SA and the indirect effect of EW, plays an important role in influencing the track at the point when Megi deflected.

7. The combined effect of midlatitude and low-latitude systems on Megi

To conduct a thorough examination on the physical processes responsible for Megi’s abnormal track, it is necessary to conduct additional studies on the combined effect of midlatitude and low-latitude mechanisms.

a. Combined sensitivity experiments design

Based on the results of the single perturbed simulations, 10 combined sensitivity (COMB) experiments were designed by perturbing several atmospheric features simultaneously, in order to reveal their joint effects on Megi’s motion. Detailed configurations of the perturbations of different atmospheric features for each experiment are shown in Table 4, in which the meaning of the labels is similar to those in Tables 2 and 3.

Table 4.

The configuration of the system perturbations in the combined sensitivity (COMB) experiments.

Table 4.

b. Results

Figure 11 presents the simulated tracks of ten COMB experiments. It is evident that compared with the control and other COMBs the simulations in COMB3 (+1C−1E−1S) and COMB4 (+1C−1E+1S) exhibit normal paths with a gradual track shift of a smaller change in the direction and acceleration of a northwestward motion after deflection. In addition, Megi makes landfall on mainland China in COMB4. The variation in the simulated tracks indicates that simultaneous initial perturbations of several systems can create conditions that are unfavorable for sudden track change, particularly in COMB3 and COMB4. Therefore, the steering factors that were changed in COMB3 and COMB4 play the critical role in determining when and where Megi began to deflect.

Fig. 11.
Fig. 11.

The simulated results of the 10 combined experiments at 6-h intervals.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

Analysis of a comparison of the perturbations configurations in COMB3 and COMB4 shows that they were all accompanied by a simultaneous strengthening of CH (C = +1) and a weakening of EW (E = −1). The simulations in the control, SCH+1, SEW−1, COMB3, and COMB4 were then put together for a comparative study (Fig. 12), and the instantaneous motion vectors during deflection were computed in each experiment (Fig. 13). Significant differences are seen to exist between these simulated tracks, particularly during the period of deflection. All the tracks take a similar northwestward direction prior to the deflection (dashed arrows in Fig. 13), but show great differences in speed, leading to significant variation in the location where the track shift occurs. The largest speed (10.06 km h−1) is presented in COMB4, which is associated with a stronger CH and SH and a weaker EW; followed by the SCH+1 and COMB3; and the slowest speed is found in the control and SEW−1. This indicates the strong westward-steering effect supplied by CH, the considerable westward steering from SH (which generates a difference of about 1 km h−1 between COMB3 and COMB4), and the relatively small eastward steering from EW on Megi. These results are consistent with the previous conclusion that the predominant westward-steering flow comes from the midlatitude system.

Fig. 12.
Fig. 12.

The simulated results of the control, COMB3, COMB4, SCH+1, and SEW−1 at 6-h intervals.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

Fig. 13.
Fig. 13.

The motion vectors (km h−1) of Megi before (dashed arrows) and after deflection (solid arrows) in the control, SCH+1, SEW−1, COMB3, and COMB4, with the translation speed labeled at the tips of the arrows. The solid points show the location of deflection.

Citation: Monthly Weather Review 142, 7; 10.1175/MWR-D-13-00283.1

When Megi’s track deflects (solid arrows in Fig. 13), the differences between the simulations are quite significant, not only in direction but also in speed. The change of the moving direction in the control exceeds 90° with the track turning from northwestward to northeastward, while the amplitude of the track change is almost halved for the single weakening EW or strengthening CH; and it further decreases under the combined effect of the weakened EW and strengthened CH. In terms of speed, the acceleration effect from the stronger CH and SH overwhelms the deceleration effect from EW, although the speed is sensitive to both of them.

In summary, the abnormal track is only transformed to a normal one when the CH and EW are perturbed simultaneously, although separate systems can affect the track to various extents. In other words, the combined effect of CH and the low-latitude circulation (the direct steering effect from SA influenced by EW) determines the formation of the sharp deflection track of Megi, particularly in relation to the time, location, and the amplitude of the deflection, which is critical for the precision of track forecast. Therefore, it is necessary to consider the midlatitude and low-latitude mechanisms as one indivisible whole, if track forecasts are to be improved in similar typhoon cases. In particular, more emphasis needs to be placed on the small system that provides a relatively weak steering effect, owing to its underlying interaction with other systems.

8. Discussion and conclusions

The mechanism of the abnormal track deflection in the WNP of Supertyphoon Megi (2010) is examined using several effective methods such as the PV diagnosis and numerical simulation. The PV diagnosis attributes the northward track shift of Megi mainly to the effect of the midlatitude circulation. Initially, the Pacific subtropical high (SH) and the continental high (CH) joined together as a high pressure belt, providing a strong easterly wind to maintain Megi’s westward motion, while the midlatitude trough (TR) provided a stable and strong westerly wind that decelerated the westward motion. Owing to the significant reduction of the easterly wind component, and the strengthening of the northward wind component caused by the break in the high pressure belt, the midlatitude circulation provided a favorable condition for the track deflection.

A series of numerical simulations are then conducted to compare their effects in steering Megi. Although SH and TR contributed greatly to the steering flow, their initial perturbations within a reasonable range could not produce any significant track variation. However, the relatively weak system, CH, played an important role in affecting the timing, location, and amplitude of the track deflection, because of its critical status in maintaining the high pressure belt.

However, the lack of an eastward component compared with Megi’s actual motion suggests some mechanism that has not been accounted for in the analysis. At tropical latitudes, a small closed anticyclonic circulation (SA) formed southeast of Megi about 10 h before track deflection, and was responsible for the eastward advection. Although a northeastward wind of only 3.5 m s−1 was provided by SA, it was found to play a critical role in transforming the total steering flow from northwestward to northeastward. The results of sensitivity experiments show that SA was critical in steering Megi in a northwestward direction after deflection, although it did not determine whether or not the track deflected.

Previous studies have shown a similar influence of an analogous peripheral anticyclone on the poleward steering of typhoon motion. However, the low-latitude steering associated with SA produced a different impact in terms of its formation process and its interaction with the cyclone to its east. SA was the outcome of the TCED and the easterly wave (EW), and by interacting with SH, EW promoted the formation of SA rather than precluded it. Based on the sensitivity experiment SEW, the apparent variation in the simulated tracks associated with an initial perturbed EW validates the positive effect of EW on the development of SA.

Further combined sensitivity studies with simultaneous perturbations of two or three systems are also conducted to investigate the combined effect of these crucial atmospheric features. These results indicate that no matter how much the effect of a single system can have on the track, it is the combined effect of CH and SA (due to the vigorous development of EW) that determined the abnormal track of Megi, particularly in relation to the timing, location, and amplitude of the track deflection. The conclusion confirms that more emphasis needs to be placed on the small-scale systems of weak, individual steering flows (such as SA in this study), because of the possible substantial influence on typhoon motion via subtle interactions with other systems. It is believed that the combined effect of the midlatitude and low-latitude systems highlighted in this study will be helpful for accurately and comprehensively capturing the processes responsible for similar abnormal tracks, and will thereby improve the track forecast accuracy of typhoons.

The results in this study are obtained using only the analysis of Megi; it is thus considered that more cases are necessary to ensure validation. Although the individual effects of the important systems CH and SA, and their critical combined effect on the abrupt track deflection have been explicitly demonstrated, further in-depth study on their evolution is still necessary, and will be discussed elsewhere. For example, the retraction of CH is a complicated process that may have been related to the influence of TR and the strength of Megi, and although the formation of SA is found to be associated to both TCED and EW, more systematic proofs and discussions would still be valuable. In addition, the initial strengths of CH and EW, which are capable of inducing the track deflection, have not been quantitatively defined in this paper. Successive works that will be of great importance in improving track forecasting are therefore necessary.

Acknowledgments

This work is supported by the National Public Benefit (Meteorology) Research Foundation of China Grant GYHY201106004, and the National Natural Science Foundation of China Grants 41005029, 41105065, and 41230421. We thank the editor, Patrick A. Harr, and three anonymous reviewers for their helpful comments and suggestions.

REFERENCES

  • Berger, H., , R. Langland, , C. S. Velden, , C. A. Reynolds, , and P. M. Pauley, 2011: Impact of enhanced satellite-derived atmospheric motion vector observations on numerical tropical cyclone track forecasts in the western North Pacific during TPARC/TCS-08. J. Appl. Meteor. Climatol., 50, 23092318, doi:10.1175/JAMC-D-11-019.1.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 1990: Observational evidence for predictions of tropical cyclone propagation relative to environmental steering. J. Atmos. Sci., 47, 542548, doi:10.1175/1520-0469(1990)047<0542:OEFPOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 1998: Objective diagnosis of binary tropical cyclone interactions for the western North Pacific basin. Mon. Wea. Rev., 126, 17341740, doi:10.1175/1520-0493(1998)126<1734:ODOBTC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 2000a: Dynamical tropical cyclone track forecast errors. Part I: Tropical region error sources. Wea. Forecasting, 15, 641661, doi:10.1175/1520-0434(2000)015<0641:DTCTFE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , and R. L. Elsberry, 2000b: Dynamical tropical cyclone track forecast errors. Part II: Midlatitude circulation influences. Wea. Forecasting, 15, 662681, doi:10.1175/1520-0434(2000)015<0662:DTCTFE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carr, L. E., , M. A. Boothe, , and R. L. Elsberry, 1997: Observational evidence for alternate modes of track-altering binary tropical cyclone scenarios. Mon. Wea. Rev., 125, 20942111, doi:10.1175/1520-0493(1997)125<2094:OEFAMO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Chan, J. C., , and W. Gray, 1982: Tropical cyclone movement and surrounding flow relationships. Mon. Wea. Rev., 110, 13541374, doi:10.1175/1520-0493(1982)110<1354:TCMASF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Charney, J. G., 1955: The use of primitive equations of motion in numerical prediction. Tellus, 7, 2226, doi:10.1111/j.2153-3490.1955.tb01138.x.

    • Search Google Scholar
    • Export Citation
  • Davis, C. A., , and K. A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev.,119, 1925–1953, doi:10.1175/1520-0493(1991)119<1929:PVDOC>2.0.CO;2.

  • Dong, K. Q., , G. J. Holland, , and B. C. Diehl, 1991: Typhoon recurvature and environmental circulation features. WMO Tech. Doc. WMO/TD-472, World Meteorological Organization, Geneva, Switzerland, 16–22.

  • Dudhia, J., 1989: Numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107, doi:10.1175/1520-0469(1989)046<3077:NSOCOD>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Elsberry, R. L., 1995: Tropical cyclone motion. Global Perspectives on Tropical Cyclones, R. L. Elsberry, Ed., WMO Tech. Doc. WMO/TD-693, World Meteorological Organization, Geneva, Switzerland, 106–197.

  • Galarneau, T. J., , and C. A. Davis, 2013: Diagnosing forecast errors in tropical cyclone motion. Mon. Wea. Rev., 141, 405–430, doi:10.1175/MWR-D-12-00071.1.

    • Search Google Scholar
    • Export Citation
  • Harr, P. A., , and R. L. Elsberry, 1991: Tropical cyclone track characteristics as a function of large-scale circulation anomalies. Mon. Wea. Rev., 119, 14481468, doi:10.1175/1520-0493(1991)119<1448:TCTCAA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Harr, P. A., , and R. L. Elsberry, 1995: Large-scale circulation variability over the tropical western North Pacific. Part I: Spatial patterns and tropical cyclone characteristics. Mon. Wea. Rev., 123, 12251246, doi:10.1175/1520-0493(1995)123<1225:LSCVOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., , Y. Noh, , and J. Dudhia, 2006: A new vertical di¤usion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341, doi:10.1175/MWR3199.1.

    • Search Google Scholar
    • Export Citation
  • Japan Meteorological Agency, 2012: Annual Report on the Activities of the RSMC Tokyo-Typhoon Center 2011. Japan Meteorological Agency, 82 pp. [Available online at http://www.jma.go.jp/jma/jma-eng/jma-center/rsmc-hp-pub-eg/AnnualReport/2011/Text/Text2011.pdf.]

  • Jian, G.-J., , and C.-C. Wu, 2008: A numerical study of the track deflection of Supertyphoon Haitang (2005). Mon. Wea. Rev., 136, 598615, doi:10.1175/2007MWR2134.1.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181, doi:10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kehoe, R. M., , M. A. Boothe, , and R. L. Elsberry, 2007: Dynamical tropical cyclone 96- and 120-h track forecast errors in the western North Pacific. Wea. Forecasting, 22, 520538, doi:10.1175/WAF1002.1.

    • Search Google Scholar
    • Export Citation
  • Kieu, C., , T. Nguyen, , M. Hoang, , 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, 17941810, doi:10.1175/JTECH-D-12-00020.1.

    • Search Google Scholar
    • Export Citation
  • Komaromi, W. A., , S. J. Majumdar, , and E. D. Rappin, 2011: Diagnosing initial condition sensitivity of Typhoon Sinlaku (2008) and Hurricane Ike (2008). Mon. Wea. Rev., 139, 32243242, doi:10.1175/MWR-D-10-05018.1.

    • Search Google Scholar
    • Export Citation
  • Kurihara, Y., , M. A. Bender, , R. E. Tuleya, , and R. J. Ross, 1990: Prediction experiments of Hurricane Gloria (1985) using a multiply nested movable mesh model. Mon. Wea. Rev., 118, 21852198, doi:10.1175/1520-0493(1990)118<2185:PEOHGU>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kurihara, Y., , M. A. Bender, , and R. J. Ross, 1993: An initialization scheme of hurricane models by vortex specification. Mon. Wea. Rev., 121, 20302045, doi:10.1175/1520-0493(1993)121<2030:AISOHM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Lander, M. A., 1996: Specific tropical cyclone track types and unusual tropical cyclone motions associated with a reverse-oriented monsoon trough in the western North Pacific. Wea. Forecasting, 11, 170186, doi:10.1175/1520-0434(1996)011<0170:STCTTA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Li, T., , and B. Fu, 2006: Tropical cyclogenesis associated with Rossby wave energy dispersion of a preexisting typhoon. Part I: Satellite data analyses. J. Atmos. Sci., 63, 13771389, doi:10.1175/JAS3692.1.

    • Search Google Scholar
    • Export Citation
  • Luo, Z. X., , N. E. Davidson, , F. Ping, , and Weican Zhou, 2011: Multiple-scale interactions affecting tropical cyclone track change. Adv. Mech. Eng., 2011, 19, doi:10.1155/2011/782590.

    • Search Google Scholar
    • Export Citation
  • 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 66316 682, doi:10.1029/97JD00237.

    • Search Google Scholar
    • Export Citation
  • Neumann, C. J., 1992: Final report: Joint Typhoon Warning Center (JTWC92) model. SAIC Contract Rep. N00014-90-C-6042 (Part 2), 83 pp.

  • Patla, J. E., , D. Stevens, , and G. M. Barnes, 2009: A conceptual model for the influence of TUTT cells on tropical cyclone motion in the Northwest Pacific Ocean. Wea. Forecasting, 24, 12151235, doi:10.1175/2009WAF2222181.1.

    • Search Google Scholar
    • Export Citation
  • Pike, A. C., 1985: Geopotential heights and thicknesses as predictors of Atlantic tropical cyclone motion and intensity. Mon. Wea. Rev., 113, 931940, doi:10.1175/1520-0493(1985)113<0931:GHATAP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Rappaport, E. N., and Coauthors, 2009: Advances and challenges at the National Hurricane Center. Wea. Forecasting, 24, 395419, doi:10.1175/2008WAF2222128.1.

    • Search Google Scholar
    • Export Citation
  • Riehl, H., , and R. J. Shafer, 1944: The recurvature of tropical storms. J. Meteor., 1, 4254, doi:10.1175/1520-0469(1944)001<0001:TROTS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Thompson, G., , P. R. Field, , R. M. Rasmussen, , and W. D. Hall, 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Mon. Wea. Rev., 136, 50955115, doi:10.1175/2008MWR2387.1.

    • Search Google Scholar
    • Export Citation
  • Torn, R. D., , and C. A. Davis, 2012: The influence of shallow convection on tropical cyclone track forecasts. Mon. Wea. Rev., 140, 21882197, doi:10.1175/MWR-D-11-00246.1.

    • Search Google Scholar
    • Export Citation
  • Velden, C. S., , and L. M. Leslie, 1991: The basic relationship between tropical cyclone intensity and the depth of the environmental steering layer in the Australian region. Wea. Forecasting, 6, 244253, doi:10.1175/1520-0434(1991)006<0244:TBRBTC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-C., , and K. A. Emanuel, 1995a: Potential vorticity diagnostics of hurricane movement. Part I: A case study of Hurricane Bob (1991). Mon. Wea. Rev., 123, 6992, doi:10.1175/1520-0493(1995)123<0069:PVDOHM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-C., , and K. A. Emanuel, 1995b: Potential vorticity diagnostics of hurricane movement. Part II: Tropical Storm Ana (1991) and Hurricane Andrew (1992). Mon. Wea. Rev., 123, 93109, doi:10.1175/1520-0493(1995)123<0093:PVDOHM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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, 12891300, doi:10.1175/1520-0493(2003)131<1289:ANLATB>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • 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, 20842093, doi:10.1175/1520-0493(2004)132<2084:PVDOTK>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wu, C.-C., , S. G. Chen, , C. C. Yang, , P. H. Lin, , and S. D. Aberson, 2012: Potential vorticity diagnosis of the factors affecting the track of Typhoon Sinlaku (2008) and the impact from dropwindsonde data during T-PARC. Mon. Wea. Rev., 140, 26702688, doi:10.1175/MWR-D-11-00229.1.

    • Search Google Scholar
    • Export Citation
  • Wu, L., , H. Zong, , and J. Liang, 2011: Observational analysis of sudden tropical cyclone track changes in the vicinity of the East China Sea. J. Atmos. Sci., 68, 30123031, doi:10.1175/2010JAS3559.1.

    • Search Google Scholar
    • Export Citation
Save