Topographic Influence on the Motion of Tropical Cyclones Landfalling on the Coast of China

OCTOBER 2016 C H E N A N D W U 1615

Topographic Inﬂuence on the Motion of Tropical Cyclones Landfalling on the Coast of China

XIAOYU CHEN Key Laboratory of Meteorological Disaster, Ministry of Education/Joint International Research Laboratory of Climate and Environment Change/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/Paciﬁc Typhoon Research Center, Nanjing University of Information Science and Technology, Nanjing, China

LIGUANG WU Key Laboratory of Meteorological Disaster, Ministry of Education/Joint International Research Laboratory of Climate and Environment Change/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/Paciﬁc Typhoon Research Center, Nanjing University of Information Science and Technology, Nanjing, and State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing, China

(Manuscript received 17 March 2016, in ﬁnal form 10 August 2016)

ABSTRACT

A few recent numerical studies investigated the inﬂuence of a continent, which is much larger in size than islands, on the track of a landfalling tropical cyclone. It has been found that land surface roughness and temperature tend to make a tropical cyclone move toward the continent. In this study, the effect of the elevated terrain on the western part of mainland China on landfall tracks is examined through idealized numerical experiments, in which an initially axisymmetric vortex is embedded into a monsoon trough and takes a landfall track across China’s mainland. It is found that the effect of the elevated terrain on the western part of mainland China can enhance the southerly environmental ﬂows in the low and midtroposphere, leading to an additional landward component of tropical cyclone motion, suggesting that forecasts for tropical cyclone tracks over land should take into consideration the change in thermal condition over the continent. The study suggests that the increased duration of tropical cyclone overland tracks observed in China may be associated with the reduction of the thermal forcing over the Tibetan Plateau.

1. Introduction roughness and temperature can affect landfalling TC motion (Wong and Chan 2006, 2007; Au-Yeung and Economic damage and casualties associated with Chan 2010; Szeto and Chan 2010; Li et al. 2014; Xu et al. tropical cyclones (TCs) occur mostly during and after 2013). So far, few studies have examined the influence of landfall (Zhang et al. 2009, 2010) and thus it is very im- continental terrain on landfalling TC motion. portant to predict the timing and location of landfall in Early studies paid much attention to TC track changes operational TC forecasts. While many studies have fo- associated with topography. Idealized numerical and cused on the influence of islands on TC activity (e.g., real case studies suggest that island topography like Chang 1982; Bender et al. 1987; Yeh and Elsberry 1993a, that of Taiwan can significantly affect environmental b; Lin et al. 2005; Jian and Wu 2008; Yang et al. 2008; flows and TC tracks during landfall, leading to track Huang et al. 2011; Wang et al. 2012; Xie and Zhang 2012), deflection or looping tracks (Chang 1982; Yeh and attention has been paid recently to the effects of conti- Elsberry 1993a,b; Wu 2001; Jian and Wu 2008; Wang nents that are much larger than the TC circulation on the et al. 2012; Xie and Zhang 2012). For instance, it is horizontal scale. It has been found that land surface suggested that the unusual tracks of Typhoons Nari (2001) and Krosa (2007) were closely related to the terrain height of the island of Taiwan (Yang et al. 2008; Corresponding author address: Prof. Liguang Wu, Pacific Ty- phoon Research Center, Nanjing University of Information Sci- Huang et al. 2011). Additionally, the slowdown of Ty- ence and Technology, Nanjing 210044, China. phoon Morakot (2009) when crossing Taiwan was mainly E-mail: [email protected] attributed to the asymmetric latent heating induced by the

DOI: 10.1175/WAF-D-16-0053.1

Ó 2016 American Meteorological Society Unauthenticated | Downloaded 09/28/21 07:44 PM UTC 1616 WEATHER AND FORECASTING VOLUME 31 topography and monsoonal flows (Liang et al. 2011; Wu et al. 2011; Wang et al. 2012). It is indicated that the island topographic effect can affect lower-level flows and induce significant asymmetric diabatic heating that will result in TC track changes (Yeh and Elsberry 1993a,b; Wang et al. 2012). A few numerical studies found that TC asymmetric flows induced by land surface roughness can lead to a landward component of TC motion during landfall in a resting environment (Wong and Chan 2006, 2007; Au- Yeung and Chan 2010; Szeto and Chan 2010; Li et al. 2014). Wong and Chan (2006) for the first time performed idealized numerical experiments that investigated the effects of land surface roughness and moisture flux on TC motion. They found that the TC can drift toward the 2 continent at a speed of about 1 m s 1 in a resting environment on an f plane as a result of the induced TC asymmetric flows at lower levels; this process is now called land drift. Land drift was further demonstrated in the idealized numerical experiments with a triangle- FIG. 1. Domains for the numerical experiments and the initial 2 shaped coastal delta on the f plane and the numerical background flow (m s 1) at 850 hPa. The terrain height (m) is experiments on the b plane (Au-Yeung and Chan 2010; shaded and the TC track in OCN is plotted. Szeto and Chan 2010). Uniform environmental flows were added in numerical experiments and the land drift work differs from the previous numerical studies in the was also found by Li et al. (2014).However,wenotice following aspects. First, we use a surface roughness that the surface roughness length was set to be 0.5 m in length that is much smaller than the land drift experi- these land drift experiments, which is much larger than ments; this approach is more realistic for eastern main- those for common land surface conditions of eastern land China. Second, we use a relatively realistic China when using the Advanced Research version of the environmental flow including the typical monsoon Weather Research and Forecasting (WRF) Model trough and subtropical high over the western North (ARW). However, the influences of environmental flows Pacific for the embedded TC, which was derived from and continental topography have not been fully in- the low-frequency fields of Typhoon Matsa (2005). Fi- vestigated in previous studies. nally, the influence of the continental-scale terrain is Xu et al. (2013) conducted a 60-yr (1950–2009) sur- included. vey on TCs that made landfall on China’s mainland and suggested that supertyphoons (maximum sustained 2 2. Experiment design 1-min surface winds above 65 m s 1) tend to take a track toward the warmer part of the mainland. Based on The numerical experiments conducted in this study their observational analysis, Xu et al. (2013) further include five two-way interactive domains using the WRF performed a set of numerical experiments to show the Model (version 2.2). The horizontal resolutions are 27, 9, influence of land surface temperature on landfalling TC 3, 1, and 1/3 km, respectively, with 40 levels in the vertical tracks. In their experiments, they modified the land and a top of 50 hPa. The fifth domain, with subkilometer surface temperature and found that the simulated ty- resolution, is used for better simulation of the inner-core phoon gravitates to the region with higher land surface structure during TC landfall (Moon and Nolan 2015). In temperature. They proposed that the warmer land sur- fact, a sensitivity experiment was conducted without the face, with intense moisture supply from the summer subkilometer domain. While the simulated tracks are Asian monsoon, is conducive to deep convection that very similar, the intensity evolution is different as a result attracts supertyphoons. of the different horizontal resolutions. Two eyewall re- In this study, we also examine the influence of the placement processes occur in the subkilometer run. To- continent on the landfalling TC motion through ideal- pography data with a resolution of 30 s (about 1 km) are ized numerical experiments. The investigation of used in the study. The center of the outermost domain is changes in the motion of TCs making landfall is closely located at 30.08N, 132.58E, and the three innermost do- associated with TC track and damage prediction. Our mains move with the TC (Fig. 1).

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC OCTOBER 2016 C H E N A N D W U 1617

TABLE 1. Summary of the numerical experiments.

Expt Terrain Land use Roughness length (m) TC vortex inserting OCN No Water Smooth Yes OCNB No Water Smooth No FLT Flat terrain Irrigated cropland and pasture 0.07 Yes FLTB Flat terrain Irrigated cropland and pasture 0.07 No TOP Topography Irrigated cropland and pasture 0.07 Yes TOPB Topography Irrigated cropland and pasture 0.07 No

The physics parameterization schemes are adopted as mainland China (Fig. 1). The TC tracks in our experi- follows. The WRF single-moment 3-class scheme and the ments are very similar to that of Typhoon Matsa. Kain–Fritsch scheme (Kain and Fritsch 1993) are used in Two sets of numerical experiments are discussed in the outmost domain for microphysics and cumulus pa- this study and each one consists of three runs (Table 1). rameterization schemes, respectively, while the WRF Both the sea surface temperature and the land surface single-moment 6-class microphysics scheme (Hong and temperature are set to 298C to eliminate the possible Lim 2006) and no cumulus parameterization scheme are influences of the initial temperature contrast between used in the four inner domains. The other physics options land and sea. The uniform land use is set to exclude include the Rapid Radiative Transfer Model (RRTM) possible influences of different types of land use, as the longwave radiation scheme (Mlawer et al. 1997), the typical type of land use in the eastern plain of mainland Dudhia shortwave radiation scheme (Dudhia 1989), the China is the irrigated cropland and pasture and a crop- Yonsei University scheme for planetary boundary layer land/grassland mosaic. Therefore, we retain the uniform parameterization (Noh et al. 2003), and the Noah land roughness length of 0.07 m in our designed landfall surface model. experiments (FLT, TOP, FLTB, and TOPB), as pre- The National Centers for Environmental Prediction scribed for these types of land use in Chen and Dudhia (NCEP) Final (FNL) operational global analysis dataset (2001), which is about one order smaller than the length with a resolution of 1.0831.08 at every 6 h is used to (0.5 m) used in Wong and Chan (2006), but is reasonable initialize the simulation. The FNL data are first filtered for the type of land use in the eastern plain of mainland with the Lanczos method to obtain the 20-day low-pass China. As described in Table 1, the three experiments in components (Duchon 1979). The low-frequency fields of the first set are conducted with different lower boundary Typhoon Matsa from 0000 UTC 5 August to 0000 UTC conditions. While the first experiment is run without 9 August 2005 are selected as the background of our land [the ocean experiment (OCN)], the other two runs experiments. At 0000 UTC 5 August, Typhoon Matsa include the flat terrain experiment (FLT) and the ele- was located to the northeast of Taiwan with maximum vated topography experiment (TOP). FLT is designed 2 surface winds of 45 m s 1. During the following 3 days, to examine the land drift induced by land surface Typhoon Matsa moved northwestward and then made roughness, and TOP investigates the effect of conti- landfall on China’s mainland at 1940 UTC 5 August. nental topography on the landfalling TC motion. In the The low-frequency environmental flows represent the second set, the experiments are the same as the corre- typical large-scale circulation pattern for TC activity sponding experiments in the first set, but they are initi- over the western North Pacific (Fig. 1), including a ated only with the background environment (without monsoon trough and a subtropical high in the lower inserting the TC vortex) to demonstrate the influences troposphere. of land and terrain on the environmental flows (OCNB, A warm-core vortex with the radial profile of tan- FLTB, and TOPB, respectively). The background gential wind used in Wang (2007) is put into the low- experiments designed in the second set are used to frequency background. It is initially axisymmetric with examine the evolution of the environmental flows. 2 maximum surface winds of 30 m s 1. After 12 h of spinup on an f plane, the vortex shows a structure typical of 3. Tracks and intensity of the simulated TCs mature TCs. The minimum sea level pressure is 965 hPa 2 and the maximum surface wind is 40 m s 1. The storm is The simulated TC tracks with at least tropical storm placed to the northeast of Taiwan (25.48N, 123.08E), intensity are shown in Fig. 2a. The TC centers in this moving along the southeasterly steering flows between study are determined at 1-h intervals. A variational the monsoon trough and the subtropical high and then approach is used to locate the TC center until the making landfall along the southeastern coastal areas of maximum azimuthal mean tangential wind speed is

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC 1618 WEATHER AND FORECASTING VOLUME 31 found (Wu et al. 2006). The TC centers at 700 hPa are used to avoid the influence of topography on low-level circulation in this figure. The simulated TCs generally move in a similar direction (Fig. 2a). The storms make landfall at 39 and 24 h in the FLT and TOP runs, respectively. The storm in OCN reaches the coastline in FLT and TOP at 41 h. Figure 2a indicates that the TCs in TOP move much faster than those in OCN and FLT, suggesting the topographic influence on the landfalling TC motion over China’s mainland. In OCN, the vortex intensifies during the first 12 h and 2 then remains around 40 m s 1 (Fig. 2b). The storms in FLT and TOP start to weaken about 3 h before their landfall. When the TCs in FLT and TOP make landfall, 2 the maximum wind speeds are 31.0 and 31.3 m s 1,re- 2 spectively, and then drop to 21.9 and 19.5 m s 1 about 6 h after landfall. The TCs in FLT and TOP weaken rapidly after landfall and the maximum wind speeds fall below tropical storm intensity at 13 and 10 h after landfall, respectively.

4. The difference in environmental steering To estimate the influence of environmental steering, the TC motion speed in this study is averaged between 850 and 300 hPa in all experiments. We focus mainly on the period from 24 h before to 12 h after making landfall in the experiments. During the 36-h period, the TC in 2 OCN moves at a mean speed of 2.67 m s 1 while the 2 mean speeds are 3.13 and 4.62 m s 1 in FLT and TOP, respectively (Fig. 3a). The TC accelerates at about 10 h before landfall in FLT and then moves faster than that in 21 OCN. It seems that the presence of land and the to- FIG. 2. Simulated TC (a) tracks and (b) intensity (m s ) in OCN (black), FLT (blue), and TOP (red) from 24 h before to 12 h after pography really leads to a landward component, com- landfall. The symbols indicate the TC locations at 12-h intervals. pared to OCN. In TOP, the landward component The landfall time is shown by the dashed vertical line. 2 (1.95 m s 1) accounts for an increase of 73% in the TC motion speed. respectively. Although the deviation of the TC motion Previous studies have suggested that environmental from the steering increases with the complication of the steering is dominant in TC motion (e.g., Chan and Gray lower boundary, we can see that the steering can gen- 1982; Wang and Holland 1996; Wu and Wang 2000). Our erally account for the differences in the TC translation further analysis indicates that the difference in the speed during the 36-h period. translation speed results mainly from the different in- To examine the influence of the land surface roughness fluences of steering in the three TC experiments. The on TC motion, we also calculated the environmental steering is calculated within a radius of 400 km from the steering in the corresponding background experiments TC center and is then vertically averaged (Fig. 3b). We without TCs embedded (Fig. 3c). The TC center positions examined the different average sizes for calculating the are used from the three TC experiments. The average 2 steering and found that the 400-km radius can minimize background steering in OCNB is 1.41 m s 1 during the the root-mean-square error (RMSE) between the TC 36-h period. In the background experiments with flat land translation speeds and environmental steering. The (FLTB) and terrain (TOPB), the mean steerings are 1.63 2 mean steerings in OCN, FLT, and TOP are 2.51, 2.71, and 3.40 m s 1 during the same period. After removing 2 and 4.23 m s 1, respectively. The RMSEs between the the background steering (Fig. 3d), the translation speeds 2 TC translation speed and the environmental steerings are 1.26, 1.50, and 1.11 m s 1 in OCN, FLT, and TOP, 2 are 0.27, 0.56, and 0.67 m s 1 in OCN, FLT, and TOP, respectively. Based on Wong and Chan (2006),theland

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC OCTOBER 2016 C H E N A N D W U 1619

21 FIG. 3. Time series of TC (a) translation speed (m s ), (b) environmental steering, (c) background steering derived from the background experiments, and (d) translation speed with background steering flow removed in OCN (black), FLT (blue), and TOP (red) from 24 h before to 12 h after landfall. The landfall time is shown by the dashed vertical line. drift is the difference in the remaining translation speeds turning southward. On the other hand, the midlatitude 2 between OCN and FLT. It is only 0.24 m s 1,whichis westerly flows are absent in TOPB because of the smaller than those in the previous land drift experiments. blocking of the topography in the northwest of the do- Moreover, the remaining speed in TOP is slower than main. The westerly monsoon flows from the Indochina that in OCN. Two possible reasons are responsible for the peninsula turn northward over the South China Sea and results. One is that the roughness length in our experi- extend northward over China’s mainland. Figure 4 ments is much smaller than that in the previous land drift clearly indicates that the blocking of the elevated ter- experiments. The other is that this component includes rain leads to a very different flow pattern over China’s the beta drift and the asymmetric flows resulted from the mainland, which enhances the landward motion of the interaction between the TC and environmental flows. TC in TOP. Based on our above analysis, we conclude that the dif- As shown in Fig. 1, the model domain actually covers ferent translation speeds in OCN, FLT, and TOP are part of the Tibetan Plateau. In summer the Tibetan Pla- mainly due to the differences in the environmental teau receives strong solar radiation at the surface, while steering (Fig. 3c). the rest of Asia is in the relatively cold midtroposphere. It has been demonstrated that the Tibetan Plateau acts as an elevated heat source with strong sensible heating (Yeh 5. The topographic effect et al. 1957; Flohn 1957; Wu et al. 2007). In TOP, the What causes the enhanced landward steering in the thermal effect of the elevated terrain is represented by presence of the topography? Figure 4 shows the simu- the high land surface temperature, which is initially set to lated 850-hPa wind fields in the three background ex- be the same as in the eastern plain. In other words, the periments during the first 12 h of integration. We can see land surface temperature in the mountainous western that the evolution of the 850-hPa wind is very similar in areas is initially the same as that in the eastern plain. In OCNB and FLTB, suggesting that the flat terrain has TOP and TOPB, the thermal effect can lead to in- little influence on the environmental flows. The westerly tensification of southerly winds in the eastern plain. The monsoon flows that are dominant south of 208N come mechanism is similar to the thermal effect of the Tibetan from the Indochina peninsula and meet the trade winds Plateau on the enhancement of the summertime East over the western North Pacific. In the midlatitudes (308– Asian monsoon (Chou 2003; Wu et al. 2007; Wu et al. 408N), a branch of westerly flow comes into the domain 2012). We think this mechanism may also work in TOP from the western lateral boundary in OCNB and FLTB. and TOPB. The midlatitude westerly flows form two branches over Figure 5 shows the wind difference between TOPB China’s mainland: one moving eastward and the other and OCNB during the 24-h period of TC landfall. The

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC 1620 WEATHER AND FORECASTING VOLUME 31

21 FIG. 4. The 850-hPa winds (m s ) at (left) 6 and (right) 12 h in (a) OCNB, (b) FLTB, and (c) TOPB.

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC OCTOBER 2016 C H E N A N D W U 1621

revealed that land surface roughness and temperature tend to make a TC move toward the continent (Wong and Chan 2006, 2007; Au-Yeung and Chan 2010; Szeto and Chan 2010; Li et al. 2014; Xu et al. 2013), the topographic effects of the elevated terrain on the western part of mainland China on the motion of landfalling TCs is examined in this study through idealized numerical experiments, in which an initially axisymmetric vortex is embedded into a monsoon trough and takes a landfall track toward China’s mainland. It is found that the effects of the elevated terrain on the western part of mainland China can enhance the southerly environmental flows, leading to a landward component of TC motion. We suggest that the effect of the elevated terrain in our experiments is similar to that of the Tibetan Plateau on the enhancement of the summertime East Asian monsoon (Chou 2003; Wu et al. 2007; Wu et al. 2012). Both the mechanic and thermal effects play a role in enhancing the southerly flow over mainland China. The numerical results from this study may have implications for interpreting the observed TC track changes on an interdecadal time scale (Wu et al. 2005; Chen et al. 2011). Chen et al. (2011) found that the an- nual average duration of TCs over China’s mainland increased significantly during the period 1975–2009. The increasing overland duration resulted from both the decreasing translation speed on the interdecadal scale and the decreasing vertical wind shear. Duan and Wu (2008) demonstrated that the sensible heat (SH) flux over the Tibetan Plateau has exhibited a significant decreasing trend since the mid-1980s, mainly because of decreases in the surface wind speed. The weakening trend in the thermal forcing of the Tibetan Plateau is associated with global warming (Duan and Wu 2008; Liu et al. 2012). Our numerical experiments suggest that the

21 reduced thermal forcing can decrease the northward FIG. 5. Differences in the 24-h mean winds (m s ) between TOPB and OCNB during landfall at (a) 500 and (b) 850 hPa. The motion of TCs over mainland China by reducing 48-h TC track in TOP is plotted. southerly winds in the eastern plain. enhanced southerly winds at 850 and 500 hPa in TOPB Acknowledgments. This research was jointly sup- are very clear in the eastern plain, extending from the ported by the National Basic Research Program of South China Sea to a latitude of about 408N. The en- China (Grants 2013CB430103 and 2015CB452803), the hanced southerly winds make the TC move northward National Natural Science Foundation of China (Grant faster in TOP. We also examine the wind difference 41275093), the project of the specially appointed pro- between TOPB and FLTB, and the wind difference fessorship of Jiangsu Province, and the Natural Science pattern is very similar to that shown in Fig. 5. Note that Foundation of the Jiangsu Higher Education In- the southerly winds can be well developed in TOPB stitutions of China (Grant 14KJA170005). during the ﬁrst several hours of integration (Fig. 4). REFERENCES 6. Conclusions and discussion Au-Yeung, A. Y. M., and J. C.-L. Chan, 2010: The effect of a river delta and coastal roughness variation on a landfalling tropi- The continent, which is much larger in size than islands, cal cyclone. J. Geophys. Res., 115, D19121, doi:10.1029/ can affect the landfalling TC motion. While it has been 2009JD013631.

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC 1622 WEATHER AND FORECASTING VOLUME 31

Bender, M. A., R. E. Tuleya, and Y. Kurihara, 1987: A numerical Liu, Y., G. Wu, J. Hong, J. Dong, B. Duan, Q. Bao, and study of the island terrain on tropical cyclones. Mon. Wea. L. Zhou, 2012: Revisiting Asian monsoon formation and Rev., 115, 130–155, doi:10.1175/1520-0493(1987)115,0130: changes associated with Tibetan Plateau forcing. II: ANSOTE.2.0.CO;2. Change. Climate Dyn., 39, 1183–1195, doi:10.1007/ Chan, J. C.-L., and W. M. Gray, 1982: Tropical cyclone motion s00382-012-1335-y. and surrounding flow relationship. Mon. Wea. Rev., Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. 110, 1354–1374, doi:10.1175/1520-0493(1982)110,1354: Clough, 1997: Radiative transfer for inhomogeneous atmo- TCMASF.2.0.CO;2. sphere: RRTM, a validated correlated-k model for the long- Chang, S. W.-J., 1982: The orographic effects induced by an island wave. J. Geophys. Res., 102, 16 663–16 682, doi:10.1029/ mountain range on propagating tropical cyclones. Mon. Wea. 97JD00237. Rev., 110, 1255–1268, doi:10.1175/1520-0493(1982)110,1255: Moon, Y., and D. Nolan, 2015: Spiral rainbands in a numerical TOEIBA.2.0.CO;2. simulation of Hurricane Bill (2009). Part I: Structures and Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface– comparisons to observations. J. Atmos. Sci., 72, 164–190, hydrology model with the Penn State–NCAR MM5 modeling doi:10.1175/JAS-D-14-0058.1. system. Part I: Model description and implementation. Mon. Noh, Y., W. G. Cheon, S.-Y. Hong, and S. Raasch, 2003: Im- Wea. Rev., 129, 569–585, doi:10.1175/1520-0493(2001)129,0569: provement of the K-profile model for the planetary boundary CAALSH.2.0.CO;2. layer based on large-eddy simulation data. Bound.-Layer Chen, X., L. Wu, and J. Zhang, 2011: Increasing duration of trop- Meteor., 107, 401–427, doi:10.1023/A:1022146015946. ical cyclones over China. Geophys. Res. Lett., 38, L02708, Szeto, K. C., and J. C. L. Chan, 2010: Structural changes doi:10.1029/2010GL046137. of a tropical cyclone during landfall: b-plane simu- Chou, C., 2003: Land–sea heating contrast in an idealized Asian lations. Adv. Atmos. Sci., 27, 1143–1150, doi:10.1007/ summer monsoon. Climate Dyn., 21, 11–25, doi:10.1007/ s00376-009-9136-x. s00382-003-0315-7. Wang, C.-C., H.-C. Kuo, Y.-H. Chen, H.-L. Huang, C.-H. Chung, Duan, A., and G. Wu, 2008: Weakening trend in the atmospheric and K. Tsuboki, 2012: Effects of asymmetric latent heating on heat source over the Tibetan Plateau during recent decades. typhoon movement crossing Taiwan: The case of Morakot Part I: Observations. J. Climate, 21, 3149–3164, doi:10.1175/ (2009) with extreme rainfall. J. Atmos. Sci., 69, 3172–3196, 2007JCLI1912.1. doi:10.1175/JAS-D-11-0346.1. Duchon, C. E., 1979: Lanczos filtering in one and two Wang, Y., 2007: A multiply nested, movable mesh, fully com- dimensions. J. Appl. Meteor., 18, 1016–1022, doi:10.1175/ pressible, nonhydrostatic tropical cyclone model—TCM4: 1520-0450(1979)018,1016:LFIOAT.2.0.CO;2. Model description and development of asymmetries without Dudhia, J., 1989: Numerical study of convection observed during explicit asymmetric forcing. Meteor. Atmos. Phys., 97, 93–116, the Winter Monsoon Experiment using a mesoscale two- doi:10.1007/s00703-006-0246-z. dimensional model. J. Atmos. Sci., 46, 3077–3107, doi:10.1175/ ——, and G. J. Holland, 1996: Tropical cyclone motion and evolu- 1520-0469(1989)046,3077:NSOCOD.2.0.CO;2. tion in vertical shear. J. Atmos. Sci., 53, 3313–3332, doi:10.1175/ Flohn, H., 1957: Large-scale aspects of the ‘‘summer mon- 1520-0469(1996)053,3313:TCMAEI.2.0.CO;2. soon’’ in South and East Asia. J. Meteor. Soc. Japan, 75, Wong, M. L.-M., and J. C.-L. Chan, 2006: Tropical cyclone motion 180–186. in response to land surface friction. J. Atmos. Sci., 63, 1324– Hong, S.-Y., and J.-O. J. Lim, 2006: The WRF single-moment 1337, doi:10.1175/JAS3683.1. 6-class microphysics scheme (WSM6). J. Korean Meteor. Soc., ——, and ——, 2007: Modeling the effects of land–sea roughness 42, 129–151. contrast on tropical cyclone winds. J. Atmos. Sci., 64, 3249– Huang, Y.-H., C.-C. Wu, and Y. Wang, 2011: The influence of is- 3264, doi:10.1175/JAS4027.1. land topography on typhoon track deflection. Mon. Wea. Rev., Wu, C.-C., 2001: Numerical simulation of Typhoon Gladys (1994) 139, 1708–1727, doi:10.1175/2011MWR3560.1. and its interaction with Taiwan terrain using the GFDL hur- Jian, G.-J., and C.-C. Wu, 2008: A numerical study of the track ricane model. Mon. Wea. Rev., 129, 1533–1549, doi:10.1175/ deflection of Supertyphoon Haitang (2005) prior to land- 1520-0493(2001)129,1533:NSOTGA.2.0.CO;2. fall in Taiwan. Mon. Wea. Rev., 136, 598–615, doi:10.1175/ Wu, G., and Coauthors, 2007: The influence of mechanical and 2007MWR2134.1. thermal forcing by the Tibetan Plateau on Asian climate. Kain, J. S., and J. M. Fritsch, 1993: Convective parameterization for J. Hydrometeor., 8, 770–789, doi:10.1175/JHM609.1. mesoscale models: The Kain–Fritch scheme. The Representa- Wu, L., and B. Wang, 2000: A potential vorticity tendency di- tion of Cumulus Convection in Numerical Models, Meteor. agnostic approach for tropical cyclone motion. Mon. Wea. Monogr., No. 46, Amer. Meteor. Soc., 165–170. Rev., 128, 1899–1911, doi:10.1175/1520-0493(2000)128,1899: Li, Y., K. K.-W. Cheung, and J. C.-L. Chan, 2014: Numerical study APVTDA.2.0.CO;2. on the development of asymmetric convection and vertical ——, ——, and S. Geng, 2005: Growing typhoon influence on wind shear during tropical cyclone landfall. Quart. J. Roy. East Asia. Geophys. Res. Lett., 32, L18703, doi:10.1029/ Meteor. Soc., 140, 1866–1877, doi:10.1002/qj.2259. 2005GL022937. Liang, J., L. Wu, X. Ge, and C.-C. Wu, 2011: Monsoonal ——, S. A. Braun, J. Halverson, and G. Heymsfield, 2006: A nu- influence on Typhoon Morakot (2009). Part II: Numeri- merical study of Hurricane Erin (2001). Part I: Model ver- cal study. J. Atmos. Sci., 68, 2222–2235, doi:10.1175/ ification and storm evolution. J. Atmos. Sci., 63, 65–86, 2011JAS3731.1. doi:10.1175/JAS3597.1. Lin, Y.-L., S.-Y. Chen, C. M. Hill, and C.-Y. Huang, 2005: Control ——, J. Liang, and C.-C. Wu, 2011: Monsoonal influ- parameters for the influence of a mesoscale mountain range on ence on Typhoon Morakot (2009). Part I: Observational cyclone track continuity and deflection. J. Atmos. Sci., 62, analysis. J. Atmos. Sci., 68, 2228–2221, doi:10.1175/ 1849–1866, doi:10.1175/JAS3439.1. 2011JAS3730.1.

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC OCTOBER 2016 C H E N A N D W U 1623

Wu, Z., J. Li, Z. Zhang, and T. Ma, 2012: Modulation of the Tibetan Rev., 121, 3193–3212, doi:10.1175/1520-0493(1993)121,3193: Plateau snow cover on the ENSO teleconnections: From the IOTWTT.2.0.CO;2. East Asian summer monsoon perspective. J. Climate, 25, ——, and ——, 1993b: Interaction of typhoons with the orog- 2481–2489, doi:10.1175/JCLI-D-11-00135.1. raphy. Part II: Continuous and discontinuous tracks across Xie, B., and F. Zhang, 2012: Impacts of typhoon track and island the island. Mon. Wea. Rev., 121, 3213–3233, doi:10.1175/ topography on the heavy rainfalls in Taiwan associated with 1520-0493(1993)121,3213:IOTWTT.2.0.CO;2. Morakot (2009). Mon. Wea. Rev., 140, 3379–3394, doi:10.1175/ ——, S. W. Lo, and P. C. Chu, 1957: The wind structure and heat MWR-D-11-00240.1. balance in the lower troposphere over Tibetan Plateau and its Xu, X., and Coauthors, 2013: Does warmer China land attract more surrounding. Acta Meteor. Sin., 28, 108–121. super typhoons? Sci. Rep., 3, 1522, doi:10.1038/srep01522. Zhang, J., L. Wu, and Q. Zhang, 2010: Tropical cyclone damages in Yang, M.-J., D.-L. Zhang, and H.-L. Huang, 2008: A modeling study China under the background of global warming (in Chinese). of Typhoon Nari (2001) at landfall. Part I: Topography effects. J. Trop. Meteor., 27, 442–454. J. Atmos. Sci., 65, 3095–3115, doi:10.1175/2008JAS2453.1. Zhang, Q., Q. Liu, and L. Wu, 2009: Tropical cyclone damages Yeh, T.-C., and R. L. Elsberry, 1993a: Interaction of typhoons with in China 1983–2006. Bull. Amer. Meteor. Soc., 90, 489–495, the orography. Part I: Upstream track deﬂections. Mon. Wea. doi:10.1175/2008BAMS2631.1.

Unauthenticated | Downloaded 09/28/21 07:44 PM UTC