PUBLICATIONS
Journal of Geophysical Research: Atmospheres
RESEARCH ARTICLE Dependence of the relationship between the tropical 10.1002/2015JD023716 cyclone track and western Pacific subtropical Key Points: high intensity on initial storm size: • Simulated TC track and WPSH are sensitive to initial storm size A numerical investigation • Large TC tends to decrease WPSH intensity and thus force TC turn Yuan Sun1,2,3, Zhong Zhong2,4, Lan Yi5, Tim Li1,3, Ming Chen6, Hongchao Wan2, Yuxing Wang2, northward 2 • Differences are attributed to the and Kai Zhong fl fl in ow mass ux entering into 1 TC region International Laboratory on Climate and Environment Change, and Key Laboratory of Meteorological Disaster of Ministry of Education, Nanjing University of Information Science and Technology, Nanjing, China, 2College of Meteorology and Oceanography, PLA University of Science and Technology, Nanjing, China, 3IPRC/AORC and Department of Atmospheric Supporting Information: 4 • Sciences, University of Hawaii at Manoa, Honolulu, Hawaii, USA, Jiangsu Collaborative Innovation Center for Climate Figures S1-S15 and Data Sets S1-S3 5 captions Change and School of Atmospheric Sciences, Nanjing University, Jiangsu, China, Chinese Academy of Meteorological • Figures S1 Sciences, Beijing, China, 6National Center for Atmospheric Research, Boulder, Colorado, USA • Figures S2 • Figures S3 • Figures S4 Abstract A suite of numerical experiments were conducted to investigate the sensitivity of the tropical • Figures S5 — fi • Figures S6 cyclone (TC) motion western Paci c subtropical high (WPSH) intensity relationship to initial storm size. • Figures S7 Two TC cases, Songda (2004) and Megi (2010), were studied. It was found that with the increase of initial • Figures S8 storm size, the main body of the WPSH tends to withdraw eastward and the TC tends to turn northward • Figures S9 • Figures S10 earlier. The involved physical mechanism was investigated. Rather than the change of the beta effect due to • Figures S11 storm size change, it is the change of the geopotential height in the TC outer region that is critical for the • Figures S12 different TC tracks between the sensitivity experiments. Due to increase of the initial storm size, the inflow • Figures S13 fl fi • Figures S14 mass ux entering the TC region increases, leading to a signi cant decrease in 500 hPa geopotential height • Figures S15 in the TC outer region after 2–3 day integration. As a result, the simulated intensity of the WPSH over its fringe • Data Set S1 closetotheTCdecreasesnotablywhentheWPSHedgeiswithin the TC outer region. Such a decrease leads to a • Data Set S2 • Data Set S3 break of WPSH. Subsequently, the TC turns northward toward the break of the subtropical high. This further weakens the intensity of the WPSH over the region close to the TC. The result helps us better understand the Correspondence to: relationship between the TC track and WPSH intensity. It also indicates that a proper representation of initial storm Z. Zhong, size is important for realistic prediction of TC track and the change of the WPSH. [email protected]
Citation: Sun, Y., Z. Zhong, L. Yi, T. Li, M. Chen, 1. Introduction H. Wan, Y. Wang, and K. Zhong (2015), Dependence of the relationship In the past 20 years, regional climate models (RCMs) have been widely applied for high-resolution climate between the tropical cyclone track simulations. RCMs are taken as a dynamical downscaling tool for global circulation models (GCMs) [e.g., and western Pacific subtropical high Giorgi, 2006]. Compared with GCMs, RCMs can provide high-resolution (typically 50 km or less) regional intensity on initial storm size: A numerical investigation, J. Geophys. weather and climate simulations at a more reasonable computational cost [Lucas-Picher et al., 2011; Diallo Res. Atmos., 120, 11,451–11,467, et al., 2013] and play an increasingly important role in addressing scientific issues associated with regional doi:10.1002/2015JD023716. climate variability and change [Giorgi and Mearns, 1999; Giorgi et al., 2001; Leung et al., 2003]. While RCMs have demonstrated great skills in downscaling GCM simulations or reanalysis, there still exist some model Received 27 MAY 2015 fi Accepted 27 OCT 2015 de ciencies that have not been solved yet [e.g., Takle et al., 1999; Leung et al., 1999; Roads et al., 2003]. For Accepted article online 29 OCT 2015 example, RCMs often do not perform well in simulating the East Asian monsoon system [McGregor, 1997; Published online 21 NOV 2015 Wang and Wang, 2001; Zhong, 2006; Zhou et al., 2008, 2009; Zou and Zhou, 2013]. Recent studies have suggested that large biases appear in East Asian monsoon simulation when tropical cyclones (TCs) are active over the western North Pacific (WNP), and the departure of the simulated TC track from its observed position may be one reason for RCMs’ failure in simulating the WPSH and thus the large-scale atmospheric circulation [Zhong, 2006; Sun et al., 2014a, 2015a, 2015b].
Due to limited capabilities for realistic simulation of convective activities in the tropics, RCMs exhibit a relatively low skill in the East Asian summer monsoon simulation during the TC-prone seasons on one hand
©2015. American Geophysical Union. [e.g., Giorgi et al., 1999; Lee and Suh, 2000; Zhong, 2006; Zhong and Hu, 2007; Kubota and Wang, 2009; All Rights Reserved. Fudeyasu et al., 2010; Cha et al., 2011]. On the other hand, TC activity is related to variation in the western
SUN ET AL. SENSITIVITY OF TC AND WPSH TO STORM SIZE 11,451 Journal of Geophysical Research: Atmospheres 10.1002/2015JD023716
Pacific subtropical high (WPSH), which is either overestimated or underestimated in most RCM simulations [Giorgi et al., 1999; Lee and Suh, 2000]. The WPSH-related large-scale forcing plays a crucial role in successful simulation of both the East Asian monsoon climate and TC activities over the WNP [Zhong, 2006]. Thereby, insufficient knowledge of interactions between TC and the WPSH is probably one important reason for the failure of RCMs in both the WPSH and TC track simulations. Zhong [2006] proposed that the erratic departure of the simulated TC track from its observed position is possibly a primary reason for RCMs’ failure in simulating WPSH and thus East Asian summer monsoon. Recently, Sun et al. [2014a] and Sun et al. [2015b] attributed the failure in RCM simulations of the TC track and the WPSH to overestimation of anvil clouds in the cumulus parameterization scheme. The simulated anvil clouds extend far away from the TC center and reach upper troposphere over the WPSH. Above the freezing level, condensation of the falling hydrometeors in the anvil clouds heats the upper troposphere in the WPSH. Below the freezing level, the evaporative cooling of the hydrometeors cools the middle and lower troposphere. Such a pattern of vertical heating profilecausesaweakeningoftheWPSH,whichinturncontributes to the early recurvature of the TC. Sun et al. [2015a] suggested that this is also the physical mechanism that explains different performances of some microphysical parameterization schemes in the simulation of TC and the WPSH. TC impacts are highly correlated with the storm size, yet the importance of storm size has not received enough attention. The TC size is an important structure parameter not only because it determines the extent of the damage caused by the TC [Hsu and Blanchard, 2008; Maclay et al., 2008] but also because it has great impacts on the motion of the TC [Lester and Elsberry, 1997, 2000; Hill and Lackmann, 2009]. Theoretically, the storm size could affect storm motion by influencing the extension and intensity of anvil clouds [Bu et al., 2014] or by influencing the outer wind structure. Yet which influence plays a major role in the storm motion remains unknown. As discussed in Sun et al. [2014a] and Sun et al. [2015a], the anvil clouds could change the microphysical latent heating over the edge of the WPSH, which subsequently affects the WPSH and TC motion. Meanwhile, the TC movement often deviates from the large-scale steering flow due to the beta-effect propagation (BEP), which depends on the mean relative angular momentum and thus is highly sensitive to the outer wind structure of a TC [Holland, 1983; Fiorino and Elsberry, 1989; Carr and Elsberry, 1997]. For this reason, the movement of large storms may differ from that of smaller ones due to the more pronounced beta drift [Hill and Lackmann, 2009]. Observational analyses have also confirmed the relationship between the TC track and TC size. Lee et al. [2010] calculated the size of 145 TCs in the western North Pacific during 2000–2005 based on the QuikSCAT oceanic winds and the best tracks of the TCs from the Joint Typhoon Warning Center (JTWC). Their results indicate that the 18 persistently large TCs mostly have northwestward or north-northwestward tracks, while the 16 persistently small TCs mostly move westward to northwestward (see their Figure 4). However, due to the lack of observations with a wide variety and a high spatial-temporal resolution, it is hard to reveal the mechanism behind the observed phenomena. While most previous studies on TC size have focused on TC internal processes and interactions between the TC and environmental circulations and their impact on TC size [Lee et al., 2010], the present study explores the impact of TC size on the interaction between the TC track and WPSH instead of changes in the TC size itself. Emanuel [1986] and Rotunno and Emanuel [1987] proposed that the size of the initial disturbance is a key factor in determining the TC size. Following their studies, here we assume that the storm size is closely related to its initial size. We will further explore the impact of TC size on the simulations of TC motion and the WPSH by changing the initial size of the storm. The objective of this study is twofold. We will first investigate the impact of the initial storm size on TC motion and the WPSH intensity and then explore the involved physical processes and possible mechanisms. Case studies of Tropical Cyclones Songda (2004) and Megi (2010) are performed in this study. This paper is organized as follows. Section 2 describes the numerical model used in this study and the experimental design. Section 3 shows the simulation results with different initial size of TC. The sensitivity of simulated TC motion to initial TC size and the involved physical processed are discussed in section 4. Conclusions and discussion are given in the final section.
2. Model Configuration and Experimental Design To illustrate the impact of initial TC size on TC track and WPSH simulations, we have performed two case studies on TC Songda (2004) and TC Megi (2010). Both Songda (2004) and Megi (2010) are characterized
SUN ET AL. SENSITIVITY OF TC AND WPSH TO STORM SIZE 11,452 Journal of Geophysical Research: Atmospheres 10.1002/2015JD023716
by high intensity, long duration, and fast development with a typical turning track. Their motions and sudden turnings are closely related to the withdrawal and extension of the WPSH. The track information of Songda (2004) and Megi (2010) are provided by Regional Specialized Meteorological Center. Typhoon Songda (2004) is among the strongest typhoons that made landfall on the main islands of Japan in the past 50 years. It caused extensive damages to Japan due to its strong winds. The storm formed in Marshall Islands on 28 August 2004 and rapidly intensified while moving northwestward over the WNP. Because of the weakening subtropical high, Songda turned to the northeast direction over the East China Sea at 1200 UTC 6 September and made landfall on Kyushu Island, south of the main island of Japan, at 0000 UTC 7 September (see Figure 3). Typhoon Megi (2010) is one of the most intense TCs on record and is the only supertyphoon in 2010. Megi formed over the WNP (11.9°N, 141.4°E) at 0000 UTC, 13 October 2010. Due to the influence of the subtropical ridge and the favorable environmental condition, Megi started moving westward after its formation and continued to gain strength. It has reached its peak intensity while making landfall over Isabela Province, Philippines at 0325 UTC 18 October. Megi became weak when passing Sierra Madre due to the effects of the land surface but rapidly regained strength over the South China Sea. Later on 19 October, Megi turned northwestward and moved slowly since the subtropical ridge weakened due to a deep midlatitude shortwave trough that was approaching. On 23 October, Megi weakened to a tropical storm as it made landfall at Zhangpu in Fujian Province, China. Megi further downgraded to a tropical depression later on 23 October (see Figure 4). The model used in this study is the Advanced Research version of Weather Research and Forecasting Model, version 3.3.1 (WRF-ARW V3.3.1) developed at the National Center for Atmospheric Research [Skamarock et al., 2008]. WRF-ARW is a three-dimensional, fully compressible, nonhydrostatic model formulated in a terrain- following mass coordinate in the vertical. The National Center for Environmental Prediction (NCEP) global final analysis data at 1° × 1° latitude-longitude grids with 6 h interval are used to provide initial and lateral boundary conditions for the WRF-ARW model. The model configuration for the simulation of TC Megi (2010) is identical to that in our previously study [Sun et al., 2015a] except that the initial time is different, as follows in the next two paragraphs. A 20 km resolution domain with 36 vertical levels is set up for the simulations of both Songda and Megi. Note that the model domains and simulation time for the two cases are different. For the case of Songda, the model domain is centered at (28°N, 137.5°E) with 206 (north-south) × 222 (east-west) grid points and the simulation is initialized at 0000 UTC 31 August and ends at 0600 UTC 07 September 2004, covering a total of 174 h. For the case of Megi, the model domain is centered at (22°N, 122°E) with 160 (north-south) × 180 (east-west) horizontal grid points and the simulation is initialized at 0000 UTC 16 October and ends at 0000 UTC 24 October 2010, with a total of 192 h integration. The domains of the two cases all extend far enough south to allow simulations of the WPSH withdrawal and the recurvature of the TCs. The model physics used in this study include (i) the single-moment three-class microphysics scheme [Hong et al., 2004]; (ii) the Grell-Dévényi cumulus parameterization scheme [Grell and Dévényi, 2002]; (iii) the Mellor-Yamada-Janjić boundary layer scheme [Mellor and Yamada, 1982; Janjić, 2002] with the Monin-Obukhov surface layer scheme [Monin and Obukhov, 1954; Janjić, 1996, 2002]; (iv) the five-layer thermal diffusion scheme for land surface processes [Skamarock et al., 2008]; and (v) the Goddard scheme for shortwave radiation calculation[Chou and Suarez, 1994] and Rapid Radiative Transfer Model for longwave radiation calculation [Mlawer et al., 1997]. For each TC case, three experiments with different initial storm sizes are conducted to investigate the response of TC track and the WPSH to changes in initial TC size. In this study, the TC Bogus scheme in the WRF model is used to change the maximum radius from TC center at the initial time [Skamarock et al., 2008]. In these experiments, the maximum radius outward from the TC center is set to 60, 120, and 180 km at the initial time, respectively. For convenience, we define the three experiments as the one with a small-sized storm (ES), the one with a medium-sized storm (EM), and the experiment with a large-sized storm (EL) in order of increasing size, respectively. All other physical schemes and model settings are the same in the three experiments described above.
3. Simulation Results 3.1. Storm Size In an operational setting, storm size is described by the area of the outermost closed isobar (ACI) in the surface level. For both Songda and Megi in this study, the value of the outermost closed isobar is about
SUN ET AL. SENSITIVITY OF TC AND WPSH TO STORM SIZE 11,453 Journal of Geophysical Research: Atmospheres 10.1002/2015JD023716
Figure 1. Temporal evolution of ACI in the sensitivity experiments in (a) Songda case and (b) Megi case.
1000 hPa. Figure 1 shows the temporal evolutions of ACI in the cases of Songda (2004) and Megi (2010). It clearly shows that in both Songda and Megi cases, the ACI is highly sensitive to the initial vortex size determined by the TC Bogus scheme and increases significantly as the initial vortex size increases, especially as the initial vortex size increases from small size in the ES to medium size in the EM. This is consistent with the idealized model results of Xu and Wang [2010], which indicated that a storm with a large initial size usually has strong outer winds and large surface entropy fluxes outside the eyewall. They are accompanied by active spiral rainbands, leading to fast increase in the inner core size. In addition, the ACI in the EL decreases significantly after 2000 UTC 04 September 2004 for Songda case. This is probably caused by the landfall of the storm in the EL, since the time of the decrease in ACI is basically consistent with the landfall time of the storm simulated in the EL (see Figure 3). To further provide a picture of typical precipitation associated with the
Figure 2. Module-simulated radar reflectivity (unit: dBZ) for the sensitivity experiments at (a–c) 0000 UTC 3 September 2004 in the case study of Songda (2004) and (d–f) 0000 UTC 18 October 2010 in the case study of Megi (2010), respectively.
SUN ET AL. SENSITIVITY OF TC AND WPSH TO STORM SIZE 11,454 Journal of Geophysical Research: Atmospheres 10.1002/2015JD023716
simulated TC, Figure 2 presents two snapshots of the model-simulated radar reflectivity at 0000 UTC 3 September 2004 for the case of Songda (2004) and 0000 UTC 18 October 2010 for the case of Megi (2010). Compared with that in the ES run, a wider and broader eyewall is evident in the EM and EL runs, along with larger area of precipitation in the outer spiral rainbands in both Songda and Megi cases. This is consistent with our hypothesis and further indicates that the size of the simulated TCs is highly dependent on the initial vortex size, i.e., the larger the initial vortexes, the larger the storms will be later. 3.2. Storm Track Figure 3. The model domain and simulated storm tracks in the sensitivity Previous studies have indicated that experiments with different initial TC size in the case of Songda (2004). The the storm size affects its motion not observed best track at 6 h intervals (black dotted line) is overlaid. only by changing the large-scale envir- onmental flow but also by influencing the BEP [Holland, 1983; Fiorino and Elsberry, 1989; Carr and Elsberry, 1997; Bu et al., 2014; Sun et al., 2014a, 2015a]. Figure 3 compares the storm track simulated in the sensitivity experiments with the JTWC best track of Songda. It indicates that the simulated storm track is highly sensitive to the initial size of the storm and large differences between the results of the three experiments occur about 2 days after the model integration starts. The simulated storm in the EM and EL turns northward earlier than observation and makes landfall in Japan at about 1200 UTC 06 September 2004 and 1800 UTC 05 September 2004, respectively, whereas the simulated storm in the ES continues to move westward and turns northward later than observation and did not make landfall before 0600 UTC 07 September 2004. Figure 4 compares the storm track simulated in the three experiments with the JTWC best track of Megi. The model with the small initial storm size can well reproduce the track of Megi, but it performs not so well in the experi- ments with the medium and large initial size. All experiments realistically simulate the northwestward movement of Megi before 0000 UTC 17 October 2010, and the west-southwestward movement along the southern periph- ery of the WPSH until the storm crossed the Luzon Island. Large differences between results of the three experiments occur after 1800 UTC 18 October 2010. Similar to the results in Songda case, the simulated storm in the EM and EL turns northward earlier than observation, whereas in ES it continues to move westward and turns northward over South China Sea at about 1800 UTC 19 Figure 4. The same as Figure 3 but for the case of Megi (2010). October 2010. Apparently, the simulated
SUN ET AL. SENSITIVITY OF TC AND WPSH TO STORM SIZE 11,455 Journal of Geophysical Research: Atmospheres 10.1002/2015JD023716
storm track is sensitive to the initial size of the storm. It is worth noting that a TC with a larger initial size turns northward earlier in both Megi and Songda cases. Feedback and interaction between the TC and WPSH are interwoven in these simulations, making it a challenging issue to address what is the root cause of the large biases in both the WPSH and TC simulations. Note that the unrealistic withdrawal and extension of the simulated WPSH is responsible for the failure in RCM simulations of TC motions. The erratic departure of the simulated TC track from its observed position contributes greatly to the RCM’s failure in simulating the WPSH [Zhong, 2006]. Thereby, the TC track simulation is a key factor that affects the WPSH simulation. In the following section, we will discuss in detail the possible reasons for the difference in TC track simulation between these experiments.
4. Possible Reasons for the Differences in TC Tracks 4.1. Potential Vorticity Tendency Diagnosis Previous studies suggested that the environmental flow and the TC structure are two key factors determining the TC motion over the WNP [e.g., Chan and Gray, 1982; Holland, 1983; Fiorino and Elsberry, 1989; Wu and Wang, 2000; Wu et al., 2005; Zhong, 2006]. Theoretically, the storm size can influence the TC motion in two ways. First, the storm size can modulate the large-scale environmental flow near the TC and thus affect TC motion by influencing the withdrawal and extension of the WPSH. Second, the storm size can affect the TC motion by modifying the thermodynamic and dynamic structures of the TC. As suggested by Carr and Elsberry [1990] and Holland [1993], the large-scale environmental flow is defined as the layer-mean (850–300 hPa) flow averaged over a 5°–7° latitude band of the TC center. In the following paragraph, we will discuss which factor is dominant in the differences between the simulated TC motions of the three sensitivity experiments. To estimate contributions of the TC structure and environmental flow to TC motion, the potential vorticity tendency (PVT) diagnosis technique is applied in this study [Chan, 1984; Wu and Wang, 2000, hereafter WW00]. Simulations of an ideal TC in WW00 indicated that a baroclinic TC moves toward the region where the azimuthal wave number 1 of the maximum PVT is located. WW00 suggested that the PVT results from horizontal PV advection (HA), vertical PV transportation (VT), and diabatic heating (DH), while the contribution of individual physical process to the TC motion is equivalent to its contribution to the wave number 1 component of the PVT. Equation (1) directly links the TC motion with PVT, and equation (2) estimates the contribution of individual physical process to the PV in the pressure coordinates. ∂ P ¼ ∇ ∂ VPV PS (1) t 1 "# ! ∂ ∂ P ¼ Λ ∇ ω P ∇ Q þ ∇θ ∂ 1 V P ∂ g 3 π q F (2) t 1 p CP
where P represents the PV, PS is the symmetric PV component of PV, V is the horizontal wind speed, VPV is the velocity of vortex motion estimated from the wave number 1 component of the PVT, p is the pressure, q is the
three-dimensional absolute vorticity vector, ∇3 is the three-dimensional gradient, Λ1 denotes an operator to ω obtain the wave number one component, g is the gravitational acceleration, is the vertical velocity in the pressure coordinates, θ is potential temperature, and F and Q denote friction and diabatic heating rate, respectively. As suggested by WW00, in the derivation of equation (1), we have assumed that the wave number 1 component of the PVT is negligibly small compared with the symmetric component of the PVT.
We apply equation (1) to each grid point (denoted by subscript i) and compute the gradient of PS, ∂ ∂ ∂ P ¼ PS PS ∂ cx ∂ cy ∂ (3) t 1i x i y i
From equation (3) the zonal (cx) and meridional (cy) components of the vortex moving speed at each level can be determined. Considering a specific region that is within a radius of 360 km from the TC center, we use the
least squares method to calculate cx and cy by minimizing ∂ ∂ ∂ ∑ P þ PS þ PS ∂ cx ∂ cy ∂ (4) i ≤ N t 1i x i y i Where N denotes the number of total grid points in the specified region. The effects of the large-scale steering flow and the BEP are all included in HA (first term on the right-hand side of equation (2)), while
SUN ET AL. SENSITIVITY OF TC AND WPSH TO STORM SIZE 11,456 Journal of Geophysical Research: Atmospheres 10.1002/2015JD023716