TMCnet News
A Study of Two Propagating Heavy-Rainfall Episodes near Taiwan during SoWMEX/TiMREX IOP-8 in June 2008. Part II: Sensitivity Tests on the Roles of Synoptic Conditions and Topographic Effects [Monthly Weather Review](Monthly Weather Review Via Acquire Media NewsEdge) ABSTRACT This study is the second of a two-part series to investigate two rainfall episodes in the Hovmöller space near Taiwan during the eighth intensive observing period (IOP-8, 12-17 June 2008) of the Southwest Monsoon Experiment/Terrain-influenced Monsoon Rainfall Experiment (SoWMEX/TiMREX). The first episode moved eastward and the second westward, and both caused heavy rainfall in Taiwan. The goal of Part I was to better understand the mechanism and controlling factors for the organization and propagation of the episodes. Here in Part II, the detailed roles played by synoptic conditions and terrain effects are further examined. Three sensitivity tests (at 2.5-km grid spacing) are designed to include only the effects of synoptic evolution (SNP), and those from land-sea distribution-diurnal variations on top of a mean background with/ without topography (DIU/DNT). As the benchmark, the control (CTL) experiment captures the 6-day event successfully and is validated in Part I. In SNP, the two episodes are reproduced with overall similarity to CTL and the observation, and this confirms that the general location/time of rainfall are mainly controlled by synoptic forcing in this case, in contrast to typical warm-season conditions in the central United States. Even so, diurnal effects can still exert discernible impacts and modulate local convective development in many instances, particularly an afternoon enhancement over terrain, and the averaged diurnal cycle in CTL over southeastern China resembles those in DIU/DNT rather than that in SNP (with no land). The steep topography of Taiwan is especially important for its rainfall distribution, including the heavy rainfall on 16 June through processes as postulated by Xu et al. 1. Introduction a. Literature review In the continental United States, organized warm- season precipitation episodes exhibit coherent behavior of eastward propagation to the lee of the Rockies in the longitude-time (Hovmöller) space (e.g., Maddox 1980; Laing and Fritsch 1997; Carbone et al. 2002). Under relatively weak synoptic forcing, their diurnal cycle is closely tied to elevated terrain (e.g., Dai et al. 1999; Ahijevych et al. 2004; Carbone and Tuttle 2008). In re- sponse to stronger solar heating during daytime, the mountain-plain solenoid (MPS) develops and induces upslope winds to frequently initiate convection over the eastern slopes of the Rockies in local afternoon (e.g., Holton 1967; Wallace 1975; Tripoli and Cotton 1989; Carbone and Tuttle 2008). Subsequent organization into mesoscale convective systems (MCSs) and downstream propagation along certain latitudinal corridors occur through cold-pool dynamics in a sheared environment (e.g., Rotunno et al. 1988; Tuttle and Carbone 2004; Trier et al. 2006) or other mechanism such as gravity waves (e.g., Fovell et al. 2006; Pritchard et al. 2011; Wang et al. 2011a). The reversal of the MPS and the development of low-level jet (LLJ) at night, on the other hand, both help maintain the MCSs and their further propagation (e.g., Maddox 1980; Tuttle and Davis 2006; Carbone and Tuttle 2008). In a case studied by Trier et al. (2006), zonally prop- agating episodes across the Great Plains occurred suc- cessively over a one-week period, and were captured by the Weather Research and Forecasting (WRF) Model at 4-km grid spacing with explicit cloud microphysics (e.g., Davis et al. 2003). Moreover, a 10-day WRF simulation driven only by repeated mean diurnal variations on top of a monthly averaged background (for July) was carried out in Trier et al. (2010) to isolate the effects of terrain and diurnal components. In this experiment, many sa- lient features similar to those observed in the climatol- ogy, including the afternoon precipitation maximum over the Rockies and the progression of rainfall systems across the Great Plains along narrow latitudinal corri- dors overnight, are successfully reproduced. Thus, Trier et al. (2010) concluded that transient synoptic distur- bances such as mobile fronts and short waves are not crucial to the general distribution and diurnal cycle of midsummer precipitation in the United States, even though they may help enhance and focus rainfall in in- dividual cases. Therefore, warm-season precipitation episodes in the central United States are highly con- trolled by topographic effects and local diurnal distur- bances such as the nocturnal LLJ (e.g., Zhong et al. 1996; Trier et al. 2006, 2010). During the past decade, the behavior of rainfall epi- sodes to the lee of the Tibetan Plateau (TP) in East Asia are also studied using both satellite (Wang et al. 2004, 2005a; Johnson 2011; Bao et al. 2011; Xu and Zipser 2011; Wang et al. 2011b) and rain gauge data (e.g., Yu et al. 2007; Zhou et al. 2008; Chen et al. 2010). Similar to the events in the United States, the roles of the MPS between eastern TP and its lee and the diurnal variation of the LLJ over southern-central China have been the foci of much investigation (e.g., Huang et al. 2010; Chen et al. 2009, 2013). Recently, Bao et al. (2011) show that besides the MPS near eastern TP, there exist two other weaker solenoids farther downstream along the Yangtze River valley (278-358N): one near 1108E with a mean terrain drop of nearly 1 km, and the other near 1208E between the land/coastal hills and the East China Sea (ECS). Using a strategy similar to Trier et al. (2010), Sun and Zhang (2012) performed WRF simulation and sensitivity tests to examine the impact of the MPS on the diurnal cycle of rainfall along the mei-yu front over east- ern China in early July 2007. In the mean diurnal cycle, the propagation and evolution of MCSs are reasonably cap- tured in model runs forced only with 10-day-averaged diurnal variations at lateral boundaries or even fixed boundary conditions (BCs) without temporal variations, since diurnal forcing still exists in model interior. Over the southern and eastern China, the synoptic environments within which the rainfall episodes develop and evolve may play a much more significant role than in the United States, because the mei-yu front and south- westerly monsoon in early summer are prominent fea- tures that often dictate rainfall distribution (e.g., Lau and Li 1984; Tao and Chen 1987; Ding 1992; Chen 1994; Ding and Chan 2005), as illustrated in many earlier studies (e.g., Wang and Orlanski 1987; Wang et al. 1993; Chen 2004; Chen et al. 2004). Wang et al. (2012) iden- tified several synoptic patterns favorable and unfavor- able to the development and propagation of episodes to the lee of the TP. Although strong upper-level forcing is indeed beneficial, their results also point to the impor- tance of the LLJ in the maintenance and propagation of episodes, as it is the one among all necessary ingredients that differs the most between long-lived and short-lived episodes. Note also that in Sun and Zhang (2012), the observations show larger differences in rainfall patterns among successive days without temporal averaging, and such variations are only reproduced in experiments using real analyses (i.e., those containing synoptic evolution during the case period) as BCs (cf. their Figs. 5 and 6). b. Motivation and study objectives The studies reviewed above have greatly improved our understanding about the behavior of rainfall epi- sodes and MCSs and their relationships with elevated terrain in the diurnal cycle. However, case studies are still few compared to climatological studies, especially in East Asia. Thus, how exactly some of the factors in- cluding synoptic conditions and local circulations in- fluence real episode cases remain unclear. One focus is the LLJ, recognized as a crucial factor for system orga- nization, maintenance, and propagation (e.g., Trier et al. 2006, 2010; Chen et al. 2013; Sun and Zhang 2012) and controlled largely by synoptic conditions (e.g., Chen 2004; Wang et al. 2012), but few studies have addressed its roles in real cases, at least not in the context of rain- producing episodes. The present study focuses on the region near and up- stream from Taiwan. Adapted from Wang et al. (2005a) as an example, Fig. 1 shows that the eastward movement of convection across the Taiwan Strait overnight is quite visible in the climatology in mei-yu season (also Fig. 8 of Johnson 2011). Coherent with this phase progression, long-term rainfall records over the coastal plains in western Taiwan also reveal a secondary early-morning peak in addition to the primary afternoon peak (e.g., Chen et al. 1999; Chen et al. 2005; Kerns et al. 2010). Shown in Fig. 2,ourdomainofHovmöller calculation (208-268N, 112.58-122.58E) covers the coastal region of southeastern China (SEC), the Taiwan Strait, and Taiwan, and is just south of the region of the MPS at the lower Yangtze River valley (YRV; near 1208E) studied by Bao et al. (2011). The topography of the Nanling Moun- tains, Wuyi Mountains, and coastal hills in SEC is on the order of 500-750 m with isolated peaks over 1.5 km. The eighth intensive observing period (IOP-8, 14- 17 June) of the Southwest Monsoon Experiment/ Terrain-influenced Monsoon Rainfall Experiment (SoWMEX/TiMREX) in 2008 (Lee et al. 2009; Jou et al. 2011) is selected as our study target. The Hovmöller di- agram produced using the Tropical Rainfall Measuring Mission (TRMM) 3B42 rain rates and the National Centers for Environmental Prediction (NCEP) analyses (to be described in section 2) during our case period shows two rainfall episodes in close association with the LLJ (Fig. 3a). The first moves eastward from 113.98 to 122.28E over 0300 UTC 13 June-0400 UTC 14 June at a zonal speed of ;9.4 m s21, with a phase across the Taiwan Strait consistent with Fig. 1, and the second moves westward from 122.18 to 113.68Eat;7.3 m s21 over 2100 UTC 15 June-0600 UTC 17 June, slightly leading the jet. While the second, westward-traveling episode is unique, only two major episodes in six days (including one moving west- ward) are in contrast to the typical daily occurrences in the United States and somewhatimplyamoredominant role of synoptic evolution. The first part of this study (Wang et al. 2014, hereafter Part I) investigated the controlling factor(s) on the organization and propagation of the two rainfall epi- sodes during IOP-8, with an emphasis on the role played by the LLJ at synoptic and regional scale (cf. Fig. 3a, and also Part I), and its results will be briefly described in section 3. In this second part of the study, the relative importance of, or the roles played by, synoptic condi- tions and terrain-associated local circulations-diurnal variations and finescale aspects in the SoWMEX IOP-8 case are further examined, mainly through a comparison among the control simulation (CTL) and three sensi- tivity tests. The roles of topography include those in SEC and Taiwan (cf. Fig. 2), and both thermodynamic effects such as local solenoidal circulations and dynamic effects due to the presence of the terrain itself. Davis and Lee (2012) and Xu et al. (2012) have also studied the IOP-8 case from an observational standpoint, and will be referred to later. The remaining sections are arranged as the following. The data and methodology are described in section 2. Section 3 provides an overview on synoptic evolution and the main results in Part I.Insection 4, the model configuration and sensitivity tests are described. The model results are presented in section 5 and compared and discussed, and finally the conclusions and summary are given in section 6. 2. Data and methodology In this study (Part II), the NCEP gridded final (FNL) analyses, produced by the Global Forecast System (GFS) every 6 h (at 0000, 0600, 1200, and 2800 UTC, where LST 5 UTC 1 8 h) at a horizontal resolution of 18318 latitude-longitude at 26 pressure (p) levels (from 1000 to 10 hPa, surface level included), during our case period are used both for the synoptic description and model experiments as initial condition (IC) and BCs. Infrared (IR) cloud imageries from the Japanese Multi- functional Transport Satellite (MTSAT) at 1-h inter- vals are also used to document the convective activities. For overall rainfall distribution in the study area (cf. Fig. 2), the TRMM 3B42 rain rates on a 0.25830.258 grid every 3 h (Huffman et al. 2007) are employed. For more detailed rainfall over Taiwan, hourly accumula- tion data from a network of nearly 400 automatic gauges (Hsu 1998) are utilized. The observed rainfall distribu- tions will be compared with model results later. Beside the CTL experiment, three sensitivity tests are performed to isolate the effects of synoptic conditions and diurnal circulations here in Part II. The results of CTL are validated in Part I with the addition of radar observations and the readers are referred to Part I for details. Through a comparison among results in CTL and the sensitivity tests, effects of synoptic conditions, diurnal circulations, land-sea contrast, and topography can be isolated and discussed. Below, we briefly review the synoptic evolution and Part I results. 3. Overview on synoptic evolution and results in Part I a. Overview of synoptic evolution The synoptic environment and evolution during IOP-8 (12-17 June 2008) are described in detail in section 3a of Part I, and therefore are only briefly reviewed here. In Fig. 4, the MTSAT IR cloud imageries near 0000 UTC 12-17 June are shown, overlaid with surface mei-yu front, wind-shift lines/troughs at 850, 700, and 500 hPa, and the area of 850-hPa LLJ ($12 m s21), all determined using NCEP FNL analyses. During the entire period, the sur- face mei-yu front advanced at most to the coast of SEC and never reached Taiwan. The convection was generally to the south and east of the front (i.e., prefrontal) and largely coincided with the LLJ (Fig. 4), consistent with TRMM data (Part I). Thus, the southwesterly LLJ is found to have much control over the general region of convection and thus dictates the overall rainfall pattern in the Hovmöller space, as shown in Fig. 3a. During 12-14 June (Figs. 4a-c), the front, wind-shift line, and trough in the lower to midtroposphere over SEC developed concurrently. This front-trough system exhibited a northwestward tilt with height and moved eastward on 13-14 June, together with the LLJ and convection. On 15 June, the surface mei-yu front reached the SEC coast and the LLJ also moved south and east of Taiwan (Fig. 4d). When another 500-hPa trough approached on 16 June (Fig. 4e), a front-trough system again deepened around 17 June and caused the mei-yu front, LLJ, and convection to shift northwest- ward (i.e., backward from their previous location; Fig. 4f) as discussed in Part I. During our case period, 500-hPa shortwave troughs are also often found to col- locate with deep convection near the SEC coast, and thus might be a response to latent heating there. How- ever, this is beyond the focus of our current study and will not be further elaborated. The synoptic analysis above (and in Part I) indicates that the LLJ developed, migrated, and extended eastward and northeastward during 12-15 June, as summarized in Fig. 5a, mainly in response to the rapid deepening and eastward movement of a frontal disturbance across SEC (cf. Figs. 4a-d). Similarly, the LLJ retreated northwestward over 16-18 June (Fig. 5b) together with the area of convection when another front-trough system devel- oped over SEC (cf. Figs. 4e,f). The deep structure and northwestward tilt (with height) of this system at 1200 UTC 16 June is depicted in Fig. 6, and its devel- opment in response to upper-level forcing caused the LLJ to appear to the west of its previous location (cf. Figs. 4 and 5b). Since the convection was closely tied to the LLJ during IOP-8, the two episodes traveled in op- posite directions in the Hovmöller diagram (cf. Fig. 3a). b. Overview of results in Part I In Part I, except for synoptic discussion, the CTL simulation using the Cloud-Resolving Storm Simulator (CReSS; Tsuboki and Sakakibara 2002, 2007) is also validated and used to study the structure and organiza- tion of convection and episodes. Below, the findings in Part I in these aspects are also overviewed to provide more complete background information on our case and assist later discussion. First, the structure and organi- zation of convection embedded in the two episodes are shown to be different and were primarily controlled by the vertical wind shear associated with the LLJ (Figs. 8, 11, and 16 of Part I). The first episode was associated with stronger shear and consisted of better-organized, squall-line-type MCSs. On the other hand, the vertical shear was not as strong during the second episode, which comprised scattered and less-organized cells that trav- eled north-northeastward with the background flow. Overall, the dominant role of synoptic conditions found during SoWMEX IOP-8 in Part I is in contrast with and complimentary to those obtained by Trier et al. (2010) for the central United States and by Sun and Zhang (2012) for the YRV in China to a lesser degree. At the local scale, the observed daily rainfall distri- butions over Taiwan from the gauges during 13-16 June are shown in Figs. 7a-d. It is evident that the two epi- sodes each caused heavy rainfall (.200 mm) in Taiwan, on 14 and 16 June, respectively. The event on 14 June was more widespread and covered western Taiwan south of 24.58N(Fig. 7b). The one on 16 June occurred mainly over the southwestern plains while a similar rainfall pattern was also observed on 15 June but with less amount (Figs. 7c,d). Even though the synoptic evo- lution and LLJ determined the general pattern and or- ganization of the episodes at the regional scale, detailed convective activities at the local scale are clearly also subject to control of mesoscale to cloud-scale processes, and the roles of MPS and its extent of influence in the IOP-8 case are still unclear. Therefore, here in Part II, we further study the local aspects of our case and the roles of synoptic conditions and terrain-associated circulations in greater detail. 4. Model configuration and sensitivity tests The CReSS model and its configuration used in this study (in CTL) are described in details in Part I, and thus only necessary information on the experiment is given here. The full-physics simulation of CTL utilizes the 6-hourly NCEP FNL analyses from 0000 UTC 12 June to 0000 UTC 18 June 2008 as IC and lateral BCs (LBCs), with a horizontal grid spacing of 2.5 km, a single domain of 1500 3 1200 3 20 km3 (cf. Fig. 2), and 40 vertical levels to reproduce the event and serve as the bench- mark for comparison (Tables 1 and 2). At the lower boundary, real terrain and observed weekly mean sea surface temperature (SST; Reynolds et al. 2002) are provided (Fig. 8). In all experiments, no additional data assimilation, except for that used in NCEP GFS during the production of the gridded analyses, is performed. A total of three sensitivity tests are designed to isolate the effects associated with synoptic conditions and diurnal variations induced by land-sea distribution and topog- raphy (Table 2). The first is the diurnal (DIU) run that includes only diurnal variations. From NCEP FNL analyses during 0000 UTC 12 June-0000 UTC 18 June 2008 (i.e., the integration period), mean fields at 0000, 0600, 1200, and 1800 UTC are computed respectively by averaging data at the same UTC time. These mean fields thus contain an averaged diurnal cycle overlaid on top of a mean, time-invariant synoptic background, and are used as IC/LBCs in DIU. In other words, the IC in DIU is the mean fields at 0000 UTC and the same cycle is re- peated every day (for six times) in the LBCs (Table 2). Similar to Trier et al. (2010), the DIU run is designed to isolate only the effects from diurnal circulations, and thus both land-sea properties and terrain are kept as in CTL. The second sensitivity experiment is named diurnal- no terrain (DNT) and tests the effect of topography in the diurnal cycle. This is done by removing all terrain above 10m in the model (i.e., topography .10 m is set to 10 m), and all other settings are identical to DIU including the IC and LBCs (Table 2). Thus, the DNT experiment contains only the effects from land-sea distributions in model interior, and its differences from DIU are attributable to the topography. A similar run is adopted in Sun and Zhang (2012). The third and final test is the synoptic (SNP) run, designed to include only the synoptic evolution during our case period without diurnal variations or land-sea con- trast. As mentioned, the 6-day mean synoptic background during our case period (0000 UTC 12-18 June) is ob- tained through time averaging as shown in Fig. 9. Then, the averaged departures (i.e., anomalies) in the NCEP analyses at 0000, 0600, 1200, and 1800 UTC are then computed, respectively, as the differences between the mean fields at each time in the diurnal cycle (i.e., the LBCs used in DIU) and the 6-day mean. Finally, these averaged anomalies, depicted in Fig. 10 at 850 hPa, at the corresponding time (within the diurnal cycle) are subtracted from the original NCEP analyses to construct the IC/LBCs used in SNP (Table 2). With mean di- urnal anomalies removed, these IC/LBCs thus contain a smoother synoptic evolution. For consistency, the underlying surface in SNP is set to ocean only with SST fixed at 299.6 K, which is the mean value inside the model domain (cf. Fig. 8), to remove all driving mech- anisms of local circulations. In Fig. 10, the mountainous regions in SEC tend to be cooler at 1800 and 0000 UTC (0200 and 0800 LST) and warmer at 0600 and 1200 UTC (1400 and 2000 LST), and more so at 1000 hPa (not shown). The southerly winds over the SCS reach a maxi- mum at 1800 UTC, also consistent with Chen et al. (2009, 2013). 5. Results of model experiments a. Overall results and control experiment Using the configuration described earlier, results of the CTL experiment are first briefly described here, while further details can be found in Part I. For easy comparison with the Hovmöller diagram in Fig. 3a, rain rates are aggregated into resolutions of 25 km and 3 h and wind speeds at 1594 m (close to 850 hPa) every 6 h from the CTL (Fig. 3b), while the diagram at the original resolution of rain rates (2.5 km and 15 min) is shown in Fig. 11a. The two rainfall episodes during IOP-8 are well captured in CTL. A remarkable similarity exists be- tween Figs. 3a and 3b, indicating a highly successful re- production of the rainfall systems (and low-level winds) in the Hovmöller domain. Even so, the phase speeds of the two episodes in CTL, at about 7.2 ms21 eastward and 4.6 m s21 westward, respectively, are somewhat slower than those determined from the TRMM data (9.4 and 7.3 m s21). Nonetheless, it should be noted that without smoothing, the streaks corresponding to episode 1 in Fig. 11a appears faster and thus more comparable to the observation. The 24-h rainfall distributions over Taiwan during 13-16 June as simulated in CTL (Figs. 7e-h) also agree closely with the observations (Figs. 7a-d). Both heavy- rainfall events during IOP-8 are well captured, with widespread heavy rainfall over the windward slopes of the Central Mountain Range (CMR) and peak values greater than 300 mm on 14 June and over the southwestern plains (also peaking near 300 mm) on 16 June (Figs. 7b,d). Note also that the rainfall along the southwestern coast of Taiwan on 15 June is cap- tured as well (Fig. 7c). Overall, although differences inevitably exist between the observations and model results, the CTL simulation is highly successful, as verified in Part I. b. The synoptic experiment As described, the SNP experiment contains only the synoptic evolution during our case period without the diurnal variations in the IC/LBCs, and a flat ocean surface is specified at the bottom with a fixed SST. Except for these differences, all other settings in SNP are the same as in CTL (Tables 1 and 2). First, the Hovmöller dia- gram of model rain rates and wind speeds in SNP using identical domain (and resolution) is compared with that in CTL (Figs. 11a,b). The overall rainfall pattern in SNP resembles that in CTL, and the two episodes are still reproduced at similar locations and coincide with the LLJ without diurnal variations. This result indicates the dominant role played by synoptic forcing in rainfall production in space and time in the present case, and is not surprising for the following reasons. First, synoptic conditions (particularly the strong and persistent LLJ) of the case must be very favorable due to its heavy-rainfall nature (cf. Figs. 4 and 7). Second, only two rainfall epi- sodes appeared on specific days and one traveled west- ward in the Hovmöller space, and such characteristics are in contrast to the more-or-less regular daily occurrences under the control of the terrain-induced MPS (Trier et al. 2010; Sun and Zhang 2012). The result also confirms the finding in Part I that the overall distribution of convection in this case was largely dictated by the LLJ, and the synoptic evolution caused the westward movement of the second episode. When Figs. 11a and 11b are compared more closely, however, some differences in details of convection still exist between SNP and CTL and are attributable to local circulations arising from land-sea contrast and/or to- pography (since they are present in CTL but absent in SNP), finer-scale processes, or to differences due to the nonlinearity in the development and evolution of con- vection when certain conditions are changed in SNP. For example, embedded in or near episode 1, the streaks labeled as C1, C2, and C3 differ to some extent in their location and span-duration, although they all propagate eastward at similar speeds (Figs. 11a,b). Figure 12 compares the convection and surface winds in plain views between CTL and SNP at 0600 UTC (1400 LST) on 13 and 17 June when the two episodes moved over SEC. Although the overall distribution of convection may not seem vastly different, the convective systems in CTL tend to be more active and better organized over or near the terrain in SEC on both occasions, compared to those in SNP (Fig. 12). Thus, some modulation effects by local solenoids associated with topography and land-sea contrast, similar to those found by Huang et al. (2010) near the eastern TP but likely significantly weaker, are still discernible in the model even though the hills over SEC are relatively flat and only about 0.5-1 km in ele- vation (e.g., Bao et al. 2011). One other region that exhibits larger differences be- tween CTL and SNP during 13-16 June is near and immediately upstream from Taiwan (Figs. 11a,b, pink dashed ovals), which has steep and complex terrain (cf. Fig. 2). The rain rates in CTL tend to be larger and slightly more upstream (mostly west of 1218E), com- pared to SNP without the island and its topography. The daily rainfall distributions over Taiwan during 13-16 June in SNP (Figs. 7i-l) in fact differ quite significantly from CTL and thus the observations (Figs. 7a-h). The heavy rainfall on 14 June (from episode 1) is on the windward slopes of the CMR and along the western plains in reality (and CTL), but shifted eastward in SNP to the interior (and southeast) of Taiwan (Figs. 7b,f,j). The local enhancement of convection offshore from southwestern Taiwan in CTL, most likely due to terrain blocking (e.g., Wang et al. 2005b; Ruppert et al. 2013), is reduced in SNP even though a substantial amount of rainfall is still received over southwestern Taiwan from episode 1. On 15 June, the rainfall in SNP is similarly shifted to eastern Taiwan, and the rain offshore from western and northwestern Taiwan on 13 June is also much reduced (Fig. 7). The biggest difference, however, occurs on 16 June from episode 2, where the heavy rainfall over the southwestern plains is completely missed in SNP (Figs. 7d,h,l) without the island. Thus, below we further investigate the cause of such a large difference on 16 June. Based on observations, Xu et al. (2012) analyzed the heavy-rainfall events in Taiwan during 14-16 June. For the one on 16 June, they found a local, mesoscale cold pool (by 28-48C) in the lowest 500 m over southwestern Taiwan extending about 50 km into the upstream ocean (also Davis and Lee 2012), most likely formed as a result of evaporation of previous, persistent precipitation and trapped by the terrain in Taiwan from spreading or moving. Xu et al. (2012) hypothesized that the cold pool lifted the oncoming flow so that convective clusters de- veloped offshore then moving onshore repeatedly to cause the heavy rainfall through a ''back-building-quasi- stationary'' process. The model-simulated surface wind and potential temperature (u) and upward motion near 1.6 km from 2100 UTC 15 June to 0600 UTC 16 June in CTL and SNP are shown in Fig. 13. In CTL, convective cells in the south-southwesterly flow are repeatedly generated within a short distance upstream from Taiwan's south- western coast, then moved onshore (Figs. 13a-d)asin the observation (also Part I and Xu et al. 2012). Linked to this prolonged rainfall and the rain occurred earlier on 15 June (details not shown, cf. Figs. 7c,g), cool air with u below 298 K appears over the western and southwestern plains and led to repeated initiation of new convection in the oncoming flow near shore, in agree- ment with Xu et al. (2012). Thus, heavy rainfall over the southwestern plains of Taiwan resulted on 16 June. Vertical cross sections of total mixing ratios of pre- cipitating hydrometeors (rain, snow, plus graupel), wind vectors, and convergence from the CTL run, on the plane along line B-B0 in Fig. 13a, are presented in Fig. 14 at 30-min intervals over 0230-0800 UTC 16 June, to- gether with the 300-K isentrope. With corresponding cells labeled using 15-min outputs, it is clear that the convection in CTL are very active and exhibit multi- cellular characteristics in early 16 June, and the same is true since late 15 June (cf. Fig. 15 in Part I). The cells are also frequently initiated just offshore (roughly within 228-22.58N), above the sloping cold air (with u # 300 K). In Fig. 14, the surface-based cold air often coincides and becomes enhanced by precipitation (thickening), in agreement with Xu et al. (2012) and Davis and Lee (2012). Hence, this heavy-rainfall event on 16 June in Taiwan was tied to the local topography and also very controlled by mesoscale to cloud-scale processes. With no data assimilation, the reproduction of convection and rainfall evolution in Taiwan during 15-16 June in CTL (Figs. 7c,d and 7g,h), on the fourth and fifth days into the simulation, is quite remarkable and indicates the capa- bility of cloud-resolving models like CReSS, given rea- sonable IC/BCs. In Figs. 13e-h from SNP, frequent initiation of new cells upstream from southwestern Taiwan does not occur, even though the surface air above the ocean surrounding southern Taiwan is generally cooler (u ; 298-299 K). In SNP, most cells develop over southern Taiwan Strait west of 1208E, and thus the highest rainfall amount occurs just offshore from western Taiwan (cf. Fig. 7l). Thus, our re- sults here from the CTL and SNP experiments support the hypothesis made by Xu et al. (2012). c. The diurnal experiments with and without terrain The Hovmöller diagrams from DIU and DNT, in which the IC/LBCs contain the same, repeated diurnal cycle on top of a fixed background (i.e., no synoptic evolution) with and without topography (cf. Table 2), are shown in Figs. 11c,d. In these tests designed to isolate the effects associated with diurnal circulations, more- or-less similar patterns occur each day throughout the 6-day integration period as expected, except perhaps for the first 12 h, and such characteristics are similar to those found by Trier et al. (2010), and by Sun and Zhang (2012) to a lesser degree. Thus, the two major rainfall episodes are absent in both DIU and DNT regardless whether the topography is retained, and the overall patterns are dissimilar to that in CTL. Since similar cycles occur each day in DIU and DNT, they are averaged through the 6-day period excluding the first and last 12 h. In DIU (Fig. 15a), the rainfall sig- nals in the Hovmöller domain tend to start near 113.58E around 1800 UTC (0200 LST) then reach 116.58E about 15 h later (near 0900 UTC), at a zonal speed of ;5.7 m s21 (dashed arrow line), somewhat slower than episode 1 in CTL (cf. Fig. 3b). The convective devel- opment around 113.58E is mainly near the SEC coast (not shown) and likely linked to a stronger LLJ and thus the induced convergence from ocean to land near 1800 UTC, in agreement with Chen et al. (2009, 2013). A weak signal continues eastward (dotted arrow line) and may reach 1208E (near the western coast of Taiwan) around 2200 UTC (0600 LST) and produce the morning signal (cf. Fig. 1). Over and immediately upstream from Taiwan, however, rainfall occurs predominately during the afternoon and early evening (about 0200-1500 UTC; Fig. 15a), also consistent with climatology (e.g., Chen et al. 1999; Chen et al. 2005). When the topography is removed in DNT, the aver- aged diurnal cycle remain very similar to that in DIU (Fig. 15), suggesting that the land-sea contrast alone can lead to the repeated pattern without topography in our case, given its synoptic background. However, more overall rainfall is produced and the signals tend to last longer near Taiwan in DIU when its steep terrain is re- tained (Fig. 15a), and this is also true over SEC although the differences are less obvious. Without the terrain, the rainfall signals in DNT also tend to travel faster (Fig. 15b), likely linked to a reduced blocking effect (e.g., Yeh and Chen 2002; Wang et al. 2005b). The surface streamlines and convergence and daily rainfall near Taiwan in DIU and DNT, also averaged over 1200 UTC 12-17 June 2008, are shown in Fig. 16.In DIU, the terrain forces the prevailing south-southwesterly winds to move around it and the air west of the CMR is deflected toward the north, leading to enhanced con- vergence with the environmental flow (from the SSW) over the Taiwan Strait (Fig. 16a), in agreement with Li and Chen (1998), Yeh and Chen (2002), Wang et al. (2005b), and Ruppert et al. (2013). In DNT, a similar deflection is not seen and the flow passes through the island without much interruption (Fig. 16b). Note, how- ever, that stronger convergence still appears near the coast of southwestern Taiwan in DNT, likely due to increased friction from ocean to land (e.g., Wang et al. 2005b)and the convergence between land breeze and prevailing flow at night (e.g., Ruppert et al. 2013). Although the rainfall pattern does not seem to be very different, there is indeed more rain over the Taiwan Strait and Taiwan's western and southwestern plains, but less rain to the north and east of CMR in DIU. Such differences between DIU and DNT are also consistent with those found between CTL and SNP (i.e., the topography of Taiwan tends to increase the total amount of rainfall and shift its location slightly upstream; cf. Figs. 7 and 11). d. Discussion In the previous subsections, the similarity of rainfall signals between CTL and SNP in the Hovmöller space, and the dissimilarity between CTL and DIU/DNT, confirm the dominant roles of synoptic conditions on the general patterns of the episodes in the present case. However, many instances of the initiation and/or en- hancement of convection and rainfall systems at similar location and time are still found in CTL and DIU (Figs. 11a,c, brown circles), and tend to concentrate over 1138- 1148E, or near 1168 or 1208E (i.e., near regions with higher terrain; cf. Figs. 2 and 15). Since the Hovmöller plots from DIU and DNT highly resemble each other, most of those instances also appear in Fig. 11d (with dif- ferent details). Thus, diurnal circulations associated with topography and land-sea contrast still exhibit discernible impacts and can modulate details in the development and propagation of convective systems during IOP-8, both over SEC and Taiwan (Figs. 7, 11, 12,and15). To further clarify the effects from land-sea contrast and those from topography on rainfall in SEC, time se- ries of rain rates averaged over 22.58-26.58N, 1138-1188E (Fig. 2, dashed box) during 12-17 June, as well as their mean diurnal cycles from all four experiments are plotted in Fig. 17. In CTL, consistent with Fig. 3, more rain is produced on 12-13 June over SEC then the rain gradually diminishes on 14 June as the MCSs move to the east (Fig. 17a). When the second episode moves back into SEC, the rainfall increases again on 16-17 June. Although large rainfall variations occur from one day to the next (linked to synoptic evolution), the mean diurnal cycle of rain rates in CTL exhibits a range of 1.2-2.1 mm h21, with maximum (minimum) around 0500 (1500) UTC [i.e., 1300 (2300) LST, Fig. 17a]. In SNP, there are large variations in daily rainfall similar to CTL as expected, with significant amounts on 12, 13, the first part of 14, and 17 June (Fig. 17b). Without land and topography, however, the mean di- urnal cycle in SEC is very flat throughout the day, with a range of only 1.5-1.75 mm h 2 1. On the contrary, except for the first day, the diurnal cycles of rainfall in all other days and their mean cycles are similar to each other over SEC in both DIU and DNT(Figs. 17c,d). While all their daily series possess a clear diurnal cycle, the deviations on different days are small, often within 0.5 mm h21 from the mean cycle. Moreover, the mean cycles in DIU/ DNT resemble that in CTL, with maximum (minimum) near 0500 (1400) UTC. This result again indicates that the land-sea contrast is responsible for much of the di- urnal circulations in SEC during this case, and the to- pography may be important only locally. In DIU, the averaged rain rates during daytime (0000-1200 UTC) are higher than those in CTL, by up to almost 0.5 mm h21 (Figs. 17a,c), suggesting a slight overall suppressing effect on rainfall over SEC from synoptic evolution, pre- sumably mostly during unfavorable (or less favorable) periods. In DNT, the mean cycle has less rainfall at night (1200-2200 UTC) compared to DIU (by 0.2-0.3 mm h21), which likely indicates premature decay of rainfall systems around early evening without terrain or the effects from the absence of a downslope mountain breeze. To explore the association of rainfall with topography over SEC, zonal variations of rain rates at 0000, 0600, 1200, and 1800 UTC and terrain height, averaged over 22.58-26.58Nand1200 UTC 12-17 June(as in Fig.17), in CTL and SNP are shown in Fig. 18. After north-south averaging, relatively high topography of about 550 and 420 m exists near 1148 and 1178E. On average, the peak rain rate in CTL occurs near 113.18E at 1800 UTC, near 114.68E at 0000 UTC, and around 116.68E later at 1200 UTC. These maxima, together with a minor peak near 115.78E at 0600 UTC in Fig. 18a, correspond well to the main propagation signals in DIU (Fig. 15a, the lon- gest dashed plus dotted arrows). Thus, the mean diurnal cycle in CTL again contains clear effects from local circulations over SEC as those in DIU. At 0600 UTC, two rainfall peaks, by far the largest in Fig. 18a, occur near 1178 and 117.58E. With a phase coherent with Fig. 1, these peaks appear to tie closely to the coastal hills and are likely enhanced by upslope/onshore winds of local MPS (sea and valley breezes) and possibly also near- surface destabilization during the daytime. Thus, local influence by the topography in SEC is also apparent in our case. Without land and terrain, similar rainfall peaks near 1178E at 0600 UTC do not exist in the mean diurnal cycle of SNP (Fig. 18b), consistent with the examples in Fig. 12, and the phenomenon of a coherent eastward progression of rainfall signals with time also disappears. 6. Conclusions and summary In this two-part series, rainfall episodes in the Hovmöller (longitude-time) space near and upstream from Taiwan during the SoWMEX/TiMREX IOP-8 (12-17 June 2008) are investigated. Two well-defined episodes oc- curred during this 6-day period in a prefrontal environ- ment: the first moved eastward during 13-14 June while the second moved westward on 16 June, and each pro- duced heavy rainfall (.200 mm) in Taiwan, on 14 and 16 June, respectively. The first part (Part I) seeks to un- derstand the main controlling factor(s) on the organi- zation and propagation of episodes, while in the present paper (Part II), we further examine the relative impor- tance of synoptic conditions and diurnal local circula- tions in the IOP-8 case in greater details mainly through model sensitivity tests. Using the Nagoya University CReSS model at 2.5-km horizontal grid spacing, the control simulation (CTL) successfully reproduced the 6-day event with high realm, including the two episodes, as well as the more detailed development and evolution of convection and the resulted rainfall in the study domain, and is validated in Part I. Here, three sensitivity tests are designed to iso- late effects of synoptic conditions (SNP) only, and those from diurnal circulations from land-sea contrasts with (DIU) and without the topography (DNT) on top of a time-invariant background. In the SNP experiment, the two episodes are repro- duced and the overall distribution of rainfall in the Hovmöller diagram is highly similar to that from CTL (and the observation). The general location and timing of rainfall are thus primarily controlled by synoptic forc- ing, especially the LLJ. On the other hand, the rainfall signals in DIU and DNT runs exhibit repeated cycles that resemble each other every day, and the distributions are thus different from CTL. These results indicate that the synoptic forcing dominated over diurnal effects in determining the overall rainfall distribution at regional scale during SoWMEX IOP-8, confirming the findings in Part I but complementary to the situations studied by Trier et al. (2010) and Sun and Zhang (2012). Even so, diurnal and terrain effects can still exert discernible impacts and modulate the initiation, de- velopment, and evolution of rainfall systems at local scale in many instances commonly found in CTL and DIU/DNT, especially over mountainous regions. In CTL, the hilly region in SEC, after averaging through the case period, exhibits a rainfall diurnal cycle similar to the mean cycle and those occurring each day in DIU/DNT, with daytime maximum and nighttime minimum (near 1300 and 2200 LST). Such a cycle (and local enhancement of rainfall in the afternoon) is absent in SNP where the underlying surface is set to ocean only in the model (i.e., no land or topography). It is perhaps worthy to note that an additional test similar to the SNP may be performed to drive the model with fixed (time averaged) surface fluxes while retaining the land-sea distribution and terrain (i.e., removing only the diurnal cycle of local circulations), and it is suggested as a possible approach for future studies. For rainfall over and near Taiwan, its steep terrain is found to be essential in the present case, for the distri- bution is vastly different in SNP compared to CTL. In general, the dynamical (blocking) and thermal effects of Taiwan's topography produces deflection of the pre- vailing low-level southwesterly flow and convergence upstream, which thus acts to shift the rainfall toward the upstream and increase the amount, consistent with previous studies. Captured nicely in CTL, the heavy- rainfall event over the southwestern plains of Taiwan on 16 June does not occur at all in SNP without the terrain. Detailed analysis of model results indicate that a surface cold pool from evaporation of earlier precipitation formed near the coast of southwestern Taiwan and acted to repeatedly initiate new convection just offshore for a lengthy period to cause the heavy rainfall. Our mod- eling results thus confirm the hypothesis of Xu et al. (2012) based on observations. Acknowledgments. The authors thank Prof. Ben Jong- Dao Jou of National Taiwan University, Taiwan, Prof. Richard H. Johnson of Colorado State University, and Prof. Fuqing Zhang of The Pennsylvania State Uni- versity, as well as the anonymous reviewers for their constructive comments and valuable suggestions. This study was supported jointly by the National Science Council of Taiwan under Grants NSC-99-2111-M-003- 004-MY3 and NSC-102-2119-M-003-003, the Korea Meteorological Administration Research and Devel- opment Program under Grant CATER 2012-2071, and by the National Research Foundation of Korea through a grant provided by the Korean Ministry of Education, Science, and Technology in 2014 (Grant 200603874). REFERENCES Ahijevych, D. A., C. A. Davis, R. E. Carbone, and J. D. Tuttle, 2004: Initiation of precipitation episodes relative to elevated terrain. J. Atmos. Sci., 61, 2763-2769, doi:10.1175/JAS3307.1. Bao, X., F. Zhang, and J. Sun, 2011: Diurnal variations of warm-season precipitation east of the Tibetan Plateau over China. Mon. Wea. Rev., 139, 2790-2810, doi:10.1175/ MWR-D-11-00006.1. Carbone, R. E., and J. D. Tuttle, 2008: Rainfall occurrence in the U.S. warm season: The diurnal cycle. J. Climate, 21, 4132- 4146, doi:10.1175/2008JCLI2275.1. _____, _____, D. Ahijevych, and S. B. Trier, 2002: Inferences of predictability associated with warm season precipita- tion episodes. J. Atmos. Sci., 59, 2033-2056, doi:10.1175/ 1520-0469(2002)059,2033:IOPAWW.2.0.CO;2. Chen, G. T.-J., 1994: Large-scale circulations associated with the East Asian summer monsoon and the mei-yu over south China and Taiwan. J. Meteor. Soc. Japan, 72, 959-983. _____, 2004: Research on the phenomena of meiyu during the past quarter century: An overview. The East Asian Monsoon, C.-P. Chang, Ed., World Scientific, 357-403. _____, C.-C. Wang, C.-S. Chang, and T.-S. Wang, 2005: Character- istics of mesoscale climatology of precipitation and heavy rainfall events over central Taiwan in Mei-yu season (in Chi- nese with English abstract). Atmos. Sci., 33, 49-76. Chen, G., W. Sha, and T. Iwasaki, 2009: Diurnal variation of pre- cipitation over southeastern China: 2. Impact of the diurnal monsoon variability. J. Geophys. Res., 114, D21105, doi:10.1029/2009JD012181. _____, _____, M. Sawada, and T. Iwasaki, 2013: Influence of summer monsoon diurnal cycle on moisture transport and pre- cipitation over eastern China. J. Geophys. Res. Atmos., 118, 3163-3177, doi:10.1002/jgrd.50337. Chen, H., R. Yu, J. Li, W. Yuan, and T. Zhou, 2010: Why nocturnal long-duration rainfall presents an eastward-delayed diurnal phase of rainfall down the Yangtze River Valley. J. Climate, 23, 905-917, doi:10.1175/2009JCLI3187.1. Chen, T.-C., M.-C. Yen, J.-C. Hsieh, and R. W. Arritt, 1999: Di- urnal and seasonal variations of the rainfall measured by the Automatic Rainfall and Meteorological Telemetry System in Taiwan. Bull. Amer. Meteor. Soc., 80, 2299-2312, doi:10.1175/ 1520-0477(1999)080,2299:DASVOT.2.0.CO;2. _____, S.-Y. Wang, W.-R. Huang, and M.-C. Yen, 2004: Variation of the East Asian summer monsoon rainfall. J. Climate, 17, 744-762, doi:10.1175/1520-0442(2004)017,0744:VOTEAS.2.0.CO;2. Dai, A., F. Giorgi, and K. E. Trenberth, 1999: Observed and model- simulated diurnal cycles of precipitation over the contiguous United States. J. Geophys. Res., 104, 6377-6402, doi:10.1029/ 98JD02720. Davis, C. A., and W.-C. Lee, 2012: Mesoscale analysis of heavy rainfall episodes from SoWMEX/TiMREX. J. Atmos. Sci., 69, 521-537, doi:10.1175/JAS-D-11-0120.1. _____, K. W. Manning, R. E. Carbone, S. B. Trier, and J. D. Tuttle, 2003: Coherence of warm-season continental rainfall in numerical weather prediction models. Mon. Wea. Rev., 131, 2667-2679, doi:10.1175/1520-0493(2003)131,2667: COWCRI.2.0.CO;2. Ding, Y., 1992: Summer monsoon rainfalls in China. J. Meteor. Soc. Japan, 70, 373-396. _____, and J. C.-L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117-142, doi:10.1007/ s00703-005-0125-z. Fovell, R. G., G. L. Mullendore, and S.-H. Kim, 2006: Discrete propagation in numerically simulated nocturnal squall lines. Mon. Wea. Rev., 134, 3735-3752, doi:10.1175/MWR3268.1. Holton, J. R., 1967: The diurnal boundary layer wind oscillation above sloping terrain. Tellus, 19A, 199-205, doi:10.1111/ j.2153-3490.1967.tb01473.x. Hsu, J., 1998: ARMTS up and running in Taiwan. Väisälä News, 146, 24-26. Huang,H.-L., C.-C. Wang,G. T.-J. Chen,andR.E. Carbone, 2010: The role of diurnal solenoidal circulation on propagating rainfall episodes near the eastern Tibetan Plateau. Mon. Wea. Rev., 138, 2975-2989, doi:10.1175/2010MWR3225.1. Huffman, G. J., and Coauthors, 2007: The TRMM Multisatellite Precipitation Analysis: Quasi-global, multiyear, combined- sensor precipitation estimates at fine scale. J. Hydrometeor., 8, 38-55, doi:10.1175/JHM560.1. Johnson, R. H., 2011: Diurnal cycle of monsoon convection. The Global Monsoon System: Research and Forecast, C.-P. Chang et al., Eds., World Scientific, 257-276. Jou, B. J.-D., W.-C. Lee, and R. H. Johnson, 2011: An overview of SoWMEX/TiMREX and its operation. The Global Monsoon System: Research and Forecast, 2nd ed. C.-P. Chang, Ed., World Scientific, 303-318. Kerns, B. W. J., Y.-L. Chen, and M.-Y. Chang, 2010: The diurnal cycleof winds, rain, and cloudsover Taiwan during the mei-yu, summer, and autumn rainfall regimes. Mon. Wea. Rev., 138, 497-516, doi:10.1175/2009MWR3031.1. Laing, A. G., and J. M. Fritsch, 1997: The global population of mesoscale convective complexes. Quart. J. Roy. Meteor. Soc., 123, 389-405, doi:10.1002/qj.49712353807. Lau, K.-M., and M.-T. Li, 1984: The monsoon of East Asia and its global associations-A survey. Bull. Amer. Meteor. Soc., 65, 114-125, doi:10.1175/1520-0477(1984)065,0114: TMOEAA.2.0.CO;2. Lee, W.-C., B. J. D. Jou, C.-R. Chen, and J.-A. Moore, 2009: Overview of SoWMEX/ TiMREX. 34th Conf. on Radar Me- teorology, Williamsburg, VA, Amer. Meteor. Soc., 9B.2. [Available online at http://ams.confex.com/ams/34Radar/ techprogram/paper_156254.htm.] Li, J., and Y.-L. Chen, 1998: Barrier jets during TAMEX. Mon. Wea. Rev., 126, 959-971, doi:10.1175/1520-0493(1998)126,0959: BJDT.2.0.CO;2. Maddox, R. A., 1980: Mesoscale convective complex. Bull. Amer. Meteor.Soc.,61,1374-1387,doi:10.1175/1520-0477(1980)061,1374: MCC.2.0.CO;2. Pritchard, M. S., M. W. Moncrieff, and R. C. J. Somerville, 2011: Orogenic propagating precipitation systems over the United States in a global climate model with embedded explicit convection. J. Atmos. Sci., 68, 1821-1840, doi:10.1175/ 2011JAS3699.1. Reynolds, R. W., N. A. Rayner, T. M. Smith, D. C. Stokes, and W. Wang, 2002: An improved in situ and satellite SST anal- ysis for climate. J. Climate, 15, 1609-1625, doi:10.1175/ 1520-0442(2002)015,1609:AIISAS.2.0.CO;2. Rotunno, R., J. B. Klemp, and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci., 45, 463-485, doi:10.1175/1520-0469(1988)045,0463:ATFSLL.2.0.CO;2. Ruppert, J. H., Jr., R. H. Johnson, and A. K. Rowe, 2013: Di- urnal circulations and rainfall in Taiwan during SoWMEX/ TiMREX (2008). Mon. Wea. Rev., 141, 3851-3872, doi:10.1175/ MWR-D-12-00301.1. Sun, J., andF. Zhang, 2012:Impacts ofmountain-plains solenoid on diurnal variations of rainfalls along the mei-yu front over the East China plains. Mon. Wea. Rev., 140, 379-397, doi:10.1175/ MWR-D-11-00041.1. Tao, S., and L. Chen, 1987: A review of recent research on the East Asian summer monsoon in China. Monsoon Meteorology, C.-P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 60-92. Trier, S. B., C. A. Davis, D. A. Ahijevych, and M. L. Weisman, 2006: Mechanisms supporting long-lived episodes of propa- gating nocturnal convection within a 7-day WRF model sim- ulation. J. Atmos. Sci., 63, 2437-2461, doi:10.1175/JAS3768.1. _____, _____, and _____, 2010: Environmental controls on the sim- ulated diurnal cycle of warm-season precipitation in the con- tinental United States. J. Atmos. Sci., 67, 1066-1090, doi:10.1175/2009JAS3247.1. Tripoli, G. J., and W. R. Cotton, 1989: Numerical study of an observed orogenic mesoscale convective system. Part I: Simulated genesis and comparison with observations. Mon. Wea. Rev., 117, 273-304, doi:10.1175/1520-0493(1989)117,0273:NSOAOO.2.0.CO;2. Tsuboki, K., and A. Sakakibara, 2002: Large-scale parallel com- puting of cloud resolving storm simulator. High Performance Computing, H. P. Zima et al., Eds., Springer, 243-259. _____, and _____, 2007: Numerical prediction of high-impact weather systems. 17th IHP Training Course in 2007, HyARC, Nagoya University and UNESCO, 273 pp. Tuttle, J. D., and R. E. Carbone, 2004: Coherent regeneration and the role of water vapor and shear in a long-lived convec- tive episode. Mon. Wea. Rev., 132, 192-208, doi:10.1175/ 1520-0493(2004)132,0192:CRATRO.2.0.CO;2. _____, and C. A. Davis, 2006: Corridors of warm season pre- cipitation in the central United States. Mon. Wea. Rev., 134, 2297-2317, doi:10.1175/MWR3188.1. Wallace, J. M., 1975: Diurnal variations in precipitation and thunderstorm frequency over the conterminous United States. Mon. Wea. Rev., 103, 406-419, doi:10.1175/ 1520-0493(1975)103,0406:DVIPAT.2.0.CO;2. Wang, B., and I. Orlanski, 1987: Study of a heavy rain vortex formed over the eastern flank of the Tibetan Plateau. Mon. Wea. Rev., 115, 1370-1393, doi:10.1175/1520-0493(1987)115,1370: SOAHRV.2.0.CO;2. Wang, C.-C., G. T.-J. Chen, and R. E. Carbone, 2004: A climatol- ogy of warm-season cloud patterns over East Asia based on GMS infrared brightness temperature observations. Mon. Wea. Rev., 132, 1606-1629, doi:10.1175/1520-0493(2004)132,1606: ACOWCP.2.0.CO;2. _____, _____, and _____, 2005a: Variability of warm-season cloud episodes over East Asia based on GMS infrared brightness temperature observations. Mon. Wea. Rev., 133, 1478-1500, doi:10.1175/MWR2928.1. _____, _____, T.-C. Chen, and K. Tsuboki, 2005b: A numerical study on the effects of Taiwan topography on a convective line during the mei-yu season. Mon. Wea. Rev., 133, 3217-3242, doi:10.1175/MWR3028.1. _____, _____, and S.-Y. Huang, 2011a: Remote trigger of deep con- vection by cold outflow over the Taiwan Strait in the mei-yu season: A modeling study of the 8 June 2007 case. Mon. Wea. Rev., 139, 2854-2875, doi:10.1175/2011MWR3613.1. _____, H.-L. Huang, J.-L. Li, T.-M. Leou, and G. T.-J. Chen, 2011b: An evaluation on the performance of the CWB NFS model in the prediction of warm-season rainfall distribution and prop- agation over the East Asian continent. Terr. Atmos. Oceanic Sci., 22, 49-69, doi:10.3319/TAO.2010.07.13.01(A). _____, G. T.-J. Chen, H.-L. Huang, R. E. Carbone, and S.-W. Chang, 2012: Synoptic conditions associated with propagating and nonpropagating cloud/rainfall episodes during the warm sea- son over the East Asian continent. Mon. Wea. Rev., 140, 721- 747, doi:10.1175/MWR-D-11-00067.1. _____, J. C.-S. Hsu, G. T.-J. Chen, and D.-I. Lee, 2014: A study of two propagating heavy-rainfall episodes near Taiwan during SoWMEX/TiMREX IOP-8 in June 2008. Part I: Synoptic evo- lution, episode propagation, and model control simulation. Mon. Wea. Rev., 142, 2619-2643, doi:10.1175/MWR-D-13-00331.1. Wang, W., Y.-H. Kuo, and T. T. Warner, 1993: A diabatically driven mesoscale vortex in the lee of the Tibetan Plateau. Mon. Wea. Rev., 121, 2542-2561, doi:10.1175/1520-0493(1993)121,2542: ADDMVI.2.0.CO;2. Xu, W., and E. J. Zipser, 2011: Diurnal variations of precipitation, deep convection, and lightning over and east of the eastern Tibetan Plateau. J. Climate, 24, 448-465, doi:10.1175/ 2010JCLI3719.1. _____, _____, Y.-L. Chen, C. Liu, Y.-C. Liou, W.-C. H. Lee, and B. J.-D. Jou, 2012: An orography-associated extreme rainfall event during TiMREX: Initiation, storm evolution, and maintenance. Mon. Wea. Rev., 140, 2555-2574, doi:10.1175/ MWR-D-11-00208.1. Yeh, H.-C., and Y.-L. Chen, 2002: The role of offshore conver- gence on coastal rainfall during TAMEX IOP 3. Mon. Wea. Rev., 130, 2709-2730, doi:10.1175/1520-0493(2002)130,2709: TROOCO.2.0.CO;2. Yu, R., T. Zhou, A. Xiong, Y. Zhu, and J. Li, 2007: Diurnal vari- ations of summer precipitation over contiguous China. Geo- phys. Res. Lett., 34, L01704, doi:10.1029/2006GL028129. Zhong, S., J. D. Fast, and X. Bian, 1996: A case study of the Great Plains low-level jet using wind profiler network data and a high- resolution mesoscale model. Mon. Wea. Rev., 124, 785-806, doi:10.1175/1520-0493(1996)124,0785:ACSOTG.2.0.CO;2. Zhou, T., R. Yu, H. Chen, A. Dai, and Y. Pan, 2008: Summer precipitation frequency, intensity, and diurnal cycle over China: A comparison of satellite data with rain gauge observations. J. Climate, 21, 3997-4010, doi:10.1175/2008JCLI2028.1. CHUNG-CHIEH WANG AND JASON CHIEH-SHENG HSU Department of Earth Sciences, National Taiwan Normal University, Taipei, Taiwan GEORGE TAI-JEN CHEN Department of Atmospheric Sciences, National Taiwan University, Taipei, Taiwan DONG-IN LEE Department of Environmental Atmospheric Sciences, Pukyong National University, Busan, South Korea (Manuscript received 9 October 2013, in final form 8 March 2014) Corresponding author address: Prof. Chung-Chieh Wang, De- partment of Earth Sciences, National Taiwan Normal University, No. 88, Sec. 4, Ting-Chou Rd., Taipei, 11677 Taiwan. E-mail: [email protected] (c) 2014 American Meteorological Society |