WANG Lei, and GUO Zhiliang

1) State Key Laboratory of Tropical Oceanography, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou 510301, P. R. China

2) Department of Atmospheric Sciences, School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, P. R. China

Modulation of Tropical Cyclogenesis over the South China Sea by ENSO Modoki During Boreal Summer

WANG Lei1),2),*, and GUO Zhiliang2)

1) State Key Laboratory of Tropical Oceanography, South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangzhou 510301, P. R. China

2) Department of Atmospheric Sciences, School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, P. R. China

This study examines the modulation of tropical cyclogenesis over the South China Sea (SCS) by the El Niño-Southern Oscillation (ENSO) Modoki during the boreal summer. Results reveal that there were more tropical cyclones (TCs) formed over the SCS during central Pacific warming years and less TC frequency during central Pacific cooling years. How different environmental factors (including low-level relative vorticity, mid-level relative humidity, vertical wind shear, and potential intensity) contribute to this influence is investigated, using a genesis potential (GP) index developed by Emanuel and Nolan. Composite anomalies of the GP index are produced for central Pacific warming and cooling years separately, which could account for the changes of TC frequency over the SCS in different ENSO Modoki phases. The degree of contribution by each factor is determined quantitatively by producing composites of modified indices in which only one of the contributing factors varies, with the others set to climatology. The results suggest that the vertical wind shear and low-level relative vorticity, which are associated with the ENSO Modoki-induced anomalous circulations in Matsuno-Gill patterns, make the largest contributions to the ENSO Modoki modulation of tropical cyclogenesis over the SCS as implied by the GP index. These results highlight the important roles of dynamic factors in the modulation of TC frequency over the SCS by the ENSO Modoki during the boreal summer.

tropical cyclone; South China Sea; ENSO Modoki; genesis potential index

1 Introduction

It is well known that tropical cyclone (TC) activity in most ocean basins is strongly influenced by various modes of natural climate variability (Carmargo et al., 2010). For interannual variations, the El Niño-Southern Oscillation (ENSO) has been found to be able to exert significant influences on TC activity in various ocean basins (e.g., Chan, 2000; Chia and Ropelewski, 2002; Wang and Chan, 2002; Chen et al., 2006; Li and Zhou, 2012; Dowdy et al., 2012). The canonical ENSO event typically develops along the west coast of South America and subsequently propagates westward along the equator (Rasmusson and Carpenter, 1982). Recent studies show that in addition to a canonical El Niño with its major sea surface temperature (SST) anomalies in the equatorial Pacific cold-tongue region, a different type of El Niño with its major action center shifted to the warm-pool edge has emerged and become common during the past two decades (Ashok et al., 2007; Yeh et al., 2009). This phenomenon has been viewed as a different ‘flavour’ of El Niño, with warming around the dateline rather than its east (Ashok and Yamagata, 2009). Recent studies revealed that this new type of El Niño could also modulate TC activities in various ocean basins (e.g., Kim et al., 2009; Chen and Tam, 2010; Kim et al., 2011).

Different names have been used to describe this new type of tropical Pacific phenomenon. Larkin and Harrison (2005) termed these events ‘Dateline El Niño’ and suggested that they should be treated differently from the conventional El Niño events due to different temperature and precipitation anomalies around the globe induced by these events. Ashok et al. (2007) introduced the term‘ENSO Modoki’ to refer to this new type of ENSO. Kao and Yu (2009) further named these two flavors of ENSO as the Eastern-Pacific ENSO and the Central-Pacific ENSO. Kug et al. (2009) introduced the term ‘Warm Pool El Niño’ to refer to the non-conventional El Niño occurring in the central Pacific and the term ‘Cold Tongue El Niño’ to refer to the conventional type of El Niño occurring in the eastern Pacific. Despite different definitions in different studies, the Dateline El Niño, Central-Pacific ENSO, ENSO Modoki and Warm Pool El Niño all refer to essentially the same phenomenon, the new type ofENSO that develops first in the central Pacific. The Eastern-Pacific ENSO and Cold Tongue El Niño refer to the conventional type of ENSO that develops first in the eastern Pacific. What is more, Yeh et al. (2009) found that the ratio of the frequency of the Central-Pacific to Eastern-Pacific type of El Niño events is projected to increase under a global warming scenario in the Coupled Model Intercomparison Project phase 3 (CMIP3) model results, suggesting that the frequency of Central-Pacific type of El Niño tends to increase in the future warming climate.

The influence of this new type of ENSO on TC activities has been paid attention to by some investigators in recent studies. Kim et al. (2009) have found that the two forms of tropical Pacific Ocean warming have substantially different impacts on the frequency and track of North Atlantic TCs. In contrast to conventional El Niño events, El Niño Modoki episodes are associated with a greater-than-average frequency and increasing landfall potential along the coasts of Gulf of Mexico and Central America. Chen and Tam (2010) found that the western North Pacific TC frequency is markedly different due to the two different forms of Pacific warming events, suggesting that the TC frequency is significantly positively correlated with ENSO Modoki index. Kim et al. (2011) extended previous studies to include the whole North Pacific, revealing that the TC activity is shifted to the west and is extended through the northwestern part of the western Pacific in El Niño Modoki years. Hong et al. (2011) revealed different impacts of the two types of El Niño on TC tracks over the western North Pacific.

The South China Sea (SCS), which is a large semienclosed marginal sea in the western Pacific Ocean, is one region where TCs occur frequently (e.g., McGregor, 1995; Wang et al., 2007a, 2007b, 2009; Wang et al., 2007c, 2008; Zuki and Lupo, 2008; Goh and Chan, 2010; Lin and Lee, 2011; Chen et al., 2012; Huang and Guan, 2012; Zhang et al., 2012; Yan et al., 2012; Yang et al., 2012). Chen (2011) revealed that the above-normal TC frequency over the SCS occurs during the boreal summer for the El Niño Modoki years, which could be explained by the anomalous cyclonic circulation, the anomalies of mid-tropospheric relative humidity and vertical wind shear. However, how do these environmental factors contribute to the influence of El Niño Modoki on TC frequency over the SCS and which factor is most important compared with other factors? Emanuel and Nolan (2004) developed an empirical index called the genesis potential (GP) index to relate tropical cyclogenesis to several environmental factors. This index provides one quantitative way to measure relative contributions of different environmental factors to tropical cyclogenesis. Using this index, Camargo et al. (2007a) have examined how different environmental factors contribute to the ENSO effects on global tropical cyclogenesis, identifying that specific factors have more influence than others in different geographic regions. For example, relative humidity and vertical wind shear are important for the reduction in genesis in the Atlantic basin, and relative humidity and vorticity are important for the eastward shift in the mean location in the western North Pacific during El Niño years; vertical wind shear and mid-level relative humidity are most important in the Southern hemisphere. A natural extension of the study from Camargo et al. (2007a) is an investigation of the impact of ENSO Modoki on tropical cyclogenesis over the SCS using the GP index. Furthermore, only Pacific warming events are examined in the study of Chen (2011). Since there is preliminary evidence that the central Pacific cooling events also have a different teleconnection on the globe (Cai and Cowan, 2009; Shinoda et al., 2011), we will examine the effects of central Pacific cooling on tropical cyclogenesis over the SCS in this study, which was not investigated in the study made by Chen (2011). Since the frequency of central Pacific type of El Niño tends to increase in the future warming climate (Yeh et al., 2009), it is instructive to understand the modulation of the ENSO Modoki on tropical cyclogenesis over the SCS, which is an important step toward improving risk assessment and reducing losses of life and property due to TCs.

This paper is structured as follows. In Section 2, various datasets and methods used in this study are outlined. Section 3 presents the TC frequency over the SCS during central Pacific warming and cooling events. Composite environmental factors related with tropical cyclogenesis during central Pacific warming and cooling years are investigated in Section 4. The relative importance of individual environmental factors in the GP index is assessed in Section 5. Conclusions and discussions are given in Section 6.

2 Data and Methodology

The best-track dataset (including 6 hourly TC position and intensity) from the Joint Typhoon Warning Center (JTWC) for the period of 1975–2010 is used to analyze the frequency of TC activities. Genesis location and time of TCs are defined as the first position and time for a TC that attains tropical depression strength.

Understanding the influence of large-scale environmental factors on tropical cyclogenesis is one important problem. Empirical methods have been used to represent the relationship between large-scale environmental factors and tropical cyclogenesis due to the absence of a comprehensive theory. Emanuel and Nolan (2004) developed an empirical index called the GP index to relate tropical cyclogenesis to several environmental factors. The GP index, which is a combination of large-scale oceanic and atmospheric factors that affect tropical cyclogenesis, is defined as

where η is the absolute vorticity (s−1) at 850 hPa, H is the relative humidity (%) at 600 hPa, Vpotthe potential intensity (m s−1), and Vshearthe magnitude of the verticalwind shear (m s−1) between 850 and 200 hPa. The absolute vorticity and vertical wind shear are considered as dynamic components, and the relative humidity and potential intensity are thermodynamic components. More detailed information about the GP index can be found in Emanuel and Nolan (2004) and Camargo et al. (2007a, b). Such an empirical index is helpful in understanding the influence of large-scale factors on tropical cyclogenesis, and it also provides an empirical quantification of the relative contributions of different environmental factors to tropical cyclogenesis. The GP index has already been used in many studies related with tropical cyclogenesis (e.g., Vechi and Soden, 2007; Yokoi et al., 2009; Zhang et al., 2010; Evan and Camargo, 2011; Jiang et al., 2012; Yanase et al., 2012). In this study, the GP index was calculated using monthly mean atmospheric data from the National Centers for Environmental Prediction-National Center for Atmospheric Research (NCEP-NCAR) reanalysis (Kalnay et al., 1996) and monthly mean SST data from the Extended Reconstruction Sea Surface Temperature (ERSST) (Smith and Reynolds, 2004; Smith et al., 2008) during the period of 1975–2010. The monthly gridded rainfall data at a resolution of 2.5°×2.5° are from the Global Precipitation Climatology Project (GPCP) version 2.2 combination data (Adler et al., 2003) (only available from 1979 to the present). The outgoing longwave radiation (OLR) is from the NOAA interpolated OLR data set with a resolution of 2.5° (http://www.esrl.

noaa.gov/psd/data/gridded/data.interp_OLR.html). More details about the OLR data sets can be found in Liebmann and Smith (1996). The temporal coverage of the OLR data set is from 1974 to the present, and data sets during 1975–2010 are used in this study.

So far, there are several definitions to identify the two types of El Niño and La Niña (e.g., Ashok et al., 2007; Kao and Yu, 2009; Kug et al., 2009; Yeh et al., 2009; Kim et al., 2009). In this study, we define central Pacific warming events when the normalized Niño-4 SST (160°E–150°W, 5°S–5°N) is greater than one standard deviation and exceeds the normalized Niño-3 SST (90°–150°W, 5°S–5°N). Similar definitions have been used in the studies of Kim et al. (2009, 2011). Anomalous Niño-3 and Niño-4 indices were directly obtained from the website (http://www.cpc.noaa.gov/data/indices/sstoi.indices) of the Climate Prediction Center. Central Pacific cooling events are also defined in the same way except for negative SST anomalies: Niño-4 SST is cooler than one standard deviation and Niño-3 SST. Totally, four central Pacific warming events (1977, 1994, 2002 and 2004) and five central Pacific cooling events (1975, 1989, 1998, 1999 and 2008) are selected for composite analyses. Through a simple transformation of the Niño3 and Niño4 indices, Ren and Jin (2011) devised two new indices to indentify the two types of ENSO events separately, which could delineate the main patterns of the two types of ENSO, different ENSO phase propagations, and ENSO regime changes. The NWPindex for the new type of ENSO (Warm Pool ENSO) proposed by Ren and Jin (2011) is defined as follows:

here, N3and N4denote Niño3 and Niño4 indices, respectively; α is the parameter of transformation. The NWPindex is used in this study as a single index to characterize the year-to-year variation and time series of the Warm Pool ENSO.

3 Tropical Cyclogenesis During Central Pacific Warming and Cooling Events

Fig.1a shows the time series of the NWPindex and TC number anomaly over the SCS during the boreal summer for the period of 1975–2010. The correlation coefficient between the NWPindex and TC number anomaly is 0.28, which is significant at 90% confidence level but is not significant at 95% confidence level. This result suggests that the TC number formed over the SCS during boreal summer tends to be increased in the year when the NWPindex is high, although the correlation coefficient is not very high and only significant at 90% confidence level. To add to more information about the year-to-year variation of the new type of ENSO and the convection over the northern SCS during boreal summer, we also provide the time series of mean OLR anomaly (Fig.1b) and rainfall anomaly (Fig.1c) averaged over the northern SCS region (100°–120°E, 10°–20°N). The correlation coefficient between the NWPindex and OLR anomaly is −0.29, which is also significant at 90% confidence level but is not significant at 95% confidence level. Negative (Positive) OLR anomalies are indicative of enhanced (suppressed) convections. The results suggest that convections over the northern SCS tend to be enhanced when the NWPindex is high, which is consistent with the tendency of more TCs formed over the northern SCS in the year with high NWPindex. For the relationship between the NWPindex and rainfall anomaly, the correlation coefficient between them is 0.44, which is significant at 95% confidence level. This significant positive correlation coefficient reveals that more rainfall over the northern SCS during boreal summer is observed in the year when the NWPindex is high. Similar results were also obtained in the study of Pradhan et al. (2011), revealing that an interesting feature during El Nino Modoki summers is the occurrence of surplus rainfall anomalies over the SCS and off-equatorial tropical Northwest Pacific. Because the interannual variations of TC numbers occurring over the SCS during boreal summer may be influenced by many other climatic processes, the correlation coefficient between the NWPindex and TC number anomaly could be reduced under the influences of other climatic processes excluding the ENSO Modoki. Previous results have revealed that TC activities over the SCS and the northwest Pacific could be influenced by other climatic factors including the tropical Indian Ocean SST (Du et al., 2011; Wang et al., 2013), the Asian summer monsoon (Wang et al., 2012b), the ENSO(Zhan et al., 2011) and the monsoon trough variability (Wu et al., 2012). These influences from other climatic processes may explain the correlation coefficient between the NWPindex and TC number anomaly is only significant at 90% confidence level but is not significant at 95% confidence level.

Fig.1 The time series of NWPindex and (a) TC number anomaly, (b) OLR anomaly, (c) rainfall anomaly over the northern SCS during boreal summer for the period of 1975–2010. The GPCP rainfall datasets are only available after 1979.

Table 1 shows the numbers of TCs formed over the SCS during boreal summer in each of central Pacific warming and cooling years. It is seen that out of the 4 central Pacific warming events, 3 were above normal. Noticeably, the year of 1994 with the largest NWPindex was associated with the maximum TC numbers, and the year of 2002 with the second largest NWPindex corresponded to the second maximum TC numbers. Using the bootstrap resampling method and the two-sample permutation procedure to conduct statistical tests, Chen (2011) also revealed that the above-normal TC frequency over the SCS occurs during June–August (JJA) for the El Niño Modoki years. For the central Pacific cooling years, 4 out of the 5 central Pacific cooling events were below normal. Spatial distributions of all TC genesis locations over the SCS during JJA from 1975–2010 are presented in Fig.2. Totally 13 TCs were formed during four central Pacific warming years, while only three TCs were formed during five central Pacific cooling years. The average TC number over the SCS during JJA per year was 3.25 for central Pacific warming years and 0.60 for central Pacific cooling years, while the climatological mean value during 1975–2010 was 1.47. The average TC number per year in the central Pacific warming years was about 5.4 times larger than that in the central Pacific cooling years. The above results reveal that there were more TC activities over the SCS during JJA in central Pacific warming years and less TC activities in central Pacific cooling years, which is significant at the 95% confidence level.

Table 1 TC frequencies over the SCS during boreal summer for the central Pacific warming (CPW) and central Pacific cooling (CPC) events

Fig.2 Geographical distribution of tropical cyclogenesis events over the SCS during boreal summer for the period 1975–2010. Totally 13 TCs formed during four central Pacific warming (CPW) events (indicated by red five- pointed stars), while only three TCs formed during five central Pacific cooling (CPC) events (indicated by six- pointed blue stars).

4 Environmental Factors Related to TC Genesis

To relate changes in TC numbers over the SCS to oceanic and atmospheric environmental conditions, several large-scale environmental parameters that influence TC genesis defined in the GP index will be examined in thefollowing. Because the TCs considered were formed in the region north of 10°N over the SCS during the boreal summer (Fig.2), figures in this section will be defined for this region.

4.1 Relative Vorticity

A large low-level cyclonic vorticity is considered as one critical environmental factor for cyclogenesis (Gray, 1968). The low-level relative vorticity is also found to be important to TC formation over the SCS (Wang et al., 2007d). Relative vorticity fields calculated at 850 hPa during JJA are presented in Fig.3 for different phases of ENSO Modoki. Positive relative vorticity with values greater than 5×10−6s−1dominated over most areas of the northern SCS in both central Pacific warming and cooling years (Fig.3a and b), which could be explained by the effects of the monsoon trough and coastal mountains (Wang et al., 2007d). The difference of 850-hPa relative vorticity between central Pacific warming and cooling years (Fig.3c) showed that stronger positive (cyclonic) vorticity was observed over the northern SCS in central Pacific warming years, with the maximum value of vorticity difference exceeding 5×10−6s−1. Results from Chen (2011) also revealed that an anomalous cyclonic circulation dominated in a large fraction of regions over the SCS during central Pacific warming years. Stronger cyclonic lowlevel relative vorticity observed in central Pacific warming years was more favorable for cyclogenesis, which was consistent with the fact that more TCs were formed over the northern SCS in central Pacific warming years.

Fig.3 Mean relative vorticity (10−6s−1) at 850 hPa over the northern SCS during JJA for (a) central Pacific warming years, (b) central Pacific cooling years, and (c) the difference between central Pacific warming and cooling years with the shading indicating areas where the statistical test is significant at the 90% level.

In central Pacific warming years, warm SST anomalies greater than one Celsius degree comparing with central Pacific cooling years were observed in large areas of the central Pacific extending from about 160°E to 130°W (Fig.4a). Low level westerlies anomalies dominated in the tropical western North Pacific and SCS, which resulted in a large scale cyclonic anomaly with the center at around 142°E and 15°N (Fig.4a). This cyclonic anomaly was also accompanied with negative sea level pressure anomalies of about −1.5 hPa. Over the northern SCS, the combination of the tropical westerly anomalies and the northeasterly anomalies from the China mainland was found (Fig. 4a), which initiated a cyclonic shear in favor of TC genesis.

Diabatic heating associated with precipitation in the tropics can generate a remote response in the atmosphere by exciting equatorial waves. Gill (1980) used a simple analytical model to generalize how large-scale atmospheric circulations response to diabatic heating in the tropics, based on the theory of Matsuno (1966) for the forced shallow-water equations in the tropics. Results from Gill (1980) suggest that a symmetric heating centered on the equator induces low-level easterlies via propagation of a Kelvin wave and low-level westerlies via propagation of a Rossby wave. The low-level westerly return flow toward the equator was manifested as a cyclonic circulation over the low pressure systems that formed on the western margins of the heating area as partof a Rossby wave (Fig.4a, b). The westerlies anomalies observed in the tropical western North Pacific and the SCS region in Fig.4a were consistent with Matsuno-Gill patterns. According to Gill’s solutions (Gill, 1980), if the heating is asymmetric about the equator and centered north of the equator, a high pressure system is found in the subtropical region of the southern hemisphere. In Fig. 4a, one high sea level pressure center with anticyclonic circulation anomalies over the Australian region extending from about 15°S to 30°S was also observed, which may be caused by atmospheric responses to asymmetric heating according to Gill’s solutions.

Fig.4 Differences between central Pacific warming and cooling years during JJA for (a) SST (shading), 850-hPa wind (vectors), and sea level pressure (dash/solid contours indicating negative/positive values; contour interval 0.5 hPa, zero contours omitted) and (b) rainfall (shading), 200-hPa wind (vectors), and OLR (dash/solid contours indicating negative/positive values; contour interval 5 W m−2, zero contours omitted).

4.2 Vertical Wind Shear

Strong vertical wind shear is known to be a major impediment to TC formation and intensification (e.g., Gray, 1968; McBride and Zehr, 1981; Tuleya and Kurihara, 1981). Weak vertical shear is necessary for tropical cyclogenesis because the shear breaks a warm core structure of TC (a ventilation effect). Physically, low vertical wind shear is required so that enthalpy and moisture can be accumulated in a vertical air column. A system with weak vertical wind shear will lose little moisture and heat energy and thus will be more likely to organize itself into a vertically stacked tropospheric system such as a tropical storm. In operational situation, the vertical wind shear is often calculated between 850 and 200 hPa levels. Because tropical data sources tend to be at upper levels (especially satellite cloud-drift winds and jet aircraft reports) and lower levels (satellite cloud-drift winds and surface reports extrapolated upward), the vertical wind shear is often expressed as the 200-mb-minus-850-mb wind difference (Elsberry and Jefferies, 1996). Observational studies suggest that the shear must be below some threshold value for a TC to develop. Zehr (1992) used the Australia Bureau of meteorology real-time analyses to estimate a threshold of 12.5 m s−1for vertical wind shear beyond which western North Pacific Ocean TC did not form. Gallina and Velden (2002) found the critical vertical wind shear for tropical cyclone development in the Atlantic to be 7–8 m s−1and 9–10 m s−1in the western North Pacific, using lower-layer (700–925 mb) and upper-layer (150–350 mb) mass-weighted layer-mean wind fields. The results of Ritchie (2002) suggested that the critical shear value in different ocean basins could be different. Black et al. (2002) revealed that hurricanes could still intensify or maintain their intensity in substantial shear if the SST was high according to the airborne radar and in situ observations in two hurricanes.

In this study, the vertical wind shear is also defined as the magnitude of the difference between the zonal and meridional wind at the 200 and 850 hPa. Results of analysis are shown for central Pacific warming and cooling years respectively in Fig.5. During the boreal summer,southwest wind was observed at 850 hPa (Fig.6a) and the upper-tropospheric winds were easterlies (Fig.6b) in the climatological mean wind fields. The magnitudes of vertical wind shear tended to increase with the decrease of latitude, because both the lower-tropospheric westerlies and upper-tropospheric easterlies increased with the decrease of latitude. The difference of vertical wind shear between central Pacific warming and cooling years showed that weaker vertical wind shear was found north of 16°N in central Pacific warming years, while stronger vertical wind shear was observed south of 16°N (Fig.5c). During central Pacific warming years, easterlies anomalies at 200 hPa dominated in the tropical western North Pacific extending from about 140°W to 120°E, which resulted in a large scale anti-cyclonic anomaly in the tropical western North Pacific (Fig.4b). These circulation anomalies were also consistent with Matsuno-Gill-type response to the tropical heating source over the central Pacific during central Pacific warming years. Surplus rainfall and negative OLR anomalies were also found over the northern SCS during central Pacific warming years as shown in Fig.4b, which was consistent with the time series of regional mean rainfall and OLR anomalies (Fig.1b and c). Over the northern SCS, the westerly anomalies were found north of 16°N in central Pacific warming years (Fig.4b), which tended to weaken the upper-level easterly flows in the area (Fig.6b). The observed decrease of the vertical wind shear north of 16°N was mainly dueto the weakening of the upper-level easterlies. The decrease of vertical wind shear in this area was favorable for TC genesis, which may explain partly why more TCs were observed to form over the northern SCS during central Pacific warming years. Over the area south of 16°N, the observed increase of the vertical wind shear was mainly due to the strengthening of the lower-level westerlies caused by the tropical westerly anomalies (Fig.4a).

Fig.5 Mean vertical wind shear (m s−1) over the northern SCS during JJA for (a) central Pacific warming years, (b) central Pacific cooling years, and (c) the difference between central Pacific warming and cooling years with the shading indicating areas where the statistical test is significant at the 90% level.

Fig.6 Climatological mean wind vectors and speeds (color shading) during JJA for the period 1975–2010 at (a) 850 hPa and (b) 200 hPa.

4.3 Potential Intensity

Climatologically, TCs can only form over waters of 26℃ or higher (Wendland, 1977). The underlying warm ocean is an energy source for the formation and development of TCs in the form of surface heat fluxes. Potential intensity is one thermodynamical parameter used in the GP index including the effects of SSTs and moist static instability on tropical cyclogenesis. The definition of potential intensity is given by Emanuel (1995) and modified by Bister and Emanuel (1998). The formula for the potential intensity (Vpot) is

where ckis the exchange coefficient for enthalpy, cDis the drag coefficient, Tsis the SST and T0stands for the mean outflow temperature. The convective available potential energy (CAPE) is the vertical integral of parcel buoyancy as a function of parcel temperature, pressure and specific humidity. CAPE*refers to the value of CAPE for an air parcel at the radius of maximum winds and CAPEbis the value of CAPE for ambient boundary layer air. The potential intensity is obtained from sea-level pressure, SST, and vertical profiles of atmospheric temperature and mixing ratio, with dissipative heating taken into account as discussed in Bister and Emanuel (1998).

Over the SCS, oceanic conditions in terms of SSTs above 26℃ (Gray, 1968) were satisfied during both central Pacific warming and cooling years, which could possibly support the occurrence of tropical cyclogenesis. The SST differences in the SCS region between central Pacific warming and cooling years were not significant (Fig.4a), suggesting that local SSTs may not play a large role in defining the ENSO Modoki-dependent changes in TC numbers over the SCS. The differences of the potential intensity between central Pacific warming and cooling years were also not significant in the SCS (Fig.7c). The changes in the potential intensity are not easily interpreted in order to explain more observed TC activity during central Pacific warming years.

Fig.7 Mean potential intensity (m s−1) over the northern SCS during JJA for (a) central Pacific warming years, (b) central Pacific cooling years, and (c) the difference between central Pacific warming and cooling years, which is not significant at the 90% level.

4.4 Mid-Level Relative Humidity

High relative humidity is one of the key factors influencing TC formation and development as it provides sufficient moisture at the lower level and the midtroposphere (Gray, 1968). High values of relative humidity are necessary to overcome the negative effects of entrainment on convection during the TC development stage. Geographical distributions of mid-tropospheric relative humidity at 600 hPa in JJA over the northern SCS are pre-sented for central Pacific warming years in Fig.8a and for cooling years in Fig.8b separately. During the boreal summer, relative high values of mid-level relative humidity greater than 50% were observed over the northern SCS in both central Pacific warming and cooling years, which could possibly support the occurrence of tropical cyclogenesis. The magnitudes of mid-level relative humidity in central Pacific warming years were found to be about 1%–2% larger than those in central Pacific cooling years, providing more favorable environmental conditions for TC genesis in central Pacific warming years; however, the differences of mid-level relative humidity between central Pacific warming and cooling years were not significant at 90% confidence level (Fig.8c).

Fig.8 Mean relative humidity (%) at 600 hPa over the northern SCS during JJA for (a) central Pacific warming years, (b) central Pacific cooling years, and (c) the difference between central Pacific warming and cooling years, which is not significant at the 90% level.

4.5 The GP Index

The GP index is defined to represent the influence of large-scale environmental factors on tropical cyclogenesis. Four factors contribute to the GP index: low-level vorticity, mid-level relative humidity, vertical wind shear and potential intensity. The differences of the GP index between central Pacific warming and cooling years are presented in Fig.9. Higher values of the GP index were observed in central Pacific warming years than those in central Pacific cooling years, which was consistent with the fact that more TCs were formed over the northern SCS in central Pacific warming years. How different environmental factors contribute to the influence of ENSO Modoki on tropical cyclogenesis quantitatively over the SCS during the boreal summer and which factor is most important compared with other factors? This question will be analyzed quantitatively using the GP index in the next section.

Fig.9 Difference of the GP index between central Pacific warming and cooling years, using unmodified interannual varying variables.

5 Factors Influencing ENSO Modoki Effects on the GP Index over the SCS

In the following, the relative importance of the individual four variables in determining the differences in TC formation over the SCS during the boreal summer between central Pacific warming and cooling years will be examined. Following the method used in Camargo et al. (2007a, 2009) and Chand and Walsh (2010), we calculated the modified GP indices using the unmodified in-terannual varying values for only one of the contributing factor, with the others set to long-term mean climatology. For example, when calculating the modified GP indices for varying low-level vorticity, we used interannual varying values for the low-level vorticity and long-term mean climatology (without interannual variations) for other three factors (mid-level relative humidity, vertical wind shear and potential intensity) in Eq. (1). The differences between central Pacific warming and cooling years were then recalculated in all four cases for varying each factor in the GP index, which were compared with the differences in Fig.9 when four unmodified interannual varying variables are used in the GP index. Considering that the GP index has weights that appropriately quantify the contributions of the different factors in TC genesis and that the nonlinearities are not large, the degree of contribution by each variable to the GP index may be quantitatively determined (Camargo et al., 2007a, 2009).

Differences in these modified GP indices between central Pacific warming and cooling years are shown in Fig.10. A more quantitative measure of the contribution of each of the variables to the total variations of the GP index is given in Table 2. In this table, the regression coefficients between the difference of the GP index and differences of the modified GP indices with varying individual variable are calculated. The vertical wind shear has the largest regression coefficient with the value of 0.586, indicating the largest role of this variable. The relative vorticity has the second largest regression coefficient with the value of 0.203, pointing to the secondary role of the variable. The potential intensity and mid-level humidity have small regression coefficients, implying the weakinfluences of these two variables. These regressions show that the vertical wind shear and relative vorticity are the most important factors contributing to the interannual variations of TC genesis frequency over the SCS during the boreal summer associated with the ENSO Modoki.

Fig.10 Differences of the modified GP indices between central Pacific warming and cooling years for varying (a) low-level vorticity, (b) vertical wind shear, (c) mid-level humidity, and (d) potential intensity while setting the other variables as the long-term mean climatology.

Table 2 Regression coefficients between difference of the GP index and differences in the modified GP indices between central Pacific warming and cooling years with varying different variables

Fig.11 Sums of differences of the modified GP indices for four individual variables between central Pacific warming and cooling years.

To assess the degree of nonlinearity in the GP index, sums of the modified GP indices for four factors (Fig.11) are compared with the total GP composite difference in Fig.9. Since the sum of all four factors (Fig.11) has a very similar pattern and amplitude to those in Fig.9, the nonlinearity effect is weak and the above analysis with the assumed linearity of the GP index is meaningful.

6 Conclusion and Discussion

In this study, we examine the modulation of tropical cyclogenesis over the SCS by the ENSO Modoki during the boreal summer when TC activities are most active in the year over this region. Results reveal that there were more TCs formed over the SCS during central Pacific warming years and less TC frequency during central Pacific cooling years. How different environmental factors contribute to these influences is investigated using the GP index. Composite anomalies of the GP index are produced for central Pacific warming and cooling years separately, which could account for the changes of TC frequency over the SCS in different ENSO Modoki phases. The degree of contribution by each factor is determined quantitatively by producing composites of modified indices in which only one of the contributing factors varies, with the others set to climatology. Results suggest that vertical wind shear makes the largest contribution to the ENSO Modoki modulation of tropical cyclogenesis over the SCS, with the low-level relative vorticity playing the second role.

Based on our recent another analysis for seasonal variations of tropical cyclogenesis over the northern SCS (Wang and Pan, 2012), results revealed that potential intensity makes the largest contribution to the seasonal variation of tropical cyclogenesis. In this study, the vertical wind shear and relative vorticity are found to make the largest contribution to the interannual variation of tropical cyclogenesis over the northern SCS associated with the ENSO Modoki, with the potential intensity making a small contribution. Dynamic factors (vertical wind shear and low-level vorticity) play more important roles for the interannual variations in tropical cyclogenesis associated with the ENSO Modoki, while thermodynamics factor (potential intensity) is more important in the seasonal variation of TC genesis frequency over the northern SCS. These results suggest the different roles of large scale environmental factors in TC genesis over the northern SCS on seasonal and interannual time scales.

Chen (2011) revealed that El Niño Modoki and canonical El Niño events could modulate TC frequency over the SCS in different seasons: the above-normal TC frequency over the SCS occurs during boreal summer for the El Niño Modoki years, whereas the below-normal TC frequency is significant during September-November (SON) for the canonical El Niño years. For the ENSO modulation of tropical cyclogenesis over the southern SCS, results from our recent another analysis (Wang et al., 2012a) reveal that the mid-level relative humidity makes the largest contribution to this ENSO modulation, which is mainly due to the ENSO-induced anomalous Walker circulations. During El Niño years, downward anomalies of vertical motions inhibited the upward transports of water vapor and led to less water vapor in the middle troposphere over the southern SCS region, which depressed TC formations over this region. For the ENSO Modoki modulation of tropical cyclogenesis over the northern SCS during the boreal summer, results in this study reveal that the vertical wind shear and low-level relative vorticity make the largest contribution, which is mainly due to the ENSO Modoki-induced anomalous circulations in Matsuno-Gill patterns. These results highlight the different roles of large scale environmental factors and different physical processes in ENSO and ENSO Modoki modulations of tropical cyclogenesis over the SCS.

Acknowledgements

The authors thank the anonymous reviewers for their helpful comments. This study was funded by the Strategic Priority Research Program of the Chinese Academy of Sciences with Grant No. XDA11010000, the National Natural Science Foundation of China (No. 41205026), the National Basic Research Program of China (2011CB403500) and the Innovation Group Program of State Key Laboratory of Tropical Oceanography (LTOZZ1201). Dr. Lei Wang was also sponsored by the Knowledge Innovation Program of the Chinese Academy of Sciences (SQ201208), the foundation for returned scholars of Ministry of Education of China, the specialized research fund for the doctoral program of Higher Education for Youths, the foundation of Guangdong Educational Committee for Youths (2012 LYM_0008), and the open fund of the Key Laboratory of Ocean Circulation and Waves of Chinese Academy of Sciences (KLOCAW1309).

Adler, R. F., Huffman, G. J., Chang, A., Ferraro, R., Xie, P. P., Janowiak, J., Rudolf, B., Schneider, U., Curtis, S., Bolvin, D., Gruber, A., Susskind, J., Arkin, P., and Nelkin, E., 2003. The Version 2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). Journal of Hydrometeorology, 4: 1147-1167.

Ashok, K., and Yamagata, T., 2009: Climate change: The El Niño with a difference. Nature, 461: 481-484.

Ashok, K., Behera, S. K., Rao, S. A., Weng, H., and Yamagata, T., 2007. El Niño Modoki and its possible teleconnection. Journal of Geophysical Research, 112, C11007, DOI: 10.1029/2006JC003.

Bister, M., and Emanuel, K. A., 1998. Dissipative heating and hurricane intensity. Meteorology and Atmospheric Physics, 52: 233-240.

Black, M. L., Gamache, J. E., Marks, F. D., Samsury, C. E., andWilloughby, H. E., 2002. Eastern Pacific Hurricanes Jimena of 1991 and Olivia of 1994: The effect of vertical shear on structure and intensity. Monthly Weather Review, 130: 2291-2312.

Cai, W., and Cowan, T., 2009. La Niño Modoki impacts Australia autumn rainfall variability. Geophysical Research Letters, 36, L12805, DOI: 10.1029/2009GL037885.

Camargo, S. J., Emanuel, K. A., and Sobel, A. H., 2007a. Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis. Journal of Climate, 20: 4819-4834.

Camargo, S. J., Sobel, A. H., Barnston, A. G., and Emanuel, K. A., 2007b. Tropical cyclone genesis potential index in climate models. Tellus A, 59 (4): 428-443.

Camargo, S. J., Sobel, A. H., Barnston, A. G., and Klotzbach, P. J., 2010. The influence of natural climate variability on tropical cyclones, and seasonal forecasts on tropical cyclone activity. Global Perspectives on Tropical Cyclones: From Science to Mitigation. Chan, J. C. L., and Kepert, J. D., eds., World Scientific Series on Asia-Pacific Weather and Climate, Vol. 4, World Scientific Publishing, 325-360.

Camargo, S. J., Wheeler, M. C., and Sobel, A. H., 2009. Diagnosis of the MJO modulation of tropical cyclogenesis using an empirical index. Journal of Atmospheric Sciences, 66: 3061-3074.

Chan, J. C. L., 2000. Tropical cyclones activity over the western North Pacific associated with El Niño and La Niña events. Journal of Climate, 13: 2960-2972.

Chand, S. S., and Walsh, K. J. E., 2010. The influence of the Madden-Julian oscillation on tropical cyclone activity in the Fiji region. Journal of Climate, 23: 868-886.

Chen, G., 2011. How does shifting Pacific Ocean warming modulate on tropical cyclone frequency over the South China Sea? Journal of Climate, 24: 4695-4700.

Chen, G., and Tam, C. Y., 2010. Different impacts of two kinds of Pacific Ocean warming on tropical cyclone frequency over the western North Pacific. Geophysical Research Letters, 37, L01803, DOI: 10.1029/2009GL041708.

Chen, J. P., Wu, R. G., and Wen, Z. P., 2012. Contribution of South China Sea tropical cyclones to an increase in southern China summer rainfall around 1993. Advances in Atmospheric Sciences, 29 (3): 585-598.

Chen, T. C., Wang, S. Y., and Yen, M. C., 2006. Interannual variation of tropical cyclone activity over the western North Pacific. Journal of Climate, 19: 5709-5720.

Chia, H. H., and Ropelewski, C. F., 2002. The interannual variability in the genesis location of tropical cyclones in the northwest Pacific. Journal of Climate, 15: 2934-2944.

Dowdy, A. J., Qi, L., Jones, D., Ramsay, H., Fawcett, R., and Kuleshov, Y., 2012. Tropical cyclone climatology of the South Pacific Ocean and its relationship to El Niño-Southern Oscillation. Journal of Climate, 25: 6108-6122.

Du, Y., Yang, L., and Xie, S.-P., 2011. Tropical Indian ocean influence on Northwest Pacific tropical cyclones in summer following strong El Niño. Journal of Climate, 24: 315-322.

Elsberry, R. L., and Jefferies, R. A., 1996. Vertical wind shear influences on tropical cyclone formation and intensification during TCM-92 and TCM-93. Monthly Weather Review, 124: 1374-1387.

Emanuel, K. A., 1995. Sensitivity of tropical cyclones to surface exchange coefficients and a revised steady-state model incorporating eye dynamics. Journal of Atmospheric Sciences, 52: 3669-3976.

Emanuel, K. A., and Nolan, D. S., 2004. Tropical cyclone activity and global climate. 26th Conference on Hurricane and Tropical Meteorology, American Meteorological Society, Miami, FL, 240- 241.

Evan, A. T., and Camargo, S. J., 2011. A climatology of Arabian Sea cyclonic storms. Journal of Climate, 24: 140-158.

Gallina, G. M., and Velden, C. S., 2002. Environmental vertical wind shear and tropical cyclone intensity change utilizing enhanced satellite derived wind information. 25th Conference on Hurricanes and Tropical Meteorology, American Meteorological Society, San Diego, CA.

Gill, A. E., 1980. Some simple solution for heat-induced tropical circulation. Quarterly Journal of the Royal Meteorological Society, 106: 447-462.

Goh, Z. C., and Chan, J. C. L., 2010. Interannual and interdecadal variations of tropical cyclone activity in the South China Sea. International Journal of Climatology, 30: 827-843.

Gray, W. M., 1968. Global view of the origin of tropical disturbances and storms. Monthly Weather Review, 96: 669-700.

Hong, C. C., Li, Y. H., Li, T., and Lee, M. Y., 2011. Impacts of central Pacific and eastern Pacific El Niños on tropical cyclone tracks over the western North Pacific. Geophysical Research Letters, 38, L16712, DOI: 10.1029/2011GL048821.

Huang, Q., and Guan, Y. P., 2012. Does the Asian monsoon modulate tropical cyclone activity over the South China Sea? Chinese Journal of Oceanology and Limnology, 30 (6): 960-965.

Jiang, X., Zhao, M., and Waliser, D. E., 2012. Modulation of tropical cyclones over the Eastern Pacific by the intraseasonal variability simulated in an AGCM. Journal of Climate, 25: 6524-6538.

Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha, S., White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, R., Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K. C., Repelewski, C., Wang, J., Jenne, R., and Joseph, D., 1996. The NCEP/NCAR 40-Year Reanalysis Project. Bulletin of the American Meteorological Society, 77: 437-471.

Kao, H. Y., and Yu, J. Y., 2009. Contrasting eastern Pacific and central Pacific types of ENSO. Journal of Climate, 22: 615-631.

Kim, H. M., Webster, P. J., and Curry, J. A., 2009. Impact of shifting patterns of Pacific Ocean warming on North Atlantic tropical cyclones. Science, 325: 77-80.

Kim, H. M., Webster, P. J., and Curry, J. A., 2011. Modulation of north Pacific tropical cyclone activity by three phases of ENSO. Journal of Climate, 24: 1839-1849.

Kug, J. S., Jin, F. F., and An, S. I., 2009. Two types of El Niño events: Cold-tongue El Niño and warm-pool El Niño. Journal of Climate, 22: 1499-1515.

Larkin, N. K., and Harrison, D. E., 2005. Global seasonal temperature and precipitation anomalies during El Niño autumn and winter. Geophysical Research Letters, 32, L16705, DOI: 10.1029/2005GL022860.

Li, R. C. Y., and Zhou, W., 2012. Changes in western Pacific tropical cyclones associated with the El Niño-Southern Oscillation cycle. Journal of Climate, 25: 5864-5878.

Liebmann, B., and Smith, C. A., 1996. Description of a complete (interpolated) outgoing longwave radiation dataset. Bulletin of the American Meteorological Society, 77: 1275-1277.

Lin, Y.-L., and Lee, C.-S., 2011. An analysis of tropical cyclone formations in the South China Sea during the late season. Monthly Weather Review, 139: 2748-2760.

Matsuno, T., 1966. Quasi-geostrophic motions in the equatorial area. Journal of the Meteorological Society of Japan, 44: 25-43.

McBride, J. L., and Zehr, R. M., 1981. Observational analysis of tropical cyclone formation. Part II: Comparison of nondeveloping versus developing systems. Journal of Atmospheric Sciences, 38: 1132-1151.

McGregor, G. R., 1995. The tropical cyclone hazard over the South China Sea, 1970–1989. Applied Geography, 15 (1): 35-52.

Pradhan, P. K., Preethi, B., Ashok, K., Krishnan, R., and Sahai, A. K., 2011. Modoki, Indian Ocean dipole, and western North Pacific typhoons: Possible implications for extreme events. Journal of Geophysical Research, 116, D18108, DOI: 10. 1029/2011JD015666.

Rasmusson, E. M., and Carpenter, T. H., 1982. Variations in tropical sea surface temperature and surface wind fields associated with the Southern Oscillation-El Niño. Monthly Weather Review, 110: 354-384.

Ren, H. L., and Jin, F. F., 2011. Niño indices for two types of ENSO. Geophysical Research Letters, 38, L04704, DOI: 10. 1029/2010GL046031.

Ritchie, E. A., 2002. Topic 1.2: Environmental effects. Topic Chairman and Rapporteur Reports of the Fifth WMO International Workshop on Tropical Cyclones (IWTC-V), WMO Tech. Doc. WMO/TD 1136, WMO, Geneva, Switzerland.

Shinoda, T., Hurlburt, H. E., and Metzger, E. J., 2011. Anomalous tropical ocean circulation associated with La Niña Modoki. Journal of Geophysical Research, 116, DOI: 10.1029/ 2011JC007304.

Smith, T. M., and Reynolds, R. W., 2004. Improved extended reconstruction of SST (1854–1997). Journal of Climate, 17: 2466-2477.

Smith, T. M., Reynolds, R. W., Peterson, T. C., and Lawrimore, J., 2008. Improvements to NOAA’s historical merged landocean surface temperature analysis (1880–2006). Journal of Climate, 21: 2283-2296.

Tuleya, R. E., and Kurihara, Y., 1981. A numerical study on the effects of environmental flow on tropical storm genesis. Monthly Weather Review, 109: 2487-2506.

Vecchi, G. A., and Soden, B. J., 2007. Effect of remote sea surface temperature change on tropical cyclone potential intensity. Nature, 450: 1066-1070.

Wang, B., and Chan, J. C. L., 2002. How strong ENSO events affect tropical storm activity over the western North pacific. Journal of Climate, 15: 1643-1658.

Wang, G. H., Ling, Z., and Wang, C. Z., 2009. Influence of tropical cyclones on seasonal ocean circulation in the South China Sea. Journal of Geophysical Research, 114, C10022, DOI: 10.1029/2009JC005302.

Wang, G. H., Su, J., Ding, Y., and Chen, D., 2007a. Tropical cyclone genesis over the South China Sea. Journal of Marine Systems, 68 (3): 318-326.

Wang, G. H., Wang, H., and Qi, Y. Q., 2007b. Seasonal variability of tropical cyclones generated over the South China Sea. Acta Oceanologica Sinica, 26 (4): 20-28.

Wang, L., and Pan, X., 2012. Diagnosis of seasonal variation of tropical cyclogenesis over the South China Sea using a genesis potential index. Journal of Ocean University of China, 11 (3): 267-278.

Wang, L., Fung, C. H., and Lau, K. H., 2007c. The upper ocean thermal structure and the genesis locations of tropical cyclones in the South China Sea. Journal of Ocean University of China, 6 (2): 125-131.

Wang, L., Huang, R., and Wu, R., 2013. Interdecadal variability in tropical cyclone frequency over the South China Sea and its association with the Indian Ocean sea surface temperature, Geophysical Research Letters, 40: 768-771, DOI: 10.1002/ grl.50171.

Wang, L., Lau, K. H., Fung, C. H., and Gan, J. P., 2007d. The relative vorticity of ocean surface winds from the QuikSCAT satellite and its effects on the geneses of tropical cyclones in the South China Sea. Tellus A, 59: 562-569.

Wang, L., Lau, K. H., Zhang, Q. H., and Fung, C. H., 2008. Observation of non-developing and developing tropical disturbances over the South China Sea using SSM/I satellite. Geophysical Research Letters, 35, L10802, DOI: 10.1029/ 2008GL033446.

Wang, L., Zhang, Q.W., and Li, W. B., 2012a. Diagnosis of the ENSO modulation of tropical cyclogenesis over the southern South China Sea using a genesis potential index. Acta Oceanologica Sinica, 31 (5): 54-68.

Wang, X., Zhou, W., Li, C., and Wang, D., 2012b. Effects of the East Asian summer monsoon on tropical cyclone genesis over the South China Sea on an interdecadal time scale. Advances in Atmospheric Sciences, 29 (2): 249-262.

Wedland, W. M., 1977. Tropical storm frequencies related to sea surface temperatures. Journal of Applied Meteorology, 16: 477-481.

Wu, L., Wen, Z., Huang, R., and Wu, R., 2012. Possible linkage between the monsoon trough variability and the tropical cyclone activity over the western North Pacific. Monthly Weather Review, 140: 140-150.

Yan, Y. F., Qi, Y. Q., and Zhou, W., 2012. Variability of tropical cyclone occurrence date in the South China Sea and its relationship with SST warming. Dynamics of Atmospheres and Oceans, 55: 45-59.

Yanase, W., Satoh, M., Taniguchi, H., and Fujinami, H., 2012. Seasonal and intraseasonal modulation of tropical cyclogenesis environment over the Bay of Bengal during the Extended Summer Monsoon. Journal of Climate, 25: 2914-2930.

Yang, L., Du, Y., Xie, S. P., and Wang, D. X., 2012. An interdecadal change of tropical cyclone activity in the South China Sea in early 1990s. Chinese Journal of Oceanology and Limnology, 30 (6): 953-959.

Yeh, S. W., Kug, J. S., Dewitte, B., Kwon, M. H., and Kirtman, B. P., 2009. El Niño in a changing climate. Nature, 461: 511-514, DOI: 10.1038/nature08316.

Yokoi, S., Takayabu, Y. N., and Chan, J. C. L., 2009. Tropical cyclone genesis frequency over the western North Pacific simulated in medium-resolution coupled general circulation models. Climate Dynamics, 33: 665-683.

Zehr, R. M., 1992. Tropical Cyclogenesis in the Western North Pacific. NOAA Tech. Rep. NESDIS 61, 181pp.

Zhan, R., Wang, Y., and Lei, X., 2011. Contributions of ENSO and East Indian Ocean SSTA to the interannual variability of Northwest Pacific tropical cyclone frequency. Journal of Climate, 24: 509-521.

Zhang, C., Yao, H. L., Zhang, Q. H., and Wang, L., 2012. Prediction of tropical disturbance development over the South China Sea using SSM/I data. Journal of Tropical Meteorology, 18 (2): 242-248.

Zhang, Y., Wang, H. J., Sun, J., and Drange, H., 2010. Changes in the tropical cyclone genesis potential index over the Western North Pacific in the SRES A2 scenario. Advances in Atmospheric Sciences, 27: 1246-1258.

Zuki, Z. M., and Lupo, A. R., 2008. Interannual variability of tropical cyclone activity in the southern South China Sea. Journal of Geophysical Research, 113, D06106. DOI: 10.1029/ 2007JD009218.

(Edited by Xie Jun)

(Received August 1, 2012; revised September 28, 2012; accepted May 7, 2013)

© Ocean University of China, Science Press and Springer-Verlag Berlin Heidelberg 2014

* Corresponding author. Tel: 0086-020-89023199

E-mail: wanglei.hkust@gmail.com