Climatological relationships among the tropical cyclone frequency ...

Article Atmospheric Science

November 2010 Vol.55 No.33: 3818–3824 doi: 10.1007/s11434-010-4168-2

SPECIAL TOPICS:

Climatological relationships among the tropical cyclone frequency, duration, intensity and activity regions over the Western Pacific LI WeiBiao1*, DU QinBo1,2 & CHEN ShuMin1 1 2

Department of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275, China; NanAo Observatory of Shantou Meteorological Bureau, Shantou 515900, China

Received April 26, 2010; accepted July 23, 2010

Climatological relationships among the tropical cyclone (TC) frequency, duration, intensity and activity regions over the Western Pacific are explored based on long-term best track data. Frequent TC occurrence does not necessarily imply a long duration of TCs in the same periods. Three types of relationship between TC number and duration in the period 1945–2007 were identified in this study: low frequency and short duration during 1945–1955 (Period I); high frequency and short duration in the 1960s (Period II); and high frequency and long duration in the 1990s (Period III). TC activity regions differed among the three periods. During Period I, the main activity regions were over the ocean east of the Philippines (120°–140°E). During period II, two prevailing storm tracks extended west-northwest between 110° and 147°E. During period III, TCs had an extensive activity region from 110° to 160°E. TC intensity is related closely to activity regions. Most strong TCs developed over the ocean far from the Philippines, and had a northwestward track. Our results also show that the relationships between TC frequency, duration and their active regions are modulated strongly by broad-scale vertical motion, geopotential height and horizontal wind anomalies. tropical cyclone, frequency, duration, intensity, activity regions Citation:

Li W B, Du Q B, Chen S M. Climatological relationships among the tropical cyclone frequency, duration, intensity and activity regions over the Western Pacific. Chinese Sci Bull, 2010, 55: 3818−3824, doi: 10.1007/s11434-010-4168-2

Tropical cyclones (TCs) are one of the most destructive natural disasters affecting China. Each year coastal south China is impacted seriously by TCs originating in the Western Pacific. During the past several decades, a number of studies have investigated the genesis, tracks and impacts of TCs over the Western Pacific [1,2]. Other studies have focused on the effects of large-scale circulations, TC structure itself, terrain and other factors on TC evolution (see the review by Meng et al. [3]). Recently, a great deal of research has been conducted to address the issue of climatological variation in the frequency of TC occurrence and intensity. Maloney and Hartmann [4] suggested that the intraseasonal variation in TC activity is attributable to the Madden–Julian oscillation (MJO). Chen [5] demonstrated that the interdecadal variation in tropical cyclone activity is *Corresponding author (email: [email protected])

© Science China Press and Springer-Verlag Berlin Heidelberg 2010

associated with the summer monsoon and sea surface temperature over the western North Pacific. Several large-scale atmospheric and oceanic modes such as the Quasi-Biennial Oscillation (QBO) [6], El Niño/Southern Oscillation (ENSO) [7,8] North Pacific Oscillation (NPO) [9] Antarctic Oscillation (AAO) [10,11], Asian-Pacific Oscillation (APO) [12] and Hadley circulation [13,14] correlate well with TC variation on the same timescales. Recently, a new prediction model for typhoon frequency over the western North Pacific was developed based on several new factors including sea ice cover in the North Pacific and the NPO [15]. In addition, other studies have suggested an increase in the frequency of occurrence of intense TCs in the past 30 years, and related this increase to a concomitant increase in sea surface temperature (SST) that may or may not be caused by global warming [16–20]. Moreover, many recent studies argue that the apparent increase of intense TCs in recent years is an active phase in csb.scichina.com

www.springerlink.com

LI WeiBiao, et al.

Chinese Sci Bull

decadal cyclic TC activity [21,22] controlled by atmospheric variability and atmosphere-ocean interactions. The above-mentioned research has revealed the characteristics of intraseasonal, interannual and decadal TC variability, and long-term trends in TC frequency and intensity. In addition to frequency and intensity, the TC duration and activity regions are also important aspects of TC climatology. However, there is a lack of detailed descriptions and understanding of the climatological relationships among the TC frequency, duration, and prevailing activity regions over the Western Pacific. In this study, we focus on characterizing climatological features of relationships between TC frequency, duration, intensity and their active regions over the Western Pacific based on TC best track data from the US Navy’s Joint Typhoon Warning Center (JTWC) for the period 1945–2007.

1

Data and methods

The tropical cyclone data were derived from the best track archives of the Joint Typhoon Warning Center (JTWC) for the period 1945–2007 (http://www.npmoc.navy.mil/jtwc/ best_tracks/wpindex.html). For this study, tropical cyclones were classified into four groups based on the JTWC criteria: tropical depressions (maximum sustained winds less than 17 m s–1), tropical storms (maximum sustained winds between 17 and 32 m s–1), typhoons (maximum sustained winds between 32 and 67 m s–1) and super typhoons (sustained winds greater than 67 m s–1). We explore four different, but related, basin-wide measures of Western Pacific TC activity: TC frequency, duration, intensity and activity regions. Annual TC frequency is the annual number of tropical cyclones in the Western Pacific; annual average TC duration is the average number of hours for each TC in each year; the TC activity regions are delineated by TC track density obtained by calculating the total number of TCs, obtained from JTWC best track data, for each 1° latitude × 1° longitude grid cell in the Western Pacific Basin. TC track density, which represents TC tracks or genesis locations, has been adopted by a number of previous authors [23] to analyze the activity features of TCs. To examine large-scale environmental parameters, we used the National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) Reanalysis dataset [24], which has a 2.5° latitude/longitude resolution.

November (2010) Vol.55 No.33

3819

erage duration and normalized annual frequency and duration in the Western Pacific during the 63-a period 1945–2007. Both TC frequency (Figure 1(a)) and duration (Figure 1(b) over the Western Pacific increased (95% significance level) over the interval 1945–2007. There were substantial interannual and decadal variabilities in both annual TC number and duration. However, TC frequency and duration exhibited different features in their interannual and decadal variations, i.e. highly frequent occurrence does not imply necessarily long TC duration in the same period. Figure 1(a) indicates that both the 1960s and the early 1990s were TC active phases, and both the 1950s and the mid-1970s to the early 1980s were TC inactive phases. This agrees with the results from recent studies [25,26]. The decadal variability in TC duration is quite different from that of their annual frequency. Figure 1(b) indicates that significant short and long duration periods were in the 1960s and 1990s, respectively. Based on this, we identified three types of relationships between TC frequency and duration (Figure 1(c)): low frequency and short duration during 1945–1955 (Period I); high frequency and short duration in the 1960s (Period II), high frequency and long duration in the 1990s (Period III). To investigate the spatial features of TC activity for the above three periods, we obtained TC track density by calculating the total number of TCs for each 1° latitude × 1° longitude grid cell over the Western Pacific Basin from the JTWC best track data. Figure 2 shows TC track density for the Period I (a), Period II (b) and Period III (c), respectively. The high-value areas in Figure 2 represent the prevailing TC activity regions. During Period I (Figure 2 (a)), the primary TC activity regions were located in the ocean basin east of the Philippines (120°–140°E), and were relatively close to the islands and west coast of the Eurasian continent. The high-value areas in Figure 2(a) suggest that TCs in Period I travelled mainly northwestward, extending from the ocean basin east of the Philippines to the east coast of China. Tropical cyclone activity regions in Period II are quite different from those in Period I. Period II had two prevailing storm tracks, both tracking west-northwestward (Figure 2(b)). The northern track extended west-northwestward between 124° and 147°E; the southern one began at 135°E and extended west-northwestward to the South China Sea and south-east coast of China. Tropical cyclone activity regions in Period III were spatially varied compared with those in Periods I and II, extending from 110° to 160°E. The axes of storm tracks in Period III were also west-northwestward, similar to Period II. Tropical cyclones developing and traversing the open ocean far from the Western Pacific Coast had longer durations.

2 Interdecadal variability of annual TC frequency and duration, and its relationship with TC activity regions

3 Relationships between TC intensity and activity regions

Figure 1 shows the time series of TC annual frequency, av-

Figure 3 shows the track density of weak TCs (tropical de-

3820

Figure 1 Pacific.

LI WeiBiao, et al.

Chinese Sci Bull


Time series of annual tropical cyclone frequency (a), average duration (b), and normalized annual frequency and duration (c) over the Western

pressions and tropical storms) and strong TCs (typhoons and super typhoons). Weak TCs formed and developed mainly over the ocean basin close to the south-east coast of China (Figure 3(a)). However, strong TCs formed and developed mainly over the ocean east of the Philippines, relatively far from the China’s coast (Figure 3(b)). The prevailing storm tracks of strong TCs are northwestward, extending from the ocean basin east of the Philippines to the east coast of China. The TCs that formed over open ocean moved northwestward and avoided direct landfall on the Philippines, which favored development into typhoons or super typhoons.

4 Broad-scale atmospheric circulation anomalies related to TC climatology Anomalies in TC climatology are linked intimately to large-scale circulation anomalies. The relationships among

wind shear and TC intensity, motion and structure have been documented by numerous previous studies [27–29]. Here we present results on broad-scale vertical motion, geopotential heights and horizontal wind velocity field anomalies related to TC climatology. Figure 4((a)–(c)) shows composite low-level wind anomalies at 1000 hPa from June to September for years of high frequency TC occurrence (1961, 1962, 1964, 1965, 1966, 1967, 1971, 1993, 1994, 1996) and low frequency TC occurrence (1946, 1949, 1950, 1951, 1954, 1955, 1956, 1957, 1973, 1977). These indicate that there were substantial differences broad-scale low-level winds between the high and low frequency occurrence years. During the high frequency years, the western North Pacific was dominated by anomalous cyclonic circulation which provided favorable conditions for TC formation. In addition, there was a strong anomalous westerly over the North Pacific between 5° to 20°N. A westerly burst in the tropics over the western

LI WeiBiao, et al.

Chinese Sci Bull


3821

Figure 3 Track density of tropical depressions and tropical storms (a) and typhoons and super typhoons (b) during the period 1945–2007.

Figure 2 TC track density (total number of TCs for each longitude/latitude grid cell) for the Period I (a), Period II (b) and Period III (c).

North Pacific can be conducive to triggering cyclone genesis, and has been identified in previous research [30]. During the low frequency years, the western North Pacific was occupied by a giant anomalous anti-cyclonic circulation which was unfavorable for TC development. Figure 4(d) shows differences in geopotential height anomalies at 500 hPa from June–September between long duration years (1972, 1976, 1987, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997) and short duration years (1945, 1947, 1948, 1950, 1961, 1964, 1965, 1966, 1969, 1970, 1971). During the long duration years, there were large positive geopotential height

anomalies over the Philippine and South China seas. Cyclone genesis occurred mainly over the ocean basin far from islands and coasts, which was favorable for long-lived TCs. However, during the short duration years, there were large negative geopotential height anomalies over the Philippine and South China seas, suggesting that cyclone genesis occurred mainly over the ocean basin close to islands and coasts, resulting in the anomalous short TC duration. Figure 5 shows vertical velocity features related to TC frequency and duration anomalies. During high-frequency years, the descending zone was located mainly in the Philippine Islands. The South China Sea and ocean basin east of the Philippines were dominated by ascending flow. This kind of vertical velocity background is favorable for TC genesis. On the other hand, during low-frequency years the South China Sea (100°–120°E) and ocean basin east of the Philippines (140°–165°E) were dominated by descending flow. Coincidentally, the averaged first occurrence positions of tropical storms in the western North Pacific Basin were focused on the two regions 100°–120°E and 130°–165°E [31]. Therefore the vertical velocity pattern shown in Figure 6(a) was not conducive to TC occurrence. Moreover, as shown in Figure 6(b) the ascending branch of the anomalous vertical circulation spread over a broad region between 100°E–180° during the long-duration years of the Western

3822

LI WeiBiao, et al.

Chinese Sci Bull


Figure 4 Composite of 1000 hPa wind anomalies from June to September for high TC frequency years (a) and low frequency years (b), differences in 1000 hPa wind from June to September between high and low frequency years (c), and differences in 500 hPa geopotential height from June–September between long and short duration years (d). The shaded areas are significant differences at the 90% confidence level. Units: m s–1 for winds; gpm for geopotential height.

Pacific TCs. An opposite anomalous zonal circulation with a descending branch located between 100° and 160°E and an ascending branch located between 170° and 150°W occurred during the short-duration years. An ascending atmospheric background appears to be crucial to TC development and persistence. In section 3, we identified three types of relationships between TC number and duration during the period 1945–2007: low frequency and short duration from 1945 to 1955 (Period I), high frequency and short duration in the 1960s (Period II), and high frequency and long duration in the 1990s (Period III). It is important to document and understand how the atmospheric circulation background changes during these periods, and for each we examined features of vertical motion, geopotential heights and horizontal wind velocity. Figure 6 shows the cross section along 0°–20°N for the composite vertical velocity anomaly for Period I (Figure 6(a)), Period II (Figure 6 (b)) and Period III (Figure 6(c)). During Period I, the ascending zone was located mainly in the eastern Pacific. The ocean basin east of the Philippines was dominated by descending flow which is

not conducive to TC genesis and development. During Period II, there was a significant descending flow over the Philippines. Weak ascending zones located over the South China Sea and ocean basin east of the Philippines were favorable for TC genesis. However, the descending zone over the Philippines was not conducive to TC persistence as they tracked westward along the southern prevailing storm tracks (Figure 2(b)). During, Period III the ocean basins over the Western Pacific (110°–170°E) were dominated by ascending flow, which was probably an important factor causing high TC frequency and long TC duration in this period. An examination of 1000 hPa wind fields (not shown) and 500 hPa geopotential height (not shown) also indicated that there were substantial differences in background atmospheric circulation from the low to middle troposphere among Periods I, II and III.

5

Conclusions

We examined climatological relationships between TC vari-

LI WeiBiao, et al.

Chinese Sci Bull


3823

Figure 5 Cross section along 0°–20°N for composite vertical velocity anomalies from June to September for highly TC frequency years subtracting low frequency years (a), and long duration years subtracting short duration years (b). The shaded areas are significant differences at the 90% confidence. Unit: Pa s–1.

ability in frequency, duration, intensity and activity regions in the Western Pacific based on JTWC best track and auxiliary atmospheric data. Periods of high TC frequency occurrence do not imply necessarily long TC duration in the same periods. There are three types of relationship between TC number and duration during the period 1945–2007: low frequency and short duration during 1945–1955 (Period I), high frequency and short duration in the 1960s (Period II), and high frequency and long duration in the 1990s (Period III). The TC activity regions differ between the three periods. During Period I the main activity regions were over the ocean basin east of the Philippines (120°–140°E). During period II, there were two prevailing storm tracks both extending west-northwestward between 110°E and 147°E. During period III, TC activity regions were spatially variable extending from 110° to 160°E. Tropical cyclone intensity was related closely to the activity regions. Most of strong TCs developed over the ocean basin far from the Philippines and tracked northwestward, which ensured time for TC development prior to landfall. Analysis of broad-scale atmospheric circulation anomalies suggests that relationships between TC frequency, duration and active regions are modulated strongly by vertical motion, geopotential height and horizontal wind anomalies.

Figure 6 Cross section along 0°–20°N for composite vertical velocity anomalies for Period I (a), Period II (b) and Period III (c). The shaded areas are significant differences at the 90% confidence. Unit: Pa s–1.

This work was supported by the National Natural Science Foundation of China (40875020), the NSFC-Guangdong Joint Fund Program (U0733002) and the National Basic Research Program of China (2009CB421404).

1

2

3 4

5

6

7

Chen L S, Ding Y H. An Introduction to the Western Pacific Typhoons (in Chinese). Beijing: Science Press, 1979. 77–82, 103, 104, 107–109, 247, 248 Chan J C L, Shi J E, Lam C M. Seasonal forecasting of tropical cyclone activity over the western North Pacific and the South China Sea. Weather Forecast, 1998, 13: 997–1004 Meng Z, Chen L, Xu X. Recent progress on tropical cyclone research in China. Adv Atmos Sci, 2002, 19: 103–110 Maloney E D, Hartmann D L. Modulation of eastern North Pacific hurricanes by the Madden-Julian oscillation. J Clim, 2000, 13: 1451– 1460 Chen G H. Interdecadal variation of tropical cyclone activity in association with summer monsoon, sea surface temperature over the western North Pacific. Chinese Sci Bull, 2009, 54: 1417–1421 Chan J C L. Tropical cyclone activity in the western North Pacific in relation to the stratospheric quasi-biennial oscillation. Mon Weather Rev, 1995, 123: 2567–2571 Chan J C L. Tropical cyclone activity over the western North Pacific associated with El Niño and La Niña events. J Clim, 2000, 13: 2960–2972

3824 8 9

10

11

12

13

14

15

16 17

18

19

LI WeiBiao, et al.

Chinese Sci Bull

Wang B, Chan J C L. How strong ENSO events affect tropical storm activity over the western North Pacific. J Clim, 2002, 15: 1643–1658 Wang H J, Sun J Q, Fan K. Relationships between the North Pacific oscillation and the typhoon/hurricane frequencies. Sci China Ser D-Earth Sci, 2007, 50: 1409–1416 Ho C H, Kim J H, Kim H S, et al. Possible influence of the Antarctic Oscillation on tropical cyclone activity in the western North Pacific. J Geophys Res, 2005, 110, D19104, doi: 10.1029/2005JD005766 Wang H J, Fan K. Relationships between the Antarctic oscillation and the typhoon frequencies in the West Pacific. Chinese Sci Bull, 2006, 51: 2910–2914 Zhou B T, Cui X, Zhao P. Relationship between the Asian-Pacific oscillation and the tropical cyclone frequency in the western North Pacific. Sci China Ser D-Earth Sci, 2008, 51: 380–385 Zhou B, Cui X. Hadley circulation signal in the tropical cyclone frequency over the western North Pacific. J Geophys Res, 2008, 113, D16107, doi: 10.1029/2007JD009156 Zhou B, Cui X. Modeling the influence of spring Hadley circulation on the summer tropical cyclone frequency in the west Northern Pacific (in Chinese). Chinese J Geophys, 2009, 52: 2958–2963 Fan K. New predictors and a new prediction model for the typhoon frequency over western North Pacific. Sci China Ser D-Earth Sci, 2007, 50: 1417–1423 Emanuel K A. Increasing destructiveness of tropical cyclones over the past 30 years. Nature, 2005, 436: 686–688 Webster P J, Holland G J, Curry J A, et al. Changes in tropical cyclone number, duration, and intensity in a warming environment. Science, 2005, 309: 1844–1846 Webster P J, Curry J A, Liu J, et al. Response to “Comment on changes in tropical cyclone number, duration, and intensity in a warming environment”. Science, 2006, 311: 1713 Hoyos C D, Agudelo P A, Webster P J, et al. Deconvolution of the


20

21 22 23

24 25

26

27 28 29 30

31

factors contributing to the increase in global hurricane intensity. Science, 2006, 312: 94–97 Holland G J, Webster P J. Heightened tropical cyclone activity in the North Atlantic: Natural variability or climate trend? Phil Trans R Soc A, 2007, 365: 2695–2716 Chan J C L. Interannual variations of intense typhoon activity. Tellus, 2007, 59A: 455−460 Chan J C L. Decadal variations of intense typhoon occurrence in the western North Pacific. Proc R Soc A, 2008, 464: 249−272 Huang R H, Chen G H. Research on interannual variation of tracks of tropical cyclones over northwest pacific and their physical mechanism (in Chinese). Acta Meteorol Sin, 2007, 65: 684–694 Kalnay E, Kanamitsu M, Kistler R, et al. The NCEP/NCAR 4-year reanalysis project. Bull Amer Meteorol Soc, 1996, 77: 437−471 Chan H C L, Shi J. Long-term trends and interannual variability in tropical cyclone activity over the western North Pacific. Geophys Res Lett, 1996, 23: 2765–2767 Yumoto M, Matsuura T. Interdecadal variability of tropical cyclone activity in the western North Pacific. J Meteorol Soc Jpn, 2001, 79: 23–35 Emanuel K, DesAutels C, Holloway C, et al. Environmental control of tropical cyclone intensity. J Atmos Sci, 2004, 61: 843–858 Frank W M, Ritchie E A. Effects on environmental flow upon tropical cyclone structure. Mon Weather Rev, 1999, 127: 2044–2069 Wang Y, Holland G J. Tropical cyclone motion and evolution in vertical shear. J Atmos Sci, 1996, 53: 3313–3332 Ritchie E A, Holland G J. Large-scale pattern associated with tropical cyclogenesis in the western Pacific. Mon Weather Rev, 1999, 127: 2027–2043 Stowasser M, Wang Y, Hamilton K. tropical cyclone changes in the western North Pacific in a global warming scenario. J Clim, 2007, 20: 2378–2396