Statistical Analysis of Typhoons in China and the Western Pacific

Ocean

Yuchen Gao

Department of Atmospheric Sciences, School of Atmospheric Sciences, Nanjing University, Nanjing, China

Keywords: Typhoon, Sea Surface Temperature, ENSO, SOI.

Abstract: Typhoon has the characteristics of sudden strong, destructive power. Typhoons can cause huge economic

losses and casualties to coastal countries and regions, so it is considered to be one of the most serious natural

disasters in the world. The impact of typhoons has gradually increased in recent years. Therefore, the accurate

analysis and prediction of typhoon is particularly important. This paper starts with statistical methods, obtains

relevant data in the last hundred years, analyzes and summarizes some important factors that may affect

typhoon. Indicators such as barometric pressure and sea surface temperature are most likely to affect typhoons.

Therefore, the study started from the subtropical high and SST, collected many years of data and research,

and analyzed the relationship between them and typhoons by statistical methods. By analyzing the data of the

last hundred years and synthesizing the previous studies, the influence of the Western Pacific subtropical high

and ENSO (El Niño-Southern Oscillation) on the Pacific typhoon is discussed. Finally, the correlation

between typhoon and these two factors is obtained through analysis.

1 INTRODUCTION

Typhoons are tropical cyclones occurring in the

northwest Pacific Ocean and the South China Sea with

central sustained winds of 17.2m/s or above, including

tropical storms, severe tropical storms, typhoons,

severe typhoons, and super typhoons (Wang et al 2021).

Typhoon has the characteristics of sudden strong,

destructive power. Typhoon can cause huge economic

losses and casualties to coastal countries and regions,

so it is considered one of the most serious natural

disasters in the world (Kossin et al 2013 & Ying et al

2014).

Favorable atmospheric circulation, sea

temperature field, and other thermal and dynamic

environmental conditions are the direct background of

typhoon development. Traditionally, low-level

cyclone disturbance, warm and humid SST, and small

vertical shear of tropospheric zonal wind are the three

basic environmental conditions for typhoon

occurrence (Zou 2009). A large number of typhoon

case diagnoses and numerical simulation tests have

expanded our understanding of the environmental

factors affecting the change of typhoon intensity, such

as circulation and underlying surface. Cold air, high-

altitude jet stream, westerly trough, small and medium

scale system, topographic action, and sea surface

spray may also have an impact on typhoon intensity.

In addition, the degeneration of tropical cyclones by

mid-latitude circulation systems will also affect their

intensity and track. Therefore, the sea-air environment

conditions of China's offshore waters will directly

affect the active degree of typhoons entering China's

offshore waters. The sea-air environmental conditions

in the South China Sea and the tropical West Pacific

also restrict the number and distribution of typhoons

generated in the sea area, thus indirectly affecting the

offshore typhoons in China (Zheng 2015).

The generation and development of typhoons

cannot be separated from the heat transfer of the ocean

to its system. In the context of the long time scale of

global warming, the activity path and influence range

of typhoons have changed significantly in the

northwest Pacific Ocean.Sea surface temperature (sea

surface temperature, SST) and air-sea flux is one of the

important factors affect the air-sea interaction (Hu

2013). At the same time, since SST and air-sea

interaction are significantly affected by the El Nino-

Southern Oscillation (ENSO), the ENSO event is also

one of the factors affecting typhoons. It is of great

significance to use numerical models to study the

impact of ENSO events on typhoons, to forecast and

defend typhoon disasters, and to assess the impact of

decadal changes of SST on typhoons (Zheng et al

2013).

Gao, Y.

Statistical Analysis of Typhoons in China and the Western Paciﬁc Ocean.

DOI: 10.5220/0012801800003885

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Data Analysis and Machine Learning (DAML 2023), pages 81-86

ISBN: 978-989-758-705-4

2 INFLUENCE OF

INTERDECADAL VARIATION

OF SUBTROPICAL HIGH ON

TYPHOON VARIATION

Northwest Pacific subtropical high (Western North

Pacific Subtropical High WNPSH), hereinafter

referred to as subtropical high. It is a permanent high-

pressure circulation system over the Pacific Ocean,

generally represented by the area surrounded by the

5880 gpm line on a 500 hpa elevation chart over the

western Pacific (Chai et al 2023). As an important

factor affecting the weather system, the inter-annual

variation of subtropical high has an important

influence on various factors of typhoons.

2.1 Research Data and Methods

Typhoon data are obtained from IBTRACS, the best

archive for International Climate Management. The

data includes the relative physical quantities of

tropical storm locations and intensities, with an

interval of 6 h. It records most tropical cyclones

developing in the Northwest Pacific Ocean from 1861

to 2020. It is used to study the annual average

duration, annual frequency, and longitude and latitude

of typhoon path. The 500 hPa altitude field data was

used by the Beijing Climate Centre, BCC-CSM2-

MR_historical_r3i1p1f1 and the French National

Centre for Meteorological Research, CNRM) CNRM

_historical_r2i1p1f2 CM6-1-2 sets of CMIP6 data

with ERA5.

2.2 Influence of Interannual Variation

of the Western Pacific Subtropical

High on Typhoon Characteristics

2.2.1 Statistical Characteristics of Minor

Heights

The Northwest Pacific Subtropical High is one of the

important systems that affect typhoons and other

weather processes. The correlation between the

500hPa altitude field and the northernmost latitude

and westernmost longitude reached by typhoons is

roughly bounded by 40°N, while the relationship

between the north and the south is opposite. The

correlation coefficient between the maximum latitude

reached by typhoon and the 500hPa altitude field is

negative north of 40°N and positive south of 40°N.

This is because the area south of 40°N is the main

active area of the subtropical high, and when the

500hPa altitude field increases and the subtropical

high strengthens, it is conducive to the northbound

typhoon (Chai et al 2023).

In general, it is of practical significance and

research value to study the typhoon activity range by

the decadal variation of the subtropical high, west

ridge and ridge position.

2.2.2 Correlation Between Typhoon Activity

Characteristics and Subtropical High

The region where WNPSH is located is the region

where the sinking branch of Hadley circulation is

located, and the stronger the sinking branch of Hadley

circulation is, the more conducive it is to the sub-high

intensification. The ascending branch of Walker

circulation is located at the equator between 120 and

160°E. When the West Pacific warm pool is strong

and the monsoon trough is more active, the ascending

branch of Walker circulation on the equator is also

stronger, which will make the subtropical high

position to the west and north, which is conducive to

typhoon influence on China (Mu et al 2001). A v-w

wind synthesis analysis of Hadley circulation at

120~160°E and 0~ 40°N was conducted to obtain

the average condition of Hadley circulation at 0~ 40

°N under the zonal average of 120~160°E during

1861 ~ 2020, as shown in figure 1.

Similarly, to study the ascending branch of the

Walker circulation, a synthetic analysis of the average

u-w wind over the equator at 120 to 160°E from 1861

to 2020 was also performed, as shown in figure 2.

Figure 1: V-W wind synthesis analysis of the zonal mean of

120°~160 °E and 0 °~40 °N from 1861 to 2020 (Picture

credit: Original).

DAML 2023 - International Conference on Data Analysis and Machine Learning

Figure 2: Average u-w wind synthesis analysis of 120 °

~160 ° E over the equator from 1861 to 2020 (Picture

credit: Original).

The ascending branch of the Walker circulation at

the equator divides into two branches at 120-160 °E.

One is located in the region above 850hPa from 122 to

127 °E, and the other is located in the region above

850hPa from 150 to 155 °E. Calculate the change in

the mean value of subsidence motion over 160 years

for the large value region of the sinking branch of the

Hadley circulation (the region represented by the blue

box in Figure 1a), defined as wh; And the change in

the mean 160-year upward motion of the high-value

regions of the two Walker circulation rising branches

(the regions represented by the blue boxes in figure 2,

defined as ww (Gong and Wang 2000).

The average 500hPa height field h in the key

WNPSH region is significantly positively correlated

with the vertical velocity wh of the sinking branch of

Hadley circulation and negatively correlated with the

average vertical velocity ww of the rising branch of

Walker circulation. The annual frequency of typhoons

in East China is significantly correlated with h. The

annual frequency affected by typhoons in South China

is significantly correlated with wh and ww. As shown

in figure 3, 4 and 5.

Figure 3: Example of Annual variation of wh and simulation results (Picture credit: Original).

Figure 4: Annual variation of ww and simulation results (Picture credit: Original).

Figure 5: Annual variation of h(c) and simulation results (orange dot line in c) (Picture credit: Original).

Statistical Analysis of Typhoons in China and the Western Paciﬁc Ocean

2.3 Result Analysis

The average 500hPa height field h in the key WNPSH

region is significantly positively correlated with the

vertical velocity wh of the sinking branch of Hadley

circulation and negatively correlated with the average

vertical velocity ww of the rising branch of Walker

circulation. The annual frequency of typhoons in East

China is significantly correlated with h. The annual

frequency affected by typhoons in South China is

significantly correlated with wh and ww. Using t and

h as independent variables, the curve estimation can

successfully simulate the annual frequency of typhoon

influence in East China. Using t, wh, and ww as

independent variables, the curve estimation can

successfully simulate the annual frequency of typhoon

influence in South China.

3 ANALYSIS OF THE

INFLUENCE OF EL NINO-

SOUTHERN OSCILLATION

(ENSO) ON TYPHOONS

Sea surface temperature (SST), as an exogenous

forcing, has an important effect on the atmosphere.

The formation and development of tropical cyclones

are closely related to the size and distribution of SST.

This is because the wide ocean is the main source of

typhoon energy, and the high-temperature seawater

provides energy for the formation and development of

typhoons in the form of sensible heat flux and latent

heat flux. On the other hand, when a tropical cyclone

moves over the sea, the SST is reduced by bringing

cold water from deep layers to the mixed layer through

physical processes such as enrolling and suction. At

the same time, since SST is an important component

of El Nino-Southern Oscillation (ENSO), and the most

intuitive manifestation of ENSO is the inter-annual

variation of SST, ENSO is also one of the important

factors affecting typhoons (Zheng et al 2013).

3.1 Data Sources

The IBTRACS (http://ibtracs.unca.edu/) dataset

includes estimates of typhoon location and intensity at

intervals of 6h, starting from 1884 in the Northwest

Pacific Ocean. The data records 4193 tropical

cyclones that have developed in the Western Pacific

Ocean since 1884. Among them, the period of the best

path data is 1884-2020, and the period of the lowest

central pressure and the maximum wind speed near the

center is 1946-2020. Select the time node to compare

the time of typhoon occurrence. Based on the time

series of typhoon locations provided by IBTRACS

data, the changes of the northernmost latitude,

westernmost longitude, typhoon frequency, frequency

of strong typhoons in East and South China, as well as

the changes of the minimum central pressure, and the

maximum wind speed near the center since 1884 were

analyzed. Based on the changes of WNPSH, West

Pacific warm Pool, ENSO, flow field, radiation flux,

and other fields, the long-term climate change since

1880 is investigated, and the correlation analysis is

made. Based on the changes of 500hPa warm pool,

radiant flux, ENSO, and related physical quantities,

the influence of long-scale changes such as ENSO and

radiant flux on tropical cyclones in the Northwest

Pacific Ocean in the past 100 years was explored, and

comparative analysis was made.

3.2 Influence of ENSO on Typhoon

Activity

Previous analyses have shown that physical quantities

such as vertical wind shear, vorticity, radiation, and

water vapor all have an impact on typhoons, and

ENSO activities may further affect typhoon activities

through the influence of these physical quantity fields.

The main areas of typhoon generation and

development in East Asia and Southeast Asia (105°-

180°E) can be calculated. The time series of the

average vertical wind shear, vorticity, SDLR, and

specific humidity of 5° -30° N, to calculate the

correlation coefficient between these variables and the

Southern Oscillation index (SOI), and calculate the

correlation distribution diagram between the physical

quantity field that passes the 95% significance test and

the typhoon activity frequency at the corresponding

location (Zheng et al 2013).

The SOI index is a representative index to measure

ENSO activities. The correlation coefficient between

the frequency of typhoons in the Northwest Pacific

Ocean since 1884 and the SOI index has reached

0.1337, passing the 90% significance test. The sliding

correlation analysis between the SOI index and the

frequency of typhoons in the Northwest Pacific Ocean

since 1884 is carried out. Thus, the influence of ENSO

activities on the frequency of typhoon generation in

different years was explored (Liang 2021).

As shown in figure 6, 7, 8, in this region, the

typhoon generation frequency in the LaNina year is

the highest, and the typhoon generation frequency in

ElNino year is the lowest, and the typhoon generation

probability in ElNino year is high in the ocean, while

the typhoon generation probability in LaNina year is

high in the offshore.

DAML 2023 - International Conference on Data Analysis and Machine Learning

Figure 6: Annual frequency distribution of typhoons in the

main typhoon-generating regions of the tropical Northwest

Pacific perennial LaNina years (Picture credit: Original).

Figure 7: Annual frequency distribution of typhoons in the

main typhoon-generating regions of the tropical Northwest

lNino (Picture credit: Original).

Figure 8: Annual frequency distribution of typhoons in the

main typhoon-generating regions of the tropical Northwest

Pacific LaNina years (Picture credit: Original).

As shown in figure 9, the positive correlation

between SOI and the frequency of typhoon generation

in the Northwest Pacific Ocean is on the rise, and since

1980, except for a short negative correlation in the early

21st century, the remaining time has been positively

correlated, and the positive correlation between the two

has been significantly enhanced since around 2008.

This positive correlation enhancement trend indicates

that when the SOI index is higher, the frequency of

typhoon generation in the Northwest Pacific Ocean is

higher. When the LaNina event occurs, the frequency

of typhoon generation in the northwest Pacific tends to

be higher (Zhang and Lv 2012).

Figure 9: Analysis of the smooth-sliding correlation between

SOI index and typhoon frequency in the Northwest Pacific

Ocean since 1884 (Picture credit: Original).

Combined with figure 6, 7, 8 and 9, it is found that

ENSO activity can affect the location of typhoon

generation, but the influence on typhoon frequency is

uncertain to a certain extent, and further study on the

influence of ENSO on typhoon activity is needed.

4 CONCLUSION

By analyzing the data of the last hundred years and

synthesizing the previous studies, the influence of the

Western Pacific subtropical high and ENSO on the

Pacific typhoon is discussed.

In the past hundred years, the duration and annual

frequency of typhoons after 1930 have increased in all

regions. The average annual impact duration of

typhoons in the Northwest Pacific during 1975-2020

has decreased compared with that during 1930-1975,

but the annual frequency of typhoons has no

significant change. There is no significant change in

the annual influence duration and frequency in South

China. However, the annual average duration and

annual frequency of typhoons in the East China coastal

area have increased significantly, which is

significantly related to the change of WNPSH in the

past 160 years. The annual frequency of typhoons in

East China is significantly correlated with h. The

annual frequency affected by typhoons in South China

is significantly correlated with wh and ww. Moreover,

Statistical Analysis of Typhoons in China and the Western Paciﬁc Ocean

using t and h as independent variables to estimate the

curve, the annual frequency of typhoon influence in

East China can be simulated successfully. Using t, wh,

and ww as independent variables, the curve estimation

can successfully simulate the annual frequency of

typhoon influence in South China.

There is a general causal relationship between

radiation flux and ENSO-related variables, some of

which are directly caused and some are indirectly

related. In general, changes in radiative fluxes can

affect changes in ocean heat, and changes in ocean

heat can affect changes in atmospheric circulation,

thus creating a causal relationship between radiative

fluxes and ENSO-related variables. In ElNino years,

typhoons tend to form ocean-going, and the typhoon

tracks tend to be east-north, with large intensity

variance. In LaNina year, typhoons tend to generate

near the coast, the typhoon path is west-south, the

intensity variance is small, and the average impact

time of southeast coastal areas of China is long.

There are still some areas that need to be further

explored in this study. First, the intensity of the

physical quantity field involved in this paper can also

be defined from different angles. How to define the

radiation amount, the intensity of ENSO activity and

the intensity of typhoon activity from various angles,

so as to study the interaction between these elements is

a very worthwhile problem. Second, more

comprehensive studies can be conducted from the

perspective of more physical quantities, such as ENSO

changes over long time scales. There are still many

related physical quantities and the interaction between

ENSO worthy of further study. Moreover, the various

ENSO-related physical quantities mentioned in this

paper, such as radiation flux, are also worthy of further

study, so as to have a deeper understanding of the effect

of ENSO changes on typhoons over a long time scale.

REFERENCES

J. Wang, J. Y. Dai, Y. J. Yang, and Y. Y. Lv, paper title

“Statistical characteristics of typhoons in China from

1977 to 2018”. Marine limnology Bulletin. 2021.

J. P. Kossin, T. L. Olander, and K. R. Knapp, paper title

“Trend analysis with a new global record of tropical

cyclone intensity”. Journal of Climate, vol. 26, no. 24,

pp. 9960-9976. 2013.

M. Ying, W. H. Zhang, X. Lu, J. Feng, Y. Fan, and D. Chen,

paper title “An overview of the China Meteorological

Administration tropical cyclone database”. Journal of

Atmospheric and Oceanic Technology, vol. 31, no. 2,

pp. 287-301. 2014.

Y. Zou, paper title “The relationship between offshore

tropical cyclone activity in China and interannual

variation of East Asian subtropical monsoon

Doctoraldissertation”. Chinese Academy of

Meteorological Sciences, 2009.

F. Zheng, paper title “A Study on Sudden Intensification

and Decline of Typhoons in Offshore China”. Chinese

Academy of Meteorological Sciences, 2015.

Y. H. Hu, paper title “Numerical Experimental Analysis of

SST influence on Typhoon”. Ocean University of

China, 2013.

H. Y. Zheng, L. Wu, Z. P. Wen, and T. M Wang, paper title

“Anomaly of typhoon frequency in Northwest Pacific

Ocean and its relationship with air-sea flux”. Journal

of Oceanography, 2013.

B. Y. Chai, F. Xu, J. J. Xu, L. G. Han, S. Q. Chen, Y. J. Li,

and S.J. Zhang, paper title “Interdecadal variation of

the Northwest Pacific Subtropical High in the last 160

years and its influence on the interannual variation of

typhoons”. Journal of Tropical Meteorology, vol. 39,

no. 2, pp. 1004-4965, 2023.

Q. Z. Mu, S. W. Wang, J. H. Zhu, and D.Y. Gong, paper

title “Changes of subtropical high over the Western

Pacific in summer during the last hundred years”.

Journal of Atmospheric Sciences, vol. 25, no. 6, pp.

787-797. 2001.

D. Y. Gong, and S. W. Wang, paper title “An experiment to

recover the 500hPa altitude field in the Northern

Hemisphere in the last hundred years”. Journal of

Tropical Meteorology, vol. 16, no. 2, pp. 148-154.

2000.

X. S. Liang, paper title “Normalized multivariate time series

causality analysis and causal graph reconstruction”.

Entropy, vol. 23, no. 6, pp. 679. 2021.

L. F. Zhang, and Q.P. Lv, paper title “Joint EOF analysis of

SSTA and SLPA in the North Pacific Ocean during

winter”. S3 focuses on climate change and exploring

a low-carbon future. 2012.

DAML 2023 - International Conference on Data Analysis and Machine Learning