A strong typhoon forecasting method based on multi-time scale climate factor cooperation in Hainan Island

Abstract

The application discloses a Hainan strong typhoon prediction method based on multi-time scale climate factor cooperation, mainly comprises the following steps: establishing an ENSO index model, establishing a PDO index model, establishing a QBO index model, using the ENSO index and the PDO index to preliminarily construct a cooperative influence index model, and finally constructing a multi-scale climate factor cooperative prediction model. The application can provide a relatively accurate prediction basis for the prediction of the strong typhoon which seriously influences Hainan, and provide decision support for the disaster prevention and reduction of the strong typhoon in Hainan.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of weather prediction and its application, and particularly relates to a Hainan Island strong typhoon prediction method based on multi-time scale climate factor cooperation. BACKGROUND

[0002] Typhoon is one of the main meteorological disasters in China. With the improvement of typhoon prevention measures in recent years, the governments and meteorological departments at all levels have developed emergency plans and measures for different levels of typhoons. Therefore, it is necessary to improve the accurate understanding and prediction of typhoon generation and development.

[0003] Typhoon disaster is the main meteorological disaster in Hainan, among which the strong typhoon affecting Hainan is the most typical. They often bring very serious natural disasters to Hainan, causing huge economic losses and serious threats to people's life and property safety. In recent years, strong typhoon events that seriously affect Hainan Island have occurred from time to time, causing major property and economic losses to Hainan Island. Research on the activity law, climate characteristics and influencing factors of strong typhoons that seriously affect Hainan is of great significance to the disaster prevention and mitigation services in Hainan and the disaster prevention and mitigation work in South China and the South China Sea.

[0004] A large number of studies have shown that tropical cyclone activity is affected by various time scale climate factors, including interdecadal scale (such as PDO), ENSO cycle, quasi-biennial oscillation (QBO) and the like. Pacific decadal oscillation (PDO) as a climate variability signal is widely used in the study of Pacific typhoons. As a strong climate variability signal on the interdecadal time scale, the cold and warm phases of PDO are important backgrounds of interannual variability. It not only has a significant impact on the climate of the Pacific and North America, but also affects the climate of the Yangtze River and South China during the flood season, and has an important influence on the generation frequency, intensity or moving path of the Northwest Pacific typhoon.

[0005] ENSO (El Nino-Southern Oscillation) as an important influencing factor of climate change is the strongest sea-air interaction event affecting global climate, which is a short name for El Nino in the tropical Pacific and Southern Oscillation in the tropical atmosphere, and is an important modulating factor of interannual variation of tropical Pacific cyclone activity, which has an impact on the frequency, intensity, location and moving path of tropical cyclones. Research shows that in warm event years, the frequency of tropical cyclones landing in China is less, and in cold event years, the frequency of tropical cyclones landing is more. The stronger the El Nino, the fewer the number of tropical cyclones landing, and the longer the duration, and the opposite is true for La Nina events.

[0006] The quasi-biennial oscillation (QBO) is the main mode of interannual variability in the tropical stratosphere, and is a periodic variation of the wind field in the lower stratosphere near the equator. Although the QBO phenomenon only occurs in the tropics, it can modulate the circulation outside the tropics through various channels, thereby having a non-negligible impact on global weather and climate. The QBO has a certain modulating effect on the track of TCs in the Northwest Pacific.

[0007] The prediction of strong typhoon events affecting Hainan Island is the focus and difficulty of Hainan flood season climate prediction and service. However, there are few studies on the climate prediction of typhoon intensity at present, and there is still a lack of effective prediction methods in business. At present, the main research on the prediction of strong typhoons affecting Hainan Island is the role of single time scale climate factors such as ENSO, PDO, QBO, and the application research of sea-air configuration and synergy of different time scales in typhoon prediction is relatively less. Therefore, it is still not possible to provide accurate prediction basis for strong typhoons in Hainan Island and to provide decision support for disaster prevention and reduction of strong typhoons in Hainan Island. SUMMARY

[0008] In order to overcome the defects existing in the prior art, the application provides a Hainan strong typhoon prediction method based on multi-time scale climate factor synergy, establishes an ENSO index, a PDO index, a QBO index and a multi-index synergy prediction model, and can provide a more accurate prediction basis for the prediction of strong typhoons seriously affecting Hainan, and provide decision support for disaster prevention and reduction of strong typhoons in Hainan Island.

[0009] To achieve the above object, the technical scheme provided by the application is as follows:

[0010] A Hainan strong typhoon prediction method based on multi-time scale climate factor synergy mainly comprises the following steps:

[0011] Step 1, dividing the ENSO state into strong, weak and medium intensity, selecting the applicable ENSO threshold value through the internal relationship between the ENSO state and the occurrence of strong typhoons, and establishing a winter ENSO and flood season ENSO index model;

[0012] Step 2, selecting the applicable PDO threshold value through the internal relationship between the PDO state and the occurrence of strong typhoons, and establishing a PDO index model;

[0013] Step 3, parameter adjustment and design test of the models in steps 1 and 2 to obtain the optimal weight coefficient between the winter ENSO index, the flood season ENSO index and the PDO index;

[0014] Step 4, constructing an ENSO, PDO synergy influence index F(x) model according to steps 1, 2 and 3;

[0015] Step 5, the optimal QBO threshold is selected by the intrinsic relationship between QBO state and the occurrence of strong typhoon, and a winter QBO index QBO win model is constructed.

[0016] Step 6, the difference between summer QBO and winter QBO is constructed QBO dif model.

[0017] Step 7, the QBO comprehensive index QBO 综合 model is constructed according to step 5 and step 6.

[0018] Step 8, finally, the multi-time scale climate factor cooperative prediction model is constructed according to step 4 and step 7.

[0019] Further, before 1989, when the average SSTA of Nino3.4 area in winter is greater than 1 standard deviation or less than-1 / 3 standard deviation, the ENSO index is defined as "1", otherwise "-1"; when the average SSTA of Nino3.4 area in the flood season is greater than 1 standard deviation or less than-1 / 3 standard deviation, the ENSO index is defined as "1", otherwise "-1", that is, the step 1 ENSO winter index model is,

[0020]

[0021] Wherein, σ is the standard deviation of the average SSTA of Nino3.4 area in winter (December of the previous year to February of the current year) in all years;

[0022] The ENSO flood season index model is,

[0023]

[0024] Wherein, σ is the standard deviation of the average SSTA of Nino3.4 area in the flood season (May to October) in all years;

[0025] After 1989, when the average SSTA of Nino3.4 area in winter is between 0 and 1 / 3 standard deviation, the ENSO index is defined as "1", otherwise "-1"; when the average SSTA of Nino3.4 area in the flood season is between 0 and 1 / 3 standard deviation, the ENSO index is defined as "1", otherwise "-1": that is, the step 1 ENSO winter index model is,

[0026]

[0027] Wherein, σ is the standard deviation of the average SSTA of Nino3.4 area in winter (December of the previous year to February of the current year) in all years;

[0028] The ENSO flood season index model is,

[0029]

[0030] wherein σ is the standard deviation of the average Nino3.4 region SSTA in the flood season (May-October) in the past years;

[0031] The above x i is the average Nino3.4 region SSTA in winter or in the flood season of different years, and the ENSO state is generally divided by calculating the standard deviation σ of the Nino3.4 region SSTA, wherein the ENSO state intensity is defined as follows: the years with a standard deviation greater than 1 are strong El Nino or strong La Nina state years; the years with a standard deviation less than 1 / 3 are weak El Nino or weak La Nina state years; and the years with a standard deviation between 1 / 3 and 1 are moderate intensity El Nino state or moderate intensity La Nina state years.

[0032] When the PDO index of a certain year is negative, the PDO index is defined as “1”, otherwise as “-1”, and the step 2 PDO index model is as follows:

[0033]

[0034] Further, in the period from 1949 to 2009 for constructing the index, the years in which the winter ENSO index and the severe strong typhoon event affecting Hainan Island are of the same sign are 38 years, the years in which the flood season ENSO index and the severe strong typhoon event affecting Hainan Island are of the same sign are 49 years, and the years in which the PDO index and the severe strong typhoon event affecting Hainan Island are of the same sign are 34 years, so after parameter adjustment and weight design test, if the sum of the winter ENSO index, the flood season ENSO index and the PDO index is 1, the optimal weight coefficients of the three indexes are finally selected as 0.3, 0.4 and 0.3 respectively, that is, the step 4 ENSO, PDO cooperative influence index model is as follows:

[0035] F(x) = 0.3*f(x) + 0.4*g(x) + 0.3*h(x)

[0036] wherein f(x) refers to the winter ENSO index, g(x) refers to the flood season ENSO index, and h(x) refers to the PDO index.

[0037] Further, since the sum of the number of years greater than 10 and the number of years less than -20 is close to 5% of the total 60 years of data from 1950 to 2019, that is, 10 and -20 are extreme critical values, when the average QBO index in winter is greater than 10, the index model is defined as “-1”, indicating that it is not conducive to the occurrence of strong typhoon; when it is less than -20, the index model is defined as “1”, indicating that it is conducive to the occurrence of strong typhoon; and the rest is defined as “0”, indicating that it cannot be determined, and the step 5 winter QBO index model is as follows:

[0038]

[0039] The difference between the summer QBO and the winter QBO is used to represent the sharp transition between the east wind phase and the west wind phase throughout the year, and two critical values of 30 and -30 are selected as the critical values of the difference between the summer QBO and the winter QBO according to the test, and when the difference between the summer QBO and the winter QBO is greater than 30, the index model is defined as "-1", indicating that it is not conducive to the occurrence of strong typhoon; when it is less than -30, the index model is defined as "1", indicating that it is conducive to the occurrence of strong typhoon; the rest is defined as "0", indicating that the influence is uncertain, and the step 6 difference between the summer QBO and the winter QBO model is as follows:

[0040]

[0041] Further, the step 7 QBO comprehensive index model is:

[0042]

[0043] The model shows that: when QBO win and QBO dif both have obvious promoting effect on the strong typhoon event seriously affecting Hainan Island, then the QBO index finally has promoting effect on the strong typhoon event seriously affecting Hainan Island in the current year; both have no obvious promoting or inhibiting effect on the strong typhoon event seriously affecting Hainan Island, then the QBO index finally has no obvious promoting or inhibiting effect on the strong typhoon event seriously affecting Hainan Island in the current year; one has no obvious promoting or inhibiting effect on the strong typhoon event seriously affecting Hainan Island, and the other has obvious inhibiting effect on the strong typhoon event seriously affecting Hainan Island, then the QBO index model finally has obvious inhibiting effect on the strong typhoon event seriously affecting Hainan Island in the current year; both have obvious inhibiting effect on the strong typhoon event seriously affecting Hainan Island, then the QBO index model finally has obvious inhibiting effect on the strong typhoon event seriously affecting Hainan Island in the current year.

[0044] Further, the step 8 multi-time scale climate factor cooperative prediction model is:

[0045]

[0046] The model shows that when the QBO comprehensive index is 0, the correction of the cooperative influence index is not considered, otherwise the cooperative influence index is corrected, that is, when the QBO comprehensive index has obvious promoting (inhibiting) effect on the strong typhoon event seriously affecting Hainan Island in the current year, no matter the state of Nino3.4 region sea surface temperature and PDO phase, due to the influence of strong east wind (west wind) shear in the lower stratosphere, the tropical deep convection activity is (promoted) inhibited, which is still conducive (not conducive) to the occurrence of Hainan Island strong typhoon event.

[0047] The Hainan strong typhoon prediction method based on multi-time scale climate factor synergy provided by the application has at least the following beneficial effects compared with the prior art:

[0048] Since most of the current researches are respectively analyzed from different time scales to affect the tropical cyclone activity, the common action cannot be discussed from the perspective of the multi-time scale synergy, and the service demand of the Hainan tropical cyclone prediction and prediction cannot be met. The application utilizes the modulation effect of different time scale climate factors (PDO, ENSO, QBO) on the strong typhoon event seriously affecting Hainan Island, constructs a synergy influence index, and the test shows that the prediction accuracy of the synergy influence index on the strong typhoon event seriously affecting Hainan Island in the past 10 years reaches 100%. In addition, the conceptual model of the strong typhoon event seriously affecting Hainan Island before 1989 and after 1989 is constructed, and the prediction of the future strong typhoon event has certain indicative significance, the physical mechanism is analyzed from different time scales, the conceptual model is established, and technical support is provided for the Hainan disaster prevention and reduction service work. BRIEF DESCRIPTION OF DRAWINGS

[0049] The drawings described herein are used to provide further understanding of the application, constitute a part of the application, the illustrative embodiments of the application and the description thereof are used to explain the application, and do not constitute improper limitation on the application. In the drawings:

[0050] Figure 1 The flow chart of the Hainan strong typhoon prediction method based on multi-time scale climate factor synergy of the embodiment of the application is shown in the figure.

[0051] Figure 2 The synergy influence index value of 1950-2019 for model test of the embodiment of the application is shown in the figure.

[0052] The implementation of the object of the application, the functional characteristics and the advantages will be further explained with reference to the embodiments and the drawings. DETAILED DESCRIPTION

[0053] It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

[0054] The application will be further described below with reference to the drawings.

[0055] English interpretation:

[0056] SSTA (sea surface temperature anomaly): average value of sea surface temperature anomaly.

[0057] For the prediction of strong typhoon events affecting Hainan Island, previous studies have shown that there is no significant correlation between strong typhoon events affecting Hainan and different phases of PDO, ENSO and QBO, and that ENSO is mainly modulated under different PDO backgrounds, while the modulation of QBO is based on ENSO. Based on this understanding, an index of the coordinated influence of PDO, QBO and ENSO is constructed.

[0058] The tropical cyclone data used in the method analysis from 1949 to 2018 is mainly from the tropical cyclone best path data set compiled by the Shanghai Typhoon Institute and the Typhoon Yearbook compiled by the China Meteorological Administration. The study selects the equatorial central and eastern Pacific ocean surface temperature anomaly (Nino 3.4 SSTA) as the characteristic value of ENSO, and the data of these characteristic values come from the monthly equatorial Pacific Nino 3.4 (5°N-5°S, 120°-170°W) SSTA (1950-2018) published by the Climate Prediction Center (CPC) of the U.S. National Weather Service (NOAA). The Pacific Decadal Oscillation (PDO) index is taken from the NOAA data; the QBO index is taken as the average zonal wind in the equatorial region (5°S-5°N) at 30 hPa. The time series used for constructing the coordinated influence index is from 1949 to 2009, and the period from 2010 to 2019 is used to test the accuracy of the coordinated influence index.

[0059] Referring to the severe impact area in the typhoon business impact area map of Hainan Province, the severe impact Hainan Island strong typhoon event is defined as: after entering the severe impact area, the intensity is above 14 (strong typhoon), and the typhoon reaches 1 observation time and above, which is defined as a strong typhoon event that severely affects Hainan Island.

[0060] The embodiment of the application provides a Hainan Island strong typhoon prediction method based on multi-time scale climate factor coordination, which comprises the following steps: Figure 1 The method mainly comprises the following steps:

[0061] I. ENSO index model establishment

[0062] The ENSO index is a global sea-air interaction event, and sea surface temperature is the most important characteristic element. The equatorial central and eastern Pacific ocean surface temperature anomaly (Nino 3.4 SSTA) is usually used as the characteristic value of ENSO.

[0063] The analysis of the ENSO index usually uses the SSTA of the Nino 3.4 region in the northwest Pacific. The standard deviation of the SSTA of the Nino 3.4 region is calculated to divide the ENSO state. The standard deviation is shown as follows:

[0064]

[0065] σ is the standard deviation.

[0066] In the design process, the years greater than 1 standard deviation are defined as strong El Nino or strong La Nina state years; the years less than 1 / 3 standard deviation are defined as weak El Nino or weak La Nina state years; and the years between 1 / 3 and 1 standard deviation are defined as moderate intensity El Nino state or moderate intensity La Nina state years.

[0067] Previous studies have shown that the sea surface temperature in the Northwest Pacific Ocean has undergone a mutation around 1989, so when processing the ENSO index, it is necessary to analyze before and after 1989 respectively. The construction of the ENSO index model also uses the SSTA of the Nino3.4 area in winter and the flood season to comprehensively consider the changes in the sea surface temperature state. Therefore, the SSTA values of the Nino3.4 area in winter and the flood season are respectively processed separately before and after 1989.

[0068] Before 1989, when the average SSTA of the Nino3.4 area in winter is greater than 1 standard deviation or less than -1 / 3 standard deviation, the ENSO index is defined as "1", otherwise it is "-1"; when the average SSTA of the Nino3.4 area in the flood season is greater than 1 standard deviation or less than -1 / 3 standard deviation, the ENSO index is defined as "1", otherwise it is "-1", that is, the step 1 ENSO winter index model is,

[0069]

[0070] Wherein, σ is the standard deviation of the average SSTA of the Nino3.4 area in winter (December of the previous year to February of the current year);

[0071] The ENSO flood season index model is,

[0072]

[0073] Wherein, σ is the standard deviation of the average SSTA of the Nino3.4 area in the flood season (May to October);

[0074] After 1989, when the average SSTA of the Nino3.4 area in winter is between 0 and 1 / 3 standard deviation, the ENSO index is defined as "1", otherwise it is "-1"; when the average SSTA of the Nino3.4 area in the flood season is between 0 and 1 / 3 standard deviation, the ENSO index is defined as "1", otherwise it is "-1"; that is, the step 1 ENSO winter index model is,

[0075]

[0076] Wherein, σ is the standard deviation of the average SSTA of the Nino3.4 area in winter (December of the previous year to February of the current year);

[0077] ENSO flood season index model is,

[0078]

[0079] where σ is the standard deviation of the average Nino3.4 region SSTA in the past years (May-October) flood season;

[0080] The above x i is the average Nino3.4 region SSTA in winter or flood season of different years,

[0081] In the design process of this method, "1" indicates that it is conducive to the occurrence of strong typhoon, "-1" indicates that it is not conducive to the occurrence of strong typhoon, and the following are designed according to this definition.

[0082] II. PDO index model

[0083] Studies have shown that PDO cold phase years are more conducive to the occurrence of strong typhoon events that seriously affect Hainan Island. Therefore, the treatment of PDO index is relatively simple. When the PDO index of a certain year is negative, define the PDO index as "1", otherwise as "-1". The model is:

[0084]

[0085] III. Preliminary construction of synergistic influence index model

[0086] First, use the ENSO index and PDO index to preliminarily construct the synergistic influence index model.

[0087] Assign weights a, b and c to the winter ENSO index, flood season ENSO index and PDO index respectively, and define the synergistic influence index as:

[0088] F(x) = a*f(x) + b*g(x) + c*h(x)

[0089] 1949-2019, if a strong typhoon event occurs in Hainan Island that year, regardless of the number of occurrences, it is recorded as "1", otherwise as "-1". From 1949 to 2009, the winter ENSO index and the strong typhoon event that seriously affected Hainan Island were the same in 38 years, the flood season ENSO index and the strong typhoon event that seriously affected Hainan Island were the same in 49 years, and the PDO index and the strong typhoon event that seriously affected Hainan Island were the same in 34 years.

[0090] In the design process of this method, after parameter adjustment and weight design test, the optimal weight coefficients are finally selected as: a=0.3, b=0.4, c=0.3.

[0091] F(x) = 0.3 * f(x) + 0.4 * g(x) + 0.3 * h(x)

[0092] Four, QBO index model building

[0093] Research shows that, in the QBO easterly phase years, more likely to occur severe typhoon events affecting Hainan Island, and in the QBO westerly phase, severe typhoon events affecting Hainan Island are relatively less. When the winter QBO index is in the extremely strong westerly phase, it will significantly inhibit the occurrence of severe typhoon events affecting Hainan Island. When the winter QBO index is in the extremely strong easterly phase, it is conducive to the occurrence of severe typhoon events affecting Hainan Island. Because the winter QBO index is not normally distributed, the absolute value of the negative index is much larger than the positive index, and the number of negative index is more than the positive index.

[0094] (1) QBO winter model building

[0095] In the design of this method, combined with statistical rules, after many tests, when the average winter QBO (X i ) index is greater than 10, define the index model as "-1", indicating that it is not conducive to the occurrence of strong typhoon; less than -20, define the index model as "1", indicating that it is conducive to the occurrence of strong typhoon; the rest is defined as "0", indicating that it cannot be determined. Among them: the sum of the number of years greater than 10 and the number of years less than -20 is close to 5% of the total 60 years of data from 1950 to 2019, that is, 10 and -20 are extreme critical values. The QBO winter model is:

[0096]

[0097] Among them, QBOwin is the winter QBO index, "1" indicates that it is conducive to the occurrence of strong typhoon, "-1" indicates that it is not conducive to the occurrence of strong typhoon, and "0" indicates that the impact is uncertain.

[0098] (2) Summer QBO and winter QBO difference model building

[0099] It is one-sided to consider only the winter QBO index. There is a possibility that the winter QBO is the westerly phase (easterly phase) and then turns into the easterly phase (westerly phase). Therefore, in the design process, in order to comprehensively consider the effect of the QBO index on the strong typhoon events that seriously affect Hainan Island, the summer QBO, which is more mature, is selected, and the difference between the summer QBO and the winter QBO is used to represent the sharp transition between the easterly phase and the westerly phase throughout the year. Combined with statistical rules, after several tests, the difference between the two critical values of 30 and -30 is selected as the critical value of the difference between the summer QBO and the winter QBO. When the difference between the summer QBO and the winter QBO is greater than 30, the index model is defined as "-1", indicating that it is not conducive to the occurrence of strong typhoons; when it is less than -30, the index model is defined as "1", indicating that it is conducive to the occurrence of strong typhoons; the rest is defined as "0", indicating that the impact is uncertain. The physical meaning is that when the QBO index sharply transitions from the extreme westerly phase in winter to the extreme easterly phase in the typhoon season, even if the previous phase is the westerly phase, the sharp increase in the easterly phase will also effectively promote the occurrence of strong typhoon events that seriously affect Hainan Island, and when the QBO index sharply transitions from the extreme easterly phase in winter to the extreme westerly phase in the typhoon season, due to the sharp increase in the westerly phase, it will also inhibit the occurrence of strong typhoon events that seriously affect Hainan Island. That is, the model is:

[0100]

[0101] QBO dif is the difference between the summer QBO and the winter QBO, representing the sharp transition between the easterly phase and the westerly phase throughout the year.

[0102] (3) QBO comprehensive index model establishment

[0103] After considering QBO win and QBO dif , the specific meaning is that when either QBO win or QBO dif has a significant promoting effect on strong typhoon events that seriously affect Hainan Island, the QBO index ultimately has a promoting effect on strong typhoon events that seriously affect Hainan Island in the current year; when neither has a significant promoting or inhibiting effect on strong typhoon events that seriously affect Hainan Island, the QBO index ultimately has no significant promoting or inhibiting effect on strong typhoon events that seriously affect Hainan Island in the current year; when one has no significant promoting or inhibiting effect on strong typhoon events that seriously affect Hainan Island, and the other has a significant inhibiting effect on strong typhoon events that seriously affect Hainan Island, the QBO index model ultimately has a significant inhibiting effect on strong typhoon events that seriously affect Hainan Island in the current year; when both have a significant inhibiting effect on strong typhoon events that seriously affect Hainan Island, the QBO index model ultimately has a significant inhibiting effect on strong typhoon events that seriously affect Hainan Island in the current year. The specific index model is:

[0104]

[0105] V. Final construction of the multi-time scale climate factor collaborative prediction model

[0106] The preliminary constructed collaborative influence index is revised by the QBO comprehensive index. When the QBO comprehensive index is 0, no revision of the collaborative influence index is considered, otherwise the collaborative influence index is revised. The physical meaning is that when the QBO comprehensive index has a significant promoting (inhibiting) effect on the strong typhoon event of Hainan Island in the current year, regardless of the Nino3.4 region sea surface temperature state and PDO phase, due to the influence of the strong easterly (westerly) shear in the lower stratosphere, the tropical deep convection activity is (promoted) inhibited, which is still conducive (not conducive) to the occurrence of the strong typhoon event of Hainan Island. That is, the multi-time scale climate factor collaborative prediction model is:

[0107]

[0108] VI. Test of the prediction model

[0109] According to the collaborative index model we obtained, the annual collaborative influence index of ENSO, PDO and QBO from 1950 to 2019 is calculated. From Figure 2 It can be seen that the accuracy of 2010-2019 (the year used to test the collaborative influence index) is significantly higher than that of 1950-2009 (the year used to construct the collaborative influence index). Among the 60 years from 1950 to 2009, the collaborative influence index correctly indicated the strong typhoon event of 46 years, and incorrectly indicated the strong typhoon event of 14 years, with a correct rate of 65.7%. Among the 10 years from 2010 to 2019, the collaborative influence index correctly indicated the strong typhoon event of 10 years, with a correct rate of 100%. It can be seen that the collaborative influence index has good indicating significance for the prediction of the strong typhoon event that seriously affects Hainan Island.

[0110] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be realized by general computing devices, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices, and alternatively, they can be realized by program codes executable by computing devices, so that they can be stored in storage devices and executed by computing devices, and in some cases, the steps shown or described can be executed in different order, or they can be manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

Claims

1. A method for forecasting severe typhoons in Hainan Island based on the synergy of multiple time-scale climate factors, characterized in that... The method includes the following steps: Step 1: Divide the ENSO status into strong, weak, and moderate intensities. Select the appropriate ENSO threshold based on the intrinsic relationship between the ENSO status and the occurrence of strong typhoons, and establish winter ENSO and flood season ENSO index models. Step 2: Select the appropriate PDO threshold based on the intrinsic relationship between PDO status and the occurrence of strong typhoons, and establish a PDO index model; Step 3: Perform parameter adjustment and design experiments on the models described in Step 1 and Step 2 to obtain the optimal weighting coefficients among the winter ENSO index, the flood season ENSO index, and the PDO index. Step 4: Based on Steps 1, 2, and 3, initially construct the synergistic impact index of ENSO and PDO. Model; Step 5: Select the optimal QBO threshold by analyzing the intrinsic relationship between QBO state and the occurrence of strong typhoons, and construct the winter QBO index QBO. win Model; Step 6: Construct the difference QBO between summer and winter QBO. dif Model; Step 7: Construct the QBO composite index QBO based on steps 5 and 6. 综合 Model; Step 8: Finally, based on steps 4 and 7, construct a multi-timescale climate factor synergistic forecasting model.

2. The method according to claim 1, characterized in that, Prior to 1989, the ENSO index was defined as "1" when the average Nino 3.4 zone SSTA during winter was greater than 1 standard deviation or less than -1 / 3 standard deviation; otherwise, it was "-1". Similarly, the ENSO index was defined as "1" when the average Nino 3.4 zone SSTA during the flood season was greater than 1 standard deviation or less than -1 / 3 standard deviation; otherwise, it was "-1". Therefore, the ENSO winter index model in step 1 was... i≤1988 ； Where σ is the standard deviation of the average Nino3.4 zone SSTA from December of the previous year to February of the current year during the winter; The ENSO flood season index model is as follows: i≤1988； Where σ is the standard deviation of the average Nino3.4 zone SSTA during the flood season of each year, i.e., from May to October; After 1989, the ENSO index was defined as "1" when the average Nino 3.4 zone SSTA during winter was between 0 and 1 / 3 standard deviations, and "-1" otherwise; the ENSO index was also defined as "1" when the average Nino 3.4 zone SSTA during the flood season was between 0 and 1 / 3 standard deviations, and "-1" otherwise: That is, the ENSO winter index model in step 1 is... i>1988； Where σ is the standard deviation of the average Nino3.4 zone SSTA from December of the previous year to February of the current year during the winter; The ENSO flood season index model is as follows: i>1988； Where σ is the standard deviation of the average Nino3.4 zone SSTA during the flood season of each year, i.e., from May to October; The above For the average Nino3.4 zone SSTA of different years during winter or flood season, the ENSO state is classified by calculating the standard deviation σ of the Nino3.4 zone SSTA. The intensity is defined as follows: years with an intensity greater than 1 standard deviation are strong El Niño or strong La Niña years; years with an intensity less than 1 / 3 standard deviation are weak El Niño or weak La Niña years; and years with an intensity between 1 / 3 and 1 standard deviation are moderate El Niño or moderate La Niña years. When the PDO index is negative in a given year, the PDO index is defined as "1"; otherwise, it is "-1". The PDO index model in step 2 is as follows: 。 3. The method according to claim 2, characterized in that, Between 1949 and 2009, the years used to construct the indices, the winter ENSO index coincided with the number of severe typhoon events affecting Hainan Island in 38 years, the flood season ENSO index coincided with the number of severe typhoon events affecting Hainan Island in 49 years, and the PDO index coincided with the number of severe typhoon events affecting Hainan Island in 34 years. Therefore, after parameter adjustment and weight design experiments, if the sum of the three indices—winter ENSO index, flood season ENSO index, and PDO index—is 1, then the optimal weight coefficients are finally selected as 0.3, 0.4, and 0.3, respectively. That is, the ENSO and PDO synergistic impact index model in step 4 is as follows: ； in, Refers to the ENSO index in winter. For the ENSO index during the flood season, This is the PDO index.

4. The method according to claim 1, characterized in that, Since the sum of the number of years with an index greater than 10 and the number of years with an index less than -20 is close to 5% of the data from the 70 years from 1950 to 2019, meaning that 10 and -20 are extreme critical values, the index model is defined as "-1" when the average winter QBO index is greater than 10, indicating that it is unfavorable for the occurrence of strong typhoons; and "1" when it is less than -20, indicating that it is favorable for the occurrence of strong typhoons; the rest are defined as "0", indicating that it cannot be clearly determined. The winter QBO index model in step 5 is as follows: ； The difference between summer and winter QBO is used to represent the dramatic shift between easterly and westerly phases throughout the year. Based on experiments, the difference between two critical values, 30 and -30, is selected as the critical values ​​for the difference between summer and winter QBO. When the difference between summer and winter QBO is greater than 30, the exponential model is defined as "-1", indicating that it is unfavorable for the occurrence of strong typhoons; when it is less than -30, the exponential model is defined as "1", indicating that it is favorable for the occurrence of strong typhoons; otherwise, it is defined as "0", indicating that the impact is uncertain. The difference model between summer and winter QBO in step 6 is as follows: 。 5. The method according to claim 4, characterized in that, The QBO composite index model in step 7 is as follows: ； This model shows that when QBO win and QBO dif If either the QBO index or the typhoon index significantly promotes the occurrence of severe typhoons affecting Hainan Island, then the QBO index will ultimately promote the occurrence of severe typhoons affecting Hainan Island in that year; if neither of the two indexes significantly promotes or inhibits the occurrence of severe typhoons affecting Hainan Island, then the QBO index will ultimately have no significant promoting or inhibiting effect on the occurrence of severe typhoons affecting Hainan Island in that year. If one of the factors has no significant promoting or inhibiting effect on severe typhoon events affecting Hainan Island, while the other has a significant inhibiting effect, then the QBO index model will ultimately have a significant inhibiting effect on severe typhoon events affecting Hainan Island in that year; if both factors have a significant inhibiting effect on severe typhoon events affecting Hainan Island, then the QBO index model will ultimately have a significant inhibiting effect on severe typhoon events affecting Hainan Island in that year.

6. The method according to claim 1, characterized in that, The multi-timescale climate factor synergistic prediction model in step 8 is as follows: ； The model shows that when the QBO composite index is 0, its correction to the synergistic impact index is not considered; otherwise, the synergistic impact index is corrected. That is, when the QBO composite index has a significant promoting effect on the severe typhoon events affecting Hainan Island in the current year, regardless of the sea surface temperature in Nino3.4 region and the PDO phase, the tropical deep convection activity is promoted due to the influence of the strong easterly shear in the lower stratosphere, which is still conducive to the occurrence of severe typhoon events in Hainan Island. When the QBO composite index has a significant inhibiting effect on the severe typhoon events affecting Hainan Island in the current year, regardless of the sea surface temperature in Nino3.4 region and the PDO phase, the tropical deep convection activity is inhibited due to the influence of the strong westerly shear in the lower stratosphere, which is still unfavorable to the occurrence of severe typhoon events in Hainan Island.

Images