CN108320055A - The determination method of multiple river mouth Storm Surge joint return periods - Google Patents

Abstract

The present invention relates to Marine Sciences and field of ocean engineering, a kind of determination method for combining the return period more particularly to multiple river mouth Storm Surges includes the following steps, S1 was counted in a period of time, the generation frequency of typhoon in institute's survey region establishes the Poisson distributions that the frequency occurs for typhoon；S2 counts the river mouth situation in institute's survey region, establishes each river mouth maximum best one-dimensional edge distribution surged caused by typhoon；S3 establishes the one-dimensional Poisson multiple malformations that the frequency occurs for each river mouth maximum compound typhoon of surging caused by typhoon；S4 establishes the multidimensional Poisson multiple malformation that the frequency occurs for all river mouths maximum compound typhoon of surging caused by typhoon in institute's survey region；S5 obtains more river mouth Storm Surges and combines the return period.The present invention proposes a kind of more rational joint probability method for calculating the extreme water level in river mouth harbour, and region resource is allocated, it is significant to realize that coastal region engineering is effectively taken precautions against natural calamities.

Description

Method for Determining the Joint Return Period of Storm Surge Increase in Multiple Estuaries

technical field

The invention relates to the technical fields of marine science and marine engineering, in particular to a method for determining a combined return period of storm surge flooding in multiple estuaries.

Background technique

The extreme high tide level in coastal engineering design refers to the tide level of hydraulic structures under abnormal working conditions, and it is one of the important parameters of structural safety design. my country's "Hydrological Standards for Seaports" stipulates that extreme high tide level is the high water level once in 50 years calculated by Gumbel distribution on the basis of the measured data of the highest tide level not less than 20 years in a row. This method treats the measured water level as a random variable for statistics. In fact, the observed annual extreme water level is usually not caused by astronomical factors alone, but by the combination of the increase and decrease of water generated by typhoons, cold waves, etc. and astronomical tides. In recent years, some scholars have begun to consider the storm surge increase as the main factor of the highest tide level, take the annual maximum increase as a sample for probability calculation, and superimpose its recurrence value on the average high tide level, so that Determine extreme orgasm levels.

In sea areas affected by typhoons, the path and frequency of typhoons are random every year, and the frequency and intensity of storm surges generated are also random. Typhoons do not occur every year, which leads to years without typhoons, which contain zero items, so the annual extreme value method cannot be used to calculate the recurrence value of typhoon wave heights.

For the same typhoon, different estuaries in the affected area have different degrees of storm surge. For an area with three estuaries, individually, each estuary will have a higher level of storm surge, but the probability of extremely large storm surge at the same time is low, that is, the storm surge in different estuaries in the same typhoon affected area. Water has a certain correlation. For the same typhoon, different estuaries in the affected area have different degrees of storm water increase. Therefore, in practical applications, if the extreme water level of each estuary is used to calculate the recurrence value, then the correlation of storm water increase in different estuaries in the same typhoon-affected area is ignored, so that the design value of the extreme high tide level is too large. It will cause a waste of resources.

Contents of the invention

The present invention aims at the problem that there is a deviation between the extreme high tide level and the actual situation in the prior art, and proposes a more reasonable joint probability method for calculating the extreme water level of estuaries and seaports, which is very important for the allocation of regional resources and the realization of effective disaster prevention in coastal areas significance.

In order to achieve the above object, the present invention adopts the following technical scheme, and the method for determining the joint return period of storm surge water increase in multiple estuaries includes the following steps,

S1. Count the frequency of typhoons in the research area within a certain period of time, and establish the Poisson distribution of typhoon frequency;

S2. Statistics of the estuaries in the research area, and establish the best one-dimensional marginal distribution of the maximum water increase caused by typhoons in each estuary;

S3. Establish the one-dimensional Poisson composite extreme value distribution of the frequency of occurrence of the maximum water-increasing composite typhoon caused by the typhoon in each estuary;

S4. Establish the multi-dimensional Poisson composite extreme value distribution of the frequency of the maximum water-increasing composite typhoon caused by typhoons in all estuaries in the study area;

S5. Obtain the joint return period of storm surge flooding in multiple estuaries.

Further, the step S2 specifically includes:

S21. Statistics on the number of estuaries within the study area;

S22. Collect the data of the maximum water increase caused by the typhoon in each estuary;

S23. Select the appropriate distribution line type to fit the maximum water increase data caused by the typhoon in each estuary;

S24. Determine the optimal one-dimensional marginal distribution of maximum typhoon-induced flooding for each estuary.

Further, in the step S23, the linear distribution includes Pearson-III type distribution, Weibull distribution, generalized extreme value distribution and lognormal distribution.

Further, in the step S24, the optimal one-dimensional marginal distribution of the maximum water increase caused by the typhoon for each estuary is determined through the K-S test, the sum of squared deviations of the observed values and estimated values, and the AIC information criterion.

Further, the step S4 specifically includes:

S41. Adopt suitable Copula function to establish the multidimensional Poisson composite extremum distribution of the maximum water-increasing composite typhoon occurrence frequency of all estuaries caused by typhoons;

S42. Determine the optimal multi-dimensional Poisson composite extreme value distribution of the typhoon-induced maximum flooding composite typhoon frequency in all estuaries.

Further, in the step S41, the Copula function includes Normal Copula, Frank Copula, Clayton Copula and Gumbel-Hougaard (G-H) Copula.

Further, in the step S34, the optimal multi-dimensional Poisson composite extreme value distribution of the frequency of occurrence of the maximum water-increasing composite typhoon in all estuaries caused by typhoon is determined by K-S test and AIC information criterion.

The present invention proposes a joint probability model that is more in line with the disaster degree of multiple estuaries in a region under the influence of storm surges, which is of great significance for the joint disaster risk analysis of multiple estuaries. At the same time, for the same typhoon, different estuaries in the affected area have different storm flooding degrees. The storm surge in estuaries in the same affected area has a certain correlation. Through the calculation of the joint probability of storm surge in different estuaries in the same typhoon affected area, especially the synchronous study of storm surge, it can be determined that the same typhoon is affected by the same typhoon in the return period. The typhoon affects the storm water increase of each estuary, which is of great significance to the engineering protection of this area. It can be targeted for regional disaster prevention and mitigation. The estuary area with high typhoon water increase will strengthen engineering protection, and it is not necessary to distribute disaster prevention materials equally in all estuary areas to achieve accurate Prevention and control, providing scientific decision-making for scientific disaster prevention and mitigation.

Description of drawings

Figure 1 is the Poisson distribution fitting of typhoon frequency;

Figure 2 shows the typhoon flooding sequence (1980-2004) at the estuaries of Nankan Creek, Suoxi Creek and Lanyang Creek;

Figure 3 is a scatter diagram of the typhoon water increase sequence at the mouth of the three streams;

Among them, (a) is the scatter diagram of Nankan River; (b) is the scatter diagram of Sulfur River; (c) is the scatter diagram of Lanyang River;

Figure 4 shows the fitting of various distribution probability densities and cumulative distributions of typhoon water increase at the mouth of the Nankan River;

Among them, (a) is the fitting result of Pearson-III; (b) is the fitting result of Weibull; (c) is the fitting result of GEV; (d) is the fitting result of Lognormal;

Figure 5 is the fitting of various distribution probability densities and cumulative distributions of typhoon water increase at the mouth of the Sulfur Creek River;

Among them, (a) is the fitting result of Pearson-III; (b) is the fitting result of Weibull; (c) is the fitting result of GEV; (d) is the fitting result of Lognormal;

Figure 6 shows the fitting of various distribution probability densities and cumulative distributions of typhoon water increase at the mouth of Lanyangxi River;

Among them, (a) is the fitting result of Pearson-III; (b) is the fitting result of Weibull; (c) is the fitting result of GEV; (d) is the fitting result of Lognormal;

Figure 7 is the isosurface of the joint return period of typhoon flooding in Sanhekou;

Figure 8 shows the combined probability of the joint distribution of the Sanhekou typhoon flood increase once in 50 years;

Among them, (a) is the 2% joint probability surface of Sanhekou typhoon water increase; (b) is the side view of the 2% joint probability of Nankan Creek and Sulfur Creek; (c) is the 2% joint probability surface of Nankan Creek and Lanyang Creek Probability side view; (d) is the 2% joint probability side view of Sulfur Creek and Lanyang Creek.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

The method for determining the combined return period of a plurality of estuary storm surge surges of the present invention comprises the following steps,

S1. Count the frequency of typhoons in the research area within a certain period of time, and establish the Poisson distribution of typhoon frequency;

Now collect the number of typhoons in the study area within a period of time, generally in units of years, and establish the Poisson distribution of typhoon frequency.

S2. Statistics on the estuary situation in the research area, establish the best one-dimensional edge distribution of the maximum water increase caused by the typhoon in each estuary; the step S2 specifically includes:

S21. Statistics on the number of estuaries within the study area;

S22. Collect the data of the maximum water increase caused by the typhoon in each estuary;

S23. Choose an appropriate distribution line type to fit the maximum water increase data caused by the typhoon in each estuary; the distribution line includes Pearson-III type distribution, Weibull distribution, generalized extreme value distribution and logarithmic normal distribution.

S24. Determine the best one-dimensional marginal distribution of the maximum water increase caused by the typhoon in each estuary; determine the best one-dimensional marginal distribution through the K-S test, the sum of squared deviations of observed values and estimated values, and the AIC information criterion.

S3. Establish the one-dimensional Poisson composite extreme value distribution of the frequency of occurrence of the maximum water-increasing composite typhoon caused by the typhoon in each estuary;

If the annual typhoon n in a certain area is a discrete random variable, and the extreme wave height in each typhoon process is set to ξ, and the extreme wave height in a typhoon-free year is set to η. Suppose ξ is a continuous random vector, and its probability distribution function is G(x). Let ξi be the i-th observation value of ξ, n is a random variable with non-negative integer value independent of ξ, and its distribution function is recorded as:

Assume random vectors:

Then F0(x) is called a one-dimensional composite extreme value distribution:

Assuming that λ represents the average number of typhoon occurrences per year, if the frequency n of the typhoon process conforms to the Poisson distribution:

From formula (3) we can get:

In the formula, if G(x) is Pearson-III type distribution, Weibull distribution, generalized extreme value (GEV) distribution or lognormal (lognormal) distribution, it can be substituted into formula (5) to get Poisson-Pearson III distribution, Poisson-Weibull distribution, Poisson-GEV distribution, or Poisson-lognormal distribution model.

S4. Establish the multi-dimensional Poisson composite extreme value distribution of the frequency of the maximum water-increasing composite typhoon caused by typhoons in all estuaries in the study area;

S41. Adopt suitable Copula function to establish the multidimensional Poisson composite extremum distribution of the maximum water-increasing composite typhoon occurrence frequency of all estuaries caused by typhoons;

The push process is:

Assume that the N marine environment random variables in each typhoon process are (ξ ₁ , ξ ₂ ,...,ξ _N ), and the extreme marine environment random variables in typhoon-free years are (ζ ₁ , ζ ₂ ,... ·,ζ _N ). Let G(x ₁ ,x ₂ ,···,x _N ) and g(x ₁ ,x ₂ ,···,x _N ) denote random variables (ξ ₁ , ξ ₂ ,···,ξ _N ) The joint distribution function and joint probability density function of , Q(x ₁ ,x ₂ ,···,x _N ) represents the joint distribution function of random variables (ζ ₁ ,ζ ₂ ,···,ζ _N ). Let (ξ _1i , ξ _2i , ξ _3i ,···,ξ _Ni ,) be the i-th observation value of N random variables respectively, and then let n denote a random variable independent of ξ with a non-negative integer value, where The probability distribution function is recorded as formula (1). Now define a random variable (X ₁ ,X ₂ ,..., X _N ):

Then the joint distribution function of (X ₁ ,X ₂ ,···,X _N ) is:

In the formula, the joint distribution function and probability density function of the multidimensional random variable (X ₁ ,X ₂ ,···,X _N ) are G(x ₁ ,x ₂ ,···,x _N ) and g(x ₁ ,x ₂ ,···,x _N ); G _X1 (u ₁ ) is the marginal distribution function of G(x ₁ ,x ₂ ,···,x _N ), namely

G _X1 (u ₁ )＝G(u ₁ ,+∞,…,+∞) (8)

Assuming that the annual typhoon frequency n obeys the Poisson distribution (Equation (4)), Equation (7) can be transformed into:

Equation (9) is the multidimensional Poisson compound extreme value distribution function of random variables (X1, X2,...,XN).

The corresponding probability density function is:

g(x ₁ ,x ₂ ,,x _N ) in formula (10) represents the original joint probability density function of the multidimensional random variable without considering the typhoon frequency. If the copula function is used to establish the original joint distribution model of the multidimensional random variable , g(x ₁ ,x ₂ ,,x _N ) can be expressed as follows:

g(x ₁ ,x ₂ ,…,x _N )=c(x ₁ ,x ₂ ,…,x _N )·f(x ₁ )·f(x ₂ )···f(x _N ) (11)

c(x ₁ ,x ₂ ,,x _N ) in formula (11) is the probability density function of copula function, and f( _xi ) represents the probability density function of univariate.

According to the multidimensional Poisson compound extreme value distribution of formula (9) and formula (10), the probability density function and probability distribution function of the two-dimensional Poisson compound extreme value distribution and the three-dimensional Poisson compound extreme value distribution can be obtained, as shown in Table 2.

Table 2 Two-dimensional and three-dimensional Poisson compound extreme value distribution models

Finally, bring in the appropriate Copula function, including Normal Copula, Frank Copula, Clayton Copula and Gumbel-Hougaard (G-H) Copula for fitting.

S42. Determine the optimal multi-dimensional Poisson composite extreme value distribution of the typhoon-induced maximum flooding composite typhoon frequency in all estuaries. The optimal multidimensional Poisson composite extreme value distribution of the typhoon-induced maximum flooding composite typhoon frequency in all estuaries is determined by the K-S test and the AIC information criterion.

S5. Obtain the joint return period of storm surge flooding in multiple estuaries.

A Poisson multi-dimensional composite probability model is established for typhoon flooding sequences in multiple estuaries, and a joint probability isosurface with a fixed return period can be drawn. At the same time, the joint return period under different typhoon intensification of multiple estuaries can be obtained.

In order to verify the reliability of the method of the present invention, the present invention selects relevant data in the northern region of Taiwan, China to verify the method of the present invention.

1. Calculation process

1. Count the frequency of typhoons in the research area within a certain period of time, and establish the Poisson distribution of typhoon frequency;

Hindcast of historical typhoons affecting northern Taiwan, China from 1980 to 2004. The typhoon conditions selected for calculation are: the distance between the typhoon track and Taiwan, China is not more than 250km. According to statistics, 36 typhoons occurred in the northern part of Taiwan from 1980 to 2004, and the statistics of the number of occurrences each year are shown in Table 1. The Poisson distribution is used to fit the annual typhoon frequency, and the estimated value of the parameter λ is 1.44. The fitting test of the Poisson distribution is shown in Table 2, and the fitting results are shown in Figure 1, and the fitting effect is very good. Therefore, the determination of frequency obeys the Poisson distribution with parameter 1.44.

Table 1 Statistics of typhoons affecting three streams and estuaries from 1980 to 2004

According to Table 1, the test statistic of the Poisson distribution of typhoon occurrences is χ2=1.8241, which is less than the critical value of the hypothesis test at the significant level of 0.05 It shows that the number of typhoon occurrences conforms to the Poisson distribution of λ=36/25=1.44.

Table 2 Statistical test of typhoon frequency

2. Statistics of the estuaries in the research area of the institute, and establish the best one-dimensional edge distribution of the maximum water increase caused by typhoons in each estuary; establish the one-dimensional Poisson composite of the maximum water increase and composite typhoon frequency of each estuary caused by typhoons Extreme value distribution, the recurrence value is obtained through the one-dimensional Poisson compound extreme value distribution.

There are three estuaries in the northern part of Taiwan, namely Nankan River (Nang-Kang), Huang River and Lan-Yang River. The typhoon flooding sequence of the estuaries of Nankan River (Nang-Kang), Huang River and Lan-Yang River from 1980 to 2004 is shown in Figure 2.

The scatter diagrams of the maximum water increase in the estuaries of Nankan River, Suo River and Lanyang River obtained from 36 storms and typhoons are shown in Figure 3 respectively.

Pearson-III distribution, three-parameter Weibull distribution, GEV distribution and Log-normal distribution were used to fit the water increase of the three sequences respectively.

a. Fitting of extreme water increase at the mouth of Nankanxi River

The typhoon water increase sequence at the mouth of the Nankan River was fitted, and the parameter estimation results are shown in Table 3. The fitting curves are shown in Fig. 4 respectively.

Table 3 Parameter estimation of typhoon water increase distribution fitting at the mouth of Nankanxi River

Note: PA, PB and PC represent the location, scale and shape parameters of each distribution, respectively.

Since the number of typhoon occurrences is not considered, the recurrence frequency mentioned on the abscissa in Figure 3 is not the reciprocal of the corresponding return period, but only the recurrence frequency in the data. The meaning of the abscissa of other similar fitting graphs is the same.

At the same time, the K-S test, the sum of squared deviations of observed values and estimated values, and the AIC information criterion were used to compare the fit of the four distributions to the extreme water increase series. Under the condition of confidence α=0.05, the calculation results of K-S statistic D^n and dispersion square RMSE and AIC are shown in Table 4. The fitting results showed that Pearson-III distribution, three-parameter Weibull distribution, GEV distribution and Log-normal distribution all passed the statistical test, and GEV distribution was the best fit, followed by Pearson-III distribution. Using one-dimensional Poisson composite distribution, the recurring values are obtained as shown in Table 5. The results showed that Weibull obtained the smallest recurring value, followed by Pearson-III.

Table 4 The K-S test and the sum of squares of the typhoon water increase distribution fitting at the Nankanxi Estuary

Table 5 Recurrence value (m) calculated by Poisson compound distribution of typhoon water increase at the mouth of Nankanxi River

b. Fitting of extreme water increase at the mouth of the Sulfur Creek

Fit the typhoon flooding sequence at the mouth of the Sulfur Creek River, and the fitting curves are shown in Figure 5. The parameter estimation results are shown in Table 6. The fitting results showed that the Pearson-III distribution, the three-parameter Weibull distribution, the GEV distribution and the Log-normal distribution all passed the statistical test, among which the Pearson-III distribution was the best fit, followed by the Weibull distribution.

Table 6 Parameter estimation of typhoon water increase distribution fitting at the mouth of the Sulfur Creek

Note: PA, PB and PC represent the location, scale and shape parameters of each distribution, respectively.

At the same time, the K-S test, the sum of squared deviations of observed values and estimated values, and the AIC information criterion were used to compare the fit of the four distributions to the extreme water increase series. Under the condition of confidence α=0.05, the calculation results of K-S statistic D^n and deviation square RMSE and AIC are shown in Table 7. The results showed that Pearson-III distribution, three-parameter Weibull distribution, GEV distribution and Log-normal distribution all passed the statistical test, among which Pearson-III fit the best, followed by Weibull. Using one-dimensional Poisson composite distribution, the recurring values are obtained as shown in Table 8. The results showed that Weibull obtained the smallest recurring value, followed by Pearson-III.

Table 7 The K-S test and the sum of squared deviations of typhoon water increase distribution fitting at the mouth of the Sulfur Creek River

Table 8 Recurrence value (m) calculated by Poisson composite distribution of typhoon water increase at the mouth of the Sulfur Creek River

c. Fitting of extreme water increase at the mouth of Lanyangxi

Fit the typhoon flooding sequence at the mouth of Lanyangxi River, the fitting curves are shown in Figure 6, and the parameter estimation results are shown in Table 9. The fitting results show that Pearson-III distribution, three-parameter Weibull distribution, GEV distribution and Log- The normal distributions all passed the statistical test, among which the Pearson-III type distribution fit the best, followed by the Lognormal distribution.

Table 9 Parameter estimation of typhoon water increase distribution fitting at the mouth of the Sulfur Creek

Note: PA, PB and PC represent the location, scale and shape parameters of each distribution, respectively.

At the same time, the K-S test, the sum of squared deviations of observed values and estimated values, and the AIC information criterion were used to compare the fit of the four distributions to the extreme water increase series. Under the condition of confidence α=0.05, the calculation results of K-S statistic D^n and dispersion square RMSE and AIC are shown in Table 10. The results show that Pearson-III has the best fit, followed by Log-normal. Using one-dimensional Poisson compound distribution, the recurring values are obtained as shown in Table 11. The results show that the return value obtained by Weibull is the smallest, followed by GEV.

Table 10 The K-S test and the sum of squared deviations of typhoon water increase distribution fitting at the mouth of the Sulfur Creek

Table 11 Recurrence value (m) calculated by Poisson composite distribution of typhoon water increase at the mouth of the Sulfur Creek River

3. Establish the multi-dimensional Poisson composite extreme value distribution of the frequency of the maximum water-increasing composite typhoon caused by typhoons in all estuaries in the study area;

When constructing the probability correlation models of typhoon water increase in the mouths of the Nankan River, Suoxi River and Lanyang River, the marginal distributions are all selected as the Pearson-III distribution, and the joint probability distribution is based on the Sklar theorem, using four commonly used ternary copulas Functions: Clayton Copula, Frank Copula, and Gumbel-Hougaard (G-H) Copula. The applicability of the model was evaluated by K-S test and AIC method, and the optimal three-dimensional joint probability model was selected.

In Table 12, the ternary G-H model has the smallest RMES and AIC values, followed by the Frank Copula model. Therefore, based on the ternary G-H, this paper established a Poisson three-dimensional P3-P3-P3 composite probability distribution model for the typhoon flooding sequence in Sanhekou, denoted as P-GH-TriP3. Combined probability of typhoon increase in 5-year, 10-year, 20-year, 50-year, 100-year, and 200-year typhoons in Nankanxi, Suoxi and Lanyangxi estuaries, etc. The value surface is shown in Figure 7. In order to observe the probability isosurfaces of different return periods more intuitively, Figure 8 shows three side views with a return period of 50 years, respectively representing the side views of two estuaries when the joint probability of the three estuaries is 2%. , which is the joint probability contour.

Table 12 Optimum selection of three-dimensional Copula model for typhoon water increase in Sanhekou

Table 13 Combined return period of typhoon water increase in Sanhekou from 1980 to 2004

Table 14 Under the given return period, the water increase design value of each estuary

According to the P-GH-TriP3 model, the joint return periods of historical typhoon flooding in the three estuaries are estimated, as shown in Table 13. It can be seen from the table that among all the 36 typhoons in the combination of typhoon water increase in the three estuaries obtained by P-GH-TriP3, only 4 typhoons caused the combined return period of the water increase in the three estuaries to exceed 10 years. Among them, Herb (1996) is the strongest typhoon with a joint return period of 66 years. The other three typhoons were Nockten (2004), Doug (1994) and Brenda (1985) in sequence, and the joint return periods were 18 years, 16 years and 12 years respectively. At the same time, the storm surge of the three estuaries under the fixed return period is obtained, and the storm surge of the three estuaries has certain differences. By establishing a joint probability model of storm surge flooding in multiple estuaries in Taiwan, it can be seen that the risk of typhoons causing disasters on the west, north, and east sides of northern Taiwan is not high at the same time, that is, the model can jointly cause disasters in multiple estuaries Risk analysis. At the same time, the storm water increase in different estuaries during the typhoon process in the joint return period was obtained, see Table 14. Strengthening protection in estuary areas with higher water increase provides a scientific basis for the allocation of disaster prevention and mitigation materials in different estuary areas in disaster-stricken areas.

It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should belong to the protection scope of the appended claims of the present invention.

Claims (7)

1. The determination method of the joint return period of storm surge increase in a plurality of estuaries is characterized in that, comprising the following steps, S1. Count the frequency of typhoons in the research area within a certain period of time, and establish the Poisson distribution of typhoon frequency; S2. Statistics of the estuaries in the research area, and establish the best one-dimensional marginal distribution of the maximum water increase caused by typhoons in each estuary; S3. Establish the one-dimensional Poisson composite extreme value distribution of the frequency of occurrence of the maximum water-increasing composite typhoon caused by the typhoon in each estuary; S4. Establish the multi-dimensional Poisson composite extreme value distribution of the frequency of the maximum water-increasing composite typhoon caused by typhoons in all estuaries in the study area; S5. Obtain the joint return period of storm surge flooding in multiple estuaries. 2. The method for determining the combined return period of a plurality of estuary storm surge surges according to claim 1, wherein said step S2 specifically comprises: S21. Statistics on the number of estuaries within the study area; S22. Collect the data of the maximum water increase caused by the typhoon in each estuary; S23. Select the appropriate distribution line type to fit the maximum water increase data caused by the typhoon in each estuary; S24. Determine the optimal one-dimensional marginal distribution of maximum typhoon-induced flooding for each estuary. 3. The method for determining the joint return period of storm surge increase in a plurality of estuaries according to claim 2, characterized in that, in the step S23, the distribution linearity includes Pearson-III type distribution, Weibull distribution, generalized extreme value distribution and the lognormal distribution. 4. The method for determining the joint return period of storm surge increase in a plurality of estuaries according to claim 2, characterized in that, in the step S24, the sum of squared deviations of the K-S test, observed values, estimated values and AIC information criterion to determine the optimal one-dimensional marginal distribution of maximum typhoon-induced flooding for each estuary. 5. The method for determining the combined return period of a plurality of estuary storm surge surges according to claim 1, wherein said step S4 specifically comprises: S41. Adopt suitable Copula function to establish the multidimensional Poisson composite extremum distribution of the maximum water-increasing composite typhoon occurrence frequency of all estuaries caused by typhoons; S42. Determine the optimal multi-dimensional Poisson composite extreme value distribution of the typhoon-induced maximum flooding composite typhoon frequency in all estuaries. 6. the joint probability method of calculating the extreme water level of estuaries and harbors according to claim 5, is characterized in that, in described step S41, described Copula function comprises Normal Copula, Frank Copula, Clayton Copula and Gumbel-Hougaard (G-H) Copula. 7. The joint probability method for calculating the extreme water level of estuaries and seaports according to claim 5, characterized in that, in the step S34, determine the occurrence of the maximum water-increasing composite typhoon caused by typhoons in all estuaries by K-S test and AIC information criterion An optimal multidimensional Poisson composite extreme value distribution for frequencies.