JP7157479B2 - Prediction method and application of Japanese flying squid abundance based on the Pacific Oscillation Index - Google Patents

Description

The present invention belongs to the technical field of squid resource forecasting, and relates to a method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index and its application.

The mica Todarodes pacificus (also known as Japanese flying squid) is an important economic cephalopod resource in the world, distributed exclusively in the Pacific Northwest and the Gulf of Alaska in the eastern Pacific. It is distributed mainly in the western Pacific Ocean at 21°-50°N, ie, the Sea of Japan, the Pacific coast of Japan, the Yellow Sea in China, and the East China Sea. It is a warm-temperate marine shallow-water species that inhabits water layers 500 m above the sea surface and has a wide suitable temperature range. According to the spawning season, growth type and migration route of mica, it is divided into three stocks: a winter-developing stock, an autumn-developing stock and a summer-developing stock. They have different life cycles but the same habits. The winter-developing stock is the most widely distributed, and before the 1970s, this stock was the largest in number, and the spawning ground was located on the outer edge of the East China Sea continental shelf in the southwestern part of Kyushu, mainly concentrated in the central and northern parts of the East China Sea. The spawning season is from January to March, and in spring and summer they travel north along both sides of the Japanese archipelago to forage, and in autumn and winter they travel south to lay eggs.

Mica is one of the earliest and most extensively developed species of cephalopods in the world. Before the 1970s, its catch accounted for 70-80% of the total cephalopod catch in Japan. According to FAO statistics, in 1968 the total mica catch exceeded 750,000 tons, a record high. However, as the fishing intensity improved, the catch decreased year by year. In 1986, at just over 120,000 tons, it was the lowest catch since 1950. After that, it increased continuously until 1996, and the annual catch reached about 700,000 tons. After that, it decreased again, and now the total mica catch is stable at 320,000 to 420,000 tons.

The main catch of the fall stock of mica is from the Sea of Japan, and the main fishing season is from May to October. Japan and South Korea are the major catching countries, and North Korea and China also catch small amounts. In coastal waters, it is mainly small squid fishing vessels (less than 30 tons), and the catch is fresh. In the open sea, medium-sized squid fishing boats (30 to 185 tons) use frozen catches. In addition to squid fishing, there are set net operations and bottom trawling operations. Japan's catch reached 300,000 tons in the latter half of the 1970s, then declined to a mere 50,000 tons in 1986, and then increased to 70,000 to 180,000 tons in the 1990s, stabilizing. In recent years, the cumulative catches of Japan and Korea are 100,000 to 200,000 tons.

Winter stocks of mica are mainly caught by Japan and Korea, and operations mainly include fishing, bottom trawling, set nets, purse seines, and so on. The main fishing grounds start from the Joban-Sanriku Pacific coast in July, move to the Pacific coast of Hokkaido from September to November, move to the Sea of Japan side from November, and reach the end of the fishing season (December to the following year). February) in the northwestern Kyushu area. Winter stock catches peaked in the 1950s and 1960s. The main fishing ground is the Pacific Ocean area in the eastern part of Hokkaido. After that, the catch declined sharply, reaching its lowest level in the 1980s. After 1989, the catch recovered somewhat, reaching 380,000 tons in 1996. Since then, the catch has always fluctuated significantly. Currently, the catch is 200,000 to 300,000 tons.

Mica resources are susceptible to marine environmental factors. Zhang Shuo et al. (2017) reported the catch per unit effort (CPUE) of the mica winter development stock in 2000-2010 and the spawning ground (28°-40°N, Based on sea surface temperature (SST) data from 125° to 140° E), statistically significant SSTs were selected as factors influencing stock abundance, and multivariable linear and BP neural network stock abundance, respectively. Build predictive models. According to research, the sea surface temperature in the sea areas around 30°-32°N, 135°-138°E and 37°-38°N, 129°-131°E is similar to warm currents (Kuroshio and Tsushima It represents the strength of the ocean current) and influences the resource abundance of the mica winter outbreak stock of the current year, and the created BP neural network model can be used as a prediction model for the resource abundance. Hu Fei-fei et al. (2015) reported that, based on the Japanese report of the stock assessment of mica's fall-generating stock, the sea surface temperature (SST) of the spawning grounds, and the mass concentration of chlorophyll-a (Chl-a), the spawning rate of mica during the spawning season Correlation between various environmental factors such as the ratio (PS) of the optimum monthly sea surface temperature range to the total area (PS), tSST representing the environment of the spawning ground, Chl-a, etc., and the catch per unit effort (CPUE) was calculated and analyzed, and various resource replenishment prediction models based on major environmental factors were developed. As can be seen from the above studies, scholars at home and abroad are now intensively researching the impact of the spawning ground environment on the replenishment of the autumn/winter squid stocks of Japanese flying squid, and the corresponding stock levels. A prediction model was created, but the question of how to use climate factors to predict resource abundance is still left blank.

Therefore, developing a method to predict the abundance of Japanese flying squid based on climatic factors is very realistic.

SUMMARY OF THE INVENTION It is an object of the present invention to overcome the problems and deficiencies existing in the prior art, and to provide a method and application for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index.

（４）上記最多２^ｍ－１個のスルメイカ資源豊度予測モデルのうち、統計的Ｐ値が最も小さいモデルを最適モデルとして選択するステップと、を含む。 To achieve the above objectives, the present invention provides the following technical solutions,
A method for predicting Japanese flying squid abundance based on the Pacific Oscillation Index applied to an electronic device and used to guide deep-sea fishing for Japanese flying squid, comprising:
(1) a step of obtaining monthly Pacific Oscillation Index PDO values in sea areas where Japanese common squid were distributed in the past N years, wherein the Japanese common squid is a common Japanese flying squid autumn or winter common;
(2) Using the time-series analysis method, perform a correlation analysis on the CPUE of Japanese flying squid stock and the monthly PDO values for the past N years, and select the PDO values for the months that show a statistically significant correlation. Then, the PDO values of these selected months are used as climatic factors that affect the abundance of Japanese flying squid, and the PDO values of these selected months are numbered in the order of 1, 2, 3 ... z ... m, and these the _PDO values of the months in order x ₁ , x ₂ , x ₃ . . . x _z .
(3 ⁾ For any 1 to _m climate factors among x ₁ , x ₂ , x ₃ . and calculate the statistical P-value for each prediction model, and the formula for the Japanese flying squid abundance prediction model is
CPUE= _a +b1*x1+ _b2 *x2+b3* _x3 +...+ _bz * _{xz +} ... _{+ bm} * _{xm ,}
where CPUE is the daily catch of small and medium squid fishing boats in Japan, a is a constant, b ₁ , b ₂ , b ₃ , ... b _z ..., b _n are x ₁ , x ₂ , x ₃ respectively … x _z … is the coefficient corresponding to x _m ,

(4) Selecting the model with the smallest statistical P-value from among the 2 ^m −1 largest Japanese common squid abundance prediction models as the optimum model.

According to the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index according to the present invention, the relationship between the climate factor (Pacific Oscillation Index) and the abundance of Japanese flying squid is created for the first time, and the climate factor (Pacific Oscillation Index) It can realize rapid and accurate forecasting of Japanese flying squid stock abundance, and plays an excellent guiding role in marine fishery production (catch of Japanese flying squid autumn stock/winter Japanese flying squid stock), greatly improving fishing efficiency, The cost can be reduced, the future can be expected, and the optimal model obtained by the present invention is not uniform, and the optimal model can be reacquired according to the latest data acquired in real time. Adaptability is high and future potential can be expected.

Preferred technical solutions include:
In the above method for predicting the abundance of Japanese common squid based on the Pacific Oscillation Index, after obtaining the optimum model, x ₁ , x ₂ , x ₃ ... x _z ... x _m corresponding to the optimum model are obtained and input to the optimum model. Then, when the optimum model outputs the abundance of Japanese flying squid, the prediction of the abundance of Japanese flying squid is completed.

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, when the flying squid prediction model is created for the climate factors _xz , b ₁ , b ₂ , b ₃ , . . . b _z−1 , b _z+1 , . , b _n are all 0, and similarly below, when the Japanese flying squid abundance prediction model is created for the climate factors x ₁ and x _z , b ₂ , b ₃ , ... b _z-1 , b _z+1 , . . . , _bn are all zero.

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, the statistically significant correlation is a calculated P value <0.05. P value is a parameter for judging the result of hypothesis testing in statistics, and P value (P value) is the probability of occurrence of a sample observation result or a more extreme result obtained when the null hypothesis is true. is. A very small P-value indicates that the probability of the null hypothesis situation occurring is very small; There are sufficient reasons to reject the hypothesis. In short, a smaller P-value indicates a more significant result.

The Japanese common squid fall stock is distributed in the Sea of Japan.
The winter squid population is distributed in the Pacific coast of Hokkaido.

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, the Japanese flying squid is an autumn squid stock,
In step (2), select the PDO values of two months that show a statistically significant correlation, that is, the PDO value in October of the past two years and the PDO value of October in the past one year, The correlation between stock abundance CPUE and October PDO values in the past two years was significant and showed a negative correlation, with a correlation coefficient of −0.390 (P<0.05), The correlation between the autumn squid stock abundance CPUE and the PDO value in October of the past first year was significant and negative, with a correlation coefficient of -0.4486 (P<0.05). and
According to the analysis of the correlation between the abundance CPUE of Japanese flying squid and the monthly PDO values of the same year, the correlation between the abundance CPUE of Japanese flying squid and the PDO values from January to December of the same year was not significant. do not have.

The number of prediction models referred to here and below is merely a part of the data to show the working logic of the prediction method of the present invention, and the scope of protection of the present invention is not limited thereto. Appropriate data can be selected as necessary to predict the abundance of Japanese common squid, and the number of selected monthly PDO values that exhibit a statistically significant correlation is not limited to two, and the prediction model varies depending on the number of statistically significant correlation month PDO values selected.

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, in step (3), the following three prediction models are created,
1) Prediction model I
For the climate factor PDO value x ₁₁ in October in the past two years and PDO value in October in the first year x ₁₂ , create a model for predicting the abundance of common Japanese flying squid in autumn in the Sea of Japan, and for,
CPUE=2.3463−0.1674*x ₁₁ −0.1977*x ₁₂ and
The F value is 4.9268, P = 0.0161 < 0.05, the F value is the statistic of the F test, and the F test is under the null hypothesis (H0). A test that follows the F-distribution and usually analyzes a statistical model with two or more parameters to determine whether all or some of the parameters in the model are applicable to estimate the population. is used to test a linear relationship,
2) Prediction model II
A model for predicting the abundance of Japanese flying squid in autumn in the Sea of Japan is created for the PDO value x ₁₁ in October in the past two years of the climate factor. Specifically,
CPUE= _{2.3894-0.2127} *x11,
F value is 4.4922, P = 0.0442 < 0.05,
3) Prediction model III
A model for predicting the abundance of Japanese flying squid in autumn in the Sea of Japan is created for the PDO value in October of the past first year of the climate factor x _12. Specifically,
CPUE=2.3958-0.2323* _x12 ,
The F value is 6.2984 and P=0.0189<0.05.

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, in step (4), the prediction model I is selected as the optimum model, and the optimum model is
CPUE=2.3463−0.1674*x ₁₁ −0.1977*x ₁₂ .

In the method for predicting the abundance of Japanese common squid based on the Pacific Oscillation Index, the Japanese common squid is a winter squid stock,
In step (2), 10 monthly PDO values exhibiting a statistically significant correlation: Select the PDO value for April, the PDO values for 1, 2, 3 and April of the same year,
The correlation between the winter squid stock abundance CPUE and the PDO values in October, November and December in the past two years was significant and showed a negative correlation, with a correlation coefficient of -0.4506 (P <0.05), −0.4985 (P<0.05) and −0.5878 (P<0.01);
The correlation between the winter squid abundance CPUE and the PDO values in January, February, March and April in the past year was significant and negative, with a correlation coefficient of -0.4665. (P<0.05), −0.4365 (P<0.05), −0.4295 (P<0.05) and −0.5072 (P<0.01);
The correlation between the winter squid abundance CPUE and the PDO values in January, February, March and April of the same year was significant and negative, with a correlation coefficient of -0.4746 (P< 0.05), −0.4837 (P<0.05), −0.5458 (P<0.01) and −0.5570 (P<0.01).

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, in step (3), the following five prediction models are created,
1) Prediction model 1
PDO value x ₂₁ in November of the past second year of the climate factor, PDO value x ₂₂ in December in the second year in the past, PDO value x ₂₃ in February of the same year, PDO value x ₂₄ in March of the same year and the past For the PDO value x ₂₅ in March of the first year, create a model for predicting the abundance of Japanese flying squid autumn outbreak stocks in the Sea of Japan. Specifically,
CPUE=1.2048+0.0330*x ₂₂ −0.1811*x ₂₁ −0.2260*x ₂₃ +0.0749*x ₂₄ −0.0196*x ₂₅ ,
F value is 4.5183, P = 0.0069 < 0.01,
2) Prediction model 2
PDO value x ₂₁ in November of the past second year of the climate factor, PDO value x ₂₂ in December in the second year in the past, PDO value x ₂₃ in February of the same year, and PDO value x ₂₄ in March of the same year In this way, we created a stock abundance prediction model for Japanese flying squid in autumn in the Sea of Japan.
CPUE = _1.1968 + _0.0273 * _x22 - _0.1865 *x21 - 0.2290*x23 + 0.0740*x24, and
F value is 5.9135, P = 0.0026 < 0.01,
3) Prediction model 3
PDO value x ₂₃ in February of the same year, PDO value x ₂₄ in March of the same year, and PDO value x ₂₅ in March of the first year of the past year of the climate factor Create a predictive model, specifically:
CPUE=1.3257−0.1120*x ₂₃ −0.0461*x ₂₄ −0.1307*x ₂₅ and
F value is 5.1699, P = 0.0078 < 0.01,
4) Prediction model 4
For the PDO value of February of the same year x ₂₃ and the PDO value of March of the same year x ₂₄ of the climate factor, a Japanese flying squid autumn outbreak stock abundance prediction model was created. Specifically,
CPUE= _{1.3093-0.0792} * _x23-0.1223 *x24,
F value is 5.1233, P = 0.0149 < 0.05,
5) Prediction model 5
For the climate factor PDO value x ₂₁ in November in the past two years and December PDO value x ₂₂ in the past two years, create a model for predicting the abundance of common Japanese flying squid in autumn in the Sea of Japan. In terms of
CPUE = 1.1582 + 0.0383* _x22 - _0.2125 *x21,
The F value is 5.8894 and P=0.0089<0.01.

In the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index, in step (4), the prediction model 2 is selected as the optimum model, and the optimum model is
CPUE=1.1968+0.0273*x ₂₂ −0.1865*x ₂₁ −0.2290*x ₂₃ +0.0740*x ₂₄ .

Based on common knowledge in this field, the above preferred conditions can be arbitrarily combined to obtain preferred examples of the present invention.

The invention further provides an electronic device, comprising one or more processors, one or more memories, one or more programs, and a data collection device,
The data acquisition device is used to obtain _{x 1} , x ₂ , x ₃ . . . x _z . A plurality of programs, when executed by the processor, cause the electronic device to perform a method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index.

(1) According to the method for predicting the abundance of flying squid based on the Pacific Oscillation Index according to the present invention, by realizing the prediction of the abundance of flying squid using the Pacific Oscillation Index PDO, marine fisheries In production (fishing of Japanese flying squid), it plays an excellent guiding role, greatly improves fishing efficiency, reduces fishing costs, and has a promising future.

(2) According to the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index according to the present invention, the optimum model is not uniform, and the optimum model can be reacquired according to the latest data obtained in real time, The method of the present invention is highly adaptable and promising.

(3) The electronic device according to the present invention has a simple configuration, is low in cost, can quickly predict the abundance of Japanese flying squid on the basis of the Pacific Oscillation Index PDO, and is expected to have future potential.

FIG. 1 is a flow chart of a method for predicting Japanese flying squid abundance based on the Pacific Oscillation Index according to the present invention. Figure 2 shows the yearly changes in CPUE abundance of the Japanese flying squid autumn outbreak stock from 1990 to 2016. Fig. 3 is a change distribution diagram of actual and predicted values of abundance CPUE of Japanese flying squid autumn-occurring stock in the Sea of Japan from 1990 to 2016. Figure 4 is a schematic diagram showing annual changes in CPUE abundance of the Japanese flying squid winter development stock from 1992 to 2016. FIG. 5 is a change distribution diagram of the actual and predicted values of CPUE for winter-occurring Japanese common squid stock abundance in the Sea of Japan from 1992 to 2016. FIG. 6 is a structural schematic diagram of an electronic device according to the present invention.

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the following clearly and completely describes the technical solutions of the embodiments of the present invention with reference to the drawings of the embodiments of the present invention. As discussed and apparent, the described embodiments are some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments that a person skilled in the art can come up with without creative efforts shall fall within the protection scope of the present invention.

As shown in Figure 1, the methodology for predicting Japanese flying squid abundance based on the Pacific Oscillation Index, which is used to guide deep-sea catches of flying squid, is as follows:
Obtain monthly Pacific Oscillation Index PDO values for sea areas where Japanese flying squid have been distributed over the past N years. is the sea area of the Sea of Japan, and the sea area where the Japanese flying squid winter outbreak stock is distributed is the Pacific Ocean side sea area of Hokkaido, Step 101;
Using the time-series analysis method, perform a correlation analysis on the CPUE of the Japanese flying squid stock and the monthly PDO values for the past N years. The selected PDO values of these months are used as climatic factors that affect the abundance of Japanese flying squid, and the PDO values of these selected months are numbered in the order of 1, 2, 3 ... z ... m, and the PDO values of these months are step 102, where the _PDO values are in order x ₁ , x ₂ , x ₃ . . . x _z .
Create a maximum 2 ^m −1 Japanese common squid abundance prediction model using multivariable linear equations for arbitrary 1 to m climate factors out of x ₁ , x ₂ , x ₃ … x _m . , to calculate the statistical P-value for each prediction model, and the formula for the Japanese flying squid abundance prediction model is
CPUE= _a +b1*x1+ _b2 *x2+b3* _x3 +...+ _bz * _{xz +} ... _{+ bm} * _{xm ,}
where CPUE is the daily catch of small and medium squid fishing boats in Japan, a is a constant, b ₁ , b ₂ , b ₃ , ... b _z ..., b _n are x ₁ , x ₂ , x ₃ respectively . . x _z . . . x _m are coefficients corresponding to step 103;
step 104 of selecting the model with the smallest statistical P value from among the maximum 2 ^m -1 Japanese flying squid abundance prediction models as the optimal model;
x ₁ , x ₂ , x ₃ . . . _{x z} . and step 105 .

Example 1

1. Materials and methods

(1) Data source

The autumn squid population is widely distributed in the surrounding waters of the Sea of Japan, and the main fishing grounds for fishing are distributed in the Sea of Japan. PDO). The Pacific Oscillation Index is a Pacific climate change phenomenon that varies with a decadal cycle. The conversion cycle is typically 20-30 years. PDO is characterized by abnormally high or low sea surface temperatures in the Pacific north of 20 degrees north latitude. During the ``warm phase'' (or ``positive phase'') of the Pacific decadal oscillation, the western Pacific is cold and the eastern Pacific is warm. Cold. PDOs are from the website of the University of Washington, USA (http://research.jisao.washington.edu/pdo/PDO.latest.txt), with a period from January 1988 to December 2017 (Table 1). be.

The Japanese flying squid fall-occurring stock abundance index CPUE (in tons/vessel) is from the catches of Japanese small and medium-sized squid fishing vessels for the period 1990-2016 (Table 2).

(2) Research method and steps

Using the daily catch CPUE of small and medium-sized squid fishing boats in Japan as an indicator of the abundance of Japanese flying squid autumn stock abundance, the CPUE of Japanese flying squid autumn stock abundance and 1988-2016 were analyzed using the time-series analysis method. Correlation analysis was performed on the PDO values for January-December of the year, among which the PDO values of the months showing a statistically significant correlation (P-value < 0.05) were selected, and the PDO values of these selected months were selected. are the climatic factors that affect the abundance of Japanese flying squid autumn-occurring stock resources, and the PDO values of these selected months are numbered in the order of 1, 2, 3, z, m, and the PDO values of these months are x ₁ , x ₂ , x ₃ … x _z … x _m (where m is the number of PDO values in the selected month), and
Abundance prediction of maximum 2 ^m −1 Japanese flying squid stock abundance using multivariable linear equations for arbitrary 1 to m climatic factors among x ₁ , x ₂ , x ₃ … x _m Create a model, calculate the statistical P-value for each prediction model, and the formula for the Japanese flying squid autumn outbreak stock abundance prediction model is
CPUE= _a +b1*x1+ _b2 *x2+b3* _x3 +...+ _bz * _{xz +} ... _{+ bm} * _{xm ,}
In the formula, CPUE is the catch per boat per day, a is a constant, b ₁ , b ₂ , b ₃ , ... b _z ..., b _n are x ₁ , x ₂ , x ₃ ... x _z ... x _m is the coefficient corresponding to
Select the model with the smallest statistical P value from among the above 2 ^m -1 Japanese flying squid autumn stock abundance prediction models as the optimal model,
Obtain x ₁ , x ₂ , x ₃ … x _z … x _m corresponding to the optimal model and input them to the optimal model. Prediction of resource quantity is completed.

2. Research results

(1) Changes in annual resource abundance CPUE

As can be seen from Figure 2, the autumn squid abundance CPUE showed large annual variation, with low abundance levels in 1990-1992, 2001-2002, 2004-2005, and 2015-2016. , 2001-2003 and 2008-2009 had high abundance levels.

(2) PDO value affecting resource abundance CPUE

As a result of analyzing the correlation between the resource abundance CPUE and the monthly PDO value in the past two years, the correlation between the resource abundance CPUE and the PDO value in October in the past two years was significant and negative. showing a correlation, each with a correlation coefficient of −0.390 (P<0.05);
As a result of analyzing the correlation between the resource abundance CPUE and the monthly PDO value in the past year, the correlation between the resource abundance CPUE and the PDO value in October in the past year was significant and negative. showing a correlation with a correlation coefficient of −0.4486 (P<0.05), respectively;
As a result of analyzing the correlation between the resource abundance CPUE and the monthly PDO value of the same year, the correlation between the resource abundance CPUE and the PDO value for January to December of the same year is not significant.

(3) Creation of resource abundance prediction model

1) Prediction model I
For the climate factor PDO value x ₁₁ in October in the past two years and PDO value in October in the first year x ₁₂ , create a model for predicting the abundance of Japanese flying squid in autumn in the Japan Sea, and for,
CPUE=2.3463−0.1674*x ₁₁ −0.1977*x ₁₂ and
F value is 4.9268, P = 0.0161 < 0.05,
A statistical table of actual and predicted values is shown in Table 3.

2) Prediction model II
A model for predicting the abundance of Japanese flying squid in autumn in the Sea of Japan is created for the PDO value x ₁₁ in October in the past two years of the climate factor. Specifically,
CPUE= _{2.3894-0.2127} *x11,
F value is 4.4922, P = 0.0442 < 0.05,
A statistical table of actual and predicted values is shown in Table 4.

3) Prediction model III
A model for predicting the abundance of Japanese flying squid in autumn in the Sea of Japan is created for the PDO value in October of the past first year of the climate factor x _12. Specifically,
CPUE=2.3958-0.2323* _x12 ,
F value is 6.2984, P = 0.0189 < 0.05,
A statistical table of actual and predicted values is shown in Table 5.

As a result of comparative analysis of the above three models, the prediction model I is selected as the optimal model, and the optimal model is CPUE=2.3463−0.1674*x ₁₁ −0.1977*x ₁₂ . Enter x ₁₁ and x ₁₂ corresponding to the 1990-2016 results into the best fit model to obtain the predicted values (take the 2000 results as an example, x ₁₁ is the PDO value for October 1998, x ₁₂ is Figure 3 shows the trend of change in abundance between the actual and predicted values. Yes, in other words, the method of the present invention can effectively predict the abundance of Japanese flying squid autumn-occurring stocks in the Sea of Japan.

Example 2

1. Materials and methods

(1) Data source

The winter squid population is widely distributed in the waters surrounding the Sea of Japan, and the main fishing grounds for fishing are distributed in the Pacific coastal waters of Hokkaido. Decadal Oscillation, PDO). The Pacific Oscillation Index is a Pacific climate change phenomenon that varies with a decadal cycle. The conversion cycle is typically 20-30 years. PDO is characterized by abnormally high or low sea surface temperatures in the Pacific north of 20 degrees north latitude. During the ``warm phase'' (or ``positive phase'') of the Pacific decadal oscillation, the western Pacific is cold and the eastern Pacific is warm. Cold. PDOs are from the website of the University of Washington, USA (http://research.jisao.washington.edu/pdo/PDO.latest.txt) with a period from January 1990 to December 2017 (Table 6). be.

The Japanese flying squid winter-occurring stock abundance index CPUE (in tons/vessel) is from the catches of small and medium-sized Japanese squid fishing vessels for the period 1992-2016 (Table 7).

(2) Research method and steps

Using the daily catch CPUE of small and medium-sized squid fishing boats in Japan as an indicator of the abundance of Japanese flying squid in winter, we used the time-series analysis method to compare the abundance of Japanese flying squid in winter with CPUE from 1990 to 2016. Correlation analysis was performed on the PDO values for January-December of the year, among which the PDO values of the months showing a statistically significant correlation (P-value < 0.05) were selected, and the PDO values of these selected months were selected. are the climatic factors that affect the abundance of Japanese flying squid wintering stocks, and the PDO values of these selected months are numbered in the order of 1, 2, 3, z, m, and the PDO values of these months are x ₁ , x ₂ , x ₃ … x _z … x _m (where m is the number of PDO values in the selected month), and
Abundance prediction of maximum 2 ^m −1 Japanese common squid winter-occurring stock abundance using multivariable linear equations for arbitrary 1 to m climate factors among x ₁ , x ₂ , x ₃ … x _m Create a model, calculate the statistical P-value for each prediction model, and the formula for the Japanese flying squid winter occurrence stock abundance prediction model is
CPUE= _a +b1*x1+ _b2 *x2+b3* _x3 +...+ _bz * _{xz +} ... _{+ bm} * _{xm ,}
In the formula, CPUE is the catch per boat per day, a is a constant, b ₁ , b ₂ , b ₃ , ... b _z ..., b _n are x ₁ , x ₂ , x ₃ ... x _z ... x _m is the coefficient corresponding to
Select the model with the smallest statistical P value from among the above 2 ^m -1 Japanese flying squid winter season abundance prediction models as the optimal model,
Obtain x ₁ , x ₂ , x ₃ … x _z … x _m corresponding to the optimal model and input them to the optimal model, and when the optimal model outputs the winter squid abundance, Prediction of resource quantity is completed.

2. Research results

(1) Changes in annual resource abundance CPUE

As can be seen from Figure 4, the Japanese flying squid wintering stock abundance CPUE showed large annual variation, with high abundance levels in 1992, 1996, 2007-2009 and 2011, but high abundance levels in 1998-1999 and 2006. In 2015-2016, the abundance level was low.

(2) PDO value affecting resource abundance CPUE

As a result of analyzing the correlation between the resource abundance CPUE and the monthly PDO value for the past two years, the correlation between the resource abundance CPUE and the PDO value for October-December in the past two years was significant. showing a negative correlation, with correlation coefficients of −0.4506 (P<0.05), −0.4985 (P<0.05), −0.5878 (P<0.01), respectively;
As a result of analyzing the correlation between the resource abundance CPUE and the monthly PDO value in the past year, the correlation between the resource abundance CPUE and the PDO value for January to April in the past year is significant. A negative correlation was shown, with correlation coefficients of −0.4665 (P<0.05), −0.4365 (P<0.05), −0.4295 (P<0.05), −0. 5072 (P<0.01);
As a result of analyzing the correlation between the resource abundance CPUE and the PDO value for each month of the same year, the correlation between the resource abundance CPUE and the PDO value for January to April of the same year was significant and showed a negative correlation. The correlation coefficients were −0.4746 (P<0.05), −0.4837 (P<0.05), −0.5458 (P<0.01), −0.5570 (P<0.05), respectively. 01).

(3) Creation of resource abundance prediction model

1) Prediction model 1
PDO value x ₂₁ in November of the past second year of the climate factor, PDO value x ₂₂ in December in the second year in the past, PDO value x ₂₃ in February of the same year, PDO value x ₂₄ in March of the same year and the past For the PDO value x ₂₅ in March of the first year, create a model for predicting the abundance of Japanese flying squid autumn outbreak stocks in the Sea of Japan. Specifically,
CPUE=1.2048+0.0330*x ₂₂ −0.1811*x ₂₁ −0.2260*x ₂₃ +0.0749*x ₂₄ −0.0196*x ₂₅ ,
F value is 4.5183, P = 0.0069 < 0.01,
A statistical table of actual and predicted values is shown in Table 8.

2) Prediction model 2
PDO value x ₂₁ in November of the past second year of the climate factor, PDO value x ₂₂ in December in the second year in the past, PDO value x ₂₃ in February of the same year, and PDO value x ₂₄ in March of the same year In this way, we created a stock abundance prediction model for Japanese flying squid in autumn in the Sea of Japan.
CPUE = _1.1968 + _0.0273 * _x22 - _0.1865 *x21 - 0.2290*x23 + 0.0740*x24, and
F value is 5.9135, P = 0.0026 < 0.01,
A statistical table of actual and predicted values is shown in Table 9.

3) Prediction model 3
PDO value in February of the same year x ₂₃ , PDO value in March of the same year x ₂₄ , and PDO value in March of the past first year x ₂₅ of the climate factor Create a predictive model, specifically:
CPUE=1.3257-0.1120* _x23-0.0461 * _x24-0.1307 * _x25 ,
F value is 5.1699, P = 0.0078 < 0.01,
A statistical table of actual and predicted values is shown in Table 10.

4) Prediction model 4
For the PDO value of February of the same year x ₂₃ and the PDO value of March of the same year x ₂₄ of the climate factor, a Japanese flying squid autumn outbreak stock abundance prediction model was created. Specifically,
CPUE= _{1.3093-0.0792} * _x23-0.1223 *x24,
F value is 5.1233, P = 0.0149 < 0.05,
A statistical table of actual and predicted values is shown in Table 11.

5) Prediction model 5
For the climate factor PDO value x ₂₁ in November of the past two years and December PDO value x ₂₂ in the past two years, create a model for predicting the abundance of Japanese flying squid in autumn in the Japan Sea, In terms of
CPUE = 1.1582 + 0.0383* _x22 - _0.2125 *x21,
F value is 5.8894, P = 0.0089 < 0.01,
A statistical table of actual and predicted values is shown in Table 12.

As a result of comparative analysis of the above five models, the prediction model 2 was selected as the optimum model, and the optimum model is CPUE = 1.1968 + 0.0273 * x ₂₂ - 0.1865 * x ₂₁ - 0.2290 * x ₂₃ + 0 _.0740 *x24. Input x ₂₂ , x ₂₁ , x ₂₃ and x ₂₄ corresponding to the results from 1992 to 2016 into the optimal model to obtain the predicted values (taking the 2000 result as an example, x ₂₁ is the PDO value, _x22 is the PDO value for December 1998, _x23 is the PDO value for _February 2000, x24 is the PDO value for March 2000), change in stock abundance between actual and predicted values The trend is shown in FIG. 5, and as can be seen from FIG. 5, the change trend of the predicted value and the actual value is almost the same. reasonably predictable.

Example 3

An electronic device, as shown in FIG. 6, comprising one or more processors, one or more memories, one or more programs, and a data acquisition device;
A data acquisition device is used to obtain _{x 1} , x ₂ , x ₃ . . . x _z . When executed by the processor, causes the electronic device to perform the method of predicting Japanese flying squid abundance based on the Pacific Oscillation Index described in Example 1 or Example 2.

Although specific embodiments of the present invention have been described above, those skilled in the art will appreciate that these are merely exemplary descriptions and the scope of protection of the present invention is defined in the appended claims. Persons skilled in the art can make various changes and modifications to these embodiments without departing from the principle and spirit of the present invention, and all such changes and modifications fall within the protection scope of the present invention.

Claims (11)

A method for predicting Japanese flying squid abundance based on the Pacific Oscillation Index used to guide deep-sea squid fishing by a processor executing a program, a memory storing the program, and an electronic device equipped with a data collection device . There is
(1) obtaining monthly Pacific Oscillation Index PDO values in sea areas where Japanese flying squid have been distributed for the past N years by the data collection device, and the flying squid is a winter squid stock or a winter squid stock; ,
(2) Using the time-series analysis method, the processor performs a correlation analysis on the Japanese flying squid resource abundance CPUE and the monthly PDO values for the past N years, and of those months showing a statistically significant correlation Select the PDO values of these selected months, use the PDO values of these selected months as climate factors that affect the abundance of Japanese flying squid, and number the PDO values of these selected months in the order of 1, 2, 3 ... z ... m and let the _PDO values of these months be x ₁ , x ₂ , x ₃ . . . x _z .
(3) the processor, for any 1 to _m climate factors out of ^x ₁ , x ₂ , x ₃ . Create an abundance forecast model, calculate the statistical P value of each forecast model, and the formula for the Japanese flying squid resource abundance forecast model is
CPUE = a + b ₁ * x ₁ + b ₂ * x ₂ + b ₃ * x ₃ + ... + b _z * x _z + ... + b _m * x _m is _a constant and _{b 1} , b ₂ , _{b 3 ,} . . . _{b z .}
(4) selecting , by the processor, a model with the smallest statistical P value from among the largest number of 2 ^m −1 Japanese flying squid resource abundance prediction models as an optimal model; and An index-based method for predicting the abundance of Japanese flying squid. After the _optimum model is obtained by the processor, x ₁ , x ₂ , x ₃ . . . x _z . The method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index according to claim 1, wherein the prediction of the abundance of Japanese flying squid is completed. 2. The method according to claim 1, wherein b ₁ , b ₂ , b ₃ , . . . b _{z −1} , b _{z +1} , . A method for predicting Japanese common squid abundance based on the Pacific Oscillation Index. The statistically significant correlation is a calculated P-value <0.05,
The Japanese common squid fall stock is distributed in the Sea of Japan.
The method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index according to claim 1, wherein the sea area in which the winter squid stock is distributed is the Pacific Ocean side of Hokkaido. The Japanese flying squid is a Japanese flying squid autumn outbreak stock,
In step (2), select the PDO values of two months that show a statistically significant correlation, that is, the PDO value in October of the past two years and the PDO value of October in the past one year, The correlation between the stock abundance CPUE and the PDO value in October in the past two years was significant and showed a negative correlation, with a correlation coefficient of -0.390. The correlation between the degree CPUE and the PDO value in October of the past year is significant, shows a negative correlation, and the correlation coefficient is -0.4486. How to predict resource abundance. In step (3), create a total of three prediction models below,
1) Prediction model I
For the climate factor PDO value x ₁₁ in October in the past two years and PDO value in October in the first year x ₁₂ , create a model for predicting the abundance of common Japanese flying squid in autumn in the Sea of Japan, and for,
CPUE=2.3463−0.1674*x ₁₁ −0.1977*x ₁₂ and
F value is 4.9268, P = 0.0161 < 0.05,
2) Prediction model II
A model for predicting the abundance of Japanese flying squid in autumn in the Sea of Japan is created for the PDO value x ₁₁ in October in the past two years of the climate factor. Specifically,
CPUE= _{2.3894-0.2127} *x11,
F value is 4.4922, P = 0.0442 < 0.05,
3) Prediction model III
A model for predicting the abundance of Japanese flying squid in autumn in the Sea of Japan is created for the PDO value in October of the past first year of the climate factor x _12. Specifically,
CPUE=2.3958-0.2323* _x12 ,
The method for predicting Japanese common squid abundance based on the Pacific Oscillation Index according to claim 5, wherein the F value is 6.2984 and P=0.0189<0.05. In step (4), predictive model I is selected as the optimal model, and the optimal model is
The method for predicting Japanese common squid abundance based on the Pacific Oscillation Index according to claim 6, wherein CPUE=2.3463-0.1674* _x11-0.1977 * _x12 . The Japanese flying squid is a Japanese flying squid winter development stock,
In step (2), 10 monthly PDO values exhibiting a statistically significant correlation: Select the PDO value for April, the PDO values for 1, 2, 3 and April of the same year,
The correlation between the winter squid abundance CPUE and the PDO values in October, November and December in the past two years was significant and showed a negative correlation, with correlation coefficients of -0.4506 and -0.4506, respectively. 0.4985 and -0.5878;
The correlation between the winter squid abundance CPUE and the PDO values in January, February, March and April in the past year was significant and negative, with a correlation coefficient of -0.4665. , −0.4365, −0.4295 and −0.5072,
The correlation between the winter squid abundance CPUE and the PDO values in January, February, March and April of the same year was significant and showed a negative correlation, with correlation coefficients of -0.4746 and -0, respectively. 4837, -0.5458 and -0.5570. In step (3), create a total of 5 prediction models below,
1) Prediction model 1
PDO value x ₂₁ in November of the past second year of the climate factor, PDO value x ₂₂ in December in the second year in the past, PDO value x ₂₃ in February of the same year, PDO value x ₂₄ in March of the same year and the past For the PDO value x ₂₅ in March of the first year, create a model for predicting the abundance of Japanese flying squid autumn outbreak stocks in the Sea of Japan. Specifically,
CPUE=1.2048+0.0330*x ₂₂ −0.1811*x ₂₁ −0.2260*x ₂₃ +0.0749*x ₂₄ −0.0196*x ₂₅ ,
F value is 4.5183, P = 0.0069 < 0.01,
2) Prediction model 2
PDO value x ₂₁ in November of the past second year of the climate factor, PDO value x ₂₂ in December in the second year in the past, PDO value x ₂₃ in February of the same year, and PDO value x ₂₄ in March of the same year In this way, we created a stock abundance prediction model for Japanese flying squid in autumn in the Sea of Japan.
CPUE = _1.1968 + _0.0273 * _x22 - _0.1865 *x21 - 0.2290*x23 + 0.0740*x24, and
F value is 5.9135, P = 0.0026 < 0.01,
3) Prediction model 3
PDO value in February of the same year x ₂₃ , PDO value in March of the same year x ₂₄ , and PDO value in March of the past first year x ₂₅ of the climate factor Create a predictive model, specifically:
CPUE=1.3257-0.1120* _x23-0.0461 * _x24-0.1307 * _x25 ,
F value is 5.1699, P = 0.0078 < 0.01,
4) Prediction model 4
For the PDO value of February of the same year x ₂₃ and the PDO value of March of the same year x ₂₄ of the climate factor, a Japanese flying squid autumn outbreak stock abundance prediction model was created. Specifically,
CPUE= _{1.3093-0.0792} * _x23-0.1223 *x24,
F value is 5.1233, P = 0.0149 < 0.05,
5) Prediction model 5
For the climate factor PDO value x ₂₁ in November of the past two years and December PDO value x ₂₂ in the past two years, create a model for predicting the abundance of Japanese flying squid in autumn in the Japan Sea, In terms of
CPUE = 1.1582 + 0.0383* _x22 - _0.2125 *x21,
The method for predicting Japanese flying squid abundance based on the Pacific Oscillation Index according to claim 8, wherein the F value is 5.8894 and P=0.0089<0.01. In step (4), predictive model 2 is selected as the optimal model, and the optimal model is
CPUE = 1.1968 + 0.0273 * x ₂₂ - 0.1865 * x ₂₁ - 0.2290 * x ₂₃ + 0.0740 * x ₂₄ The method for predicting Japanese flying squid abundance based on the Pacific Oscillation Index according to claim 9 . An electronic device comprising one or more processors, one or more memories, one or more programs, and a data acquisition device,
The data acquisition device is used to obtain _{x 1} , x ₂ , x ₃ . . . x _z . An electronic device characterized in that, when a plurality of programs are executed by the processor, the electronic device executes the method for predicting the abundance of Japanese flying squid based on the Pacific Oscillation Index according to any one of claims 1 to 10.

Images