Northwest Pacific Ocean squid resource abundance prediction method based on time sequence model

By constructing a seasonal multi-covariate autoregressive moving average model, combining the marine environmental factors of soft fish spawning ground and bait field, the problem of failing to effectively predict soft fish resource abundance in the existing technology is solved, and more accurate and reliable resource abundance forecasts to serve fishery production guidance.

CN119006199AActive Publication Date: 2024-11-22SHANGHAI OCEAN UNIV +2
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410893240.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-04
Publication Date
2024-11-22
Estimated Expiration
2044-07-04

AI Technical Summary

Technical Problem

The existing technology failed to effectively consider the marine environmental factors in the early life history of soft fish when predicting the abundance of soft fish resources in the Northwest Pacific, resulting in a short prediction time and great uncertainty, and poor guidance on fishery production.

Method used

Using a time series model based method, combining soft fish habitat and climate multicovariates, a seasonal multicovariates autoregressive moving average model was constructed, and the marine environmental factors of the spawning ground and the bait field were considered, and the unit fishing effort was used to measure the fishing force. Obtaining quantities is used as a benchmark indicator to establish a more accurate resource abundance prediction model.

Benefits of technology

By considering the marine environmental factors in the life history of soft fish, the accuracy and time length of resource abundance prediction are improved, the range of uncertainty is narrowed, and the guidance effect of fishery production is significantly improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119006199A_ABST
    Figure CN119006199A_ABST
Patent Text Reader

Abstract

The invention relates to a northwest Pacific Ocean squid resource abundance prediction method based on a time sequence model, and belongs to the technical field of marine fishing condition analysis and processing, and the method comprises the following steps: dividing the fishing data of northwest Pacific Ocean squid over the years into a training set and a test set, and representing the resource abundance by using the fishing yield per unit fishing effort; calculating the fishing catch of each fishing point and monthly average unit fishing effort; performing unit root test on the training set data to meet the stability requirement; establishing a seasonal single autoregressive moving average model by using the training set data; establishing a seasonal multi-covariant autoregressive moving average model by combining data of a squid spawning site and a bait rearing site; using the test set to carry out residual white noise test and fitting degree and prediction precision screening evaluation on the model to obtain an optimal model; and performing northwest Pacific Ocean squid resource abundance prediction by using the optimal model. According to the method, a more accurate resource abundance prediction model is established, and more accurate northwest Pacific Ocean squid resource abundance prediction can be realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of marine fishery situation analysis and processing, and in particular relates to a method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model. Background Art

[0002] Ommastrephes bartramii is an important economic cephalopod widely distributed in the North Pacific Ocean. The confluence of the Kuroshio warm current and the Oyashio cold current in the northwest Pacific Ocean has formed a complex marine environment. Driven by factors such as temperature and nutrients, the scallops migrate between spawning grounds in subtropical waters and feeding grounds in the subarctic waters. my country has been commercially fishing for scallops in the northwest Pacific Ocean since 1994. According to the distribution area and peak spawning period, the scallops are divided into winter-spring groups in the east and autumn groups in the west, with 170°E as the boundary. Currently, the winter-spring groups are the main groups to be fished. Due to the short capture period, there is a high demand for the prediction of the abundance of scallops resources in the northwest Pacific Ocean.

[0003] At present, scholars at home and abroad have explored the impact of marine environmental factors such as sea surface temperature (SST) and seawater chlorophyll concentration (Chloroplyll, Chl_a) on the abundance of squid resources. The relevant results reflect the mechanism of squid resource changes to a certain extent. However, the selection of these environmental factors is mostly based on the temporal and spatial distribution of the fishing location, that is, the feeding ground, without considering the marine environment of the early life history stages of squid eggs, larvae, and juveniles, that is, the spawning ground, and the delayed response of squid resources to marine environmental factors. As a species with a short life cycle, squid is more sensitive to environmental changes. The marine environment of its spawning ground in its early life history process is a key factor in determining the replenishment of squid resources. However, the current prediction method does not consider the relevant covariates of biological characteristics, the effective prediction time is short, the confidence interval is large, and the guidance effect on the actual squid fishery production is not obvious. Therefore, when predicting the abundance of squid resources in the northwest Pacific, it is necessary to simultaneously consider the change mechanism of the marine environmental factors of the feeding ground and spawning ground on the abundance of resources. Summary of the invention

[0004] The purpose of the present invention is to provide a method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model in response to the problems existing in the prior art. A time series model based on squid habitat and climate multi-covariates is constructed on the basis of the original single time series, and the marine environmental factors experienced by the squid during its life history are considered. The catch per unit effort (CPUE) that reflects the fishing situation is used as a benchmark indicator to increase the effective prediction time of squid resource abundance and reduce the uncertainty range.

[0005] The technical solution is as follows:

[0006] The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on the time series model includes the following steps:

[0007] Step 1: Divide the historical catch data of Northwest Pacific squid into training set and test set, use the catch per unit fishing effort (CPUE) to represent the abundance of squid resources in the Northwest Pacific, calculate the catch per unit fishing effort of each fishing point and the monthly average catch per unit fishing effort;

[0008] Step 2: Perform a unit root test on the training set data to meet the stationarity requirements;

[0009] Step 3: Establish a seasonal single autoregressive moving average model (SARIMA) based on the training set data;

[0010] Step 4: Based on the seasonal single autoregressive moving average model, the sea surface temperature (SST) of the squid spawning ground (S) is combined S , sea surface salinity SSS S , sea surface chlorophyll concentration Chl_a S , sea surface height SSH S , Mixed layer depth MLD S , Sea surface oxygen content O2 S and SST at the feeding ground (F) F 、SSS F , Chl_a F SSH F 、MLD F 、O2 F Values, establish seasonal multi-covariate autoregressive moving average model (SARIMAX);

[0011] Step 5: Use the test set to perform white noise test on the residuals, goodness of fit and prediction accuracy screening and evaluation on the seasonal multi-covariate autoregressive moving average model to obtain the optimal model;

[0012] Step 6: Use the optimal model to predict the abundance of squid resources in the northwest Pacific.

[0013] Furthermore, in step 1, the test set is the catch and the number of fishing vessels at 144.5°E-170°E, 35°N-49.5°N from June to November within two years of the time point when the method was used; the training set is the catch and the number of fishing vessels at 144.5°E-170°E, 35°N-49.5°N from June to November since records began to two years before the time point when the method was used.

[0014] Furthermore, the calculation method of the catch per unit fishing effort in step 1 is to first calculate the CPUE of each fishing point using the following formula:ymij value: CPUE ymij , C ymij , E ymij The unit fishing effort (tons / vessel, t / v), production, and number of operating fishing vessels for year y, month m, longitude i, and latitude j are respectively; the monthly average unit fishing effort catch is calculated using the following formula: CPUE ym is the average monthly catch per unit fishing effort in year y and month m, CPUE k(ymij) For CPUE ymij The formula is used to calculate the average catch per unit fishing effort at the kth fishing location, where M is the total amount of data in m months.

[0015] Furthermore, the unit root test in step 2 is constructed based on the relationship between the stationary sequence and the unit root to test the unit root of the time series data. If all the characteristic roots of the sequence are within the unit circle, the sequence is a stationary sequence. If the sequence has a characteristic root on or outside the unit circle, then the sequence is a non-stationary sequence.

[0016] Furthermore, the unit root test is performed using the adf.test function in the tseries package in the R language for time series analysis. If the p-value is less than 0.05, it means that the series is stationary. If the p-value is greater than 0.05, the series is not stationary. It is necessary to perform d-order differences and D-time seasonal differences in succession to stabilize the series, where d / D≤2. The order of d-order differences and D-time seasonal differences does not affect the results. When the seasonality of the series is strong, seasonal differences are used first.

[0017] Furthermore, the seasonal single autoregressive moving average model in step 3

[0018] The expression of SARIMA(p, d, q)(P, D, Q)[S] is:

[0019] Among them, p, d, q correspond to the three processes of autoregression, difference, and moving average, respectively. P, D, Q, S are seasonal autoregression terms, difference terms, moving average terms, and frequency. Y t is the observed value of the time series at time t, δ is a constant term, and B is a delay operator; (1-B) d Indicates d-order difference; (1-B s ) D represents seasonal differences with a seasonal period of S for D times; the polynomial Φ (P) (B s )=1-Φ1B s -…-Φ p B s, the polynomial acts on Y t The left side of the equation is represents the AR(p / P) component of the model, and the right side of the equation is represents the MA(q / Q) component of the model.

[0020] Furthermore, the expression of the seasonal multi-covariate autoregressive moving average model SARIMAX in step 4 is: Y t ′=Y t +βX i , where Y t ′ represents the fitted value of the seasonal multi-covariate autoregressive moving average model at time t; Y t is the fitting value of the seasonal single autoregressive moving average model at time t; X i represents the environmental covariate of the spawning ground or feeding ground; β is the coefficient vector of the covariate, describing the influence of the covariate on the dependent variable.

[0021] Furthermore, in step 5, the checkresiduals function in the Forecast package in the R language, a time series forecasting tool set, is used to perform a white noise test on the residuals of the seasonal multi-covariate autoregressive moving average model. If the p-value is greater than 0.05, the model passes the white noise test; if the p-value is less than 0.05, the model fails the test and needs to be remodeled.

[0022] Furthermore, in step 5, the Akaike information criterion (AICc) is used to evaluate the goodness of fit of the seasonal multi-covariate autoregressive moving average model, and the calculation formula is: Where L represents the likelihood function value of the model, indicating the degree of fit of the model; k is the number of parameters in the model, n is the number of samples, and the smaller the AICc value, the better the fit of the model; the root mean square error RMSE indicator is used for the prediction accuracy screening evaluation, and the calculation formula is: where Y i Indicates the actual value, Represents the predicted value, n represents the number of test set samples, and RMSE measures the average error between the model predicted value and the true value. The smaller the value, the better the model prediction performance.

[0023] Beneficial effects:

[0024] 1) The present invention constructs a time series model based on squid habitat and climate multi-covariates in combination with the marine environmental covariates on the basis of the original single time series, which can effectively capture the changes in squid resources, establish a more accurate resource abundance prediction model, and achieve a more accurate prediction of squid resource abundance in the northwest Pacific. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the process of the present invention.

[0026] Figure 2 Table 1: The optimal SARIMA model under different difference types.

[0027] Figure 3 Table 2: SARIMAX model constructed based on traversal marine environmental factors under the basic model.

[0028] Figure 4 Table 3: Prediction of CPUE of squid from June to November 2021 based on the selected optimal SARIMAX model.

[0029] Figure 5 It is a line graph of the prediction of squid resource abundance in the northwest Pacific Ocean using the optimal model in the embodiment. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. The directions or positional relationships indicated by the terms "upper", "lower", "front", "back", "left", "right", "bottom", "inside", "outside", etc. are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention:

[0031] like Figure 1 The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on the time series model includes the following steps:

[0032] Step 1: Divide the historical catch data of squid in the Northwest Pacific into a training set and a test set, use the catch per unit fishing effort to characterize the abundance of squid resources in the Northwest Pacific, and calculate the catch per unit fishing effort and the monthly average catch per unit fishing effort of each fishing point; the test set is the catch and number of fishing vessels at 144.5°E-170°E, 35°N-49.5°N from June to November within two years of using this method; the training set is the catch and number of fishing vessels at 144.5°E-170°E, 35°N-49.5°N from June to November since the record to two years before the use of this method; the method for calculating the catch per unit fishing effort is to first calculate the CPUE of each fishing point using the following formula ymij value: CPUE ymij , C ymij , E ymijThe unit fishing effort, catch, and number of fishing vessels in year y, month m, longitude i, and latitude j are respectively; the monthly average unit fishing effort catch is calculated using the following formula: CPUE ym is the average monthly catch per unit fishing effort in year y and month m, CPUE k(ymij) (for CPUE ymij The average catch per unit fishing effort at the kth fishing location calculated in the formula, M is the total data volume in m months;

[0033] Step 2: Perform a unit root test on the training set data. The unit root test is constructed based on the relationship between the stationary sequence and the unit root to test the unit root of the time series data. If all the characteristic roots of the sequence are within the unit circle, the sequence is a stationary sequence. If the sequence has a characteristic root on or outside the unit circle, then the sequence is a non-stationary sequence. The unit root test is performed using the adf.test function in the tseries package in the R language for time series analysis. If the p value is less than 0.05, it means that the sequence is stationary. If the p value is greater than 0.05, the sequence is non-stationary. It is necessary to perform d-order differences and D-time seasonal differences in sequence to make the sequence stable, where d / D≤2. The order of d-order differences and D-time seasonal differences does not affect the results. When the seasonality of the sequence is strong, seasonal differences are used first;

[0034] Step 3: Use the training set to build a seasonal single autoregressive moving average model

[0035] The expression of SARIMA(p, d, q)(P, D, Q)[S] is:

[0036] Among them, p, d, q correspond to the three processes of autoregression, difference, and moving average, respectively. P, D, Q, S are seasonal autoregression terms, difference terms, moving average terms, and frequency. Y t is the observed value of the time series at time t, δ is a constant term, and B is a delay operator; (1-B) d Indicates d-order difference; (1-B s ) D represents seasonal differences with a seasonal period of S for D times; the polynomial Φ (P) (B s )=1-Φ1B s -…-Φ p B s , the polynomial acts on Y t The left side of the equation is represents the AR(p / P) component of the model, and the right side of the equation is represents the MA(q / Q) component of the model;

[0037] Step 4: Based on the seasonal single autoregressive moving average model, the SST of the squid spawning grounds is combined S , sea surface salinity SSS S , sea surface chlorophyll concentration Chl_a S , sea surface height SSH S , Mixed layer depth MLD S , Sea surface oxygen content O2 S and SST at the feeding grounds F 、SSS F , Chl_a F SSH F 、MLD F 、O2 F The expression of establishing seasonal multi-covariate autoregressive moving average model SARIMAX is: t ′=Y t +βX i , where Y t ′ represents the fitting value of the seasonal multi-covariate autoregressive moving average model at time t; Yt is the fitting value of the seasonal single autoregressive moving average model at time t; X i represents the environmental covariate of the spawning ground or feeding ground; β is the coefficient vector of the covariate, describing the influence of the covariate on the dependent variable;

[0038] Step 5: Use the test set to perform white noise test on the residuals of the seasonal multi-covariate autoregressive moving average model, and screen and evaluate the goodness of fit and prediction accuracy to obtain the optimal model; use the checkresiduals function in the Forecast package in the R language for time series prediction tools to perform a white noise test on the residuals of the seasonal multi-covariate autoregressive moving average model. If the p value is greater than 0.05, the model passes the white noise test; if the p value is less than 0.05, the model fails the test and needs to be remodeled; the goodness of fit evaluation uses the Akaike information criterion, and the calculation formula is: Where L represents the likelihood function value of the model, indicating the degree of fit of the model; k is the number of parameters in the model, n is the number of samples, and the smaller the AICc value, the better the fit of the model; the root mean square error RMSE indicator is used for the prediction accuracy screening evaluation, and the calculation formula is: where Y i Indicates the actual value, Represents the predicted value, n represents the number of test set samples, and RMSE measures the average error between the model predicted value and the true value. The smaller the value, the better the model prediction performance;

[0039] Step 6: Use the optimal model to predict the abundance of squid resources in the northwest Pacific.

[0040] Example 1: When predicting the abundance of squid in the northwest Pacific in 2021, the original data set from June to November 1995 to 2020 was first divided into a training set from 1995 to 2018 and a test set from 2019 to 2020. CPUE (tons / vessel) was used to characterize the abundance of squid resources, and the CPUE of each fishing point was calculated. Then calculate the monthly average value within the fishing area

[0041] The fishery data of the fishing location with a spatial resolution of 0.5°×0.5° were matched with the corresponding environmental data, and the marine environmental factors of the spawning grounds 130°-170°E, 20°-33°N, from December of year y-1 to January-May of year y were extracted at a resolution of 0.5°×0.5°. The adf.test function in the forecast package in R language was used to test the unit root of the training set. The result showed that the P value was 0.299, indicating that the CPUE sequence accepted the null hypothesis and the sequence was not stable. Three difference operations were tried: d=1, D=0; d=0, D=1; d=1, D=1, and three types of difference models were obtained. The optimal model with the minimum AICc of the training set and the minimum RMSE of the test set prediction was as follows: Figure 2 Table 1: Optimal SARIMA models under different difference types shows that the minimum RMSE model does not pass the residual white noise test, indicating that the model cannot extract all the effective information in the sequence. Therefore, this model is excluded and SARIMA (0, 1, 2) (0, 0, 2) [6] is selected as the basic model. Substituting into the formula Get Y t =-0.548∈ t-1 -0.405∈ t-2 +0.254∈ t-6 +0.179∈ t-1 2+ε t , ε t ~N(0,0.819).

[0042] Combined with the sea surface temperature SST of the squid spawning ground (S) S , sea surface salinity SSS S , sea surface chlorophyll concentration Chl_a S , sea surface height SSH S , Mixed layer depth MLD S , Sea surface oxygen content O2 S and SST at the feeding ground (F) F 、SSS F , Chl_a F SSH F 、MLD F 、O2 FThe length of all covariate data is consistent with the original CPUE data length. The multivariate SARIMAX models constructed are compared on the test set by traversing the marine environmental variables. The results are as follows: Figure 3 Table 2: SARIMAX model constructed based on traversal marine environmental factors under the basic model. Combining the model fitting and prediction performance, the optimal model selected is the SARIMA model combining the two covariates of spawning ground sea surface salinity and spawning ground sea surface temperature as the optimal model, and its expression is

[0043] Y t =-0.528∈ t-1 -0.325∈ t-2 +0.139∈ t-6 +0.175∈ t-12 +7.006SSS S -0.031SST S +ε t , ε t ~N(0,0.746).

[0044] Based on the optimal model, the monthly average values ​​of sea surface salinity and temperature in the spawning grounds from 1995 to 2020 were used as the monthly average values ​​of environmental covariates in 2021. The abundance of squid resources in 2021 was predicted based on this average value. The specific CPUE data and their confidence intervals for June to November 2021 are as follows: Figure 4 Table 3 shows the CPUE and confidence interval of squid from June to November 2021 predicted based on the selected optimal SARIMAX model. Figure 5 As shown in the figure: The prediction line chart of the abundance of squid resources in the northwest Pacific based on the optimal model. As can be seen from Table 2, compared with the basic model, the AICc and RMSE values ​​of most SARIMAX models combined with covariates are reduced, and the fitting effect and prediction effect are improved. However, the addition of most marine environmental factor covariates in the feeding ground only improves the model fitting effect, but the prediction accuracy on the test set is reduced. The addition of marine environmental factors in the spawning ground, especially the sea surface salinity factor, can effectively improve the fitting and prediction accuracy of the model. In general, the addition of spawning ground environmental covariates can improve the prediction effect of the model more than the feeding ground. The optimal model selected by traversing all marine environmental factors is the combination of spawning ground sea surface salinity and sea surface temperature covariates, indicating that the spawning ground environmental factors can better reflect the changes in squid resources. It also confirms that the environmental factors in the early life history stage of squid are the key factors determining its resource abundance. Accurately identifying these marine environmental factors can effectively improve the effect of squid resource abundance prediction and better serve the work of fishery forecasting.

[0045] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the principles and spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model, characterized by: The following steps are involved: Step 1: Divide the historical catch data of Northwest Pacific squid into training set and test set, use the catch per unit fishing effort to represent the abundance of squid resources in the Northwest Pacific, calculate the catch per unit fishing effort of each fishing point and the monthly average catch per unit fishing effort; Step 2: Perform a unit root test on the training set data to meet the stationarity requirements; Step 3: Establish a seasonal single autoregressive moving average model using the training set data; Step 4: Based on the seasonal single autoregressive moving average model, the SST of the squid spawning grounds is combined s , sea surface salinity SSS s , sea surface chlorophyll concentration Chl_a s , sea surface height SSH S , Mixed layer depth MLD S , Sea surface oxygen content O2 S and SST at the feeding grounds F 、SSS F , Chl_a F SSH F 、MLD F 、O2 F Values, establish seasonal multi-covariate autoregressive moving average model; Step 5: Use the test set to perform white noise test on the residuals, goodness of fit and prediction accuracy screening and evaluation on the seasonal multi-covariate autoregressive moving average model to obtain the optimal model; Step 6: Use the optimal model to predict the abundance of squid resources in the northwest Pacific.

2. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: In the step 1, the test set is the catch and the number of fishing vessels at 144.5°E-170°E, 35°N-49.5°N from June to November within two years from the time when the method is used; the training set is the catch and the number of fishing vessels at 144.5°E-170°E, 35°N-49.5°N from June to November since records began to two years before the method was used.

3. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: The calculation method of the catch per unit fishing effort in step 1 is to first calculate the CPUE of each fishing point using the following formula: ymij value: CPUE ymij , C ymij , F ymij The unit fishing effort, catch, and number of fishing vessels in year y, month m, longitude i, and latitude j are respectively; the monthly average unit fishing effort catch is calculated using the following formula: CPUE ym is the average monthly catch per unit fishing effort in year y and month m, CPUE k(ymij) For CPUE ymij The formula is used to calculate the average catch per unit fishing effort at the kth fishing location, where M is the total amount of data in m months.

4. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: The unit root test in step 2 is constructed based on the relationship between the stationary sequence and the unit root to test the unit root of the time series data. If all the characteristic roots of the sequence are within the unit circle, the sequence is a stationary sequence. If the sequence has a characteristic root on or outside the unit circle, then the sequence is a non-stationary sequence.

5. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 4, characterized in that: The unit root test is performed using the adf.test function in the tseries package for time series analysis in the R language. If the p-value is less than 0.05, it means that the sequence is stable. If the p-value is greater than 0.05, the sequence is not stable. It is necessary to perform d-order differences and D-time seasonal differences in sequence to stabilize the sequence, where d / D≤2. The order of d-order differences and D-time seasonal differences does not affect the results. When the seasonality of the sequence is strong, seasonal differences are used first.

6. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: The seasonal single autoregressive moving average model SARIMA (p, d, q) (P, D, Q) [S] in step 3 is expressed as: Among them, p, d, q correspond to the three processes of autoregression, difference, and moving average, respectively. P, D, Q, S are seasonal autoregression terms, difference terms, moving average terms, and frequency. Y t is the observed value of the time series at time t, δ is a constant term, and B is a delay operator; (1-B) d Indicates d-order difference; (1-B s ) D represents seasonal differences with a seasonal period of S for D times; the polynomial Φ (P) (B s )=1-Φ1B s -…-Φ p B s , the polynomial acts on Y t The left side of the equation is represents the AR(p / P) component of the model, and the right side of the equation is δ+θ (1) ∈ t-1 +…+θ (q) ∈ t-q +Θ (1) ∈ t-s +…+Θ (q) ∈ t-QS +∈ t represents the MA(q / Q) component of the model.

7. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: The expression of the seasonal multi-covariate autoregressive moving average model SARIMAX in step 4 is: t =Y t +βX i , where Y′ t represents the fitted value of the seasonal multi-covariate autoregressive moving average model at time t; Y t is the fitting value of the seasonal single autoregressive moving average model at time t; X i represents the environmental covariate of the spawning ground or feeding ground; β is the coefficient vector of the covariate, describing the influence of the covariate on the dependent variable.

8. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: In step 5, the checkresiduals function in the Forecast package in the R language, a time series forecasting tool set, is used to perform a white noise test on the residuals of the seasonal multi-covariate autoregressive moving average model. If the p-value is greater than 0.05, the model passes the white noise test; if the p-value is less than 0.05, the model fails the test and needs to be remodeled.

9. The method for predicting the abundance of squid resources in the northwest Pacific Ocean based on a time series model as claimed in claim 1, characterized in that: In step 5, the seasonal multi-covariate autoregressive moving average model fit evaluation adopts the Akaike information criterion, and the calculation formula is: Where L represents the likelihood function value of the model, indicating the degree of fit of the model; k is the number of parameters in the model, n is the number of samples, and the smaller the AICc value, the better the fit of the model; the root mean square error RMSE indicator is used for the prediction accuracy screening evaluation, and the calculation formula is: where Y i Indicates the actual value, Represents the predicted value, n represents the number of test set samples, and the smaller the RMSE value, the better the model prediction performance.

Citation Information

Patent Citations

  • Method for predicting ocean fishery resource abundance by integrating multi-source satellite remote sensing and application thereof

    CN111652443A

  • Southwest Atlantic Argentinus sylvestris habitat evaluation and prediction technical method based on water temperature vertical structure

    CN111680844A