Fish roe drifting forecasting method

Abstract

The invention discloses a method for forecasting roe drifting. The method comprises the following steps: S1, acquiring a river section influence factor sequence and roe quantity in a research area in historical data; s2, according to the research area river segment influence factor sequence and the roe amount in the historical data obtained in the step S1, a prediction model of research area river segment multi-scale factor fusion is constructed; s3, performing significant inspection and correction on the prediction model of the river section multi-scale factor fusion of the research area constructed in the step S2 to obtain a corrected prediction model of the river section multi-scale factor fusion of the research area; s4, obtaining the influence factors of the river section of the daily research area, and forecasting the daily roe drifting condition of the fish by using the corrected forecasting model of the river section multi-scale factor fusion of the research area obtained in the step S3; according to the method, the fish roe drifting is predicted by constructing the multi-scale factor fusion prediction model, the prediction accuracy is improved, and meanwhile, the channel engineering construction of the research area is enabled to avoid the peak period of fish oviposition.

Description

Technical Field

[0001] The present invention relates to a method for predicting spawning, and more particularly to a method for predicting the drift of fish eggs. Background Art

[0002] The Yangtze River is the world's inland navigable river with the largest freight volume. The gross national product of the Yangtze River Basin accounts for about 36% of the country's total. The waterways within the jurisdiction of Chongqing undertake 70% of the freight volume from the upper reaches to the middle and lower reaches of the Yangtze River. The golden waterway in the upper reaches of the Yangtze River is a key support for the twin-city economic circle in the Chengdu-Chongqing region and the strategy of a powerful transportation country. There is an urgent need to build a high-grade waterway network of "one main, two branches, and six lines", with great development needs.

[0003] For the waterway projects of the Chaotianmen-Fuling section and the Fuling-Fengdu section located in the Three Gorges Reservoir area, their construction time is greatly restricted by the drift of fish eggs. Specifically, during the "fishing moratorium" from March to June, although the construction conditions are good, there are strict regulations at the legal level; during the flood season from July to August, it is not easy to construct due to poor water flow conditions; during the water storage period from September to March, the construction water level exceeds 30m, which is also not conducive to construction. Therefore, the waterway projects in the Three Gorges Reservoir area and other water-related projects face the situation of prohibiting construction during the fishing moratorium, which delays the construction progress and increases the construction cost. Summary of the Invention

[0004] In view of the above deficiencies in the prior art, the present invention provides a method for predicting the drift of fish eggs. By predicting the peak period of fish egg drift and making full use of the "low water level" during the fishing moratorium, the construction of the waterway project in the research area can avoid the peak period of fish spawning, so as to shorten the construction process of the waterway project in the research area.

[0005] In order to achieve the above invention purpose, the technical solution adopted by the present invention is as follows:

[0006] A method for predicting the drift of fish eggs, comprising the following steps:

[0007] S1. Obtain the sequence of influencing factors and the amount of fish eggs in the river section of the research area from historical data;

[0008] S2. According to the sequence of influencing factors and the amount of fish eggs in the river section of the research area obtained in step S1, construct a prediction model for multi-scale factor fusion of the river section of the research area;

[0009] S3. Conduct a significant test correction on the prediction model for multi-scale factor fusion of the river section of the research area constructed in step S2 to obtain a corrected prediction model for multi-scale factor fusion of the river section of the research area;

[0010] S4. Obtain the influencing factors of the river section of the research area every day, and use the corrected prediction model for multi-scale factor fusion of the river section of the research area obtained in step S3 to predict the daily fish egg drift situation of fish.

[0011] Furthermore, step S1 specifically includes:

[0012] S11. Obtain the daily fish egg quantity of the river section in the research area monitored in the historical data to obtain the fish egg quantity of the river section in the research area in the historical data;

[0013] S12. According to the fish egg quantity of the river section in the research area in the historical data obtained in step S11 and the corresponding daily water flow, water level, water temperature, and water flow velocity data of the river section in the research area monitored in the historical data, obtain the influencing factor sequence of the river section in the research area in the historical data.

[0014] Furthermore, step S2 specifically includes:

[0015] S21. According to the fish egg quantity of the river section in the research area in the historical data obtained in step S1, perform binary conversion and classification on whether fish spawn in the historical data. If the fish egg quantity is greater than 0, it means that fish spawning occurs and is marked as 1; otherwise, it means that fish spawning does not occur and is marked as 0, to obtain the fish spawning result of binary classification processing;

[0016] S22. Based on the fish spawning result of binary classification processing obtained in step S21 and the influencing factor sequence of the river section in the research area in the historical data obtained in step S1, establish a generalized linear equation based on the theory of multivariate non-linear functions, and construct a prediction model for multi-scale factor fusion of the river section in the research area.

[0017] Furthermore, the prediction model for multi-scale factor fusion of the river section in the research area constructed in step S22 is:

[0018]

[0019] Among them, represents the predicted daily fish spawning situation. When Y = 1, it means that fish spawning occurs, and when Y = 0, it means that fish spawning does not occur. μ Y represents the probability of fish having fish eggs when given the influencing factor value, and 1 - μ Y represents the situation where fish do not have fish eggs when given the influencing factor value. α0 represents the fish spawning situation when all influencing factors are 0, p represents the number of days, β j represents the independent variable coefficient on the jth day, χ j represents the water flow of the river section in the research area on the jth day, γ j represents the independent variable coefficient on the jth day, ψ j represents the average flow velocity of the river section in the research area on the jth day, δ j represents the independent variable coefficient on the jth day, φ j represents the water level of the river section in the research area on the jth day, ε jRepresents the independent variable coefficient on the j-th day, ω j Represents the water temperature on the j-th day in the river section of the study area.

[0020] Furthermore, step S3 specifically includes:

[0021] S31. Assume that the sequence of influencing factors in the river section of the study area has no significant impact on fish spawning, that is:

[0022]

[0023] Among them, Represents the fish spawning probability after considering the sequence of influencing factors on the j-th day in the river section of the study area, Represents the fish spawning probability when not considering the sequence of influencing factors on the j-th day in the river section of the study area;

[0024] S32. Assume that the sequence of influencing factors in the river section of the study area has a significant impact on fish spawning, that is:

[0025]

[0026] S33. Calculate the expected value of the sequence of influencing factors in the river section of the study area on fish spawning, that is:

[0027]

[0028] Among them, Represents the expected value of the sequence of influencing factors on the j-th day in the river section of the study area for fish spawning, and n represents the total number of historical data;

[0029] S34. According to the expected value of the sequence of influencing factors in the river section of the study area on fish spawning calculated in step S33, calculate the standard error of the expected value of the sequence of influencing factors in the river section of the study area on fish spawning, that is:

[0030]

[0031] Among them, Represents the standard error of the expected value of the sequence of influencing factors on the j-th day in the river section of the study area for fish spawning;

[0032] S35. According to the standard error of the expected value of the sequence of influencing factors in the river section of the study area on fish spawning calculated in step S34, calculate the test statistic of the sequence of influencing factors in the river section of the study area on fish spawning;

[0033] S36. Through the standard normal distribution table, find the left-tail probability number and right-tail probability number corresponding to the test statistic of the sequence of influencing factors in the river section of the study area on fish spawning calculated in step S35, and add the left-tail probability and right-tail probability to obtain the p-value of the two-sided test;

[0034] S37. If the p - value of the two - sided test obtained in step S36 is less than 0.05, the hypothesis in step S31 does not hold, and the hypothesis in step S32 holds, that is, the influencing factor sequence of the river section in the study area has a significant impact on fish spawning;

[0035] S38. If the p - value of the two - sided test obtained in step S36 is greater than or equal to 0.05, the hypothesis in step S31 holds, and the hypothesis in step S32 does not hold, that is, the influencing factor sequence of the river section in the study area has no significant impact on fish spawning;

[0036] S39. Remove the influencing factors that have no significant impact on fish spawning in the influencing factor sequence of the river section in the study area in step S38, and re - input the influencing factors that have a significant impact on fish spawning in the influencing factor sequence of the river section in the study area into the prediction model of multi - scale factor fusion of the river section in the study area constructed in step S2 for calculation, and obtain the corrected prediction model of multi - scale factor fusion of the river section in the study area.

[0037] Further, the calculation formula for the test statistic of the influencing factor sequence of the river section in the study area on fish spawning in step S35 is:

[0038]

[0039] Among them, represents the test statistic of the influencing factor sequence of the j - th day of the river section in the study area on fish spawning, represents the total number of times of fish eggs collected on the j - th day in historical data.

[0040] Further, the calculation formula for the p - value of the two - sided test in step S36 is:

[0041] p two - sided = 2 * (left - tail probability number, right - tail probability number)

[0042] Further, step S4 specifically includes:

[0043] S41. Obtain the influencing factors such as the daily water flow, water level, water temperature, and water flow velocity of the river section in the study area, and import them into the corrected prediction model of multi - scale factor fusion of the river section in the study area constructed in step S3 to calculate the fish spawning probability;

[0044] S42. If the fish spawning probability calculated in step S41 is less than 0.5, there is no fish spawning phenomenon, otherwise, there is a fish spawning phenomenon.

[0045] The present invention has the following beneficial effects:

[0046] A prediction method for fish egg drifting proposed by the present invention fuses the influencing factors of the river section in the research area with the fish egg quantity, and quantitatively and qualitatively predicts the fish egg drifting situation by constructing a prediction model with multi-scale factor fusion, improving the prediction accuracy of fish egg drifting. At the same time, it can accurately predict the peak period of fish egg drifting, and enable the construction of the waterway project in the research area to avoid the peak period of fish spawning, thereby shortening the construction process of the waterway project in the research area and improving the construction efficiency, so as to save construction costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 FIG. is a schematic flow chart of a prediction method for fish egg drifting;

[0048] Figure 2 FIG. is a schematic diagram of the scope of the golden waterway in the upper reaches of the Yangtze River in the embodiment;

[0049] Figure 3 FIG. is a graph of the water level change of Qingxichang Hydrological Station in the fluctuating backwater area of the Three Gorges Reservoir in 2019 in the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0050] The following describes the specific implementation manners of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation manners. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0051] As Figure 1 shown, a prediction method for fish egg drifting includes the following steps S1-S4:

[0052] S1. Obtain the sequence of influencing factors of the river section in the research area and the fish egg quantity in historical data.

[0053] As Figure 2 shown, Figure 2 FIG. is a schematic diagram of the scope of the golden waterway in the upper reaches of the Yangtze River. In this embodiment, the research area is the waterway sections of Chaotianmen-Fuling and Fuling-Fengdu in the Three Gorges Reservoir area. Since the waterway regulation project is being implemented in this section of the waterway, but there are four major domestic fish protection areas and rare fish protection areas as Figure 2 shown in this section of the waterway, its construction time is restricted by the "fishing moratorium" and water level, water flow, etc. Therefore, this embodiment proposes a prediction method for fish egg drifting. By obtaining the sequence of influencing factors of the river section in the research area and the fish egg quantity in historical data, a prediction model with multi-scale factor fusion is constructed to predict the fish spawning and drifting situation. By accurately predicting the peak period of fish egg drifting, and enabling the construction of the waterway project in the research area to avoid the peak period of fish spawning, thereby shortening the construction process of the waterway project in the research area.

[0054] As Figure 3 shown Figure 3 is the water level change diagram of Qingxichang Hydrological Station in the fluctuating backwater area of the Three Gorges Reservoir Area in 2019. Figure 3 It shows the water level changes of Qingxichang in the fluctuating backwater area of the Three Gorges Reservoir Area on different days in 2019. It can be seen that it is not conducive to construction during the fish spawning period and the flood period. Through long-term in-situ observation results and related research in the upper reaches of the Yangtze River, large-scale fish egg drifting occurs from the end of May to the middle of July every year; large-scale fish egg drifting passes through the Fuling and Fengdu sections in the fluctuating backwater area of the Three Gorges Reservoir when the flow rate increase is ~40% and the water level increase is >2m; during the 3-5 day peak period of continuous rising water, the number of drifting fish eggs monitored accounts for more than 90% of the total number of fish eggs during the entire fish spawning period; within the 120-day fishing moratorium, the peak period of fish reproduction in the fluctuating backwater area of the Three Gorges Reservoir is only about 15 days, and there is basically no large-scale fish reproduction and fish egg drifting phenomenon in other periods. In order to shorten the construction process of the waterway project in the research area and make full use of the low water level during the fishing moratorium, this embodiment provides a method for predicting the peak period of domestic fish spawning in the fluctuating backwater area of the Yangtze River Three Gorges Reservoir Area, which can meet the requirement of avoiding the peak period of domestic fish spawning during the construction period.

[0055] Specifically, step S1 specifically includes S11 - S12:

[0056] S11. Obtain the daily fish egg quantity monitored in the river section of the research area in historical data to obtain the fish egg quantity in the river section of the research area in historical data.

[0057] In this embodiment, the fish egg quantity in the river section of the research area in historical data is the fish egg runoff calculated by combining the fish eggs collected through field monitoring with a formula. Specifically, the calculation process of the fish egg runoff is as follows: The scale of fish eggs collected through field monitoring is obtained, and the Yibolu survey method is used to calculate the fish egg net density per unit time, that is: where d represents the fish egg net density per unit time, s represents the net mouth area, v represents the net mouth flow velocity, t represents the collection time, and n represents the number of fish eggs entering the net during the collection period; according to the fish egg net density per unit time, calculate the fish egg runoff passing through the cross-section of the river section of the research area during a single collection period, that is: M i = d i × Q i × C, where M i represents the fish egg runoff passing through the cross-section of the river section of the research area during the i-th collection period, d i represents the fish egg net density per unit time collected during the i-th collection, Q i represents the river section cross-section flow of the research area during the i-th collection; calculate the fish egg runoff passing through the cross-section of the river section of the research area during a non-collection period, that is: where Mi,i+1 The fish egg runoff passing through the cross-section of the river section in the study area during the time interval between the i-th and (i + 1)-th collections, t i,i+1 Indicates the time interval between the i-th and (i + 1)-th collections, t i Indicates the i-th collection time, M i+1 Indicates the fish egg runoff passing through the cross-section of the river section in the study area during the (i + 1)-th collection period, t i+1 Indicates the (i + 1)-th collection time; calculate the total fish egg runoff of the river section in the study area, that is: M = ∑M i +∑M i,i+1 , where ∑M i Indicates the sum of the fish egg flows of each collection during the collection time, ∑M i,i+1 Indicates the sum of the fish egg runoffs passing through the cross-section of the river section in the study area during the non-collection time between two consecutive collections.

[0058] S12. Obtain the sequence of influencing factors of the river section in the study area from the historical data of the fish egg quantity in the river section in the study area obtained in step S11 and the daily water flow, water level, water temperature, and water flow velocity data monitored in the historical data of the river section in the study area.

[0059] In this embodiment, the rising water condition is one of the necessary conditions for the spawning of the four major Chinese carps, which usually occurs during the period from April to July. The rising water can occur under the condition of 0.5 days to 2 days. The rising water can increase the water level, water flow rate and water flow velocity, which is a stimulating signal for the parent fish with mature gonads and prompts them to start spawning. However, when the rising water stops and the water flow velocity decreases, the spawning behavior of most parent fish will also stop accordingly. In the case of a small amount of falling water and flat water, as long as the water flow velocity meets the requirements, a small number of Chinese carps will also maintain spawning for a period of time. Secondly, the change of water flow velocity is also very important for the spawning of the four major Chinese carps. Spawning requires a certain degree of water flow stimulation, and the water flow velocity range for Chinese carps to spawn is 0.33m / s to 1.5m / s. During the spawning period of Chinese carps, the water flow velocity will increase by 0.2m / s to 1m / s on this basis. At this time, the water flow acceleration can be considered as an important signal for Chinese carps to spawn. The water flow acceleration is divided into horizontal and vertical, and the vertical acceleration is much greater than the horizontal acceleration. Therefore, the vertical acceleration is an important factor for Chinese carps to spawn. Finally, temperature is also one of the important factors affecting the spawning of the four major Chinese carps. Appropriate temperature is a necessary prerequisite for fish to spawn. The gonad development in the parent fish becomes more obvious with the increase of water temperature. However, when the water temperature is lower than 18°C or higher than 30°C, the four major Chinese carps will stop spawning, and the development of fish eggs will also stagnate or die. To improve the hatching rate, the fish eggs must be kept within a certain temperature range. The speed of fish egg development is also closely related to the water temperature. The increase of water temperature will accelerate all stages of embryo development and shorten the overall development time of the embryo. Therefore, in this embodiment, according to the daily water flow rate, water level, water temperature and water flow velocity data of the river section in the study area monitored in the historical data, the influence factor sequence of the river section in the study area in the historical data is obtained.

[0060] S2. Construct a prediction model for multi-scale factor fusion of the river section in the study area according to the influence factor sequence of the river section in the study area in the historical data and the amount of fish eggs obtained in step S1.

[0061] Specifically, step S2 specifically includes S21 - S22:

[0062] S21. According to the amount of fish eggs in the river section in the study area in the historical data obtained in step S1, conduct binary conversion classification on whether the fish spawned in the historical data. If the amount of fish eggs is greater than 0, it means that the fish have spawning phenomena and are marked as 1; otherwise, it means that the fish have no spawning phenomena and are marked as 0, and the spawning results of the fish after binomial classification are obtained.

[0063] In this embodiment, the spawning situation of domestic fish is converted into a binary factor, that is, the occurrence of spawning is marked as 1, and the non-occurrence of spawning is marked as 0. At the same time, during the subsequent processing with R language, the result of the occurrence of spawning is marked as NO, and the result of the non-occurrence of spawning is marked as YES, so as to facilitate the establishment of a prediction model for multi-scale factor fusion in the follow-up.

[0064] S22. Based on the fish spawning results obtained from the binomial classification in step S21 and the sequence of influencing factors in the historical data of the river section in the study area obtained in step S1, a generalized linear equation is established based on the theory of multivariate non-linear functions to construct a prediction model for multi-scale factor fusion in the river section of the study area, that is:

[0065]

[0066] Among them, represents the predicted daily spawning situation of fish. When Y = 1, it means that fish have spawning phenomena, and when Y = 0, it means that fish have no spawning phenomena. μ Y represents the probability that fish have fish eggs when given the values of influencing factors, and 1 - μ Y represents the situation where fish have no fish eggs when given the values of influencing factors. α0 represents the spawning situation of fish when all influencing factors are 0, p represents the number of days, and β j represents the independent variable coefficient on the jth day, χ j represents the water flow of the river section in the study area on the jth day, γ j represents the independent variable coefficient on the jth day, ψ j represents the average flow velocity of the river section in the study area on the jth day, δ j represents the independent variable coefficient on the jth day, φ j represents the water level of the river section in the study area on the jth day, ε j represents the independent variable coefficient on the jth day, ω j represents the water temperature of the river section in the study area on the jth day.

[0067] In this embodiment, the fish spawning results obtained from the binomial classification and the sequence of influencing factors in the historical data of the river section in the study area are stored in an Excel table and this table is imported into R language in the "xlsx" format to establish a generalized linear equation and construct a prediction model for multi-scale factor fusion in the river section of the study area. The R language in this embodiment is a computer software.

[0068] S3. Conduct a significant test correction on the prediction model for multi-scale factor fusion in the river section of the study area constructed in step S2 to obtain a corrected prediction model for multi-scale factor fusion in the river section of the study area.

[0069] In this embodiment, the constructed prediction model for the multi-scale factor fusion of the river section in the study area is tested and corrected. Significance testing, i.e., setting the p-value to 0.05, and regression diagnosis are used for testing and correction. During the significance testing process, the independent variables that do not meet the significance standard, i.e., the influencing factors, need to be excluded, and the influencing factors that pass the significance test are retained and re-introduced into the constructed prediction model for multi-scale factor fusion for fitting calculation. The process of continuously repeating the test and observing whether the regression diagnosis of the prediction model for multi-scale factor fusion is qualified and the level of the fitting degree is carried out to obtain the corrected prediction model for the multi-scale factor fusion of the river section in the study area. The corrected prediction model for the multi-scale factor fusion of the river section in the study area in this embodiment has the characteristic of good fitting effect, and the fitting degree of the obtained prediction model for multi-scale factor fusion in this embodiment is higher than 60%.

[0070] Specifically, step S3 specifically includes S31 - S39:

[0071] S31. Assume that the influencing factor sequence of the river section in the study area has no significant impact on fish spawning, that is:

[0072]

[0073] Among them, represents the fish spawning probability after considering the influencing factor sequence on the j-th day of the river section in the study area, represents the fish spawning probability when not considering the influencing factor sequence on the j-th day of the river section in the study area.

[0074] S32. Assume that the influencing factor sequence of the river section in the study area has a significant impact on fish spawning, that is:

[0075]

[0076] S33. Calculate the expected value of the influencing factor sequence of the river section in the study area on fish spawning, that is:

[0077]

[0078] Among them, represents the expected value of the influencing factor sequence of the river section on the j-th day in the study area on fish spawning, and n represents the total number of historical data.

[0079] S34. According to the expected value of the influencing factor sequence of the river section in the study area on fish spawning calculated in step S33, calculate the standard error of the expected value of the influencing factor sequence of the river section in the study area on fish spawning, that is:

[0080]

[0081] Among them, Denote the standard error of the expected value of the influencing factor sequence on the fish spawning on the j-th day in the river section of the study area.

[0082] S35. Calculate the test statistic of the influencing factor sequence in the river section of the study area on the fish spawning according to the standard error of the expected value of the influencing factor sequence in the river section of the study area on the fish spawning calculated in step S34, that is:

[0083]

[0084] Where Denote the test statistic of the influencing factor sequence on the j-th day in the river section of the study area on the fish spawning, Denote the total number of times of fish eggs collected on the j-th day in the historical data.

[0085] S36. Find the left-tail probability number and the right-tail probability number corresponding to the test statistic of the influencing factor sequence in the river section of the study area on the fish spawning calculated in step S35 through the standard normal distribution table, and superimpose the left-tail probability and the right-tail probability to obtain the p-value of the two-sided test, that is:

[0086] p two-sided = 2 * (left-tail probability number, right-tail probability number)

[0087] S37. If the p-value of the two-sided test obtained in step S36 is less than 0.05, the hypothesis in step S31 does not hold, and the hypothesis in step S32 holds, that is, the influencing factor sequence in the river section of the study area has a significant impact on the fish spawning.

[0088] S38. If the p-value of the two-sided test obtained in step S36 is greater than or equal to 0.05, the hypothesis in step S31 holds, and the hypothesis in step S32 does not hold, that is, the influencing factor sequence in the river section of the study area has no significant impact on the fish spawning.

[0089] S39. Remove the influencing factors that have no significant impact on the fish spawning in the influencing factor sequence in the river section of the study area in step S38, and re-enter the influencing factors that have a significant impact on the fish spawning in the influencing factor sequence in the river section of the study area into the prediction model of the multi-scale factor fusion in the river section of the study area constructed in step S2 for calculation to obtain the corrected prediction model of the multi-scale factor fusion in the river section of the study area.

[0090] S4. Obtain the influencing factors in the river section of the study area every day, and use the corrected prediction model of the multi-scale factor fusion in the river section of the study area obtained in step S3 to predict the daily fish egg drift situation.

[0091] In this embodiment, historical data from previous years (i.e., data not involved in curve fitting) is used to evaluate the calibrated forecast model for the multi-scale factor fusion of the river section in the study area. Also, it should be noted that the historical data obtained here needs to have the same year and location. At the same time, the calibrated forecast model for the multi-scale factor fusion of the river section in the study area constructed in this embodiment ignores multicollinearity during testing because multicollinearity has no impact on the model fitting and predicted values in this embodiment. The calibrated forecast model for the multi-scale factor fusion of the river section in the study area constructed in this embodiment can be integrated into the fish egg drift forecast platform for practical applications. That is, by importing influencing factors such as the daily water flow, water level, water temperature, and water flow velocity of the river section in the study area into the calibrated forecast model for the multi-scale factor fusion of the river section in the study area, the fish spawning probability results can be calculated as shown in Table 1:

[0092] Table 1 Fish Spawning Probability Results

[0093]

[0094]

[0095] As can be seen from Table 1, the calibrated forecast model for the multi-scale factor fusion of the river section in the study area constructed in this embodiment can predict the number of fish eggs based on the input influencing factors, namely water flow velocity, water level, water temperature, and water flow, and calculate the fish spawning probability during this period.

[0096] Specifically, step S4 specifically includes S41 - S42:

[0097] S41. Obtain influencing factors such as the daily water flow, water level, water temperature, and water flow velocity of the river section in the study area, and import them into the calibrated forecast model for the multi-scale factor fusion of the river section in the study area constructed in step S3 to calculate the fish spawning probability;

[0098] S42. If the fish spawning probability calculated in step S41 is less than 0.5, it means that there is no fish spawning phenomenon; otherwise, there is a fish spawning phenomenon.

[0099] The calibrated forecast model for the multi-scale factor fusion of the river section in the study area constructed in this embodiment is applicable to predicting the fish egg drift situation in the main stream and tributaries. In addition, when constructing this model in this embodiment, for special water condition years such as "wet years" and "dry years", a separate fitting curve, i.e., a forecast model for multi-scale factor fusion, needs to be established to predict the fish egg drift situation in special water condition years. So that when the water condition in the predicted year is similar to that in the historical year, the spawning results of the historical year can be directly used as an auxiliary reference.

[0100] In the present invention, specific embodiments are used to illustrate the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

[0101] Those of ordinary skill in the art will realize that the embodiments described herein are for helping the reader understand the principles of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations without departing from the essence of the present invention based on the technical revelations disclosed in the present invention, and these deformations and combinations are still within the protection scope of the present invention.

Claims

1. A prediction method for fish egg drifting, characterized in that, It includes the following steps: S1. Obtain the sequence of influencing factors and the amount of fish eggs in the river section of the study area from historical data; S2. According to the sequence of influencing factors and the amount of fish eggs in the river section of the study area obtained in step S1, construct a prediction model for multi-scale factor fusion in the river section of the study area; S3. Conduct a significant test correction on the prediction model for multi-scale factor fusion in the river section of the study area constructed in step S2 to obtain a corrected prediction model for multi-scale factor fusion in the river section of the study area; S4. Obtain the influencing factors of the river section in the study area every day, and use the corrected prediction model for multi-scale factor fusion in the river section of the study area obtained in step S3 to predict the daily fish egg drifting situation of fish.

2. The prediction method for fish egg drift according to claim 1, wherein Step S1 specifically includes: S11. Obtain the daily amount of fish eggs in the river section of the study area monitored in historical data to obtain the amount of fish eggs in the river section of the study area in historical data; S12. According to the amount of fish eggs in the river section of the study area in historical data obtained in step S11 and the corresponding daily water flow, water level, water temperature, and water flow velocity data of the river section of the study area monitored in historical data, obtain the sequence of influencing factors in the river section of the study area in historical data.

3. The prediction method for fish egg drift according to claim 1, characterized in that, Step S2 specifically includes: S21. According to the amount of fish eggs in the river section of the study area obtained in step S1, conduct a binary conversion and classification on whether fish spawn in historical data. If the amount of fish eggs is greater than 0, it means that fish spawning occurs and is marked as 1. Otherwise, it means that fish spawning does not occur and is marked as 0, to obtain the fish spawning result of binary classification processing; S22. Based on the fish spawning result of binary classification processing obtained in step S21 and the sequence of influencing factors in the river section of the study area obtained in step S1, establish a generalized linear equation based on the theory of multivariate nonlinear functions to construct a prediction model for multi-scale factor fusion in the river section of the study area.

4. A prediction method for fish egg drifting according to claim 3, characterized in that, The prediction model for multi-scale factor fusion in the river section of the study area constructed in step S22 is: Among them, represents the predicted daily spawning situation of fish. When Y = 1, it indicates that fish spawning occurs; when Y = 0, it indicates that fish spawning does not occur. μ Y represents the probability of fish egg production when the given influencing factor value is given. 1 - μ Y represents the situation where fish do not produce eggs when the given influencing factor value is given. α0 represents the spawning situation of fish when all influencing factors are 0. p represents the number of days, and β j represents the independent variable coefficient on the j-th day, χ j represents the water flow on the j-th day in the river section of the study area, γ j represents the independent variable coefficient on the j-th day, ψ j represents the average flow velocity on the j-th day in the river section of the study area, δ j represents the independent variable coefficient on the j-th day, φ j represents the water level on the j-th day in the river section of the study area, ε j represents the independent variable coefficient on the j-th day, ω j represents the water temperature on the j-th day in the river section of the study area.

5. A method for predicting the drift of fish eggs according to claim 1, characterized in that, Step S3 specifically includes: S31. Assume that the sequence of influencing factors in the river section of the study area has no significant effect on fish spawning, that is: Among them, represents the fish spawning probability after considering the influencing factor sequence on the j-th day of the river section in the study area, represents the fish spawning probability without considering the influencing factor sequence on the j-th day of the river section in the study area; S32. Assume that the sequence of influencing factors in the river section of the study area has a significant effect on fish spawning, that is: S33. Calculate the expected value of the sequence of influencing factors in the river section of the study area on fish spawning, that is: Among them, represents the expected value of the impact factor sequence on fish spawning in the j-th day of the river section in the study area, and n represents the total number of historical data; S34. According to the expected value of the sequence of influencing factors in the river section of the study area on fish spawning calculated in step S33, calculate the standard error of the expected value of the sequence of influencing factors in the river section of the study area on fish spawning, that is: Among them, represents the standard error of the expected value of the influence factor sequence on the fish spawning in the study area section on the j-th day; S35. According to the standard error of the expected value of the sequence of influencing factors in the river section of the study area on fish spawning calculated in step S34, calculate the test statistic of the sequence of influencing factors in the river section of the study area on fish spawning; S36. Find the left-tail probability number and right-tail probability number corresponding to the test statistic of the sequence of influencing factors in the river section of the study area on fish spawning calculated in step S35 through the standard normal distribution table, and add the left-tail probability and right-tail probability to obtain the p-value of the two-sided test; S37. If the p-value of the two-sided test obtained in step S36 is less than 0.05, the hypothesis in step S31 does not hold, and the hypothesis in step S32 holds, that is, the influencing factor sequence of the study area river section has a significant impact on fish spawning; S38. If the p-value of the two-sided test obtained in step S36 is greater than or equal to 0.05, the hypothesis in step S31 holds, and the hypothesis in step S32 does not hold, that is, the influencing factor sequence of the study area river section has no significant impact on fish spawning; S39. Remove the influencing factors that have no significant impact on fish spawning in the influencing factor sequence of the study area river section in step S38, and re-introduce the influencing factors that have a significant impact on fish spawning in the influencing factor sequence of the study area river section into the prediction model of multi-scale factor fusion of the study area river section constructed in step S2 for calculation, and obtain the corrected prediction model of multi-scale factor fusion of the study area river section.

6. The prediction method for fish egg drifting according to claim 5, wherein The calculation formula for the test statistic of the influencing factor sequence of the study area river section on fish spawning in step S35 is: Among them, represents the test statistic of the influence factor sequence on fish spawning in the j-th day of the study area river section, represents the total number of times of fish eggs collected on the j-th day in historical data.

7. A method for predicting the drift of fish eggs according to claim 5, characterized in that, The calculation formula for the p-value of the two-sided test in step S36 is: p two-sided = 2 * (left tail probability number, right tail probability number).

8. A prediction method for fish egg drifting according to claim 1, characterized in that Step S4 specifically includes: S41. Obtain the influencing factors such as daily water flow, water level, water temperature, and water flow velocity of the study area river section, and import them into the corrected prediction model of multi-scale factor fusion of the study area river section constructed in step S3 to calculate the fish spawning probability; S42. If the fish spawning probability calculated in step S41 is less than 0.5, there is no fish spawning phenomenon, otherwise, there is a fish spawning phenomenon.

Images