El Nino index-based red tide organism species quantity prediction method and system

Abstract

The invention discloses an El Nino index-based red tide organism species quantity prediction method and system, and the method comprises the following steps: S1, obtaining Nino3.4 index historical data of a target sea area, and collecting red tide organism species quantity historical data; s2, the change rate of the number of red tide types in the El Nino year, the next year and the third year is calculated, and the El Nino year is screened; s3, analyzing a correlation coefficient between the Nino3.4 index and the number of red tide types of the next year by adopting a point double-column correlation method; s4, constructing a red tide organism species quantity prediction model; s5, introducing a segmented correction factor to optimize model output; and S6, according to the red tide biological species quantity prediction model, performing red tide species prediction and outputting. The system is directly integrated to an existing marine monitoring platform, is simple and convenient to operate and low in cost, remarkably improves the dynamic prediction capability of response to large-scale climate events, and is suitable for marine ecological protection and disaster early warning.

Description

Technical Field

[0001] The present invention relates to the field of marine environment monitoring and disaster prediction, and in particular to a method and system for predicting the number of red tide organisms based on the El Niño index. Background Art

[0002] Red tides (HBsAg) are a major environmental problem in marine ecosystems, posing a serious threat to fisheries, tourism, and marine ecological security. Existing HBsAg prediction methods are mostly based on local environmental parameters (such as temperature and nutrients), but lack dynamic response analysis of large-scale climate events. El Niño, a global climate event, significantly impacts marine environmental parameters, but the quantitative relationship between HBsAg and changes in the abundance of HBsAg species has not been effectively applied to prediction models. Currently, no patented solutions exist that combine El Niño indices (such as the Nino 3.4 index) with historical HBsAg data to construct prediction models. Summary of the Invention

[0003] The purpose of the present invention is to provide a method and system for predicting the number of red tide organisms based on the El Niño index, so as to solve the problems raised in the above background technology.

[0004] To achieve the above-mentioned object, one aspect of the present invention provides a method for predicting the number of red tide organisms based on the El Niño index, comprising the following steps:

[0005] Step S1, obtaining historical data of the Nino3.4 index of the target sea area and collecting historical data on the number of red tide organisms;

[0006] Step S2, calculating the change rate of the number of red tide species in the El Niño year, the next year, and the third year, and selecting the El Niño year;

[0007] Step S3, using the point biserial correlation method to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year;

[0008] Step S4: construct a red tide organism species quantity prediction model. The linear regression model formula is as follows:

[0009] Y = a⋅X+b,

[0010] Where Y is the predicted rate of change in the number of red tide species; X is the Nino3.4 index; a is the regression coefficient, which represents the direct impact of the El Niño event on the number of red tide species; b is the intercept term, which represents the baseline predicted value of the rate of change in the number of red tide species when the Nino3.4 index is 0;

[0011] Step S5, introducing a segmented correction factor to optimize the model output;

[0012] Step S6: predicting and outputting the red tide species based on the red tide organism species quantity prediction model.

[0013] Furthermore, the steps for obtaining the Nino3.4 index in step S1 are:

[0014] Step S101, obtaining sea surface temperature data of the Nino3.4 area from a data source;

[0015] Step S102, calculating the Nino3.4 index, including calculating the monthly average outlier value, the regional average value, and selecting the linear trend removal;

[0016] Step S103: Smoothing the Nino3.4 index.

[0017] Furthermore, the data source is NOAA's ERSST dataset.

[0018] Furthermore, in step S2, the year in which the Nino3.4 index exceeds +0.5°C for six consecutive months is regarded as the El Niño year.

[0019] Furthermore, the calculation formula of the change rate G is:

[0020] .

[0021] Furthermore, the point biserial correlation method is used to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year, including the following steps:

[0022] Step S301, performing data processing on the Nino3.4 index and red tide species count data, wherein the data processing includes time alignment and normalization;

[0023] Step S302, calculating the correlation coefficient and performing a significance test;

[0024] Step S303: Analyze and visualize the correlation coefficient results.

[0025] Furthermore, the method for introducing the segmented correction factor in step S5 includes the following steps:

[0026] Step S501, determine the segmentation point and divide the data into two stages;

[0027] Step S502: Analyze the data of the two phases respectively, calculate the correlation between the change rate of the number of red tide organisms in each phase and the Nino3.4 index, and establish a preliminary linear regression model;

[0028] Step S503, defining a correction factor as the ratio or difference of the regression parameters of the two stages according to the difference in the model parameters of the two stages;

[0029] Step S504: Modify the model output according to the stage to which the current year belongs.

[0030] Another aspect of the present invention provides a red tide species number prediction system based on the El Niño index, comprising a data acquisition module, a calculation module, a correlation coefficient module, a modeling module, a model optimization module, and an output module, wherein:

[0031] The data acquisition module is used to obtain the historical data of the Nino3.4 index of the target sea area and collect historical data on the number of red tide organisms;

[0032] The calculation module is used to calculate the change rate of red tide species in the El Niño year, the next year, and the third year, and to filter out the El Niño year;

[0033] The correlation coefficient module is used to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year using the point biserial correlation method;

[0034] The modeling module is used to build a prediction model for the number of red tide organisms. The linear regression model formula is as follows:

[0035] Y = a⋅X+b,

[0036] Where Y is the predicted rate of change in the number of red tide species; X is the Nino3.4 index; a is the regression coefficient, which represents the direct impact of the El Niño event on the number of red tide species; b is the intercept term, which represents the baseline predicted value of the rate of change in the number of red tide species when the Nino3.4 index is 0;

[0037] The model optimization module is used to introduce segmented correction factors to optimize the model output;

[0038] The output module is used to predict the red tide species based on the red tide species number prediction model and output

[0039] Compared with the prior art, the present system and method have the following advantages:

[0040] 1. This invention introduces a segmented correction factor, enabling the model to better adapt to data characteristics at different stages, thereby improving prediction accuracy. Through segmented correction, the model can adapt to the data patterns of the two stages separately, avoiding prediction bias caused by stage-specific data differences. By introducing segmented correction factors, the model can more accurately reflect the changes in the number of red tide species at different stages, providing a more reliable scientific basis for red tide disaster warnings, fishery resource management, and marine ecological protection.

[0041] 2. This invention is the first to dynamically correlate the El Niño index with the number of red tide species, filling a gap in this field. It employs a segmented correction model to address periodic discrepancies in historical data. Applicable to the East China Sea and other offshore areas affected by El Niño, it can be directly integrated into existing ocean monitoring platforms, offering ease of operation and low cost. This provides a scientific basis for red tide disaster warning, fishery resource management, and marine ecological protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of the method for predicting the number of red tide organisms based on the El Niño index.

[0043] Figure 2 This is a curve chart of the number of red tide species occurring in the East China Sea and the Nino3.4 zone index. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0045] like Figure 1 FIG. 1 is a flow chart of a method of the present invention. An embodiment of the present invention provides a method for predicting the number of red tide organisms based on the El Niño index. The specific steps are as follows:

[0046] Step S1: Obtain historical data on the Nino 3.4 index for the target sea area and collect historical data on the number of red tide organisms. This data is sourced from marine monitoring agencies. Data on the number of red tide species from the following year is obtained, along with time series data on the number of red tide species, ensuring that the temporal resolution of the data is consistent with the Nino 3.4 index.

[0047] Obtaining Nino3.4 index data includes the following steps:

[0048] Step S101 : Obtain sea surface temperature (SST) data of the Nino3.4 region (5°S-5°N, 170°W-120°W) from relevant data sources (such as NOAA's ERSST dataset).

[0049] Step S102, calculating the Nino3.4 index, including calculating the monthly average outlier, the regional average, and optionally removing the linear trend.

[0050] Step S103: Smoothing the Nino3.4 index (eg, a three-month sliding average).

[0051] Step S2: Calculate the rate of change in the number of red tide species between the El Niño year, the following year, and the third year, and select El Niño years. Select years in which the Nino3.4 index exceeds +0.5°C for six consecutive months as El Niño years. The formula for calculating the rate of change in the number of red tide species between the El Niño year, the following year, and the third year is as follows:

[0052] .

[0053] Step S3, using the point biserial correlation method to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year, includes the following steps:

[0054] Step S301: Data processing is performed on the Nino3.4 index and red tide species count data, including time alignment and standardization. Time series alignment of the Nino3.4 index and red tide species count data is ensured, particularly considering the relationship between the Nino3.4 index and the red tide species count in the following year. If lag or lead effects need to be analyzed, a time offset can be applied to one of the variables. Both variables are standardized to a mean of 0 and a standard deviation of 1 to ensure a fairer comparison.

[0055] Step S302: Calculate the correlation coefficient and perform a significance test.

[0056] Calculate the correlation coefficient using statistical software or a programming language. In Python, you can use the numpy or xarray libraries to calculate the correlation coefficient between two time series. Perform a significance test on the calculated correlation coefficient to determine whether it is statistically significant.

[0057] Step S303: Analyze and visualize the correlation coefficient results.

[0058] First, let's explain the correlation coefficient. The range of the correlation coefficient is [-1, 1]. Values closer to 1 or -1 indicate a stronger correlation between the two variables. Based on the sign of the correlation coefficient, determine whether the relationship between the Nino3.4 index and the number of red tide species is positive or negative. Plot a time series graph of the two variables to observe their trends and correlations. You can use the atplotlib library to create correlation charts to help you more intuitively understand the results.

[0059] Step S4: construct a red tide organism species quantity prediction model. The linear regression model formula is as follows:

[0060] Y = a⋅X+b,

[0061] Where Y is the predicted rate of change in the number of red tide species; X is the Nino3.4 index; a is the regression coefficient, which represents the direct impact of the El Niño event on the number of red tide species; and b is the intercept term, which represents the baseline predicted value of the rate of change in the number of red tide species when the Nino3.4 index is 0.

[0062] Step S5: Introduce a segmented correction factor to optimize the model output. The segmented correction factor is introduced to optimize the output of the red tide organism population prediction model, solve the problem of periodic differences in historical data, and thus improve the prediction accuracy of the model.

[0063] The occurrence of red tides in the East China Sea can be divided into two stages starting in 2000:

[0064] The first stage (1981-2000): There were an average of 8 red tides per year, and the average number of red tide species was 4 per year, with an overall low frequency of occurrence.

[0065] The second stage (2001-2011): The frequency of red tides was 6.4 times that of the first stage, and the average annual number of red tide-causing species was 2.8 times that of the first stage.

[0066] This stage difference indicates that the occurrence patterns and influencing factors of red tides may have changed significantly around 2000, so the model needs to be revised in sections.

[0067] The specific method of introducing the segment correction factor includes the following steps:

[0068] Step S501: Based on the above research results, the year 2000 is selected as the segmentation point, and the data is divided into two stages:

[0069] The first stage (1981-2000): used to analyze the early changes in the number of red tide species.

[0070] The second phase (2001-2011): used to analyze the changes in the number of red tide species in the later period.

[0071] Step S502 , analyzing the data of the two phases respectively, calculating the correlation between the rate of change of the number of red tide organisms in each phase and the Nino3.4 index, and establishing a preliminary linear regression model.

[0072] First-stage model: Calculate correlation coefficients and regression parameters based on data from 1981 to 2000.

[0073] Second-stage model: Calculate correlation coefficients and regression parameters based on data from 2001 to 2011.

[0074] Step S503 defines a correction factor as the ratio or difference of the regression parameters of the two stages based on the difference in model parameters between the two stages. For example, if the regression coefficient of the first stage is a1 and the regression coefficient of the second stage is a2, the correction factor can be defined as Δa = a2 − a1.

[0075] Step S504: During the prediction, the model output is modified according to the stage to which the current year belongs. For example, for data after 2000, the predicted value can be modified by the following formula: 修正 =Y+Δa⋅X,

[0076] Where Y is the predicted value of the preliminary model, X is the Nino3.4 index, and Δa is the correction factor.

[0077] After introducing the segmented correction factor, the model can better adapt to the data characteristics of different stages, thereby improving the prediction accuracy. Taking the prediction of the number of red tide species in the East China Sea as an example, assuming that 2023 is an El Niño year and the Nino3.4 index is +1.2℃. According to the linear regression model The preliminary forecast is that the change rate of red tide species in the next year will be 15.0%.

[0078] The process of introducing the segmented correction factor is as follows: Assuming the correction factor is Δa=0.1, the corrected prediction value is:

[0079] Y correction=15.0%+0.1×1.2=16.2%,

[0080] Combining the historical mean and the number of species in the current year, the predicted number of red tide species in the next year is 11.6 (confidence interval ±1.2 species).

[0081] Step S6: Input the Nino3.4 index of the current El Niño year into the constructed red tide organism species number prediction model, and output the predicted value and confidence interval of the number of red tide species in the next year.

[0082] Cross-validation of these predictions using data from 1981 to 2011 revealed an average prediction error of ≤12%. Compared to traditional temperature models, this approach improves accuracy by approximately 35%.

[0083] This invention dynamically correlates the El Niño index with the number of red tide species for the first time, filling a technological gap and addressing the issue of periodic discrepancies in historical data using a piecewise correction model. Directly integrating into existing ocean monitoring platforms, this method offers ease of operation and low cost, significantly improving the ability to dynamically predict responses to large-scale climate events and making it suitable for marine ecological protection and disaster warning.

[0084] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the number of red tide organisms based on the El Niño index, characterized in that: The following steps are involved: Step S1, obtaining historical data of the Nino3.4 index of the target sea area and collecting historical data on the number of red tide organisms; Step S2, calculating the change rate of the number of red tide species in the El Niño year, the next year, and the third year, and selecting the El Niño year; Step S3, using the point biserial correlation method to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year; Step S4: construct a red tide organism species quantity prediction model. The linear regression model formula is as follows: Y = a⋅X+b, Where Y is the predicted rate of change in the number of red tide species; X is the Nino3.4 index; a is the regression coefficient, which represents the direct impact of the El Niño event on the number of red tide species; b is the intercept term, which represents the baseline predicted value of the rate of change in the number of red tide species when the Nino3.4 index is 0; Step S5, introducing a segmented correction factor to optimize the model output; Step S6: predicting and outputting the red tide species based on the red tide organism species quantity prediction model.

2. The method for predicting the number of red tide organisms based on the El Niño index according to claim 1, characterized in that: The steps for obtaining the Nino3.4 index in step S1 are: Step S101, obtaining sea surface temperature data of the Nino3.4 area from a data source; Step S102, calculating the Nino3.4 index, including calculating the monthly average outlier value, the regional average value, and selecting the linear trend removal; Step S103: Smoothing the Nino3.4 index.

3. The method for predicting the number of red tide organisms based on the El Niño index according to claim 2, characterized in that: The data source is NOAA's ERSST dataset.

4. The method for predicting the number of red tide organisms based on the El Niño index according to claim 1, wherein: In step S2, the year in which the Nino3.4 index exceeds +0.5°C for six consecutive months is regarded as the El Nino year.

5. The method for predicting the number of red tide organisms based on the El Niño index according to claim 1, characterized in that: The calculation formula of the change rate G is: 。 6. The method for predicting the number of red tide organisms based on the El Niño index according to claim 1, characterized in that: The point biserial correlation method is used to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year, which comprises the following steps: Step S301, performing data processing on the Nino3.4 index and red tide species count data, wherein the data processing includes time alignment and normalization; Step S302, calculating the correlation coefficient and performing a significance test; Step S303: Analyze and visualize the correlation coefficient results.

7. The method for predicting the number of red tide organisms based on the El Niño index according to claim 1, characterized in that: The method for introducing the segmented correction factor in step S5 includes the following steps: Step S501, determine the segmentation point and divide the data into two stages; Step S502: Analyze the data of the two phases respectively, calculate the correlation between the change rate of the number of red tide organisms in each phase and the Nino3.4 index, and establish a preliminary linear regression model; Step S503, defining a correction factor as the ratio or difference of the regression parameters of the two stages according to the difference in the model parameters of the two stages; Step S504: Modify the model output according to the stage to which the current year belongs.

8. The red tide species number prediction system based on the El Niño index is characterized by: It includes data acquisition module, calculation module, correlation coefficient module, modeling module, model optimization module and output module, among which: The data acquisition module is used to obtain the historical data of the Nino3.4 index of the target sea area and collect historical data on the number of red tide organisms; The calculation module is used to calculate the change rate of red tide species in the El Niño year, the next year, and the third year, and to filter out the El Niño year; The correlation coefficient module is used to analyze the correlation coefficient between the Nino3.4 index and the number of red tide species in the following year using the point biserial correlation method; The modeling module is used to build a prediction model for the number of red tide organisms. The linear regression model formula is as follows: Y = a⋅X+b, Where Y is the predicted rate of change in the number of red tide species; X is the Nino3.4 index; a is the regression coefficient, which represents the direct impact of the El Niño event on the number of red tide species; b is the intercept term, which represents the baseline predicted value of the rate of change in the number of red tide species when the Nino3.4 index is 0; The model optimization module is used to introduce segmented correction factors to optimize the model output; The output module is used to predict and output red tide species based on the red tide organism species quantity prediction model.

Images