Method for making long-term weather forecast in first-batch flowering period of osmanthus fragrans

Abstract

The invention discloses a method for making a long-term weather forecast of a first-batch flowering period of osmanthus fragrans. The method comprises the following steps: S1, collecting data of an initial flowering period of osmanthus fragrans as an actually measured initial flowering period; s2, determining a low-temperature daily index and a low-temperature process effectiveness index before flowering of the sweet-scented osmanthus by using an improved gradual approaching method; s3, establishing a unary linear regression equation of the actually measured initial flowering period and the effective low-temperature process starting day; and S4, counting a first effective low-temperature process starting day sequence of each year since the weather record exists. By improving the determination method of the low-temperature daily index, the low-temperature process effectiveness index can be better determined, and missing report of the osmanthus fragrans flowering period is avoided; the long-term weather forecasting model established by simulating and constructing the flowering beginning of the osmanthus overcomes the defect of less actual measurement data in the flowering stage of the osmanthus, increases the number of samples for establishing the forecasting model, and improves the accuracy of long-term weather forecasting in the flowering beginning of the osmanthus.

Description

Technical Field

[0001] The invention relates to the technical field of plant cultivation, and in particular to a method for making a long-term weather forecast for the first batch flowering period of osmanthus fragrans. Background Art

[0002] Osmanthus is one of the top ten famous flowers. The preparation work for the Osmanthus Festival in some places requires the technical support of producing long-term weather forecasts for the beginning of the first batch of osmanthus flowering.

[0003] In 2007, Wu Xuanke et al. used the data of Liuzhou City from 1995 to 2005 and adopted the stepwise regression method to establish a prediction model for the peak flowering period of osmanthus using the main meteorological factors affecting the peak flowering period of osmanthus; in 2013, Jue Xinxin et al. compared and analyzed the flowering period and meteorological data of osmanthus in Guilin from 1999 to 2011, and established a prediction model for the flowering period of osmanthus using the stepwise regression method; in 2014, Wang Cunzhen et al. used the observation data of the beginning flowering period of osmanthus in Guilin from 1999 to 2013, investigated and analyzed the meteorological reasons, selected meteorological factors, and used statistical analysis methods to produce long-term and short-term forecast models for flowering period forecast; in 2022, Bai Yunxia et al. used the phenological observation data of osmanthus in Lishui City from 2000 to 2019 and meteorological data such as sunshine, precipitation, relative humidity, temperature, and active accumulated temperature in the same period, and studied the influence of various meteorological factors on the peak flowering period of osmanthus in Lishui through the stepwise regression method, and established a long-term prediction model for the peak flowering period of osmanthus. In 2023, Wang Yuxiang et al. studied the low temperature forecast indicators before the flowering of Osmanthus fragrans in Nanyang from 2016 to 2022. The stepwise approximation method was used to determine that the low temperature daily indicator before flowering was the maximum daily temperature <22.3℃, the effective low temperature process indicator was more than one low temperature day, and Osmanthus fragrans bloomed 4-9 days after the start of the effective low temperature process. The starting date of Osmanthus fragrans flowering and the starting date of the effective low temperature process had an extremely significant univariate linear correlation; in 2024, Zhu Shouyan et al. studied the low temperature forecast indicators before the flowering of Osmanthus fragrans in Taizhou from 2006 to 2023. The low temperature daily indicator before flowering was the maximum daily temperature <24.4℃, the effective low temperature process was <24.4℃ for two consecutive days or the maximum daily temperature was <24.4℃ for one day, and the next day was ≤27.3℃, the Osmanthus fragrans bloomed 6-9 days after the start of the effective low temperature process, and the starting date of Osmanthus fragrans flowering had an extremely significant univariate linear correlation.

[0004] The low temperature daily index determined by the stepwise approximation method is low, and sometimes the effectiveness of the low temperature process is misjudged in the forecast year, resulting in the omission of the osmanthus flowering period. Sometimes the effectiveness index of the low temperature process cannot be determined in the existing data. The effective low temperature process is used to predict the start of osmanthus flowering, which can only produce a short-term forecast. The long-term forecast is made by establishing a statistical regression forecast model based on the measured start of osmanthus flowering and the previous meteorological factors. Because the measured data of the start of osmanthus flowering are short and the sample size is small, the extrapolated forecast effect is poor and the error is large.

[0005] The purpose of the present invention is to overcome the above-mentioned technical background difficulties. According to the one-to-one correspondence between the flowering of osmanthus and the effective low-temperature process, the starting date of the flowering of osmanthus and the starting date of the effective low-temperature process are extremely significantly correlated with each other in a univariate linear manner. The method for determining the low-temperature day index and the effective low-temperature process index is improved, the first effective low-temperature process starting date sequence is statistically analyzed, and a long-term sequence of the starting date of the first batch of flowering of osmanthus is simulated and constructed. On this basis, a forecast of the starting date of the first batch of flowering of osmanthus is prepared. Therefore, a method for preparing a long-term meteorological forecast for the first batch of flowering of osmanthus is proposed. Summary of the Invention

[0006] The present invention solves the problems existing in the above-mentioned prior art through the following technical solutions, and the present invention comprises the following steps:

[0007] S1: Collect data on the onset of flowering of Osmanthus fragrans;

[0008] S2: Use the improved stepwise approach method to determine the effectiveness index of the low temperature process before the flowering of Osmanthus fragrans;

[0009] S3: Establish a linear regression equation between the flowering start date and the effective low temperature process start date;

[0010] S4: Count the starting date sequence of the first effective low temperature process in each year since meteorological records began;

[0011] S5: Simulate and construct the first batch flowering start sequence of Osmanthus fragrans over the years;

[0012] S6: Establish a long-term meteorological forecast statistical model for the first batch of osmanthus flowering in previous years;

[0013] S7: Prepare long-term weather forecasts to predict the start of the first batch of sweet osmanthus flowers.

[0014] Furthermore, the data of the flowering start date of Osmanthus fragrans is collected for n years, in which two batches bloom in the i-th year and one batch blooms in the other n-1 years. The flowering start dates of each batch are T1, T2, ..., T i1 、T i2 ,…,T n ;

[0015] First, use the stepwise approach method to determine the first candidate value of the low temperature day index. Count the minimum values ​​in the highest temperature series before flowering for each batch, then select the maximum value and add 0.1℃:

[0016] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the minimum value of the maximum temperature in the m days before the first batch of flowering in the first year is Z 11 =small(t1, t2, ..., t m , 1);

[0017] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the minimum value of the highest temperature in p days before the first batch of flowering in the second year is Z 12 =small(t1, t2, ..., t p , 1);

[0018] And so on;

[0019] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the minimum value of the maximum temperature in the q days before the first batch of flowering in year i is Z 1i1 =small(t1, t2, ..., t q , 1);

[0020] And so on;

[0021] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the minimum value of the highest temperature in the r days before the second batch of flowering in year i is Z 1i2 =small(t q+1 , t q+2 ,…,t r , 1);

[0022] And so on;

[0023] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the minimum value of the highest temperature in g days before the first batch of flowering in year n is Z 1n =small(t1, t2, ..., t g , 1);

[0024] Then the first candidate value of the low temperature day index is Z1=max(Z 11 、Z 12 ,…,Z 1i1 、Z 1i2 ,…,Z 1n )+0.1,

[0025] Furthermore, it is to judge whether the first candidate value of the low temperature day index determined by the stepwise approach method can be used to determine the low temperature process effectiveness index after statistically analyzing the low temperature process:

[0026] According to the fact that the daily maximum temperature is lower than the first value Z1 of the candidate value of the low temperature day index, all low temperature processes before the flowering of each batch of osmanthus are counted; the low temperature process with the largest correlation coefficient with the starting period of osmanthus flowering and passing the significance correlation test is screened out as the effective low temperature process by the full permutation method; the low temperature process before the effective low temperature process is the invalid low temperature process; if the minimum number of low temperature days of all effective low temperature processes is greater than the maximum number of low temperature days of all invalid low temperature processes, or the minimum number of low temperature days of all effective low temperature processes is equal to the maximum number of low temperature days of all invalid low temperature processes, and the maximum value of the highest temperature of the day after the effective low temperature process with the same number of low temperature days is less than the minimum value of the highest temperature of the day after the invalid low temperature process with the same number of low temperature days, then the difference between the number of low temperature process days and the highest temperature of the day after the low temperature process between the effective low temperature process and the invalid low temperature process is determined as the low temperature process effectiveness index, and the candidate value Z1 of the low temperature day index is determined as the low temperature day index; otherwise, the low temperature process effectiveness index cannot be determined;

[0027] Furthermore, the stepwise approach method is improved, and there are three methods to determine other candidate values ​​of the low temperature day index:

[0028] The first method is the adjacent value method:

[0029] Find the year and date of the first candidate value of the low temperature day index, select the minimum value of the highest temperature of the previous day and the next day plus 0.1℃ as the second candidate value, and the maximum value plus 0.1℃ as the third candidate value: that is, the first candidate value of the low temperature day index is Z1, which occurs on the jth day of the i-th year, and the highest temperature on the day before the jth day is t j-1 The maximum temperature on the next day is t j+1 , then the second candidate value Z of the low temperature day index of the adjacent value method is x2 =min(t j-1 , t j+1 )+0.1, the third candidate value Z of the low temperature day index of the adjacent value method x3 =max(t j-1 , t j+1 )+0.1.

[0030] The second method is the step-by-step approach:

[0031] The second smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected therefrom to determine the second candidate value. The third smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected therefrom to determine the third candidate value. And so on, the sth smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected therefrom to determine the sth candidate value. The specific steps are:

[0032] Count the second smallest maximum temperature value in the sequence of the day before flowering for each batch, and select the maximum value plus 0.1℃ to determine the second candidate value:

[0033] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the second minimum value of the maximum temperature in the first batch of flowering in the first year m days before the start of flowering is Z s21 =small(t1, t2, ..., t m , 2);

[0034] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the second minimum value of the maximum temperature in the p days before the first batch of flowering in the second year is Z s22 =small(t1, t2, ..., t p , 2);

[0035] And so on;

[0036] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the second minimum value of the maximum temperature in the q days before the first batch of flowering in the i-th year is Z s2i1 =small(t1, t2, ..., t q , 2);

[0037] And so on;

[0038] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the second minimum value of the maximum temperature in the r days before the second batch of flowering in the i-th year is Z s2i2 =small(t q+1 , t q+2 ,…,t r , 2);

[0039] And so on;

[0040] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the second minimum value of the maximum temperature in the g days before the first batch of flowering in the nth year is Z s2n =small(t1, t2, ..., t g , 2);

[0041] Then the second candidate value of the low temperature day index of the step-by-step upward approach method is Z s2 =max(Z s21 、Z s22 ,…,Z s2i1 、Z s2i2,…,Z s2n )+0.1.

[0042] Count the third smallest value in the maximum temperature sequence of each batch before flowering, and select the maximum value plus 0.1℃ to determine the third candidate value:

[0043] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the third minimum value of the maximum temperature in the m days before the first batch of flowering in the first year is Z s31 =small(t1, t2, ..., t m ,3);

[0044] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the third minimum value of the highest temperature on the p days before the first batch of flowering in the second year is Z s32 =small(t1, t2, ..., t p ,3);

[0045] And so on;

[0046] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the third minimum value of the maximum temperature in the q days before the first batch of flowering in the i-th year is Z s3i1 =small(t1, t2, ..., t q ,3);

[0047] And so on;

[0048] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the third minimum value of the highest temperature in the r days before the second batch of flowering in year i is Z s3i2 =small(t q+1 , t q+2 ,…,t r ,3);

[0049] And so on;

[0050] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the third minimum value of the highest temperature in the g days before the first batch of flowering in the nth year is Z s3n =small(t1, t2, ..., t g ,3);

[0051] Then the third candidate value of the low temperature day index of the step-by-step upward approach method is Z s3 =max(Z s31 , Z s32 ,…,Z s3i1 , Z s3i2 ,…,Z s3n )+0.1.

[0052] And so on;

[0053] Count the sth smallest value in the maximum temperature sequence before flowering for each batch, and select the maximum value plus 0.1℃ to determine the sth candidate value:

[0054] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the sth minimum value of the highest temperature in the m days before the first batch of flowering in the first year is Z ss1 =small(t1, t2, ..., t m , s);

[0055] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the sth minimum value of the highest temperature on p days before the first batch of flowering in the second year is Z ss2 =small(t1, t2, ..., t p , s);

[0056] And so on;

[0057] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the sth minimum value of the maximum temperature in the q days before the first batch of flowering in the i-th year is Z ssi1 =small(t1, t2, ..., t q , s);

[0058] And so on;

[0059] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the sth minimum value of the highest temperature in the r days before the second batch of flowering in the i-th year is Z ssi2 =small(t q+1 , t q+2 ,…,t r , s);

[0060] And so on;

[0061] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the sth minimum value of the highest temperature in the g days before the first batch of flowering in the nth year is Z ssn =small(t1, t2, ..., t g , s);

[0062] Then the candidate value of the low temperature day index of the step-by-step upward approach method is Z ss =max(Z ss1 、Z ss2 ,…,Z ssi1 、Z ssi2 ,…,Z ssn )+0.1.

[0063] The third method is to increase the number of low-temperature days in the effective low-temperature process:

[0064] The first candidate value of the low temperature day index determined by the stepwise approach method is selected from the effective low temperature processes with 1 low temperature day selected by the full permutation method by counting the minimum value of the highest temperature of the previous day and the maximum temperature of the next day, and the maximum value is selected from them as the second candidate value. The second candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature processes is selected from the effective low temperature processes with 2 low temperature days selected by the full permutation method by counting the minimum value of the highest temperature of the previous day and the maximum temperature of the next day, and the maximum value is selected from them as the third candidate value. And so on. The Xth candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature processes is selected from the effective low temperature processes with X-1 low temperature days selected by the full permutation method by counting the minimum value of the highest temperature of the previous day and the maximum temperature of the next day, and the maximum value is selected from them as the Xth candidate value.

[0065] The first candidate value of the low temperature day index determined by the stepwise approach method is the second candidate value determined by selecting the maximum value from the minimum values ​​of the maximum temperature of the previous day and the maximum temperature of the next day in the effective low temperature process with one low temperature day selected by the full permutation method:

[0066] Assume that the first candidate value of the low temperature day index determined by the stepwise approach method is h samples of the effective low temperature process with 1 low temperature day selected by the full permutation method;

[0067] The number of low temperature days in the effective low temperature process before the flowering of the first sample is 1 day, which is the wth day of the first sample year. The highest temperature on the day before the wth day is t w-1 The maximum temperature of the next day t w+1 , then their minimum value Z y21 =min(t w-1 , t w+1 );

[0068] And so on;

[0069] The number of low temperature days in the effective low temperature process before the flowering of the i-th sample is 1 day, which is the u-th day of the i-th sample year. The highest temperature on the day before the u-th day is t u-1 The maximum temperature of the next day t u+1 , then their minimum value Z y2i =min(t u-1 , t u+1 );

[0070] And so on;

[0071] The number of low temperature days in the effective low temperature process before the flowering of the hth sample is 1 day, which is the vth day of the hth sample year. The highest temperature on the day before the vth day is t v-1 The maximum temperature of the next day t v+1 , then their minimum value Z y2h =min(t v-1 , t v+1 );

[0072] Add the second candidate value Z of the low temperature day in the low temperature day number method of the effective low temperature process y2 =max(Z y21 ,…,Z y2i ,…,Z y2h )+0.1;

[0073] The second candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature process is the third candidate value. The minimum value of the maximum temperature of the previous day and the maximum temperature of the next day is selected from the statistics of all the effective low temperature processes with 2 low temperature days selected by the full permutation method.

[0074] Assume that there are f samples with 2 days of low temperature in the effective low temperature process selected by the full permutation method of the second candidate value of the low temperature day index determined by the method of increasing the number of low temperature days of the effective low temperature process;

[0075] The number of low temperature days in the effective low temperature process before the flowering of the first sample is 2 days. This process is on the wth day and w+1th day of the first sample year. The highest temperature on the day before this process is t w-1 The maximum temperature of the next day t w+2 , then their minimum value Z y31 =min(t w-1 , t w+2 );

[0076] And so on;

[0077] The number of low temperature days in the effective low temperature process before the flowering of the i-th sample is 2 days. The process is on the u-th day and u+1-th day of the i-th sample year. The highest temperature on the day before the process is t u-1 The maximum temperature of the next day t u+2, then their minimum value Z y3i =min(t u-1 , t u+2 );

[0078] And so on;

[0079] The number of low temperature days in the effective low temperature process before the flowering of the f-th sample is 2 days. This process is on the v-th day and the v+1-th day of the f-th sample year. The highest temperature on the day before this process is t v-1 The maximum temperature of the next day t v+2 , then their minimum value Z y3f =min(t v-1 , t v+2 );

[0080] Add the third candidate value Z of the low temperature day in the low temperature day number method of the effective low temperature process y3 =max(Z y31 ,…,Z y3i ,…,Z y3f )+0.1;

[0081] And so on;

[0082] According to the method of increasing the number of low-temperature days in effective low-temperature processes, the X-1 candidate value of the low-temperature day index is determined. Among all the effective low-temperature processes with X-1 low-temperature days selected by the full permutation method, the minimum value of the maximum temperature of the previous day and the maximum temperature of the next day is counted and the maximum value is selected as the X-1 candidate value:

[0083] Assume that there are L samples with X-1 days of effective low temperature process low temperature days selected by the full permutation method of the low temperature day index X-1th candidate value determined by the method of increasing the number of low temperature days of effective low temperature process;

[0084] The number of low temperature days in the effective low temperature process before the flowering of the first sample is X-1 days. The process is on the wth and w+1th days of the first sample year. The highest temperature on the day before the process is t w-1 The maximum temperature of the next day t w+2 , then their minimum value Z y31 =min(t w-1 , t w+2 );

[0085] And so on;

[0086] The number of low temperature days in the effective low temperature process before the flowering of the i-th sample is X-1 days. The process is on the u-th day and u+1-th day of the i-th sample year. The highest temperature on the day before the process is t u-1 The maximum temperature of the next day t u+2 , then their minimum value Z y3i =min(t u-1 , t u+2);

[0087] And so on;

[0088] The number of low temperature days in the effective low temperature process before the flowering of the Lth sample is X-1 days. The process is on the vth day and v+1th day of the Lth sample year. The highest temperature on the day before the process is t v-1 The maximum temperature of the next day t v+2 , then their minimum value Z y3L =min(t v-1 , t v+2 );

[0089] Add the effective low temperature process low temperature day number method low temperature day X candidate value Z yX =max(Z y31 ,…,Z y3i ,…,Z y3L )+0.1.

[0090] Furthermore, according to the method of the maximum temperature of the day being lower than the adjacent value, the second candidate value Z of the low temperature day index is x2 , the third candidate value Z x3 Instead of Z1, repeat the above process of determining whether the low temperature process effectiveness index before the sweet osmanthus blooms can be determined, and check which low temperature day index candidate values ​​can determine the low temperature process effectiveness index.

[0091] According to the candidate values ​​of low temperature day index of the method of gradually approaching the maximum temperature of the day, they are Z s2 , Z s3 ,…,Z ss Replace Z1 and repeat the above process of determining whether the effectiveness index of the low temperature process before the flowering of osmanthus can be determined until Z ss-1 Ability to determine the low temperature process effectiveness index, Z ss Low temperature process effectiveness indicators cannot be determined.

[0092] According to the method of increasing the number of low-temperature days of effective low-temperature processes according to the daily maximum temperature below the second candidate value Z of low-temperature days y2 , Z y3 ,…,Z yX Repeat the above process of determining whether the effectiveness index of the low temperature process before the flowering of osmanthus can be determined for Z1 until Z yX-1 Ability to determine the low temperature process effectiveness index, Z yX Low temperature process effectiveness indicators cannot be determined.

[0093] Among all the low-temperature process effectiveness indices that can be determined based on the candidate values ​​of the low-temperature day index, the one with the highest low-temperature day index value is selected as the final effective low-temperature process index.

[0094] Furthermore, after the effectiveness index of the low temperature process before the flowering of osmanthus is determined, a linear regression equation is established between the measured flowering start date and the starting date of the effective low temperature process.

[0095] Furthermore, after the effectiveness index of the low temperature process before the flowering of osmanthus is determined, the sequence of the starting date of the first effective low temperature process in all years since meteorological records were kept is counted, that is, the starting date of the low temperature process that reaches the effective low temperature process index for the first time in all years is counted according to the effectiveness index of the effective low temperature process.

[0096] Furthermore, a long-term series of the first batch of flowering starts of osmanthus fragrans in previous years was simulated and constructed, and the starting date of the first effective low-temperature process in previous years was substituted into the univariate linear regression equation of the measured flowering start date and the starting date of the effective low-temperature process. The calculation result was the simulated value of the first batch of flowering start date series of osmanthus fragrans in previous years.

[0097] Furthermore, a long-term meteorological forecast statistical model for the first batch of osmanthus flowering is established, and relevant and significant forecast factors are selected from the meteorological factors more than one month before the earliest date in the series simulation values ​​of the first batch of osmanthus flowering, and a regression model is established with the simulation values ​​of the first batch of osmanthus flowering.

[0098] Furthermore, the meteorological forecast factors of the forecast year are substituted into the long-term meteorological forecast statistical model for the first batch of osmanthus flowering, and the flowering start date of osmanthus is forecasted based on the calculated results.

[0099] Compared with the prior art, the present invention has the following advantages: the method for preparing a long-term meteorological forecast for the first flowering period of osmanthus fragrans determines an effective low-temperature process index by analyzing the correlation between the flowering start date of osmanthus fragrans and the low-temperature process start date statistically calculated based on different low-temperature day index candidate values, and determines the upper limit of the low-temperature day index as much as possible, so that the effective low-temperature process index can better judge the effectiveness of the low-temperature process; at the same time, combined with the method for preparing a long-term meteorological forecast for the first flowering period of osmanthus fragrans, statistics are made on the first effective low-temperature process start date of each year since meteorological records were kept, a long-term series of the first flowering period of osmanthus fragrans in previous years is simulated and constructed, and a forecast for the first flowering period of osmanthus fragrans fragrans is prepared on the basis of the long-term forecast, thereby increasing the number of samples for establishing the long-term meteorological forecast for the first flowering period of osmanthus fragrans fragrans, and improving the accuracy of the forecast for the first flowering period of osmanthus fragrans fragrans. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 This is a flow chart of the long-term weather forecast method for the first batch of flowering of sweet osmanthus of the present invention;

[0101] Figure 2 This is a flow chart of the improved stepwise approach method of the present invention for determining the low-temperature day index and the low-temperature process effectiveness index;

[0102] Figure 3This is a schematic diagram of the minimum maximum temperature on the day before flowering and the first candidate value of the low temperature day index for each batch of Nanyang Osmanthus fragrans from 2016 to 2023 of the present invention;

[0103] Figure 4 This is a schematic diagram of the one-variable linear correlation between the flowering start date of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 and the first candidate value of the low temperature day index 22.3°C effective low temperature process start date;

[0104] Figure 5 This is a comparison chart of the number of low-temperature days in the effective low-temperature process and the invalid low-temperature process of the third candidate value of the low-temperature day index of 28.2°C in the adjacent value method of Nanyang Dangui from 2016 to 2023 of the present invention;

[0105] Figure 6 This is a schematic diagram of the second minimum value of the maximum temperature on the day before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 and the second candidate value of the low temperature day index by the gradual upward approach method;

[0106] Figure 7 This is a schematic diagram of the one-variable linear correlation between the flowering start date of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 and the second candidate value of the low temperature day index of the stepwise upward approach method, 23.6°C, the effective low temperature process start date;

[0107] Figure 8 This is a schematic diagram of the third minimum value of the maximum temperature on the day before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 and the third candidate value of the low temperature day index by the gradual upward approach method;

[0108] Figure 9 This is a schematic diagram comparing the second candidate value of the low-temperature day index of 25.0°C effective low-temperature process and invalid low-temperature process by the method of increasing the number of low-temperature days of effective low-temperature process in Nanyang Dangui from 2016 to 2023 of the present invention;

[0109] Figure 10 This is a schematic diagram of the flowering start dates and daily maximum temperatures of the first two batches of Nanyang Osmanthus fragrans in 2024 of the present invention;

[0110] Figure 11 This is a schematic diagram of the sequence of the measured flowering start dates of the first batch of Nanyang Osmanthus fragrans from 2016 to 2024, constructed by simulating the flowering start dates of the present invention;

[0111] Figure 12 This is a schematic diagram comparing the simulation construction sequence of the first batch of flowering onset of Nanyang Osmanthus fragrans over the years and the back-substitution fitting value sequence of the long-term weather forecast model;

[0112] Figure 13 Schematic diagram of the minimum maximum temperature on the day before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans of the present invention and the first candidate value of the low temperature day index of 26.0°C;

[0113] Figure 14 This is a schematic diagram of the number of days during the effective low-temperature process of the first candidate value of the method for gradually approaching the low-temperature day index before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0114] Figure 15 2. This is a schematic diagram showing a comparison of the effective and invalid low-temperature process days for the second candidate value of 26.9°C using the adjacent value method before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0115] Figure 16 2. This is a schematic diagram showing a comparison of the effective and invalid low-temperature process days for the third candidate value of 27.4°C using the adjacent value method before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0116] Figure 17 2. This is a schematic diagram comparing the maximum temperature of the first day after the effective and invalid low temperature processes of the third candidate value of 27.4°C for the adjacent value method before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0117] Figure 18 This is a schematic diagram of the one-variable linear correlation between the flowering start date of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 and the third candidate value of the adjacent value method low temperature day index 27.4°C effective low temperature process start date;

[0118] Figure 19 It is a schematic diagram of the second minimum value of the maximum temperature on the day before flowering of each batch of Pucheng Zhusha Dangui of the present invention and the second candidate value of 26.9°C when the low temperature day index gradually approaches the method;

[0119] Figure 20 It is a schematic diagram of the third minimum value of the highest temperature on the day before flowering of each batch of Pucheng Zhusha Dangui of the present invention and the third candidate value of 27.1°C of the method of gradually approaching the low temperature day index;

[0120] Figure 21 This is a comparison of the number of days of effective and invalid low temperature processes for the third candidate value of 27.1°C using the method of gradually approaching upwards before flowering for each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0121] Figure 22 Schematic diagram of the fourth minimum value of the maximum temperature on the day before flowering of each batch of Pucheng Zhusha Dangui of the present invention and the fourth candidate value of 27.4°C in the method of gradually approaching the low temperature day index upward;

[0122] Figure 23 Schematic diagram of the third minimum value of the highest temperature on the day before flowering of each batch of Pucheng Zhusha Dangui of the present invention and the fifth candidate value of 28.0°C when the low temperature day index gradually approaches upward;

[0123] Figure 242. This is a schematic diagram comparing the effective and ineffective low-temperature process days of the fifth candidate value of 28.0°C for the method of gradually approaching upwards before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0124] Figure 25 This is a schematic diagram comparing the effective and invalid low temperature process days of 27.7°C, the second candidate value of the method for increasing the effective low temperature process low temperature days before flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 of the present invention;

[0125] Figure 26 This is a schematic diagram comparing the maximum temperatures of the day after the effective and invalid low-temperature processes of Pucheng Zhusha Dangui from 2010 to 2023, using the second candidate value of 27.7°C, which is a method for increasing the number of low-temperature days during the effective low-temperature process.

[0126] Figure 27 This is a schematic diagram of the unary linear correlation between the flowering start date of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 and the second candidate value of the low temperature day index of the method of increasing the number of low temperature days in the effective low temperature process, 27.7°C, the effective low temperature process start date;

[0127] Figure 28 This is a schematic diagram of the flowering start dates and daily maximum temperatures of the first two batches of Pucheng Cinnabar Osmanthus fragrans in 2024 of the present invention;

[0128] Figure 29 This is a schematic diagram of the simulated construction sequence of the first batch of flowering onset of Pucheng Cinnabar Osmanthus fragrans over the years and the measured flowering onset sequence from 2010 to 2024;

[0129] Figure 30 It is a schematic diagram comparing the simulation construction sequence of the first batch of flowering onset of Pucheng Cinnabar Osmanthus fragrans over the years and the back-substitution fitting value sequence of the long-term meteorological forecast model. DETAILED DESCRIPTION

[0130] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process. However, the protection scope of the present invention is not limited to the following embodiment.

[0131] like Figures 1 to 30 As shown, this embodiment provides a technical solution: a method for making a long-term weather forecast for the first batch of osmanthus flowering period, comprising the following steps:

[0132] S1: Collect data on the onset of flowering of Osmanthus fragrans;

[0133] S2: Use the improved stepwise approach method to determine the effectiveness index of the low temperature process before the flowering of Osmanthus fragrans;

[0134] S3: Establish a linear regression equation between the flowering start date and the effective low temperature process start date;

[0135] S4: Count the starting date sequence of the first effective low temperature process in each year since meteorological records began;

[0136] S5: Simulate and construct the first batch flowering start sequence of Osmanthus fragrans over the years;

[0137] S6: Establish a long-term meteorological forecast statistical model for the first batch of osmanthus flowering in previous years;

[0138] S7: Prepare long-term weather forecasts to predict the start of the first batch of sweet osmanthus flowers.

[0139] Furthermore, the data of the flowering start date of Osmanthus fragrans is collected for n years, in which two batches bloom in the i-th year and one batch blooms in the other n-1 years. The flowering start dates of each batch are T1, T2, ..., T i1 、T i2 ,…,T n ;

[0140] First, use the stepwise approach method to determine the first candidate value of the low temperature day index. Count the minimum values ​​in the highest temperature series before flowering for each batch, then select the maximum value and add 0.1℃:

[0141] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the minimum value of the maximum temperature in the m days before the first batch of flowering in the first year is Z 11 =small(t1, t2, ..., t m , 1);

[0142] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the minimum value of the highest temperature in p days before the first batch of flowering in the first year is Z 12 =small(t1, t2, ..., t p , 1);

[0143] And so on;

[0144] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the minimum value of the maximum temperature in the q days before the first batch of flowering in year i is Z 1i1 =small(t1, t2, ..., t q , 1);

[0145] And so on;

[0146] The second batch of flowering starts at T in year i2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the minimum value of the highest temperature in the r days before the second batch of flowering in year i is Z 1i2 =small(t q+1 , t q+2 ,…,t r , 1);

[0147] And so on;

[0148] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the minimum value of the highest temperature in g days before the first batch of flowering in year n is Z 1n =small(t1, t2, ..., t g , 1);

[0149] Then the first candidate value of the low temperature day index is Z1=max(Z 11 , Z 12 ,…,Z 1i1 , Z 1i2 ,…,Z 1n )+0.1,

[0150] Determine whether the first candidate value of the low temperature daily index determined by the stepwise approach method can be used to determine the low temperature process effectiveness index after statistically analyzing the low temperature process:

[0151] According to the fact that the daily maximum temperature is lower than the first value Z1 of the candidate value of the low temperature day index, all low temperature processes before the flowering of each batch of osmanthus are counted; the low temperature process with the largest correlation coefficient with the starting period of osmanthus flowering and passing the significance correlation test is screened out as the effective low temperature process by the full permutation method; the low temperature process before the effective low temperature process is the invalid low temperature process; if the minimum number of low temperature days of all effective low temperature processes is greater than the maximum number of low temperature days of all invalid low temperature processes, or the minimum number of low temperature days of all effective low temperature processes is equal to the maximum number of low temperature days of all invalid low temperature processes, and the maximum value of the highest temperature of the day after the effective low temperature process with the same number of low temperature days is less than the minimum value of the highest temperature of the day after the invalid low temperature process with the same number of low temperature days, then the difference between the number of low temperature process days and the highest temperature of the day after the low temperature process between the effective low temperature process and the invalid low temperature process is determined as the low temperature process effectiveness index, and the candidate value Z1 of the low temperature day index is determined as the low temperature day index; otherwise, the low temperature process effectiveness index cannot be determined;

[0152] There are three ways to improve the stepwise approach method to determine other candidate values ​​of the low temperature day index:

[0153] The first method is the adjacent value method:

[0154] Find the year and date of the first candidate value of the low temperature day index, select the minimum value of the highest temperature of the previous day and the next day plus 0.1℃ as the second candidate value, and the maximum value plus 0.1℃ as the third candidate value: that is, the first candidate value of the low temperature day index is Z1, which occurs on the jth day of the i-th year, and the highest temperature on the day before the jth day is t j-1 The maximum temperature on the next day is t j+1 , then the second candidate value Z of the low temperature day index of the adjacent value method is x2 =min(t j-1 , t j+1 )+0.1, the third candidate value Z of the low temperature day index of the adjacent value method x3 =max(t j-1 , t j+1 )+0.1.

[0155] The second method is the step-by-step upward approach:

[0156] The second smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected therefrom to determine the second candidate value. The third smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected therefrom to determine the third candidate value. And so on, the sth smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected therefrom to determine the sth candidate value. The specific steps are:

[0157] Count the second smallest value in the maximum temperature sequence of each batch before flowering, and select the maximum value plus 0.1℃ to determine the second candidate value:

[0158] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the second minimum value of the maximum temperature in the first batch of the first year before flowering is Z s21 =small(t1, t2, ..., t m , 2);

[0159] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the second minimum value of the maximum temperature in the p days before the first batch of flowering in the second year is Z s22 =small(t1, t2, ..., t p , 2);

[0160] And so on;

[0161] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the second minimum value of the maximum temperature in the q days before the first batch of flowering in the i-th year is Z s2i1=small(t1, t2, ..., t q , 2);

[0162] And so on;

[0163] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the second minimum value of the maximum temperature in the r days before the second batch of flowering in the i-th year is Z s2i2 =small(t q+1 , t q+2 ,…,t r , 2);

[0164] And so on;

[0165] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the second minimum value of the maximum temperature in the g days before the first batch of flowering in the nth year is Z s2n =small(t1, t2, ..., t g , 2);

[0166] Then the second candidate value of the low temperature day index of the step-by-step upward approach method is Z s2 =max(Z s21 、Z s22 ,…,Z s2i1 、Z s2i2 ,…,Z s2n )+0.1.

[0167] Count the third smallest value in the maximum temperature sequence of each batch before flowering, and select the maximum value plus 0.1℃ to determine the third candidate value:

[0168] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the third minimum value of the maximum temperature in the m days before the first batch of flowering in the first year is Z s31 =small(t1, t2, ..., t m ,3);

[0169] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the third minimum value of the highest temperature on the p days before the first batch of flowering in the second year is Z s32 =small(t1, t2, ..., t p ,3);

[0170] And so on;

[0171] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the third minimum value of the maximum temperature in the q days before the first batch of flowering in the i-th year is Z s3i1 =small(t1, t2, ..., t q ,3);

[0172] And so on;

[0173] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the third minimum value of the highest temperature in the r days before the second batch of flowering in year i is Z s3i2 =small(t q+1 , t q+2 ,…,t r ,3);

[0174] And so on;

[0175] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the third minimum value of the highest temperature in the g days before the first batch of flowering in the nth year is Z s3n =small(t1, t2, ..., t g ,3);

[0176] Then the third candidate value of the low temperature day index of the step-by-step upward approach method is Z s3 =max(Z s31 , Z s32 ,…,Z s3i1 , Z s3i2 ,…,Z s3n )+0.1.

[0177] And so on;

[0178] Count the sth smallest value in the maximum temperature sequence before flowering of each batch, select the maximum value and add 0.1℃ to determine it as the sth candidate value:

[0179] The highest temperatures on the day before the first batch of flowering in the first year T1 are t1, t2, ..., t m , then the sth minimum value of the highest temperature in the m days before the first batch of flowering in the first year is Z ss1 =small(t1, t2, ..., t m , s);

[0180] The maximum temperature on the day before the first batch of flowering T2 in the second year is t1, t2, ..., t p , then the sth minimum value of the highest temperature on p days before the first batch of flowering in the second year is Z ss2 =small(t1, t2, ..., t p , s);

[0181] And so on;

[0182] The first flowering period of the first batch in year i is T 1i The highest temperatures on the day before yesterday were t1, t2, ..., t q , then the sth minimum value of the maximum temperature in the q days before the first batch of flowering in the i-th year is Z ssi1 =small(t1, t2, ..., t q , s);

[0183] And so on;

[0184] The second batch of flowering starts at T in year i 2i The highest temperature the day before was t q+1 , t q+2 ,…,t r , then the sth minimum value of the highest temperature in the r days before the second batch of flowering in the i-th year is Z ssi2 =small(t q+1 , t q+2 ,…,t r , s);

[0185] And so on;

[0186] The first flowering period of the nth year is T n The highest temperatures on the day before yesterday were t1, t2, ..., t g , then the sth minimum value of the highest temperature in the g days before the first batch of flowering in the nth year is Z ssn =small(t1, t2, ..., t g , s);

[0187] Then the candidate value of the low temperature day index of the step-by-step upward approach method is Z ss =max(Z ss1 , Z ss2 ,…,Z ssi1 , Z ssi2 ,…,Z ssn )+0.1.

[0188] The third method is to increase the number of low-temperature days in the effective low-temperature process:

[0189] According to the first candidate value of the low temperature day index determined by the stepwise approach method, in the effective low temperature process with 1 low temperature day screened out by the full arrangement method, the minimum value of the highest temperature of the previous day and the maximum temperature of the next day is counted, and the maximum value is selected from them as the second candidate value. According to the second candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature process, in the effective low temperature process with 2 low temperature days screened out by the full arrangement method, the minimum value of the highest temperature of the previous day and the maximum temperature of the next day is counted, and the maximum value is selected from them as the third candidate value. And so on, according to the xth candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature process, in the effective low temperature process with x-1 low temperature days screened out by the full arrangement method, the minimum value of the highest temperature of the previous day and the maximum temperature of the next day is selected from them as the xth candidate value.

[0190] The first candidate value of the low temperature day index determined by the stepwise approach method is the second candidate value determined by selecting the maximum value from the minimum values ​​of the maximum temperature of the previous day and the maximum temperature of the next day in the effective low temperature process with one low temperature day selected by the full permutation method:

[0191] Assume that the first candidate value of the low temperature day index determined by the stepwise approach method is h samples of the effective low temperature process with 1 low temperature day selected by the full permutation method;

[0192] The number of low temperature days in the effective low temperature process before the flowering of the first sample is 1 day, which is the wth day of the first sample year. The highest temperature on the day before the wth day is t w-1 The maximum temperature of the next day t w+1 , then their minimum value Z y21 =min(t w-1 , t w+1 );

[0193] And so on;

[0194] The number of low temperature days in the effective low temperature process before the flowering of the i-th sample is 1 day, which is the u-th day of the i-th sample year. The highest temperature on the day before the u-th day is t u-1 The maximum temperature of the next day t u+1 , then their minimum value Z y2i =min(t u-1 , t u+1 );

[0195] And so on;

[0196] The number of low temperature days in the effective low temperature process before the flowering of the hth sample is 1 day, which is the vth day of the hth sample year. The highest temperature on the day before the vth day is t v-1 The maximum temperature of the next day t v+1 , then their minimum value Z y2h =min(t v-1 , tv+1 );

[0197] Add the second candidate value Z of the low temperature day in the low temperature day number method of the effective low temperature process y2 =max(Z y21 ,…,Z y2i ,…,Z y2h )+0.1;

[0198] The second candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature process is the third candidate value. The minimum value of the maximum temperature of the previous day and the maximum temperature of the next day is selected from the statistics of all the effective low temperature processes with 2 low temperature days selected by the full permutation method.

[0199] Assume that there are f samples with 2 days of low temperature in the effective low temperature process, which are screened by the full permutation method of the second candidate value of the low temperature day index determined by the method of increasing the number of low temperature days in the effective low temperature process.

[0200] The number of low temperature days in the effective low temperature process before the flowering of the first sample is 2 days. This process is on the wth day and w+1th day of the first sample year. The highest temperature on the day before this process is t w-1 The maximum temperature of the next day t w+2 , then their minimum value Z y31 =min(t w-1 , t w+2 );

[0201] And so on;

[0202] The number of low temperature days in the effective low temperature process before the flowering of the i-th sample is 2 days. The process is on the u-th day and u+1-th day of the i-th sample year. The highest temperature on the day before the process is t u-1 The maximum temperature of the next day t u+2 , then their minimum value Z y3i =min(t u-1 , t u+2 );

[0203] And so on;

[0204] The number of low temperature days in the effective low temperature process before the flowering of the f-th sample is 2 days. This process is on the v-th day and the v+1-th day of the f-th sample year. The maximum temperature on the day before this process is t v-1 The maximum temperature of the next day t v+2 , then their minimum value Z y3f =min(t v-1 , t v+2 );

[0205] Add the third candidate value Z of the low temperature day in the low temperature day number method of the effective low temperature process y3 =max(Z y31 ,…,Zy3i ,…,Z y3f )+0.1;

[0206] And so on;

[0207] According to the method of increasing the number of low-temperature days in effective low-temperature processes, the X-1 candidate value of the low-temperature day index is determined. Among all the effective low-temperature processes with X-1 low-temperature days selected by the full permutation method, the minimum value of the maximum temperature of the previous day and the maximum temperature of the next day is counted and the maximum value is selected as the X-1 candidate value:

[0208] Assume that there are L samples with X-1 days of effective low temperature process low temperature days selected by the full permutation method of the low temperature day index X-1th candidate value determined by the method of increasing the number of low temperature days of effective low temperature process,

[0209] The number of low temperature days in the effective low temperature process before the flowering of the first sample is X-1 days. The process is on the wth and w+1th days of the first sample year. The highest temperature on the day before the process is t w-1 The maximum temperature of the next day t w+2 , then their minimum value Z y31 =min(t w-1 , t w+2 );

[0210] And so on;

[0211] The number of low temperature days in the effective low temperature process before the flowering of the i-th sample is X-1 days. The process is on the u-th day and u+1-th day of the i-th sample year. The highest temperature on the day before the process is t u-1 The maximum temperature of the next day t u+2 , then their minimum value Z y3i =min(t u-1 , t u+2 );

[0212] And so on;

[0213] The number of low temperature days in the effective low temperature process before the flowering of the Lth sample is X-1 days. The process is on the vth day and v+1th day of the Lth sample year. The highest temperature on the day before the process is t v-1 The maximum temperature of the next day t v+2 , then their minimum value Z y3L =min(t v-1 , t v+2 );

[0214] Add the effective low temperature process low temperature day number method low temperature day X candidate value Z yX =max(Z y31 ,…,Z y3i ,…,Z y3L )+0.1.

[0215] According to the method of the maximum temperature of the day being lower than the adjacent value, the second candidate value Z of the low temperature day index x2 , the third candidate value Z x3 Instead of Z1, repeat the above process of determining whether the low temperature process effectiveness index before the sweet osmanthus blooms can be determined, and check which low temperature day index candidate values ​​can determine the low temperature process effectiveness index.

[0216] According to the candidate values ​​of low temperature day index of the method of gradually approaching the maximum temperature of the day, they are Z s2 , Z s3 ,…,Z ss Replace Z1 and repeat the above process of determining whether the effectiveness index of the low temperature process before the flowering of osmanthus can be determined until Z ss-1 Ability to determine the low temperature process effectiveness index, Z ss Low temperature process effectiveness indicators cannot be determined.

[0217] According to the method of increasing the number of low-temperature days of effective low-temperature processes according to the daily maximum temperature below the second candidate value Z of low-temperature days y2 , Z y3 ,…,Z yX Repeat the above process of determining whether the effectiveness index of the low temperature process before the flowering of osmanthus can be determined for Z1 until Z yX-1 Ability to determine the low temperature process effectiveness index, Z yX Low temperature process effectiveness indicators cannot be determined.

[0218] Among all the low-temperature process effectiveness indices determined based on the candidate values ​​of the low-temperature day index, the one with the highest low-temperature day index value is selected as the final effective low-temperature process index.

[0219] After the effectiveness index of the low temperature process before the flowering of Osmanthus fragrans was determined, a linear regression equation was established between the measured flowering start date and the starting date of the effective low temperature process.

[0220] After the effectiveness index of the low temperature process before the flowering of osmanthus is determined, the sequence of the start date of the first effective low temperature process in all years since meteorological records were kept is counted, that is, the start date of the low temperature process that reaches the effective low temperature process index for the first time in all years is counted according to the effectiveness index of the effective low temperature process.

[0221] A long-term series of the first batch of flowering start dates of Osmanthus fragrans over the years was simulated and constructed. The starting date of the first effective low-temperature process over the years was substituted into the univariate linear regression equation of the measured flowering start date and the starting date of the effective low-temperature process. The calculated result was the simulated value of the first batch of flowering start date series of Osmanthus fragrans over the years.

[0222] A statistical model for long-term meteorological forecast of the first batch of osmanthus flowering was established. The relevant and significant forecast factors were selected from the meteorological factors more than one month before the earliest date in the series simulation values ​​of the first batch of osmanthus flowering, and a regression model was established with the simulation values ​​of the first batch of osmanthus flowering.

[0223] Substituting the meteorological forecast factors of the forecast year into the long-term meteorological forecast statistical model for the first batch of osmanthus flowering, the flowering start date of osmanthus is predicted based on the calculated results;

[0224] Specific examples are as follows;

[0225] Implementation Case 1:

[0226] 1. Nanyang, Henan: The improved stepwise approach method was used to determine the low-temperature daily index and the low-temperature process effectiveness index, avoiding underreporting.

[0227] 1.1. The flowering start dates of each batch of Nanyang Osmanthus fragrans observed from 2016 to 2023 are: only one batch from 2016 to 2019, two batches in 2020 and 2023, four batches in 2021, and three batches in 2022 (Table 1).

[0228] Table 1: Flowering start dates of each batch of Nanyang Osmanthus fragrans from 2016 to 2023

[0229]

[0230] 1.2. Use the stepwise approach method to determine the first candidate value of the low temperature day index and determine whether the low temperature process effectiveness index can be determined;

[0231] like Figure 3 The highest and lowest temperatures before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 were 7.3 to 22.2℃. The first candidate value of the low temperature day index was determined by the stepwise approach method: Z1=22.2+0.1=22.3℃ (see Appendix Figure 3 ).

[0232] According to the first candidate value of 22.3℃ for the daily maximum temperature below the low temperature day index, the low temperature process was counted, and 14 valid low temperature processes and 0 invalid low temperature processes were screened out by the full permutation method (Table 2). The correlation coefficient between the start date of flowering of each batch in all years and the first low temperature process before flowering was the largest and passed the extremely significant correlation test. All the first low temperature processes were valid low temperature processes; the minimum number of low temperature days in the valid low temperature process was 1 day. Therefore, it can be determined that the effective low temperature process index of the daily maximum temperature below 22.3℃ is: the number of low temperature process days is ≥1 day. The relationship model between the start date X of the effective low temperature process with the daily maximum temperature below 22.3℃ and the start date Y of the flowering of Osmanthus fragrans is the univariate linear regression equation Y=1.1308X+1.7178(Appendix Figure 4 ).

[0233] Table 2: Characteristics of the low temperature process of 22.3℃, the first candidate value of the low temperature index before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023

[0234]

[0235] 1.3. Use the adjacent value method to determine the second candidate value and the third candidate value of the low temperature day index and determine whether the low temperature process effectiveness index can be determined;

[0236] The first candidate value of the low temperature day index appears on August 27, 2023. The highest temperatures of the previous day and the next day are 22.6℃ and 28.1℃ respectively. The minimum value is 22.6℃ and the maximum value is 28.1℃. The second candidate value of the low temperature day index is determined to be Z x2 =22.6+0.1=22.7℃, the third candidate value of the low temperature day index is determined to be Z x3 =28.1+0.1=28.2℃.

[0237] 1.3.1. Low-temperature processes were counted according to the second candidate value of 22.7°C for the low-temperature day indicator when the daily maximum temperature was lower than the adjacent value method. The full permutation method screened out 14 valid low-temperature processes and 0 invalid low-temperature processes (Table 3). The correlation coefficient between the start date of flowering and the first low-temperature process before flowering in all batches in all years was the largest and passed the extremely significant correlation test. All first low-temperature processes were valid low-temperature processes. The indicator for a valid low-temperature process that can determine that the daily maximum temperature is lower than 22.7°C is: the number of low-temperature days is ≥1 day.

[0238] Table 3: Characteristics of the low temperature process of 22.7°C, the second candidate value of the low temperature day index using the adjacent value method, before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023

[0239]

[0240] 1.3.2. According to the third candidate value of 28.2℃ for the low temperature day index with the daily maximum temperature lower than the adjacent value method, the low temperature process was counted. The full permutation method screened out 14 effective low temperature processes and 16 invalid low temperature processes (Table 4). The minimum number of effective low temperature process days was 2 days, and the maximum number of invalid low temperature process days was 5 days (see Appendix). Figure 5 ), therefore, the effectiveness index of the low-temperature process with the daily maximum temperature below 28.2℃ cannot be determined.

[0241] Table 4 Characteristics of the low temperature process of 28.2℃, the third candidate value of the low temperature day index using the adjacent value method, before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023

[0242]

[0243]

[0244] 1.4. Use the stepwise upward approach method to determine other candidate values ​​of the low-temperature day index and determine whether the low-temperature process effectiveness index can be determined;

[0245] 1.4.1. Determine the second candidate value Z of the low temperature day index by the stepwise upward approach method s2 :

[0246] The second smallest value of the highest temperature sequence before the flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 was selected, and the maximum value was 23.5℃ (see Appendix). Figure 6 ), the second candidate value of the low temperature day index of the gradual upward approach method; Z s2 =23.5+0.1=23.6℃. According to the low temperature process statistics when the daily maximum temperature is lower than the second candidate value of 23.6℃ of the low temperature day index of the gradual upward approach method, 14 valid low temperature processes and 1 invalid low temperature process are screened out by the full arrangement method (Table 5). The minimum number of valid low temperature process days is 1 day, and the maximum number of invalid low temperature process days is 1 day. The maximum temperature of the day after the valid low temperature process with 1 low temperature day is 25.9℃, and the minimum maximum temperature of the day after the invalid low temperature process with 1 low temperature day is 28.4℃. Therefore, it can be determined that the valid low temperature process index with the daily maximum temperature below 23.6℃ is: the first category of valid low temperature process is when the number of low temperature days is ≥2 days, the second category is when the number of low temperature days is =1 day, and the maximum temperature of the day after the low temperature process is ≤25.9℃; the invalid low temperature process is when the number of low temperature days is =1 day, and the maximum temperature of the day after the low temperature process is ≥28.4℃.

[0247] The relationship model between the effective low temperature process start date X with the flowering start date Y of Nanyang Osmanthus fragrans with the maximum temperature below 23.6℃ is the linear regression equation Y=1.113X+2.8242(Attached Figure 7 ).

[0248] Table 5 Characteristics of the low temperature process of 23.6℃, ​​the second candidate value of the low temperature day index using the stepwise upward approach method, before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023

[0249]

[0250] 1.4.2. Determine the third candidate value Z of the low temperature day index by the stepwise upward approach method s3 :

[0251] The third smallest value in the highest temperature sequence before the flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023 was selected, and the maximum value was 28.1℃ (see Appendix). Figure 8 ), the third candidate value Z of the low temperature day index of the step-by-step upward approach method s2=28.1+0.1=28.2℃. According to the third candidate value of 28.2℃ for the low temperature day index with the daily maximum temperature lower than the gradually upward approach method, the low temperature process is counted. The result is the same as the third candidate value of the adjacent value method. The effectiveness index of the low temperature process with the daily maximum temperature lower than 28.2℃ cannot be determined.

[0252] 1.5. Use the method of increasing the number of low-temperature days in the effective low-temperature process to determine other candidate values ​​of the low-temperature day index and determine whether the low-temperature process effectiveness index can be determined;

[0253] Determine the second candidate value Z of low temperature day index by increasing the number of low temperature days in effective low temperature process y2 ;

[0254] The first candidate value of the low temperature day index before the flowering of Nanyang Osmanthus fragrans was determined by the stepwise approach method as 22.3°C. There were 5 effective low temperature processes with 1 low temperature day selected by the full permutation method (Table 2). The minimum values ​​of the maximum temperature of the previous day and the maximum temperature of the next day were 23.5, 24.4, 22.8, 24.9, and 22.6°C, respectively. The maximum value of 24.9°C was selected from them. The second candidate value of the low temperature day index Z was determined by adding the number of low temperature days of effective low temperature processes. y2 =24.9+0.1=25.0℃, according to the second candidate value of 25.0℃ for the low temperature day index of the method of the number of low temperature days with the maximum temperature lower than the effective low temperature process, the low temperature process was counted, and 14 effective low temperature processes and 2 invalid low temperature processes were screened out by the full arrangement method (Table 6). The minimum number of effective low temperature process days is 1 day, and the maximum number of invalid low temperature process days is 1 day (see Appendix). Figure 9 ), compared with the maximum temperature of 27.2℃ on the day after an effective low temperature process with 1 low temperature day, the minimum maximum temperature of 26.0℃ on the day after an invalid low temperature process with 2 low temperature days and 1 day is higher. Therefore, the effectiveness index of the low temperature process with the maximum daily temperature below 25.0℃ cannot be determined.

[0255] Table 6 Characteristics of the low temperature process of 25.0℃, the second candidate value of the low temperature day index using the low temperature day number method, before flowering of each batch of Nanyang Osmanthus fragrans from 2016 to 2023

[0256]

[0257]

[0258] 1.6. Comprehensively determine the low temperature daily index and low temperature process effectiveness index;

[0259] The above analysis shows that the candidate low-temperature day index value (22.3°C) using the stepwise approach method, the second candidate low-temperature day index value (22.7°C) using the adjacent value method, and the second candidate low-temperature day index value (23.6°C) using the stepwise upward approach method can be used to determine the effectiveness of the low-temperature process. The effectiveness indicator for the low-temperature process is: the first category of a valid low-temperature process is when the number of low-temperature days is ≥2, the second category is when the number of low-temperature days is ≤1, and the maximum temperature on the day following the low-temperature process is ≤25.9°C; the invalid low-temperature process is when the number of low-temperature days is ≤1, and the maximum temperature on the day following the low-temperature process is ≥28.4°C.

[0260] 1.7. Verification of the effectiveness index of the improved stepwise approach method for low-temperature processes in the short-term forecast for 2024

[0261] Judging from the daily maximum temperature in 2024 and the flowering period of the first two batches of Nanyang Osmanthus fragrans (see Appendix Figure 10 ), the flowering starts on October 9 and October 15 respectively. If the low temperature daily index of 22.3℃ is determined according to the stepwise approach method, there is only one low temperature process with the maximum daily temperature below 22.3℃ before October 15, which is October 5-6. The starting date of this effective low temperature process predicts that the flowering start date of Nanyang Dangui coincides with the second batch flowering start date of October 15, and the first batch flowering start date of October 9 is missed.

[0262] The low temperature daily index determined by the improved stepwise approximation method was 23.6℃. Before October 15, there were two effective low temperature processes on October 2 and October 5-6, corresponding to two batches of flowering respectively. The correspondence between the starting dates of the two effective low temperature processes and the starting dates of the two batches of flowering was completely consistent with the one-variable linear regression equation established for the 2016-2023 samples.

[0263] 1.8. Simulate and construct a long-term series of the first batch of flowering of Nanyang Osmanthus fragrans and establish a long-term meteorological forecast model and test it;

[0264] According to the low temperature process effectiveness index of 23.6℃ on a low temperature day, the sequence of the first effective low temperature process start date from 1952 to 2023 was statistically constructed, and the sequence of the first batch of flowering start dates of Nanyang Osmanthus fragrans was simulated and constructed. The earliest one was August 20 (see Appendix). Figure 11 ), a long-term forecast statistical model for the first batch flowering start date and meteorological factors before July 20 was established using samples from 1954 to 2023 (Table 7). The back-substitution fitting values ​​of the long-term forecast statistical model are basically synchronized with the simulated sequence of the first batch flowering start date of Nanyang Osmanthus fragrans in previous years (Appendix Figure 12 ), in 2024, the meteorological factor values ​​before July 20 were substituted into the long-term forecast statistical model for calculation, and it was predicted that the first batch of Nanyang Osmanthus fragrans would begin to bloom around October 8, and the actual date was October 9.

[0265] Table 7 Meteorological factors and regression coefficients of the long-term forecast statistical model for the start date of the effective low temperature process in Nanyang Dangui, and the numerical values ​​and calculation results of meteorological factors in 2024

[0266] Meteorological factors Representative significance Regression coefficient 2024 values Calculation results (Intercept) intercept 976.3 976.3 X01 The average maximum temperature during the Spring Equinox -13.68 20.76 -284.00 X012 Average maximum temperature during the Spring Equinox^2 0.4011 430.98 172.87 X02 The total average maximum temperature of the four solar terms before the summer solstice of that year -50.69 31.76 -1609.69 X022 The total average maximum temperature of the four solar terms before the summer solstice of that year^2 0.8963 1008.42 903.85 X032 The total average minimum temperature of the four solar terms before Bailu last year^2 -0.1752 498.07 -87.26 X04 The average minimum temperature during the winter solstice of the previous year -4.55 -2.41 10.95 X042 The average minimum temperature during the winter solstice of the previous year^2 -0.8271 5.79 -4.79 X05 The total value of the extreme minimum temperature in the two solar terms before Bailu last year 30.97 15.9 492.42 X052 Total value of the extreme minimum temperature in the two solar terms before Bailu last year^2 -1.017 252.81 -257.11 X06 Total extreme minimum temperatures during the two solar terms before the winter solstice last year 13.02 -7.75 -100.91 X062 Total of the extreme minimum temperatures during the two solar terms before the winter solstice last year^2 0.866 60.06 52.01 X07 The lowest temperature during the Qingming solar term that year 1.017 7.4 7.53 X082 The average day difference of the Cold Dew solar term last year was 2 0.06838 150.98 10.32 X09 The average daily difference of Xiaoxue solar term last year was -6.12 10.39 -63.56 X092 The average daily difference of Xiaoxue solar term last year^2 0.2527 107.86 27.26 X10 The total value of the average daily temperature difference during the 12 solar terms before Grain Rain -28.45 9.7 -276.90 x102 The total value of the average daily temperature difference of the 12 solar terms before Grain Rain in that year^2 1.482 94.73 140.39 X11 The total precipitation value of the 13 solar terms before Grain Full in that year 0.06121 186.5 11.42 X12 Total number of precipitation days during the three solar terms before the Great Heat last year -5.37 9 -48.33 X122 Total number of precipitation days during the three solar terms before the Great Heat of the previous year^2 0.1229 81.00 9.95 X132 Total sunshine hours during the 9 solar terms before Minor Heat last year^2 -0.00002 742354.56 -16.29

[0267] Implementation Case 2:

[0268] 2. Pucheng, Fujian: The improved stepwise approach method was used to determine the low temperature daily index and low temperature process effectiveness index.

[0269] 2.1. The flowering periods of each batch of Osmanthus fragrans were observed and recorded from 2010 to 2023. Two batches of flowers bloomed in 2014, 2015, and 2018, and only one batch bloomed in other years (Table 8).

[0270] Table 8 The flowering start date of each batch of Pucheng Cinnabar Osmanthus fragrans

[0271]

[0272] 2.2. Use the stepwise approach method to determine the first candidate value of the low temperature day index and determine whether the low temperature process effectiveness index can be determined;

[0273] The highest and lowest temperatures on the day before the flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2010 to 2023 were 13.4 to 25.9℃ (see Appendix Figure 13 ).

[0274] The first candidate value of the low temperature day index for Zhusha Dangui in Pucheng was determined by the stepwise approach method as Z1=25.9+0.1=26.0℃. According to the statistics of low temperature processes with the daily maximum temperature lower than the first candidate value of the low temperature day index Z1=26.0℃, 17 valid low temperature processes and 11 invalid low temperature processes were selected by the full permutation method (Table 9). The minimum number of low temperature days for valid low temperature processes was 1 day; the maximum number of low temperature days for invalid low temperature processes was 2 days (see Appendix 9). Figure 14 The maximum number of low-temperature days in the invalid low-temperature process is greater than the minimum number of low-temperature days in the valid low-temperature process, and the low-temperature process effectiveness index cannot be determined.

[0275] Table 9 Characteristics of low temperature processes for the first candidate values ​​of the low temperature day index before flowering of each batch of Osmanthus fragrans in Pucheng from 2010 to 2023

[0276]

[0277] 2.3. Determine the second and third candidate values ​​of the low temperature day index using the adjacent value method:

[0278] The stepwise approach method determines that the first candidate value of the low temperature day index appears on October 7, 2014. The maximum temperature on the previous day is 26.8℃, and the maximum temperature on the next day is 27.3℃. Among them, the minimum value is 26.8℃ and the maximum value is 27.3℃. The second candidate value of the low temperature day index of Zhusha Dangui is Z x2 =26.8+0.1=26.9℃, the third candidate value of the Zhushadangui low temperature day index is determined to be Z x3 =27.3+0.1=27.4℃.

[0279] 2.3.1. The second candidate value Z of the low temperature day index according to the method of the maximum temperature of the day being lower than the adjacent value x2 =26.9℃ low temperature process statistics, full permutation method screened out 17 effective low temperature processes and 17 invalid low temperature processes (Table 10), the minimum number of low temperature days in effective low temperature process is 1 day; the maximum number of low temperature days in invalid low temperature process is 2 days (see Appendix Figure 15 The maximum number of low-temperature days in the invalid low-temperature process is greater than the minimum number of low-temperature days in the valid low-temperature process, and the low-temperature process effectiveness index cannot be determined.

[0280] Table 10 Characteristics of low temperature processes before flowering of each batch of Osmanthus fragrans in Pucheng from 2010 to 2023 (second candidate value of 26.9℃ using adjacent value method)

[0281]

[0282]

[0283] 2.3.2. The third candidate value Z of the low temperature day index according to the method of the maximum temperature of the day being lower than the adjacent value x3 =27.4℃ low temperature process statistics, full permutation method screened out 19 effective low temperature processes and 20 invalid low temperature processes (Table 11), the minimum number of low temperature days for effective low temperature process is 2 days, and the maximum number of low temperature days for invalid low temperature process is 2 days (see Appendix Figure 16 The maximum number of low-temperature days in the invalid low-temperature process is equal to the minimum number of low-temperature days in the valid low-temperature process. Comparing the maximum temperature of the day after the valid low-temperature process with low-temperature days = 2 days and the invalid low-temperature process, one valid low-temperature process is 28.1℃, and the six invalid low-temperature processes are all ≥29.4℃ (see Appendix). Figure 17 ). Therefore, the effectiveness index of the low temperature process with the daily maximum temperature below 27.4℃ can be determined as follows: the first category of effective low temperature process is when the number of low temperature days is ≥3 days, the second category is when the number of low temperature days is ≤2 days, and the maximum temperature on the day after the low temperature process is ≤28.1℃; the first category of invalid low temperature process is when the number of low temperature days is ≤1 day, and the second category is when the number of low temperature days is ≤1 day, and the maximum temperature on the day after the low temperature process is ≥29.4℃.

[0284] The relationship model between the effective low temperature process start date X and the flowering start date Y of Osmanthus fragrans is a linear regression equation Y = 0.8972X + 13.596 (see Appendix Figure 18 ).

[0285] Table 11 Characteristics of low temperature processes (the third candidate value of 27.4°C using the adjacent value method) for the low temperature day index before flowering of each batch of Osmanthus fragrans in Pucheng from 2010 to 2023

[0286]

[0287]

[0288] 2.4. Determining low-temperature day index and low-temperature process effectiveness index by stepwise upward approach

[0289] 2.4.1. Determining the Second Candidate Value Z of the Low-Temperature Day Index by the Stepwise Upward Approach s2 :

[0290] The second smallest value in the maximum temperature sequence before the flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2016 to 2023 was selected as the maximum value of 26.8℃ (see Appendix Figure 19 ), the second candidate value Z of the low temperature day index of the step-by-step upward approach method s2 =26.8+0.1=26.9℃, according to the daily maximum temperature lower than the second candidate value of the low temperature day index of the gradual upward approach method 26.9℃ statistical low temperature process, the result is consistent with the second candidate value of the low temperature day index of the adjacent value method Z x2 Similarly, no low-temperature process effectiveness indicators can be determined.

[0291] 2.4.2. Determine the third candidate value Z of the low temperature day index by the stepwise upward approach method s3 :

[0292] The third smallest value in the maximum temperature sequence before the flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2016 to 2023 was selected as the maximum value of 27.0℃ (see Appendix Figure 20 ), the third candidate value Z of the low temperature day index is determined by the stepwise upward approach method s3 =27.0+0.1=27.1℃. According to the third candidate value of the low temperature day index of the gradual upward approach method, 27.1℃, the low temperature process was counted. The full arrangement method screened out 17 valid low temperature processes and 20 invalid low temperature processes (Table 12). The minimum number of low temperature days in the valid low temperature process was 1 day; the maximum number of low temperature days in the invalid low temperature process was 2 days (see Appendix). Figure 21 The maximum number of low-temperature days in the invalid low-temperature process is greater than the minimum number of days in the valid low-temperature process, and the low-temperature process effectiveness index cannot be determined.

[0293] Table 12 Characteristics of low temperature processes before flowering of each batch of Osmanthus fragrans in Pucheng from 2010 to 2023 (the third candidate value of 27.1℃ by the method of gradually approaching the low temperature daily index)

[0294]

[0295]

[0296] 2.4.3. Determining the fourth candidate value Z of the low temperature day index by the stepwise upward approach method s4 :

[0297] The fourth smallest value in the maximum temperature sequence of each batch of Pucheng Cinnabar Osmanthus fragrans before flowering from 2016 to 2023 was selected as the maximum value of 27.3℃ (see Appendix Figure 22 ), the fourth candidate value of the low temperature day index of the gradual upward approach method; Z s4 =27.3+0.1=27.4℃. According to the statistics of the low temperature process (Table 8), the daily maximum temperature is lower than the fourth candidate value of the low temperature day index of 27.4℃ by the gradual upward approach method. The result is the same as the statistical low temperature process of the third candidate value of the low temperature day index of 27.4℃ by the adjacent value method, and the same low temperature process validity index is determined.

[0298] 2.4.4. Determine the fifth candidate value Z of the low temperature day index by the stepwise upward approach method s5 :

[0299] The fifth minimum value of the highest temperature sequence before the flowering of each batch of Pucheng Cinnabar Osmanthus fragrans from 2016 to 2023 was selected as the maximum value of 27.9℃ (see Appendix Figure 23 ), the fifth candidate value Z of the low temperature day index is determined by the stepwise upward approach method s5 =27.9+0.1=28.0℃, according to the third candidate value of 27.1℃ for the low temperature day index of the gradual upward approach method, the low temperature process was counted, and 17 valid low temperature processes and 34 invalid low temperature processes were screened out by the full arrangement method (Table 13). The minimum number of low temperature days in the valid low temperature process is 2 days; the maximum number of low temperature days in the invalid low temperature process is 3 days (see Appendix). Figure 24 The maximum number of low-temperature days in the invalid low-temperature process is greater than the minimum number of days in the valid low-temperature process, and the low-temperature process effectiveness index cannot be determined.

[0300] Table 13 Characteristics of low temperature processes before flowering of each batch of Osmanthus fragrans in Pucheng from 2010 to 2023 (the fifth candidate value of 28.0℃ according to the method of gradually approaching the low temperature daily index)

[0301]

[0302]

[0303]

[0304] 2.5 Add the method of low temperature days for effective low temperature process to determine the low temperature day index and low temperature process effectiveness index

[0305] 2.5.1 Add the method of low temperature days in effective low temperature process to determine the second candidate value of low temperature day index:

[0306] All the low temperature day indexes determined by the stepwise approach method with the highest daily temperature lower than the first candidate value Z1, and the full permutation method screened out three effective low temperature processes with one low temperature day. The minimum values ​​of the highest temperatures of the previous day and the next day of the three samples were 26.8℃, 26.9℃, and 27.6℃, respectively. The maximum value was 27.6℃ (Table 9), and the second candidate value of the low temperature day index of the method of increasing the number of low temperature days in the effective low temperature process was determined to be Z. y2 =27.6+0.1=27.7℃, the second candidate value of the low temperature day index Z is determined by the method of increasing the number of low temperature days in the effective low temperature process according to the daily maximum temperature. y2 =27.7℃ low temperature process statistics, full permutation method screened out 19 effective low temperature processes and 27 invalid low temperature processes (Table 14), the minimum number of low temperature days for effective low temperature process is 2 days, and the maximum number of low temperature days for invalid low temperature process is 2 days (see Appendix Figure 25 ), the maximum number of low-temperature days in the invalid low-temperature process is equal to the minimum number of low-temperature days in the valid low-temperature process. Comparing the maximum temperature of the day after the valid low-temperature process with low-temperature days = 2 days and the invalid low-temperature process, one valid low-temperature process is 28.1℃, and the eight invalid low-temperature processes are all ≥29.4℃ (Appendix Figure 26 ). Therefore, the effectiveness index of the low temperature process with the daily maximum temperature below 27.7℃ can be determined as follows: the first category of effective low temperature process is when the number of low temperature days is ≥3 days, the second category is when the number of low temperature days is ≤2 days, and the maximum temperature on the day after the low temperature process is ≤28.1℃; the first category of invalid low temperature process is when the number of low temperature days is ≤1 day, and the second category is when the number of low temperature days is ≤1 day, and the maximum temperature on the day after the low temperature process is ≥29.4℃.

[0307] The relationship model between the effective low temperature process start date X and the flowering start date Y of Osmanthus fragrans is a linear regression equation Y = 0.8973X + 13.73 (Appendix Figure 27 ).

[0308] Table 14 Characteristics of low temperature processes before flowering of each batch of Osmanthus fragrans in Pucheng from 2010 to 2023, and the second candidate value of 27.7℃ for the low temperature day index of the effective low temperature process method

[0309]

[0310] 1.6. Comprehensively determine the low temperature daily index and low temperature process effectiveness index;

[0311] The above analysis shows that the third candidate value for the low-temperature day indicator using the adjacent value method (27.4°C), the fourth candidate value for the low-temperature day indicator using the stepwise upward approach method (27.4°C), and the second candidate value for the number of low-temperature days using the effective low-temperature process method (27.7°C) can be used to determine the effectiveness of the low-temperature process. The effectiveness indicators for low-temperature processes are: The first category of effective low-temperature processes is defined as having 3 or more low-temperature days; the second category is defined as having 1 low-temperature day and a maximum temperature of 28.1°C or less on the day following the low-temperature process; and the second category is defined as having 1 low-temperature day and a maximum temperature of 29.4°C or more on the day following the low-temperature process.

[0312] 1.7. Verification of the effectiveness index of the improved stepwise approach method for low-temperature processes in the short-term forecast for 2024;

[0313] Judging from the daily maximum temperature in 2024 and the first batch of flowering of Osmanthus fragrans (see Appendix Figure 28 ), the first batch of flowering started on September 29, the low temperature day index determined by the improved stepwise approach method was 27.7℃, there was an effective low temperature process for 5 consecutive days from September 22 to 26, and the correspondence between the start date of the effective low temperature process and the start date of the first batch of flowering was completely consistent with the one-variable linear regression equation established for the 2010-2023 samples.

[0314] 1.8. Simulate and construct a long-term series of the first batch of flowering of Osmanthus fragrans in Pucheng and establish a long-term meteorological forecast model and test it;

[0315] According to the low temperature day index, the maximum temperature on the day below 27.7℃ was counted, and the starting date sequence of the first effective low temperature process in Pucheng, Fujian from 1951 to 2023 was simulated to construct the first batch of flowering start date sequence of Cinnabar Osmanthus fragrans in Pucheng, Fujian (see Appendix). Figure 29 ) is the earliest on September 1st. Using samples from 1953 to 2023, a long-term forecast statistical model for the simulated value sequence of the first batch of flowering onset and meteorological factors before August 1st was established (Table 15). The back-substitution fitting values ​​of the long-term forecast statistical model are basically synchronized with the simulated first batch flowering onset sequence of Osmanthus fragrans in previous years (Appendix Figure 30 ), in 2024, the meteorological factor values ​​before August 1 were substituted into the long-term forecast statistical model, and it was predicted that the first batch of Osmanthus fragrans in Pucheng, Fujian would begin to bloom around October 7, but the actual date was September 29.

[0316] Table 15 Meteorological factors and regression coefficients of the long-term forecast statistical model for the start date of the effective low temperature process in Zhushadangui, Pucheng, Fujian, and meteorological factors and calculation results in 2024

[0317]

[0318]

[0319] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0320] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0321] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for producing a long-term weather forecast for the first batch of osmanthus fragrans flowering period, characterized in that: The following steps are involved: S1: Collect the data of the flowering start date of Osmanthus fragrans as the measured flowering start date; S2: Use the improved stepwise approach method to determine the low temperature day index and the low temperature process effectiveness index before the flowering of Osmanthus fragrans; S3: Establish a linear regression equation between the measured flowering start date and the effective low temperature process start date; S4: Count the starting date sequence of the first effective low temperature process in each year since meteorological records began; S5: Simulate and construct the first batch flowering start sequence of Osmanthus fragrans over the years; S6: Establish a long-term meteorological forecast statistical model for the first batch of osmanthus flowering in previous years; S7: Prepare long-term weather forecasts to predict the start of the first batch of sweet osmanthus flowers.

2. a method for making the first batch flowering period long-term weather forecast of sweet osmanthus according to claim 1, characterized in that: The specific process of determining the effectiveness index of low temperature days and low temperature processes before the flowering of osmanthus fragrans by the improved stepwise approach method in S2 is as follows: S21; Determine the candidate value of the low temperature day index by an improved stepwise approach method; S22: determining whether the low-temperature process effectiveness index can be determined, and extracting the determined low-temperature process effectiveness index; S23: Selecting the low-temperature day index with the highest value from all the low-temperature process effectiveness indicators as the final low-temperature day index and the effective low-temperature process index.

3. a method for making the first batch flowering period long-term weather forecast of sweet osmanthus according to claim 2, characterized in that: The specific process of S21 using the improved stepwise approach method to determine the candidate value of the low-temperature day index is as follows: the first candidate value of the low-temperature day index is determined by the stepwise approach method, and other candidate values ​​of the low-temperature day index are determined using three improved adjacent value methods, the stepwise upward approach method, and the method of increasing the number of effective low-temperature process days.

4. a method for making the first batch flowering period long-term weather forecast of sweet osmanthus according to claim 2, characterized in that: The specific process of determining whether the low temperature process effectiveness index can be determined in S22 is as follows: S221: Count the low temperature process according to the candidate value of the low temperature day index; S222: Screen out effective low-temperature processes and invalid low-temperature processes through full permutation method; S223: comparing the low-temperature characteristics of the effective low-temperature process and the low-temperature characteristics of the invalid low-temperature process; S224: The criterion for determining whether the effectiveness index of the low-temperature process can be determined is that the minimum number of low-temperature days of the effective low-temperature process is greater than the maximum number of low-temperature days of the invalid low-temperature process, or when the minimum number of low-temperature days of the effective low-temperature process is equal to the maximum number of low-temperature days of the invalid low-temperature process, the maximum value of the highest temperature on the day after the effective low-temperature process with the same number of low-temperature days is less than the minimum value of the highest temperature on the day after the invalid low-temperature process with the same number of low-temperature days.

5. a method for making the first batch flowering period long-term weather forecast of sweet osmanthus according to claim 2, characterized in that: The improved stepwise approach method used in determining the candidate values ​​of the low temperature day index before the flowering of osmanthus includes the adjacent value method, the stepwise upward approach method, and the method of increasing the number of low temperature days in the effective low temperature process.

6. A method for making a long-term weather forecast for the first batch of flowering period of sweet osmanthus according to claim 5, characterized in that: The specific process of the adjacent value method is: Find the year and date of the first candidate value of the low temperature day index, select the minimum value from the highest temperatures of the previous day and the next day to determine the second candidate value, and the maximum value to determine the third candidate value.

7. A method for making a long-term weather forecast for the first batch of flowering period of sweet osmanthus according to claim 5, characterized in that: The specific process of the step-by-step upward approach method is as follows: The second smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected from it to determine the second candidate value. The third smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected from it to determine the third candidate value. And so on, the sth smallest value in the sequence of the highest temperature on the day before flowering of each batch is counted, and the maximum value is selected from it to determine the sth candidate value, where s≥2.

8. A method for making a long-term weather forecast for the first batch of flowering period of sweet osmanthus according to claim 5, characterized in that: The specific process of the method for increasing the number of low-temperature days in the effective low-temperature process is as follows: The first candidate value of the low temperature day index determined by the stepwise approach method is used in the effective low temperature process with one low temperature day selected by the full permutation method in S222. The minimum value of the maximum temperature of the previous day and the maximum temperature of the next day is counted, and the maximum value is selected from them to determine the second candidate value; The second candidate value of the low temperature day index determined by the method of adding the number of low temperature days in the effective low temperature process is obtained by counting the minimum value of the highest temperature of the previous day and the maximum temperature of the next day from the effective low temperature process with 2 days, and the maximum value is selected from them to be determined as the third candidate value. Similarly, the Xth candidate value of the low temperature day index determined by the method of adding the number of low temperature days in the effective low temperature process is obtained by counting the minimum value of the highest temperature of the previous day and the maximum temperature of the next day from the effective low temperature process with X-1 days.

9. A method for making a long-term weather forecast for the first batch of flowering period of sweet osmanthus according to claim 2, characterized in that: The specific process of selecting the lowest temperature day index with the highest value from all the determined low temperature process effectiveness indices in S23 as the final lowest temperature day index and the effective low temperature process index is as follows: S22 is performed respectively by the first candidate value of the stepwise approach method, the second candidate value of the adjacent value method, and the third candidate value of the adjacent value method; The second candidate value obtained by the stepwise upward approach method, the third candidate value obtained by the stepwise upward approach method, ..., the s-th candidate value obtained by the stepwise upward approach method are respectively subjected to S22 until the s-1-th candidate value by the stepwise upward approach method can determine the low-temperature process effectiveness index, and the s-th candidate value by the stepwise upward approach method cannot determine the low-temperature process effectiveness index; S22 is performed by increasing the second candidate value of the effective low temperature process days method, increasing the third candidate value of the effective low temperature process days method, ..., increasing the Xth candidate value of the effective low temperature process days method, respectively, until the X-1th candidate value of the stepwise upward approach method can determine the low temperature process effectiveness index, and the Xth candidate value of the stepwise upward approach method cannot determine the low temperature process effectiveness index; After all the candidate values ​​of the low temperature day index determined by the stepwise approach method and the improved stepwise approach method are respectively subjected to S22, the low temperature day index with the highest value among the determined low temperature process effectiveness indices is selected as the final low temperature day index and effective low temperature process index.

Images