Method for predicting influence of greenhouse effect on extreme rainfall

Abstract

The invention discloses a method for predicting the influence of a greenhouse effect on extreme rainfall in the technical field of climate change. The method comprises the following steps: (1) collecting the monthly average temperature and monthly accumulated rainfall of an area in which the influence of the greenhouse effect on the extreme rainfall needs to be predicted; (2) determining the regression relationship between the monthly accumulated rainfall and the monthly average temperature at the extreme high quantile of the region; (3) downloading a predicted value of the monthly average temperature of the region under the greenhouse effect in the sixth stage of the international coupling mode comparison plan from the internet; (4) selecting a prediction result most conforming to the regional condition from different monthly average temperature prediction results; (5) substituting the selected most conforming monthly average temperature into the regression relation obtained in the step (2) to obtain the extreme rainfall; (6) determining the influence of the greenhouse effect on the extreme rainfall; the extreme rainfall under the greenhouse effect is obtained by using the relationship between the temperature and the rainfall, and the method is easy to use in different areas.

Description

Technical Field

[0001] The present invention relates to the technical field of climate change, and in particular to a method for predicting the impact of greenhouse effect on extreme rainfall. Background Art

[0002] Predicting the impact of the greenhouse effect on extreme rainfall is urgent. Technically, there are many methods to predict rainfall due to the greenhouse effect, including machine learning and application of climate models for prediction; however, there are few studies on predicting extreme rainfall. There is a clear correlation between rainfall and temperature, and predicting the impact of the greenhouse effect on extreme rainfall by temperature is a more feasible method. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a method for predicting the impact of the greenhouse effect on extreme rainfall. The relationship between the monthly average temperature and the monthly cumulative rainfall is used to predict the impact of extreme rainfall under the greenhouse effect. The present invention is easy to obtain data and has a flexible method, and can be promoted and used in flood control and drainage planning in different regions.

[0004] The object of the present invention is achieved in this way: a method for predicting the impact of greenhouse effect on extreme rainfall, comprising the following steps:

[0005] (1) Collect the monthly average temperature and monthly cumulative rainfall in the area where the impact of greenhouse effect on extreme rainfall needs to be predicted;

[0006] (2) Determine the regression relationship between the monthly cumulative rainfall and the monthly mean temperature under the extreme high quantile in the region;

[0007] (3) Download the predicted monthly average temperature of the region under the greenhouse effect from the sixth phase of the International Coupled Model Intercomparison Project (CMIP6) from the Internet;

[0008] (4) Select the forecast result that best suits the situation in the region from the forecast results of average monthly temperatures of different months;

[0009] (5) Substituting the most suitable monthly average temperature into the regression relationship obtained in step (2) to obtain the extreme rainfall;

[0010] (6) Determine the impact of the greenhouse effect on extreme rainfall.

[0011] Furthermore, step (1) specifically includes the following steps:

[0012] a. Determine the number of meteorological stations in the area to be studied;

[0013] b. Collect the monthly average temperature T from all meteorological stations in the area. i And the corresponding monthly cumulative rainfall Pi where \(i = 1, 2, \ldots, n\); \(n\) is the number of observation values, and \(n\geq240\).

[0014] c. If the number of meteorological stations in the study area is greater than 1, each meteorological station is regarded as an independent meteorological station for collection and recording.

[0015] Further, step (2) specifically includes the following steps:

[0016] a. Select an extreme quantile \(p\), where \(0.9 < p < 1\). Using the entire data \(P\) of monthly cumulative rainfall as the dependent variable and the entire data \(T\) of monthly average temperature as the independent variable, establish the regression relationship between the two under the \(p\) - quantile:

[0017] ;

[0018] Two of the coefficients and are determined by the following steps b to e;

[0019] b. Calculate the bandwidth \(h\) adapted to the number of observation values \(n\):

[0020] ;

[0021] c. For each collected monthly average rainfall \(T\) i within the range of values, calculate and to obtain the temperature observation values \(T\) id within this range, as well as the corresponding rainfall observation values \(P\) id ; The number of qualified observation values is denoted as \(m\).

[0022] d. For each qualified \(T\) id calculate the weight using the following formula:

[0023] ;

[0024] e. For each pair of observation values \(T\) i , \(P\) i , only use the qualified observation values \(T\) id , \(P\) id to participate in the following calculation to minimize it:

[0025] ; Thus, the corresponding and are obtained. The above formula can be numerically solved by the interior - point method;

[0026] f. Repeat this step, with a step size of 0.02, to determine the regression relationship between monthly cumulative rainfall and monthly average temperature at the 0.90 - 0.98 quantiles.

[0027] Further, step (3) specifically includes the following steps:

[0028] a. Search for all global datasets with monthly cumulative rainfall and monthly average temperature on the CMIP6 website;

[0029] b. Download the historical calculation results of the monthly average temperature in previous years in each dataset; and download the datasets corresponding to the most severe greenhouse effect scenarios in these datasets, that is, the datasets corresponding to the RCP8.5-SSP5 case;

[0030] c. Extract the monthly cumulative rainfall and monthly average temperature of this region from the above datasets.

[0031] Further, step (4) specifically includes the following steps:

[0032] a. Calculate the average value T of the monthly average temperature in previous years measured at all meteorological stations in this region, a that is, average the monthly average temperature observation values of the same month at different stations;

[0033] b. Compare the above measured monthly average value with the calculated monthly average temperature of this region in the downloaded CMIP6 dataset, and calculate the root mean square error of the monthly average temperature of each dataset respectively;

[0034] c. Select the CMIP6 dataset with the smallest root mean square error as the prediction dataset that best fits the situation of this region;

[0035] d. Select the monthly average temperature corresponding to the annual extreme cumulative rainfall to be calculated from the above selected prediction dataset, denoted as , d = 1, 2, …, s, 0.90 < p < 1; s is the number of selected monthly average temperature values.

[0036] Further, step (5) specifically includes the following steps:

[0037] a. Given the quantile p, 0.90 < p < 1, select the regression relationship corresponding to it calculated in step (2);

[0038] b. Substitute the T obtained in step (4) a into the regression relationship corresponding to the p quantile; if there is no regression relationship corresponding to T in step (4), a then obtain a and and through linear interpolation in the regression relationship of the temperature closest to T a to obtain the regression relationship corresponding to T; thus obtain the extreme monthly cumulative rainfall corresponding to T a ​ ;

[0039] c. Also Substitute the regression relationship corresponding to the p quantile to obtain the corresponding extreme cumulative rainfall at the predicted temperature ;

[0040] d. Repeat steps ac for all quantiles p to obtain the corresponding extreme monthly cumulative rainfall values ​​for all quantiles.

[0041] Furthermore, the step (6) specifically includes the following steps:

[0042] a. Given a month, the predicted value of the extreme cumulative rainfall corresponding to a certain year Extract and calculate the data The maximum absolute value of the difference is denoted as DIS P :

[0043]

[0044] DIS P As the greenhouse effect, the extreme monthly cumulative rainfall of this year and month under the p quantile;

[0045] b. Repeat step a for all quantiles p to obtain the DIS corresponding to the extreme monthly cumulative rainfall under all quantiles P ;

[0046] c. Select all quantiles p under DIS P The maximum value corresponds to the extreme monthly cumulative rainfall as a measure of the impact of the greenhouse effect on extreme rainfall in that year and month:

[0047] .

[0048] In the present invention, meteorological research indicates that the greenhouse effect will cause temperature rise, while environmental research indicates that there is a definite relationship between the monthly average temperature and the monthly cumulative rainfall; on this basis, the present patent uses internationally recognized authoritative climate forecast data and regression and other methods to analyze the impact of the greenhouse effect on the monthly cumulative rainfall.

[0049] Compared with the prior art, the beneficial effect of the present invention is that the present invention can be used to analyze and predict the impact of temperature rise caused by the greenhouse effect on extreme rainfall, and can be widely used in flood control and drainage planning under future climate change. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0051] Figure 1 It is a flowchart of the present invention. Detailed implementation manners

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0053] As Figure 1 shown, a method for predicting the impact of greenhouse effect on extreme rainfall includes the following steps:

[0054] (1) Collect the monthly average temperature and monthly cumulative rainfall in the area where the impact of the greenhouse effect on extreme rainfall needs to be predicted, specifically including the following steps:

[0055] (1-1) Determine the number of meteorological stations in the research area.

[0056] (1-2) Collect the monthly average temperature T i and the corresponding monthly cumulative rainfall P i for each month in the historical time series from all meteorological stations in this area; where i = 1, 2,..., n; n is the number of observation values, and n is greater than or equal to 240.

[0057] (1-3) If the number of meteorological stations in the research area is greater than 1, each meteorological station is regarded as an independent meteorological station for collection and recording.

[0058] (2) Determine the regression relationship between the monthly cumulative rainfall and the monthly average temperature at the extreme high quantile in this area, specifically including the following steps:

[0059] (2-1) Select the extreme quantile p, 0.9 < p < 1. Using all the data P of the monthly cumulative rainfall as the dependent variable and all the data T of the monthly average temperature as the independent variable, establish the regression relationship between the two at the p quantile:

[0060] ;

[0061] Among them, the two coefficients and The decision is calculated by following steps (1-2) to (1-3);

[0062] (2-2) Calculate the bandwidth h suitable for the number of observations n:

[0063] ;

[0064] (2-3) For each collected monthly average rainfall T i In the range of values, calculate and The temperature observation value T in this range is obtained id , and the corresponding rainfall observation value P id ; The number of observations that meet the requirements is recorded as m;

[0065] (2-4) For each T that meets the requirements id The weights are calculated using the following formula:

[0066] ;

[0067] (2-5), for each observation pair T i ,P i , only use the observations T that meet the requirements id ,P id Participate in the following calculation to minimize it:

[0068] ; In this way, the corresponding as well as , the above equation can be solved numerically using the interior point method;

[0069] (2-6) Repeat this step with a step size of 0.02 to determine the regression relationship between the monthly cumulative rainfall and the monthly average temperature at the 0.90-0.98 quantile.

[0070] (3) Download the predicted monthly average temperature of the region under the greenhouse effect from the sixth phase of the International Coupled Model Intercomparison Project (CMIP6) from the Internet. The specific steps include:

[0071] (3-1) Search for all global datasets with monthly cumulative rainfall and monthly mean temperature from the CMIP6 website;

[0072] (3-2) Download the historical calculation results of the monthly average temperature in previous years in each dataset; and download the datasets corresponding to the most severe greenhouse effect in these datasets, that is, the datasets corresponding to the RCP8.5-SSP5 scenario;

[0073] (3-3) Extract the monthly cumulative rainfall and monthly average temperature of this area from the above dataset;

[0074] (4) Select the prediction result that best suits the situation of this area from the prediction results of different monthly average temperatures, which specifically includes the following steps:

[0075] (4-1) Calculate the average value T of the monthly average temperatures of previous years measured at all meteorological stations in this area, a that is, average the observed values of the monthly average temperatures of the same month at different stations;

[0076] (4-2) Compare the above-mentioned measured monthly average value with the calculated monthly average temperature of this area in the downloaded CMIP6 dataset, and calculate the root mean square error of the monthly average temperature of each dataset respectively;

[0077] (4-3) Select the CMIP6 dataset with the smallest root mean square error as the prediction dataset that best suits the situation of this area;

[0078] (4-4) Select the monthly average temperature of a certain year corresponding to the extreme cumulative rainfall to be calculated from the above-selected prediction dataset, denoted as , d = 1, 2, …, s, 0.90 < p < 1; s is the number of selected monthly average temperatures.

[0079] (5) Substitute the selected most suitable monthly average temperature into the regression relationship obtained in step (2) to obtain the extreme rainfall, which specifically includes the following steps:

[0080] (5-1) Given the quantile p, 0.90 < p < 1, select the corresponding regression relationship calculated in step (2);

[0081] (5-2) Substitute the T a obtained in step (4) into the regression relationship corresponding to the p quantile; if there is no regression relationship corresponding to T a in step (4), then obtain a and and through linear interpolation in the regression relationship of the temperature closest to T a to obtain the regression relationship corresponding to T a ; thus obtain the extreme monthly cumulative rainfall corresponding to T

[0082] (5-3) Similarly, substitute into the regression relationship corresponding to the p quantile to obtain the extreme cumulative rainfall corresponding to the predicted temperature;

[0083] (5-4) Repeat steps (5-1) to (5-3) for all quantiles p to obtain the corresponding extreme monthly cumulative rainfall values ​​for all quantiles;

[0084] (6) Determine the impact of the greenhouse effect on extreme rainfall, including the following steps:

[0085] (6-1) Given a month, the predicted value of the extreme cumulative rainfall corresponding to a certain year Extract and calculate the data The maximum absolute value of the difference is denoted as DIS P :

[0086]

[0087] DIS P As the greenhouse effect, the extreme monthly cumulative rainfall of this year and month under the p quantile;

[0088] (6-2) Repeat step (6-1) for all quantiles p to obtain the DIS corresponding to the extreme monthly cumulative rainfall under all quantiles P ;

[0089] (6-3) Select all quantiles p under DIS P The maximum value corresponds to the extreme monthly cumulative rainfall as a measure of the impact of the greenhouse effect on extreme rainfall in that year and month:

[0090] .

[0091] The present invention will be further described below in conjunction with specific examples.

[0092] (1) First, we determine that there are 8 meteorological stations with valid observation data in the city. The monthly average temperature and the corresponding monthly cumulative rainfall of each meteorological station are 240, that is, there are 1920 observation data in total.

[0093] (2) The extreme quantile is selected as 0.902, and the optimal bandwidth corresponding to 1920 observations is calculated to be 0.23. The monthly average temperature observation value T i The corresponding standard deviation is 6.03 degrees. The average temperature observations of each month belonging to T are extracted. i -4.15 degrees to T i +4.15 degrees between observations T id ; and the monthly cumulative rainfall P corresponding to these observations id ; In this example, 8-15 observations are selected for each temperature.

[0094] For each selected temperature T idThe value calculates its weight:

[0095]

[0096] For each observation pair (T i ,P i ), only use the observations that meet the requirements (T id ,P id ) participates in the following calculation to minimize it:

[0097] Solve using the interior point method; for this example, for example, when the extreme quantile is 0.902 and the temperature is 32.05 degrees, the corresponding monthly average temperature regression coefficient is 19.5, and the intercept in the regression relationship is 3.86. Note that the regression coefficient and intercept in the regression relationship are different for different temperatures; the regression coefficient and intercept are also different for different quantiles.

[0098] Repeat the above process to solve the regression coefficient and intercept corresponding to different temperatures between the quantiles of 0.904-0.980.

[0099] (3) Search for all global datasets with monthly cumulative rainfall and monthly mean temperature from the CMIP6 website, a total of 21 datasets. Download the historical calculation results of monthly mean temperature from 2000 to 2015 in each dataset; as well as the corresponding datasets under RCP8.5-SSP5; and use geographic information system software to extract the data corresponding to the city.

[0100] (4) Average the monthly average temperatures of the eight meteorological stations in the city from 2000 to 2015 according to the month; obtain the monthly average temperature value Ta for 12 months. In this example, the average temperatures of the city from January to December are 4.02, 8.53, 13.32, 18.12, 23.43, 27.18, 32.15, 33.18, 25.76, 17.13, 10.57, and 5.03, respectively, in degrees Celsius.

[0101] Then, the monthly average temperatures from 2000 to 2015 in the CMIP6 historical data sets were downloaded respectively, and the root mean square errors between them and the measured values ​​were calculated. The data set with the smallest error was selected as the data set that Nanjing University of Information Science and Technology participated in the calculation.

[0102] Finally, the monthly average temperature values ​​of the city in the above data set that need to calculate the extreme cumulative rainfall are selected. Here, the extreme cumulative rainfall in July 2050 and 2080 needs to be calculated, and the corresponding monthly average temperatures are 33.13 and 33.56 degrees Celsius.

[0103] (5) Substituting the average temperature of July, 32.15 degrees Celsius, into the regression relationship with a quantile of 0.902, we obtain the extreme cumulative monthly rainfall at this quantile to be 583.16 mm.

[0104] Similarly, substituting the average July temperatures of 33.13 and 33.56 degrees Celsius in 2050 and 2080 into the regression relationship, the extreme monthly cumulative rainfall under this quantile is obtained to be 604.55 mm and 615.37 mm respectively.

[0105] Use the above steps to calculate the monthly extreme cumulative rainfall corresponding to all quantiles for the three temperatures of 0.902-0.980.

[0106] (6) For July 2050, the predicted extreme cumulative rainfall corresponding to the monthly average temperature (i.e., 33.13 degrees Celsius) at the 0.902 quantile is extracted, and the result is 604.55 mm. The absolute value of the difference between the extreme cumulative rainfall at the 0.902 quantile and that in July is 21.39 mm.

[0107] Using the same method, the maximum absolute value of the extreme cumulative rainfall difference in July for all quantiles is 50.08 mm; therefore, the impact of the greenhouse effect on the monthly extreme rainfall in July in 2050 is 50.08 mm.

[0108] The above embodiments are only used to help understand the method and core idea of ​​the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made to the present invention without departing from the principles of the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting the impact of greenhouse effect on extreme rainfall, characterized in that: It includes the following steps: (1) Collect the monthly average temperature and monthly cumulative rainfall in the area where the impact of greenhouse effect on extreme rainfall needs to be predicted; (2) Determine the regression relationship between the monthly cumulative rainfall and the monthly average temperature at the extreme high quantile in this area; (3) Download from the Internet the predicted values of the monthly average temperature in this area under the greenhouse effect by the Coupled Model Intercomparison Project Phase 6 (CMIP6); (4) Select the prediction result that best suits the situation of this area from the prediction results of different monthly average temperatures; (5) Substitute the selected most suitable monthly average temperature into the regression relationship obtained in step (2) to obtain the extreme rainfall; (6) Determine the impact of the greenhouse effect on extreme rainfall.

2. A method for predicting the impact of greenhouse effect on extreme rainfall according to claim 1, characterized in that: Step (1) specifically includes the following steps: a. Determine the number of meteorological stations in the area to be studied; b. Collect the monthly average temperature T from all meteorological stations in the area. i And the corresponding monthly cumulative rainfall P i ; Where i=1,2,…,n; n is the number of observations, which is greater than or equal to 240; c. If the number of meteorological stations in the study area is greater than 1, each meteorological station is regarded as an independent meteorological station for collection and recording.

3. A method for predicting the impact of greenhouse effect on extreme rainfall according to claim 1 or 2, characterized in that: The said step (2) specifically includes the following steps: a. Select the extreme quantile p, where 0.9 < p < 1. Using all the data P of the monthly cumulative rainfall as the dependent variable and all the data T of the monthly average temperature as the independent variable, establish the regression relationship between the two at the p quantile: ； The two coefficients as well as The following steps b to e are used to calculate and determine; b. Calculate the bandwidth h adapted to the number of observation values n: ； c. For each collected monthly average rainfall T i In the range of values, calculate and The temperature observation value T in this range is obtained id , and the corresponding rainfall observation value P id ; The number of observations that meet the requirements is recorded as m; d. For each T that meets the requirements id The weights are calculated using the following formula: ； e. For each observation pair T i ,P i , only use the observations T that meet the requirements id ,P id Participate in the following calculation to minimize it: ; In this way, the corresponding as well as , the above equation can be solved numerically using the interior point method; f. Repeat this step, with a step size of 0.02, to determine the regression relationship between the monthly cumulative rainfall and the monthly average temperature at the 0.90 - 0.98 quantiles.

4. A method for predicting the impact of greenhouse effect on extreme rainfall according to claim 1 or 2, characterized in that: The said step (3) specifically includes the following steps: a. Search the CMIP6 website for all global datasets with monthly cumulative rainfall and monthly average temperature; b. Download the historical calculation results of the monthly average temperature in previous years in each dataset; and download the datasets corresponding to the most severe greenhouse effect situation in these datasets, that is, the datasets corresponding to the RCP8.5 - SSP5 situation; c. Extract the monthly cumulative rainfall and monthly average temperature of this area from the above datasets.

5. A method for predicting the impact of greenhouse effect on extreme rainfall according to claim 1 or 2, characterized in that: The said step (4) specifically includes the following steps: a. Calculate the average monthly temperature T measured by all weather stations in the area in previous years a , that is, averaging the monthly mean temperature observations of different stations in the same month; b. Compare the above measured monthly average values with the calculated monthly average temperature of this area in the downloaded CMIP6 datasets, and calculate the root mean square error of the monthly average temperature of each dataset respectively; c. Select the CMIP6 dataset with the smallest root mean square error as the prediction dataset that best suits the situation of this area; d. Select the monthly average temperature of a certain year corresponding to the extreme cumulative rainfall to be calculated from the selected prediction dataset, denoted as , where d = 1, 2, …, s, 0.90 < p < 1; s is the number of selected monthly average temperatures.

6. A method for predicting the impact of greenhouse effect on extreme rainfall according to claim 1 or 2, characterized in that: The said step (5) specifically includes the following steps: a. Given the quantile p, where 0.90 < p < 1, select the regression relationship corresponding to it calculated in step (2); b. T obtained in step (4) a Substitute the regression relationship corresponding to the p quantile; if there is no T in step (4) a The corresponding regression relationship is obtained by the nearest neighbor T a The regression relationship of temperature is obtained as well as Perform linear interpolation to get T a Corresponding regression relationship; thus, we get T a Corresponding extreme monthly cumulative rainfall ; c. Also Substitute the regression relationship corresponding to the p quantile to obtain the corresponding extreme cumulative rainfall at the predicted temperature ; d. Repeat steps a - c for all quantiles p to obtain the values of the extreme monthly cumulative rainfall corresponding to all quantiles.

7. A method for predicting the impact of greenhouse effect on extreme rainfall according to claim 1 or 2, characterized in that: The said step (6) specifically includes the following steps: a. Given a month, the predicted value of the extreme cumulative rainfall corresponding to a certain year Extract them and calculate the The maximum absolute value of the difference is denoted as DIS P :

8. DIS P As the greenhouse effect, the extreme monthly cumulative rainfall of this year and month under the p quantile; b. Repeat step a for all quantiles p to obtain the DIS corresponding to the extreme monthly cumulative rainfall under all quantiles P ; c. Select all quantiles p under DIS P The maximum value corresponds to the extreme monthly cumulative rainfall as a measure of the impact of the greenhouse effect on extreme rainfall in that year and month: 。

Images