POT drainage basin flood change estimation method and system under SSP-RCPs scene

Abstract

The invention belongs to the technical field of flood change estimation, and discloses a POT drainage basin flood change estimation method and system under an SSP-RCPs scene, and the method comprises the steps: obtaining CMIP6 downscaling processing data under a plurality of SSP-RCPs scenes; inputting rainfall and air temperature data, obtained after downscaling processing, of different scenes of the CMIP6 into a drainage basin distributed model, simulating a drainage basin daily runoff process, and obtaining daily runoff sequences of outlets of all sub-drainage basins and outlets of the whole drainage basin; based on the day-by-day runoff sequence of the reference period and the pre-estimation period under the selected climate change scene, constructing a flood sample by applying a POT method; applying the GP distribution function to carry out flood fitting and calculating a parameter value of the GP function; and based on the reference period GP function and the pre-estimated period GP function, comparing the flood recurrence period of the pre-estimated period with the flood recurrence period of the reference period, and analyzing the change of the future frequency of the future flood. According to the method, the CMIP6 downscaling climate data, the watershed hydrological model, the POT flood sampling and the GPD extreme value fitting method are comprehensively applied, and the method is high in universality and reliability.

Description

Technical Field

[0001] The present invention belongs to the technical field of flood change prediction, and particularly relates to a method and system for predicting POT basin flood changes under SSP-RCPs scenarios. Background Art

[0002] Climate change affects runoff by changing climate conditions such as precipitation and temperature, and may increase or decrease the risks of extreme hydrological events such as floods and droughts. Estimation of future flood intensity and flood recurrence period can provide reference information for long-term water resources planning, water conservancy project design, and formulation of strategies for coping with or adapting to climate change. Usually, the method of driving a hydrological model with the climate prediction results of a climate model is adopted to predict the basin runoff process under future climate change scenarios. Then, based on the runoff simulation results, flood samples are obtained, and the extreme value function is applied to fit the flood sample sequence, and further the possible changes of floods within a certain period under future climate change scenarios are estimated.

[0003] Global climate models are the main way to provide future climate change scenarios at present. Limited by the large simulation deviation and limited resolution of global climate models, global climate models have insufficient description of regional / basin climate characteristics and very limited simulation capabilities for individual climate phenomena and climate elements, resulting in large runoff simulation deviations when directly using the output of global climate models to drive regional hydrological models. Internationally, high-resolution and low-simulation-error climate data are usually obtained by downscaling global model simulations for driving hydrological models. The data of individual global coupled climate models in the latest released Sixth Coupled Model Intercomparison Project Phase 6 (CMIP6) internationally have a horizontal spatial resolution of more than 100 km, and the greenhouse gas emission scenarios adopted are SSP-RCPs (Shared Socioeconomic Pathway-Representative Concentration Pathways: SSP-RCPs), also known as climate change scenarios.

[0004] Basin hydrological models are an important way to carry out basin runoff simulation and provide future runoff sequences under climate change scenarios. Commonly used basin hydrological models can be divided into lumped hydrological models and distributed hydrological models according to the spatial description ability, and into conceptual hydrological models and physically based hydrological models according to the hydrological process description method. At present, hydrological models combining artificial intelligence and physical models have also been developed. Physically based distributed hydrological models are still the mainstream tools for current basin hydrological simulation.

[0005] When estimating future flood changes in a river basin, the reference period is often selected as 20 - 40 years, and the future estimation period usually ranges from 20 - 100 years. To improve the calculation accuracy of flood frequencies with longer return periods, it is necessary to obtain as much hydrological information as possible from the limited decades of hydrological data. The annual maximum value and Peak Over Threshold (POT) sampling methods are two commonly used flood sampling methods. Among them, the annual maximum value sampling may lead to missing some extreme flood events, with a small number of flood samples, affecting the accuracy of flood analysis. The Peak Over Threshold (POT) sampling can increase the flood sample sequence and better reflect flood characteristics.

[0006] Commonly used distribution functions for flood fitting include: Generalized Pareto (GP), Generalized Extreme Value (GEV), Pearson III (P - III), Exponential, and Wakeby.

[0007] Currently, when conducting future river basin flood estimation, either the CMIP5 climate model results before CMIP6 are used as the climate driver for the hydrological model, or CMIP6 climate drivers are used but without downscaling, or CMIP6 downscaled climate drivers are used but with annual maximum value flood sampling, or POT flood sampling technology is used but the generalized extreme value function is applied for flood frequency analysis. There is still a lack of a flood estimation method with strong versatility and high reliability that comprehensively optimizes multiple links such as high - resolution climate drivers, fine - process simulation of runoff, POT flood sampling, and GP extreme value fitting. Summary of the Invention

[0008] Aiming at the problems existing in the prior art, the present invention provides a POT river basin flood change estimation method and system under SSP - RCPs scenarios.

[0009] The present invention is implemented as follows. A POT river basin flood change estimation method under SSP - RCPs scenarios includes the following steps:

[0010] Step 1, obtain the downscaled data of multiple SSP - RCPs scenarios of the CMIP6 global climate model;

[0011] Step 2, input the precipitation and temperature data obtained from the CMIP6 downscaling of different SSP - RCPs scenarios into the river basin distributed model to simulate the daily runoff process of the river basin, and obtain the daily runoff sequences at the outlets of each sub - basin and the entire river basin outlet;

[0012] Step 3: Based on the daily runoff series of the reference period and the projection period at the outlets of each sub-basin and the whole basin under the selected climate change scenario, the POT method is applied to construct flood samples;

[0013] Step 4: Apply the GP distribution function to fit the flood series of the reference period and the projection period for each scenario, each time period, and each sub-basin (including the whole basin) and calculate the parameter values of the GP function;

[0014] Step 5: Based on the GP function of the reference period and the GP function of the projection period, compare the flood return periods of the projection period and the reference period, and analyze the changes in future flood frequencies.

[0015] Furthermore, in Step 1, collect and collate the daily data of precipitation, average temperature, maximum temperature, and minimum temperature with a resolution of 25 km or higher for the years 1961 - 2014 and 2015 - 2100 after CMIP6 downscaling for certain scenarios or a certain scenario among SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5.

[0016] Furthermore, in Step 2, input the precipitation and temperature data after CMIP6 downscaling under different RCP-SSPs scenarios into the distributed basin model to simulate the runoff of each sub-basin in the study area during 1961 - 2014 and the future scenarios of 2015 - 2100, analyze the model simulation results, and obtain the daily runoff series at the outlets of each sub-basin and the whole basin.

[0017] Furthermore, in Step 3, based on the daily runoff series of the reference period and the projection period at the outlets of each sub-basin and the whole basin under the selected climate change scenario, the POT method is applied to construct flood samples for each scenario, each time period, and each sub-basin (or basin) outlet at a frequency of 3 floods per year on average; the selection rule for the reference period is at least 20 consecutive years during 1961 - 2014, and the projection period is at least 20 consecutive years during 2025 - 2100.

[0018] Furthermore, in Step 4, methods such as the maximum likelihood method and the probability proportion method can be used to estimate the GP parameters.

[0019] Further, in step 5, according to the definitions of small floods, medium floods, large floods, and extreme floods, their recurrence periods are less than 5 years, greater than or equal to 5 years and less than 20 years, greater than or equal to 20 years and less than 50 years, and greater than 50 years, respectively; based on the reference period GP function, the flood volume of a certain recurrence period in the reference period is calculated for each sub-basin (including the entire basin); then, based on the GP function of the prediction period under a certain scenario, the recurrence period corresponding to this flood volume in the prediction period is calculated, and by comparing the flood recurrence periods in the prediction period and the reference period, the change in the future flood frequency is analyzed; at the same time, based on the reference period GP function, the flood volumes of the 20-year and 50-year recurrence periods in the reference period are calculated for each sub-basin (including the entire basin), and based on the flood volumes of the 20-year and 50-year recurrence periods in the prediction period under a certain scenario, the reduction in flood intensity is analyzed by comparing the difference in flood volumes between the prediction period and the reference period.

[0020] Another object of the present invention is to provide a method for predicting POT basin flood changes under SSP-RCPs scenarios and a system for predicting POT basin flood changes under SSP-RCPs scenarios, including:

[0021] A downscaling data acquisition module that acquires CMIP6 downscaled data under multiple SSP-RCPs scenarios;

[0022] A basin daily runoff process simulation module that inputs the precipitation and temperature data processed by CMIP6 downscaling under different RCP-SSPs scenarios into a basin distributed model to simulate the basin daily runoff process and obtain the daily runoff sequences at the outlets of each sub-basin and the entire basin;

[0023] A flood sample construction module that constructs flood samples using the POT method based on the reference period and prediction period daily runoff sequences at the outlets of each sub-basin and the entire basin under the selected climate change scenario;

[0024] A GP flood fitting and parameter estimation module that fits the reference period and prediction period flood sequences and calculates the parameter values of the GP function for each scenario, each time period, and each sub-basin (including the entire basin) using the GP distribution function;

[0025] A flood intensity and frequency change analysis module that compares the flood recurrence periods in the prediction period and the reference period based on the reference period GP function and the prediction period GP function to analyze the change in the future flood frequency.

[0026] Another object of the present invention is to provide a computer device. The computer device includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor executes the steps of the method for predicting POT basin flood changes under the SSP-RCP scenario.

[0027] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to execute the steps of the method for predicting POT basin flood changes in the SSP-RCPs scenario.

[0028] Another object of the present invention is to provide an information data processing terminal, the information data processing terminal including the system for predicting POT basin flood changes in the SSP-RCPs scenario.

[0029] Combined with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by the present invention are as follows:

[0030] The present invention provides a method for predicting basin floods in the SSP-RCPs scenario. Its advantage lies in the comprehensive application of CMIP6 downscaled climate data, basin hydrological models, POT flood sampling, and GPD extreme value fitting methods, forming a complete flood frequency prediction chain applicable to different climate scenarios. This method can provide a scientific basis for hydrological disaster prevention and mitigation under future climate change, and has strong versatility and reliability. Its innovation and technical advantages are mainly reflected in the following aspects.

[0031] First, this method uses the high-precision climate data after CMIP6 downscaling as the climate driving field of the hydrological model. The CMIP6 global climate models (GCMs) usually have a spatial resolution of more than 100 km. After downscaling, the resolution of the climate data can be increased to 25 km or even higher, making the spatial distribution of key climate variables such as temperature and precipitation more accurate. The Taylor score of the downscaled data is closer to 1 than the CMIP6 original data, indicating that its simulation deviation is smaller. This improved data input method improves the simulation accuracy of the basin hydrological model, reduces the uncertainty of the simulation results, and thus enhances the reliability of flood prediction.

[0032] Second, this method adopts the POT (Peak Over Threshold) flood sampling method. Compared with the traditional annual maximum (AM) method, the POT method can extract more flood samples and improve the effectiveness of statistical analysis. The annual maximum method only extracts the largest flood event every year, which may lead to instability in the estimation of flood frequencies with long return periods. The POT method extracts all flood events exceeding the threshold by setting the threshold, making the flood frequency curve more stable, especially suitable for evaluating the changing trend of flood risks under future extreme climates.

[0033] In addition, this method uses the two-parameter GP (Generalized Pareto) distribution for flood frequency fitting. In contrast, the GEV (Generalized Extreme Value distribution) and P-III distribution functions are mainly used for annual extreme flood simulation. Although the Wakeby function has high flexibility, due to its large number of parameters, it increases the estimation difficulty and computational complexity. The two-parameter GP distribution method selected by the present invention is simple to calculate and applicable to POT flood samples, which can effectively reduce the underestimation problem of large flood events, thereby improving the accuracy and applicability of flood prediction.

[0034] In addition to its technological innovation, the present invention also has significant economic and social value in practical applications. According to the "China's Water and Flood Disaster Prevention Bulletin 2023" released by the Ministry of Water Resources, in 2023, water conservancy project facilities in 30 provinces (autonomous regions, municipalities directly under the Central Government) across the country were damaged due to floods, with direct economic losses reaching up to 63.366 billion yuan. Since the accuracy of flood prediction directly affects the design of flood control projects and the optimization of flood warning systems, the present invention can provide more accurate climate-driven information and more reliable flood statistical analysis for basin flood management, improving the scientific nature and pertinence of flood control projects.

[0035] According to the national standard "Hydrological Information and Forecast Specification" (GB / T 22482-2008), accurate flood forecasting can reduce the direct economic losses caused by floods by about 5%-15%. Calculated based on the losses of national water conservancy facilities in 2023, the application of the present invention across the country is expected to reduce the direct economic losses of floods by 3.1-9.5 billion yuan annually. In addition, the present invention can also be applied to fields such as urban waterlogging management, farmland water conservancy planning, and optimization of emergency warning systems, providing high-value data support for multiple industries.

[0036] Generally speaking, the present invention not only has innovation in technical methods, but also has prominent application value. By improving the accuracy and reliability of flood prediction, the present invention can effectively serve climate change adaptation strategies, optimize flood control and disaster reduction plans, improve water resource management efficiency, and provide a solid guarantee for basin security and economic and social development. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flowchart of the method for predicting POT basin flood changes under the SSP-RCPs scenario provided by an embodiment of the present invention;

[0038] Figure 2 is a structural diagram of the system for predicting POT basin flood changes under the SSP-RCPs scenario provided by an embodiment of the present invention;

[0039] Figure 3 is a Taylor diagram and Taylor score comparison diagram of the CMIP6 global climate model and the annual average temperature of the Yellow River Basin after downscaling provided by an embodiment of the present invention;

[0040] Figure 4 It is the Taylor diagram and Taylor score comparison diagram of the annual precipitation and temperature of the CMIP6 global climate model provided by the embodiments of the present invention and the downscaled Yellow River Basin;

[0041] Figure 5 It is the schematic diagram of the return period of the future three estimated periods of the 30-year return period POT3 flood provided by the embodiments of the present invention;

[0042] Figure 6 It is the schematic diagram of the percentage change in the intensity of the 30-year return period POT3 flood relative to the reference period in the future three estimated periods provided by the embodiments of the present invention. Specific implementation manners

[0043] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0044] As Figure 1 shown, the embodiments of the present invention provide a method for predicting the change of POT basin flood under SSP-RCP scenarios, including the following steps:

[0045] Step 1, obtain the downscaled data of multiple SSP-RCPs scenarios of the CMIP6 global climate model;

[0046] Step 2, input the precipitation and temperature data after the CMIP6 downscaling under different SSP-RCPs scenarios into the basin distributed model to simulate the daily runoff process of the basin, and obtain the daily runoff sequences at the outlets of each sub-basin and the entire basin outlet;

[0047] Step 3, based on the daily runoff sequences of the reference period and the estimated period at the outlet of each sub-basin and the entire basin outlet under the selected climate change scenario, apply the POT method to construct flood samples;

[0048] Step 4, apply the GP distribution function to fit the flood sequences of the reference period and the estimated period for each scenario, each period, and each sub-basin (including the entire basin) one by one, and calculate the parameter values of the GP function;

[0049] Step 5, based on the GP function of the reference period and the GP function of the estimated period, compare the return periods of the floods in the estimated period and the reference period, and analyze the change of the future flood frequency.

[0050] This method is mainly based on climate simulation data under different SSP-RCP scenarios, combined with a watershed hydrological model and statistical methods, to evaluate the impact of future climate change on POT (Peaks Over Threshold) flood events. First, in step 1, the method obtains a variety of SSP-RCP (Shared Socioeconomic Pathway-Representative Concentration Pathway) climate scenario data provided by CMIP6 (Coupled Model Intercomparison Project Phase 6) and downscales it to improve the spatial resolution of the data, making it suitable for hydrological simulations at the watershed scale.

[0051] In step 2, the downscaled CMIP6 meteorological data (including rainfall and temperature) are input into the distributed watershed hydrological model to simulate the hydrological processes of the target watershed. The model calculates the daily runoff process based on the climate input data and generates daily runoff sequences at the sub-watershed and the outlet of the entire watershed, laying a data foundation for flood frequency analysis. This process ensures that future hydrological conditions under different SSP-RCP scenarios are reasonably simulated for subsequent flood assessments.

[0052] In step 3, this method uses the POT (Peaks Over Threshold) method to extract flood events from the simulated daily runoff sequences. The POT method selects flood data that exceed a set threshold as the standard to construct a flood sample. By selecting an appropriate threshold, the representativeness of the flood data is ensured, making the analysis of future flood trends more accurate. This step constructs flood samples for both the reference period (historical period) and the projection period (future period) for subsequent statistical analysis.

[0053] In steps 4 and 5, this method uses the Generalized Pareto (GP) distribution function to fit the flood samples and calculate their parameter values. By fitting the GP distribution to the flood events in the reference period and the projection period respectively, the statistical characteristics of flood frequency can be quantified. Further, this method compares the changes in flood return periods between the reference period and the projection period to evaluate the changes in the frequency of future flood occurrences. This analysis can reveal the changing trends of flood risks under different SSP-RCP climate scenarios, providing a scientific basis for watershed flood management and disaster prevention and mitigation.

[0054] Furthermore, in step 1, daily data of precipitation, mean temperature, maximum temperature, and minimum temperature with a resolution of 25 km or higher for the years 1961 - 2014 and 2015 - 2100, which are downscaled by CMIP6 for certain scenarios or a certain scenario among SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5, are collected and sorted.

[0055] Further, in step 2, the precipitation and temperature data after downscaling of CMIP6 under different RCP-SSPs scenarios are input into the distributed basin model to simulate the runoff of each sub-basin in the study area during 1961-2014 and the future scenario of 2015-2100, and the simulation results of the model are analyzed to obtain the daily runoff series at the outlets of each sub-basin and the entire basin outlet.

[0056] The basin hydrological model is the distributed hydrological model of the study area that has been verified by parameter calibration to ensure the accuracy and reliability of runoff simulation. The downscaled data can be sourced from publicly available datasets on the Internet or processed independently.

[0057] Further, in step 3, based on the reference period and projected period daily runoff series at the outlet of each sub-basin and the entire basin outlet under the selected climate change scenario, the POT method is applied to construct flood samples for each scenario, each time period, and each sub-basin (or basin) outlet at a frequency of 3 floods per year on average; the reference period selection rule is at least 20 consecutive years during 1961-2014, and the projected period is at least 20 consecutive years during 2025-2100; for example, the entire period of 2025-2100 can also be divided into 2025-2060 and 2065-2100, or other concerned time periods. The POT flood sampling can be carried out using the patent "A POT Flood Automatic Extraction Method and System" (ZL201810350505.9), or other POT flood sampling methods can also be adopted.

[0058] Further, in step 4, methods such as the maximum likelihood method and probability weight method can be used to estimate the GP parameters.

[0059] Further, in step 5, according to the definitions of small floods, medium floods, large floods, and extremely large floods, their recurrence intervals are less than 5 years, greater than or equal to 5 years and less than 20 years, greater than or equal to 20 years and less than 50 years, and greater than 50 years respectively; based on the reference period GP function, the flood volume corresponding to a certain recurrence interval in the reference period is calculated for each sub-basin (including the entire basin); then, based on the projected period GP function under a certain scenario, the recurrence interval corresponding to this flood volume in the projected time period is calculated, and by comparing the flood recurrence intervals in the projected period and the reference period, the change in future flood frequency is analyzed; at the same time, based on the reference period GP function, the flood volumes corresponding to recurrence intervals of 20 years, 30 years, and 50 years in the reference period are calculated for each sub-basin (including the entire basin), and based on the flood volumes corresponding to recurrence intervals of 20 years, 30 years, and 50 years in the projected period under a certain scenario, the reduction in flood intensity is analyzed by comparing the differences in flood volumes between the projected period and the reference period.

[0060] As Figure 2 shown, the embodiment of the present invention provides a POT basin flood change prediction method under SSP-RCPs scenarios and a POT basin flood change prediction system under SSP-RCPs scenarios, including:

[0061] The downscaling data acquisition module acquires the CMIP6 downscaled data under multiple SSP-RCPs scenarios;

[0062] The basin daily runoff process simulation module inputs the precipitation and temperature data obtained by CMIP6 downscaling under different RCP-SSPs scenarios into the basin distributed model to simulate the basin daily runoff process and obtain the daily runoff sequences at the outlets of each sub-basin and the entire basin outlet;

[0063] The flood sample construction module constructs flood samples by applying the POT method based on the daily runoff sequences of the reference period and the projection period at the outlets of each sub-basin and the entire basin outlet under the selected climate change scenario;

[0064] The GP flood fitting and parameter estimation module applies the GP distribution function to fit the flood sequences of the reference period and the projection period for each scenario, each time period, and each sub-basin (including the entire basin) and calculates the parameter values of the GP function;

[0065] The flood intensity and frequency change analysis module compares the flood return periods of the projection period and the reference period based on the GP functions of the reference period and the projection period, and analyzes the changes in future flood frequencies.

[0066] Based on the present invention, it is possible to estimate the changes in flood intensity, frequency and phase relative to the baseline period under future climate change scenarios for any basin, which can be used for the feasibility demonstration of basin water conservancy project design, for formulating the impact and adaptation countermeasures of climate change on the basin or water conservancy project, and for flood risk management of basin or water conservancy project climate change.

[0067] Taking the flood prediction of the Huayuankou in the Yellow River as an example, the downscaled daily precipitation and average temperature data of 13 global climate models of CMIP6 from 1961 to 2014 and three climate change scenarios of SSP1-2.6, SSP2-4.5, and SSP5-8.5 from 2015 to 2100 are used ( Figure 3 ), driving the semi-distributed hydrological model of the upper and middle reaches of the Yellow River ( Figure 4 ), simulating the daily runoff from 1961 to 2014 and from 2021 to 2100 under the three climate change scenarios. Based on these runoff simulation results, flood samples are extracted at a frequency of 3 floods per year on average, and the GP function is used to fit the floods in the reference period of 1986–2010 of the 13 climate models and in the three projection periods of 2026–2050, 2051–2075, and 2076–2100 under the three climate changes of SSP1-2.6, SSP2-4.5, and SSP5-8.5 and calculate the corresponding parameters.

[0068] Based on the GP function of the reference period, calculate the flood volumes with 20-year and 50-year return periods at the Huayuankou Station, the outlet of the basin. Then, based on the GP functions for the periods 2026–2050, 2051–2075, and 2076–2100 under SSP1-2.6, SSP2-4.5, and SSP5-8.5, calculate the return periods corresponding to the 20-year and 50-year flood volumes for the three future projection periods ( Figure 5 ). In most future scenarios, the flood frequency increases, while in a small number of scenarios, it decreases. Calculate the flood volumes with 20-year and 50-year return periods at Huayuankou during the reference and projection periods ( Figure 6 ). In most scenarios, the flood intensity increases, while in a small number of scenarios, it weakens. It can be seen that the flood risk increases in most cases, and only decreases in some scenarios.

[0069] Example 1: Basin Flood Risk Assessment and Flood Control Planning

[0070] In a flood risk assessment project for a tributary of the Yangtze River Basin, researchers used this method to predict the future flood change trends under different climate scenarios. First, obtain the downscaled meteorological data for the SSP2-4.5 and SSP5-8.5 scenarios provided by CMIP6, including the future temperature and precipitation change trends. Then, input the data into the VIC (Variable Infiltration Capacity) distributed hydrological model to simulate the daily runoff in the basin and generate the daily runoff sequences for each sub-basin.

[0071] Subsequently, the research team used the POT method to extract the flood event samples for the past 50 years and perform GP distribution fitting for the flood events in the reference period (1980 - 2010) and the projection period (2030 - 2060). By calculating the flood return periods, the research results show that under the high-emission scenario (SSP5-8.5), a once-in-a-century flood event may occur twice within 50 years in the future, and the flood frequency increases by about 30%. Based on this analysis, the water conservancy department optimized the flood control infrastructure construction plan, increased the design standards of dams and river embankments, and adjusted the urban drainage system to enhance the flood resistance of the basin.

[0072] Example 2: Agricultural Irrigation Management and Water Resources Regulation

[0073] In an agricultural water resources management project in the North China Plain, the present invention was used to analyze the impact of future climate change on agricultural water use safety. Researchers obtained the downscaled meteorological data for the two climate scenarios of SSP1-2.6 and SSP3-7.0 and, in combination with a distributed hydrological model such as the SWAT (Soil and Water Assessment Tool) model, simulated the basin runoff process to evaluate the future changes in water resources supply.

[0074] Based on the POT method, flood events of key irrigation water supply rivers in the study area were extracted, and the return period changes of future floods were analyzed by combining with the GP distribution. The results show that under the high emission scenario (SSP3-7.0), the occurrence frequency of flood events increases, while under the low emission scenario (SSP1-2.6), the occurrence frequency of flood events is relatively stable. Further analysis shows that the instability of future water resource supply increases. Especially in the case of an increase in extreme rainfall events in summer, it may lead to an overabundance or insufficiency of agricultural irrigation water supply. Based on this study, the agricultural management department optimized the reservoir operation plan, formulated adaptive irrigation strategies, and promoted intelligent water-saving irrigation technologies to improve the water resource utilization efficiency and ensure the stability of agricultural production.

[0075] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention by those skilled in the art within the technical scope disclosed by the present invention shall be covered by the protection scope of the present invention.

Claims

1. A method for predicting POT basin flood changes under SSP-RCPs scenarios, characterized in that It includes the following steps: Step 1: Obtain the downscaled data of multiple SSP-RCPs scenarios of the CMIP6 global climate model; Step 2: Input the precipitation and temperature data after downscaling of CMIP6 under different SSP-RCPs scenarios into the basin distributed model to simulate the daily runoff process of the basin, and obtain the daily runoff series at the outlets of each sub-basin and the whole basin; Step 3: Based on the daily runoff series of the reference period and the prediction period at the outlet of each sub-basin and the whole basin under the selected climate change scenario, apply the POT method to construct flood samples; Step 4: Apply the GP distribution function to fit the flood series of the reference period and the prediction period for each scenario, each time period, and each sub-basin (including the whole basin) one by one, and calculate the parameter values of the GP function; Step 5: Based on the GP function of the reference period and the GP function of the prediction period, compare the flood return periods of the prediction period and the reference period, and analyze the change of future flood frequency.

2. The method for predicting POT basin flood changes in the SSP-RCPs scenario according to claim 1, wherein In Step 1, collect and organize the daily data of precipitation, average temperature, maximum temperature, and minimum temperature with a resolution of 25 km or higher for the years 1961-2014 and 2015-2100 after downscaling of CMIP6 for certain scenarios or a certain scenario among SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.

5.

3. The method for predicting POT basin flood changes in the SSP-RCPs scenario according to claim 1, wherein, In Step 2, input the precipitation and temperature data after downscaling of CMIP6 under different RCP-SSPs scenarios into the basin distributed model to simulate the runoff of each sub-basin in the study area during 1961-2014 and the future scenario of 2015-2100, analyze the model simulation results, and obtain the daily runoff series at the outlets of each sub-basin and the whole basin.

4. The method for predicting POT basin flood changes in the SSP-RCPs scenario according to claim 1, wherein In Step 3, based on the daily runoff series of the reference period and the prediction period at the outlet of each sub-basin and the whole basin under the selected climate change scenario, apply the POT method to construct flood samples for each scenario, each time period, and each sub-basin (or basin) outlet at a frequency of 3 floods per year on average; the selection rule for the reference period is at least 20 consecutive years during 1961-2014, and the prediction period is at least 20 consecutive years during 2025-2100.

5. The method for predicting POT basin flood changes in the SSP-RCPs scenario according to claim 1, characterized in that, In Step 4, methods such as the maximum likelihood method and the probability proportion method can be used to estimate the GP parameters.

6. The method for predicting POT basin flood changes in the SSP-RCPs scenario according to claim 1, wherein, In Step 5, according to the definitions of small floods, medium floods, large floods, and extremely large floods, their return periods are less than 5 years, greater than or equal to 5 years and less than 20 years, greater than or equal to 20 years and less than 50 years, and greater than 50 years respectively; based on the GP function of the reference period, calculate the flood volume of a certain return period in the reference period for each sub-basin (including the whole basin); Then, based on the GP function of the prediction period under a certain scenario, calculate the return period corresponding to this flood volume during the prediction period, and analyze the change of future flood frequency by comparing the flood return periods of the prediction period and the reference period; at the same time, based on the GP function of the reference period, calculate the flood volumes of 20-year, 30-year, 50-year or other return periods for each sub-basin (including the whole basin), and based on the flood volumes of 20-year, 30-year, 50-year or other return periods under a certain scenario of the prediction period, analyze the reduction of flood intensity by comparing the difference in flood volumes between the prediction period and the reference period.

7. A POT basin flood change prediction system in the SSP-RCPs scenario of the method according to any one of claims 1 to 6, characterized in that, It includes: The downscaling data acquisition module acquires the CMIP6 downscaled data under multiple SSP-RCPs scenarios; The basin daily runoff process simulation module inputs the precipitation and temperature data obtained by CMIP6 downscaling under different RCP-SSPs scenarios into the basin distributed model to simulate the basin daily runoff process, and obtains the daily runoff sequences at the outlets of each sub-basin and the whole basin; The flood sample construction module constructs flood samples by applying the POT method based on the daily runoff sequences of the reference period and the projected period at the outlets of each sub-basin and the whole basin under the selected climate change scenario; The GP flood fitting and parameter estimation module applies the GP distribution function to fit the flood sequences of the reference period and the projected period scenario by scenario, period by period, and sub-basin (including the whole basin) by sub-basin, and calculates the parameter values of the GP function; The flood intensity and frequency change analysis module compares the flood return periods of the projected period and the reference period based on the GP functions of the reference period and the projected period, and analyzes the changes in future flood frequencies; 8. A computer device, characterized in that, The computer device includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor executes the steps of the POT basin flood change prediction method under the SSP-RCPs scenario according to any one of claims 1 to 6; 9. A computer-readable storage medium, characterized in that, A computer program is stored. When the computer program is executed by the processor, the processor executes the steps of the POT basin flood change prediction method under the SSP-RCPs scenario according to any one of claims 1 to 6; 10. An information data processing terminal, characterized in that, The information data processing terminal includes the POT basin flood change prediction system under the SSP-RCPs scenario according to claim 7;

Images