Method and system for evaluating human climate impact on high mountain runoff, and computer device

Abstract

The application discloses a method and system for evaluating human climate influence on runoff in high mountain areas and a computer device, wherein after a daily runoff evaluation model is trained according to to-be-measured data of a measurement area, historical simulation data affected by human influence and natural simulation data not affected by human influence are input into the daily runoff evaluation model for calculation, so that historical simulation runoff and natural simulation runoff are output, and finally, whether the runoff is affected by human influence is determined by comparing the difference between the historical simulation runoff and the natural simulation runoff. After the simulation runoff with human influence and the simulation runoff without human influence are simulated, the two are compared, whether the runoff in the measurement area is affected by human activities is determined, and the problem that the existing quantification method of the influence of the glacier basin runoff in the high mountain area does not consider the environmental influence of the climate change caused by human activities since the industrial revolution in the region is solved.

Description

Technical Field

[0001] This application relates to the field of model generation, and in particular to a method, system and computer equipment for assessing the anthropogenic climate impact of runoff in high-altitude areas. Background Technology

[0002] Quantifying the impact of human activities on glacial watershed runoff under climate change is crucial for addressing water resources and water-related disasters in high-altitude mountainous regions. However, current methods for quantifying watershed runoff primarily rely on comparing historical baseline data with runoff observations or simulations from periods of frequent human activity. These methods mainly consider local human activities, such as land use change and irrigation data, but they neglect the global nature of climate change and fail to account for the regional differences and environmental impacts of climate change caused by global human activities since the Industrial Revolution. Therefore, developing a quantitative assessment method that incorporates the impact of anthropogenic emissions on glacial watershed runoff has become an urgent problem to solve. Summary of the Invention

[0003] Based on this, it is necessary to propose a method, system, and computer equipment for assessing the anthropogenic climate impacts of high-altitude runoff in light of the above-mentioned problems, which can take into account the impact of climate change caused by anthropogenic emissions since the Industrial Revolution on runoff in glacial basins.

[0004] This application provides a method for assessing the anthropogenic climate impacts of runoff in high-altitude areas, including:

[0005] Obtain the set of test data for the test area, wherein the set of test data is a collection of test data arranged in a set time series within the test area;

[0006] A set learning model is obtained, and the learning model is trained using the test data to output a daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction based on the input meteorological variables.

[0007] Historical simulation data and natural simulation data are obtained from a set database. The historical simulation parameters are climate simulation data of the area to be tested after human disturbance, and the natural simulation parameters are climate simulation data of the area to be tested without human disturbance.

[0008] The historical simulation data is simulated based on the daily runoff simulation model to output the historical simulated runoff.

[0009] The natural simulation data is simulated based on the daily runoff assessment model to output the natural simulated runoff.

[0010] A t-test was performed on the difference between the historical simulated runoff and the natural simulated runoff, and the p-value output after the t-test was used to determine whether the difference reached a significant level.

[0011] If so, it is determined that the runoff in the area to be measured is affected by human-induced global climate change.

[0012] Furthermore, the step of acquiring the set learning model and training the learning model using the test data to output a daily runoff assessment model, which is used to predict runoff based on the input parameters, specifically includes:

[0013] The dataset to be tested is split according to the set time series to generate a training dataset and a validation dataset.

[0014] The learning model is trained using the training dataset to output a model to be evaluated.

[0015] The validation dataset is input into the model to be evaluated for iteration, and the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated are obtained.

[0016] Determine whether the determination coefficient is greater than a preset threshold;

[0017] If so, then the model to be evaluated is the daily runoff evaluation model.

[0018] Furthermore, the step of inputting the validation dataset into the model to be evaluated for iteration and obtaining the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated specifically includes:

[0019] Acquire set observation data, wherein the set observation data is observation data that is consistent with the data acquisition parameter type in the verification dataset;

[0020] The validation dataset is input into the model to be evaluated for iteration, thereby outputting model evaluation data;

[0021] Obtain the set efficiency coefficient calculation formula, and calculate the Nse efficiency coefficient between the model evaluation data and the observed data according to the efficiency coefficient calculation formula;

[0022] The output Nse efficiency coefficient is denoted as the judgment coefficient.

[0023] Furthermore, after the step of inputting the validation dataset into the model to be evaluated for iteration, thereby outputting model evaluation data, the method further includes:

[0024] Obtain the set formula for calculating the coefficient of certainty, and calculate the coefficient of certainty between the model evaluation data and the observed data based on the formula.

[0025] The output deterministic coefficients are denoted as the judgment coefficients.

[0026] Furthermore, the data to be measured includes rainfall parameters, temperature parameters, and runoff parameters of the area to be measured at each time point within the set time series;

[0027] Before the step of obtaining the set of test data for the test area, wherein the set of test data is the set of test data arranged according to a set time series within the test area, the specific steps include:

[0028] The temperature parameters are obtained, and a set degree-day model is used to convert them into meltwater depth parameters that match the temperature parameters. The degree-day model is specifically manifested as follows:

[0029]

[0030] Wherein, the meltwater depth parameter is defined as the water temperature parameter, the time step is defined as a set time step, and the degree-day factor is defined as a set time step.

[0031] The meltwater depth parameter is converted into a runoff depth parameter using a predefined flow conversion formula, which is specifically described as follows:

[0032]

[0033] Wherein, the runoff parameter is the area parameter of the area to be measured, and the runoff depth parameter is the area parameter of the area to be measured.

[0034] The meltwater depth parameter and the runoff depth parameter are added to the data to be measured to generate the data set to be measured.

[0035] Furthermore, the step of simulating the historical simulation data based on the daily runoff assessment model to output the historical simulated runoff specifically includes:

[0036] The historical simulation data are then averaged to generate the first weighted dataset.

[0037] The first weighted dataset is input into the daily runoff assessment model for simulation, thereby outputting the historical simulated runoff;

[0038] The step of simulating the natural simulation data based on the daily runoff assessment model to output the natural simulated runoff specifically includes:

[0039] The natural simulation data are then ensembled and averaged to generate a second weighted dataset.

[0040] The second weighted dataset is input into the daily runoff assessment model for simulation, thereby outputting the natural simulated runoff.

[0041] Furthermore, the step of performing a t-test on the difference between the historical simulated runoff and the natural simulated runoff, and determining whether the difference reaches a significant level by using the p-value output after the t-test, specifically includes:

[0042] The mean and standard deviation of each runoff parameter in the historical simulated runoff are obtained and denoted as the first mean and the first standard deviation.

[0043] The mean and standard deviation of each runoff parameter in the natural simulated runoff are obtained and denoted as the second mean and the second standard deviation.

[0044] The p-value is calculated using a set t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, and then outputs the corresponding p-value.

[0045] Obtain the set significance level value, and look up the matching critical value from the set critical value table based on the significance level value;

[0046] Determine whether the p value is less than the critical value;

[0047] If so, the difference between the historical simulated runoff and the natural simulated runoff is determined to be significant.

[0048] Furthermore, after the step of calculating and outputting the corresponding p-value using a predefined t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, the method further includes:

[0049] A probability distribution map is generated based on each of the calculated p values, and the probability value of each of the p values ​​exceeding the set limit value is calculated within the probability distribution map.

[0050] Determine whether the probability value exceeds a set threshold;

[0051] If so, proceed with the step of determining whether the runoff in the area to be measured is affected by human-induced global climate change.

[0052] This application also provides an anthropogenic climate impact assessment system for high-altitude runoff, including:

[0053] The first acquisition unit is used to acquire a set of test data for the test area, wherein the set of test data is a set of test data arranged in a set time series within the test area.

[0054] The training unit is used to acquire a set learning model and train the learning model using the test data to output a daily runoff assessment model, which is used to perform runoff inversion and prediction based on the input meteorological variables.

[0055] The second acquisition unit is used to acquire historical simulation data and natural simulation data from a set database. The historical simulation parameters are climate simulation data of the area to be tested after human disturbance, and the natural simulation parameters are climate simulation data of the area to be tested without human disturbance.

[0056] The first simulation unit is used to simulate the historical simulation data according to the daily runoff simulation model, thereby outputting the historical simulated runoff.

[0057] The second simulation unit is used to simulate the natural simulation data according to the daily runoff assessment model, thereby outputting the natural simulated runoff;

[0058] The test unit is used to perform a t-test on the difference between the historical simulated runoff and the natural simulated runoff, and to determine whether the difference reaches the significance level by the p-value output after the t-test.

[0059] The judgment unit is used to determine whether the runoff in the area to be measured is affected by human-induced global climate change.

[0060] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps:

[0061] Obtain the set of test data for the test area, wherein the set of test data is a collection of test data arranged in a set time series within the test area;

[0062] A set learning model is obtained, and the learning model is trained using the test data to output a daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction based on the input meteorological variables.

[0063] Historical simulation data and natural simulation data are obtained from a set database. The historical simulation parameters are climate simulation data of the area to be tested after human disturbance, and the natural simulation parameters are climate simulation data of the area to be tested without human disturbance.

[0064] The historical simulation data is simulated based on the daily runoff simulation model to output the historical simulated runoff.

[0065] The natural simulation data is simulated based on the daily runoff assessment model to output the natural simulated runoff.

[0066] A t-test was performed on the difference between the historical simulated runoff and the natural simulated runoff, and the p-value output after the t-test was used to determine whether the difference reached a significant level.

[0067] If so, it is determined that the runoff in the area to be measured is affected by human-induced global climate change.

[0068] A computer-readable medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:

[0069] Obtain the set of test data for the test area, wherein the set of test data is a collection of test data arranged in a set time series within the test area;

[0070] A set learning model is obtained, and the learning model is trained using the test data to output a daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction based on the input meteorological variables.

[0071] Historical simulation data and natural simulation data are obtained from a set database. The historical simulation parameters are climate simulation data of the area to be tested after human disturbance, and the natural simulation parameters are climate simulation data of the area to be tested without human disturbance.

[0072] The historical simulation data is simulated based on the daily runoff simulation model to output the historical simulated runoff.

[0073] The natural simulation data is simulated based on the daily runoff assessment model to output the natural simulated runoff.

[0074] A t-test was performed on the difference between the historical simulated runoff and the natural simulated runoff, and the p-value output after the t-test was used to determine whether the difference reached a significant level.

[0075] If so, it is determined that the runoff in the area to be measured is affected by human-induced global climate change.

[0076] This application, through the aforementioned method and structure, trains a daily runoff simulation model based on the data to be measured in the measurement area. It then inputs historical simulation data affected by human activities and natural simulation data unaffected by human activities into the daily runoff simulation model for calculation, thereby outputting historical simulated runoff and natural simulated runoff. Finally, by comparing the differences between the historical simulated runoff and the natural simulated runoff, it determines whether the runoff is affected by human activities. This method achieves a judgment on whether the runoff in the measurement area is affected by human activities by comparing simulated runoff with and without human influence. It solves the problem that existing methods for quantifying the impact of watershed runoff do not consider the regional differences in climate change caused by global human activities since the Industrial Revolution and its environmental impact, thus improving the accuracy of judging whether runoff is affected by global climate change. Attached Figure Description

[0077] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0078] in:

[0079] Figure 1 This is a flowchart of a method for assessing the anthropogenic climate impacts of runoff in high-altitude areas, as shown in one embodiment.

[0080] Figure 2 This is a performance graph of the daily runoff assessment model described in one embodiment;

[0081] Figure 3 This is a comparison chart of the probability of extreme runoff events in one embodiment;

[0082] Figure 4 This is a schematic diagram of the structure of an anthropogenic climate impact assessment system for runoff in high-altitude areas, as shown in one embodiment.

[0083] Figure 5 This is a structural block diagram of a computer device in one embodiment;

[0084] The labels in the diagram are as follows: 1 - First acquisition unit, 2 - Training unit, 3 - Second acquisition unit, 4 - First simulation unit, 5 - Second simulation unit, 6 - Verification unit, 7 - Judgment unit. Detailed Implementation

[0085] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0086] refer to Figure 1 This application provides a method for assessing the anthropogenic climate impacts of runoff in high-altitude areas, including:

[0087] S1. Obtain the set of test data for the test area, wherein the set of test data is a set of test data arranged in a set time series within the test area;

[0088] As described in step S1 above, the background system obtains the set of test data for the test area. It can be understood that the test area is the high-altitude mountainous area where glacier runoff prediction is required. The background system obtains the test data at each time point and arranges the test data according to the time point to output the set of test data.

[0089] S2. Obtain the set learning model and train the learning model using the test data to output the daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction for the input meteorological variables.

[0090] As described in the above embodiments, the background system inputs all the data to be measured into a set learning model for calculation, thereby outputting the corresponding daily runoff assessment model. It can be understood that the learning model is a LightGBM deep learning model built using the sklearn library of Python 3.9, and the daily runoff assessment model is used to perform runoff inversion and prediction for the input meteorological variables.

[0091] S3. Obtain historical simulation data and natural simulation data from the set database. The historical simulation parameters are climate simulation data of the area to be tested after human disturbance, and the natural simulation parameters are climate simulation data of the area to be tested without human disturbance.

[0092] As described in step S3 above, the acquisition of historical simulation data and natural simulation data from the designated database can be understood as follows: the designated database is a simulation results sharing website utilizing the CMIP6 multi-model comparison program. In the CMIP6 phase, simulations were conducted from 1850 to 2014, using the last year of the piControl experiment (a pre-industrial revolution simulation experiment) as the initial conditions. External forcing conditions were directly derived from observations, including solar activity, volcanic activity (volcanic aerosols), and atmospheric composition (greenhouse gases and aerosols) and land-use change caused by human activities. Natural simulation (hist-nat) refers to simulations that only consider natural processes and do not include anthropogenic disturbances. Therefore, collecting diurnal simulation results from multiple models with these two types of scenarios constitutes the historical simulation data and the natural simulation data.

[0093] S4. Simulate the historical simulation data according to the daily runoff simulation model, and output the historical simulated runoff;

[0094] As described in step S4 above, the background system inputs the historical simulation data into the daily runoff simulation model for simulation, thereby outputting the historical simulated runoff affected by anthropogenic global climate change.

[0095] S5. Simulate the natural simulation data according to the daily runoff assessment model, and output the natural simulation runoff;

[0096] As described in step S5 above, the background system inputs the natural simulation data into the daily runoff simulation model for simulation, thereby outputting the natural simulation data that is not affected by anthropogenic global climate change.

[0097] S6. Perform a t-test on the difference between the historical simulated runoff and the natural simulated runoff, and determine whether the difference reaches the significance level by using the p-value output after the t-test.

[0098] As described in step S6 above, the background system performs a t-test based on the historical simulated runoff and the natural simulated runoff. It is understood that the t-test is a commonly used statistical test method used to compare whether there is a significant difference between the means of two samples. It calculates a t-value based on the sample mean and standard deviation, and then, based on the degrees of freedom and significance level, looks up or calculates the corresponding critical value to determine whether there is a significant difference between the two sample means. The background system then obtains the t-value output by the t-test and determines whether the t-value is greater than the set critical value.

[0099] S7. If so, it is determined that the runoff in the area to be measured is affected by human-induced global climate change.

[0100] As described in step S7 above, when the t value is greater than the set critical value, the background system determines that the runoff in the area to be measured is affected by human activities.

[0101] It is understandable that, in this embodiment, since the area to be tested is a high-altitude mountainous region that requires glacier runoff prediction, the runoff in the area to be tested is actually glacier runoff in the high-altitude mountainous region. Therefore, the background system determines that the glacier runoff in the high-altitude mountainous region is affected by human-induced global climate change.

[0102] In summary, the greatest benefit of this application lies in the following: After training a daily runoff simulation model based on the data to be measured in the measurement area, historical simulation data affected by human activities and natural simulation data unaffected by human activities are input into the daily runoff simulation model for calculation, thereby outputting historical simulated runoff and natural simulated runoff. Finally, by comparing the differences between historical simulated runoff and natural simulated runoff, the method of determining whether the runoff is affected by human activities is achieved. This method solves the problem that existing methods for quantifying the impact of watershed runoff do not consider the regional differences in climate change caused by global human activities since the Industrial Revolution and its environmental impact, thereby improving the accuracy of judging whether runoff is affected by human global climate change.

[0103] In one embodiment, step S2 specifically includes:

[0104] S21. The set of data to be tested is split according to the set time series to generate a training dataset and a validation dataset. The learning model is trained using the training dataset to output an evaluation model. The validation dataset is input into the evaluation model for iteration, and the judgment coefficient of the runoff simulation prediction curve finally output by the evaluation model is obtained. It is determined whether the judgment coefficient is greater than a preset threshold. If so, the evaluation model is the daily runoff evaluation model.

[0105] As described in the above embodiments, the backend system splits the set time series, thereby splitting the dataset to be tested into the training dataset and the validation dataset. Then, the backend system uses the training dataset to train the learning model, thereby outputting a preliminary model to be evaluated. After obtaining the model to be evaluated, the backend system uses the validation dataset to iteratively validate the model to be evaluated, and outputs the runoff simulation prediction curve during the iterative validation process. Then, the backend system calculates the judgment coefficient of the runoff simulation prediction curve and determines whether the judgment coefficient is greater than a preset threshold. If so, the model to be evaluated is the daily runoff evaluation model.

[0106] It is understood that in this embodiment, the preset threshold can be set by the user, and this application does not limit the specific size of the preset threshold.

[0107] In one embodiment, the step of inputting the validation dataset into the model to be evaluated for iteration and obtaining the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated specifically includes:

[0108] S22. Obtain set observation data, wherein the set observation data is observation data with the same data acquisition parameter type as the verification dataset. Input the verification dataset into the model to be evaluated for iteration, thereby outputting model evaluation data. Obtain the set efficiency coefficient calculation formula, and calculate the Nse efficiency coefficient between the model evaluation data and the observation data according to the efficiency coefficient calculation formula. Record the output Nse efficiency coefficient as the judgment coefficient.

[0109] As described in the above embodiments, the backend system acquires the observation data that is consistent with the data acquisition parameter type in the verification dataset. Then, the backend system inputs the verification dataset into the model to be evaluated for iteration, thereby outputting the model evaluation data. After that, the backend system acquires the set efficiency coefficient calculation formula. It can be understood that, in this embodiment, the efficiency coefficient calculation formula is actually a Nash-Sutcliffe efficiency coefficient (NES) efficiency coefficient calculation formula. The backend system calculates the NES efficiency coefficient according to the NES efficiency coefficient calculation formula and records the NES efficiency coefficient as the judgment coefficient.

[0110] In this embodiment, the preset threshold in the above embodiment is set to Nse > 0.5. When the Nse efficiency coefficient is > 0.5, the background system determines that the model to be evaluated is the daily runoff evaluation model.

[0111] In one embodiment, after the step of inputting the validation dataset into the model to be evaluated for iteration, thereby outputting model evaluation data, the method further includes:

[0112] Obtain the set formula for calculating the coefficient of certainty, and calculate the coefficient of certainty between the model evaluation data and the observed data based on the formula.

[0113] The output deterministic coefficients are denoted as the judgment coefficients.

[0114] As described in the above embodiments, the background system obtains the set deterministic coefficient calculation formula. It can be understood that the deterministic coefficient calculation formula is a commonly used statistical formula used to evaluate the degree of fit of the regression model to the observed data. After the background system uses the deterministic coefficient calculation formula to calculate the deterministic coefficient R2 between the model evaluation data and the observed data, the output deterministic coefficient R2 is recorded as the judgment coefficient.

[0115] In this embodiment, the preset threshold in the above embodiment is set to R2 > 0.6. When the certainty coefficient R2 > 0.6, the background system determines that the model to be evaluated is the daily runoff evaluation model.

[0116] In addition, the background system can also make a judgment by combining the deterministic coefficient R2 and the Nse efficiency coefficient. Specifically, when the Nse efficiency coefficient is greater than 0.5 and the deterministic coefficient R2 is greater than 0.6, the background system will determine the model to be evaluated as the daily runoff evaluation model.

[0117] In one embodiment, the data to be measured includes rainfall parameters, temperature parameters, and runoff parameters of the area to be measured at each time point within the set time series.

[0118] Before step S1, the specific procedures include:

[0119] S11. Obtain the temperature parameters, and use the set day-per-day model to convert them into meltwater depth parameters that match the temperature parameters. The day-per-day model is specifically manifested as follows:

[0120]

[0121] Among them, the The meltwater depth parameter is T+, the air temperature parameter is Δt, the time step is a set value, and the degree-day factor is a set value.

[0122] S12. Convert the meltwater depth parameter into a runoff depth parameter using a set flow conversion formula. The flow conversion formula is specifically as follows:

[0123]

[0124] Among them, the For the runoff parameters, the Dt represents the area parameter of the region to be measured, and Dt represents the runoff depth parameter.

[0125] S13. Add the meltwater depth parameter and the runoff depth parameter to the data to be measured, thereby generating the data set to be measured.

[0126] As described in the above embodiments, the data to be measured includes rainfall parameters, temperature parameters, and runoff parameters for the area to be measured at each time point within the set time series. The background system then acquires the temperature parameters and, based on these parameters, uses a set degree-day model to convert them into meltwater depth parameters for runoff within the area to be measured. Specifically, the degree-day model is as follows:

[0127]

[0128] Among them, the The meltwater depth parameter is T+, the air temperature parameter is Δt, the time step is a set value, and the degree-day factor is a set value.

[0129] Then when the backend system outputs the daily living model... The meltwater depth parameter is denoted as the meltwater depth parameter, and converted into a runoff depth parameter using a set flow conversion formula. The flow conversion formula is specifically embodied as follows:

[0130]

[0131] Among them, the For the runoff parameters, the Dt represents the area parameter of the region to be measured, and Dt represents the runoff depth parameter.

[0132] Then when the backend system outputs the traffic conversion formula... Let be the runoff parameter, and let be the runoff parameter. and the The data is input into the test data to generate the test data set.

[0133] In one embodiment, step S4 specifically includes:

[0134] S41. The historical simulation data are aggregated and averaged to generate a first weighted dataset. The first weighted dataset is then input into the daily runoff assessment model for simulation, thereby outputting the historical simulated runoff.

[0135] Step S5 specifically includes:

[0136] S51. The natural simulation data is ensembled and averaged to generate a second weighted dataset. The second weighted dataset is then input into the daily runoff assessment model for simulation, thereby outputting the natural simulation runoff.

[0137] As described in the above embodiments, the backend system performs a cascaded average of the historical simulation data to generate a first weighted dataset, inputs the first weighted dataset into the daily runoff assessment model for simulation, and outputs the historical simulated runoff. At the same time, the backend system also performs a cascaded average of the historical simulation data to generate a first weighted dataset, inputs the first weighted dataset into the daily runoff assessment model for simulation, and outputs the historical simulated runoff.

[0138] In one embodiment, step S6 specifically includes:

[0139] S61. Obtain the mean and standard deviation of each runoff parameter in the historical simulated runoff, and record them as the first mean and the first standard deviation. Obtain the mean and standard deviation of each runoff parameter in the natural simulated runoff, and record them as the second mean and the second standard deviation. Calculate the p-value using a set t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, and output the corresponding p-value. Obtain a set significance level value, and look up the matching critical value from a set critical value table based on the significance level value. Determine whether the p-value is less than the critical value. If so, determine that the difference between the historical simulated runoff and the natural simulated runoff reaches a significant level.

[0140] Furthermore, after the step of calculating and outputting the corresponding p-value using a predefined t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, the method further includes:

[0141] S62. Generate a probability distribution map based on the calculated p values, and calculate the probability value of each p value exceeding the set limit value in the probability distribution map. Determine whether the probability value exceeds the set threshold. If so, execute the step of determining that the runoff in the area to be measured is affected by human-induced global climate change.

[0142] As described in the above embodiments, the background system obtains the mean and standard deviation of each runoff parameter in the historical simulated runoff and records them as the first mean and the first standard deviation. It also obtains the mean and standard deviation of each runoff parameter in the natural simulated runoff and records them as the second mean and the second standard deviation. Based on the first mean, the first standard deviation, the second mean, and the second standard deviation, a set t-value estimation formula is used to estimate and output the corresponding p-value. It can be understood that the t-test calculated by the t-value estimation formula is a commonly used statistical test method to compare whether there is a significant difference between the means of two samples. The t-value estimation formula is specifically expressed as t = (x1-x2) / sqrt(s1² / n1 + s2² / n2), where x1 and x2 are the means of the two samples, s1 and s2 are the standard deviations of the two samples, and n1 and n2 are the sample sizes of the two samples.

[0143] At this point, the background system uses a set t-value estimation formula to calculate the p-value based on the first mean, the first standard deviation, the second mean, and the second standard deviation, thereby outputting the corresponding p-value. Then, the background system obtains a set significance level value. In this embodiment, the significance level value is 0.05. The background system then looks up the matching critical value from a set critical value table based on the significance level value and determines whether the t-value is less than the critical value. If so, it is determined that the difference between the historical simulated runoff and the natural simulated runoff has reached a significant level.

[0144] Furthermore, after the step of calculating and outputting the corresponding p-value using a predefined t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, the method further includes:

[0145] A probability distribution map is generated based on the calculated p values, and the probability value of each p value exceeding the set limit value within the probability distribution map is calculated. It is then determined whether the probability value exceeds the set threshold. If so, the step of determining whether the runoff in the area to be tested is affected by human-induced global climate change is executed. This enables the background system to determine whether glacial runoff in high-altitude mountainous areas is affected by human-induced global climate change, either separately or simultaneously, based on the specific value of the t value and the probability value of the t value exceeding the set limit value.

[0146] Furthermore, for all embodiments mentioned in this application, the following are corresponding usage examples:

[0147] The target area was selected as a glacial watershed on the eastern slope of the Qinghai-Tibet Plateau. Limited observed runoff data from Gonghe Village at the downstream outlet from 2017 to 2020, combined with collected rainfall and temperature data, were used to construct the target data set. The backend system then split this data set into a training dataset and a validation dataset. Using the training dataset, a LightGBM-based daily runoff estimation model for the glacial watershed was constructed. The simulation results during the model commissioning and validation periods are attached. Figure 2 As shown, the attached Figure 2 'a' represents the performance of the model when built using the training dataset. Figure 2 b represents the model's performance when using the validation dataset for estimation. The results show that the constructed model can reasonably estimate daily runoff flow and its changes.

[0148] The subsequent backend system, through statistical analysis of extreme events, obtained the distribution of the annual occurrence frequency of extreme events from 1850 to 2014. Based on a comparison of runoff sequences driven by climate conditions with and without human activity influence, the output results are shown in the appendix. Figure 3 As shown;

[0149] Among them, the appendix Figure 3 Differences and trends in the ten-year average number of extreme runoff events with and without human activity (HIST) and without human activity (NAT): Appendix Figure 3 (a) and appendix Figure 3 (c) Statistics based on percentages exceeding 95%; Appendix Figure 3 (b) and appendix Figure 3 (d) Based on statistics exceeding the 90% percentage threshold, it can be seen that global human activities affect local climate, and the number of extreme runoff events shows a downward trend compared to the scenario without human activity.

[0150] refer to Figure 4 This application also provides an anthropogenic climate impact assessment system for high-altitude runoff, including:

[0151] The first acquisition unit 1 is used to acquire a set of test data for the test area, wherein the set of test data is a set of test data arranged in a set time series within the test area.

[0152] Training unit 2 is used to acquire a set learning model and train the learning model using the test data to output a daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction based on the input meteorological variables.

[0153] The second acquisition unit 3 is used to acquire historical simulation data and natural simulation data from a set database. The historical simulation parameters are climate simulation data of the area to be tested after human disturbance, and the natural simulation parameters are climate simulation data of the area to be tested without human disturbance.

[0154] The first simulation unit 4 is used to simulate the historical simulation data according to the daily runoff simulation model, thereby outputting the historical simulated runoff.

[0155] The second simulation unit 5 is used to simulate the natural simulation data according to the daily runoff assessment model, thereby outputting the natural simulation runoff;

[0156] The testing unit 6 is used to perform a t-test on the difference between the historical simulated runoff and the natural simulated runoff, and to determine whether the difference reaches the significance level by the p-value output after the t-test.

[0157] Judgment unit 7 is used to determine if the runoff in the area to be measured is affected by human-induced global climate change.

[0158] The modules mentioned above are for implementing the anthropogenic climate impact assessment system for runoff in high-altitude areas, and will not be described in detail here.

[0159] Figure 5 An internal structural diagram of a computer device in one embodiment is shown. This computer device may specifically be a server, including but not limited to high-performance computers and high-performance computer clusters. Figure 5 As shown, the computer device includes a processor, memory, and network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement the anthropogenic climate impact assessment method for high-altitude runoff. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to implement the anthropogenic climate impact assessment system for high-altitude runoff.

[0160] In one embodiment, the method for assessing the anthropogenic climate impacts of high-altitude runoff provided by the present invention can be implemented as a computer program, which can be implemented in, for example... Figure 5 The system runs on the computer device shown. The computer device's memory can store the various program templates that make up the anthropogenic climate impact assessment system for high-altitude runoff. For example: 1 – First acquisition unit, 2 – Training unit, 3 – Second acquisition unit, 4 – First simulation unit, 5 – Second simulation unit, 6 – Verification unit, 7 – Judgment unit.

[0161] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the following steps:

[0162] A set of test data for the test area is obtained. This set of test data is a collection of test data arranged according to a set time series within the test area. A set learning model is obtained, and the test data is used to train the learning model, thereby outputting a daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction for input meteorological variables. Historical simulation data and natural simulation data are obtained from a set database. The historical simulation parameters are climate simulation data of the test area after human disturbance, and the natural simulation parameters are climate simulation data of the test area without human disturbance. The historical simulation data is simulated according to the daily runoff simulation model, thereby outputting historical simulated runoff. The natural simulation data is simulated according to the daily runoff assessment model, thereby outputting natural simulated runoff. A t-test is performed on the difference between the historical simulated runoff and the natural simulated runoff, and the p-value output after the t-test is used to determine whether the difference reaches a significant level. If so, it is determined that the runoff in the test area is affected by human-induced global climate change.

[0163] As can be seen from the above embodiments, the greatest benefit of this application lies in the fact that, after training a daily runoff simulation model based on the data to be measured in the measurement area, historical simulation data affected by human activities and natural simulation data unaffected by human activities are input into the daily runoff simulation model for calculation, thereby outputting historical simulated runoff and natural simulated runoff. Finally, by comparing the differences between historical simulated runoff and natural simulated runoff, the method of determining whether the runoff is affected by human activities is realized. This method achieves the judgment of whether the runoff in the measurement area is affected by human activities by comparing simulated runoff with and without human influence. It solves the problem that the existing methods of quantifying the impact of watershed runoff do not consider the regional differences in the manifestation of climate change caused by global human activities since the Industrial Revolution and its environmental impact, thereby improving the accuracy of judging whether the runoff is affected by human global climate change.

[0164] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, the above embodiments are not described in detail.

[0165] All possible combinations of the various technical features in the examples are described; however, only if these technical features are...

[0166] There is no contradiction in the combinations, and all combinations should be considered to be within the scope of this specification.

[0167] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for assessing the anthropogenic climate impact of runoff in high-altitude areas, characterized in that, include: Obtain the set of test data for the test area, wherein the set of test data is a collection of test data arranged in a set time series within the test area; A set learning model is obtained, and the learning model is trained using the test data to output a daily runoff assessment model. The daily runoff assessment model is used to perform runoff inversion and prediction based on the input meteorological variables. Historical simulation data and natural simulation data are obtained from a set database. The historical simulation data is climate simulation data of the area to be tested after human disturbance, and the natural simulation data is climate simulation data of the area to be tested without human disturbance. The historical simulation data is simulated based on the daily runoff assessment model to output the historical simulated runoff. The natural simulation data is simulated based on the daily runoff assessment model to output the natural simulated runoff. A t-test was performed on the difference between the historical simulated runoff and the natural simulated runoff, and the p-value output after the t-test was used to determine whether the difference reached a significant level. If so, it is determined that the runoff in the area to be measured is affected by human-induced global climate change; The steps of acquiring a set learning model and training the learning model using the test data to output a daily runoff assessment model, wherein the daily runoff assessment model is used for runoff inversion and prediction based on input meteorological variables, specifically include: The dataset to be tested is split according to the set time series to generate a training dataset and a validation dataset. The learning model is trained using the training dataset to output a model to be evaluated. The validation dataset is input into the model to be evaluated for iteration, and the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated are obtained. Determine whether the determination coefficient is greater than a preset threshold; If so, then the model to be evaluated is the daily runoff evaluation model; The step of inputting the validation dataset into the model to be evaluated for iteration and obtaining the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated specifically includes: Acquire set observation data, wherein the set observation data is observation data that is consistent with the data acquisition parameter type in the verification dataset; The validation dataset is input into the model to be evaluated for iteration, thereby outputting model evaluation data; Obtain the set efficiency coefficient calculation formula, and calculate the Nse efficiency coefficient between the model evaluation data and the observation data based on the efficiency coefficient calculation formula; The output Nse efficiency coefficient is denoted as the judgment coefficient.

2. The method for assessing the anthropogenic climate impact of runoff in high-altitude areas as described in claim 1, characterized in that, After the step of inputting the validation dataset into the model to be evaluated for iteration, thereby outputting model evaluation data, the method further includes: Obtain the set formula for calculating the coefficient of certainty, and calculate the coefficient of certainty between the model evaluation data and the observed data based on the formula. The output deterministic coefficients are denoted as the judgment coefficients.

3. The method for assessing the anthropogenic climate impact of runoff in high-altitude areas as described in claim 1, characterized in that, The data to be measured includes rainfall parameters, temperature parameters, and runoff parameters of the area to be measured at each time point within the set time series. Before the step of obtaining the set of test data for the test area, wherein the set of test data is the set of test data arranged according to a set time series within the test area, the specific steps include: The temperature parameters are obtained, and a set degree-day model is used to convert them into meltwater depth parameters that match the temperature parameters. The degree-day model is specifically manifested as follows: ； Among them, the The meltwater depth parameter, T + The temperature parameter is defined as Δt, which is a set time step, and the DDF is a set degree-day factor. The meltwater depth parameter is converted into a runoff depth parameter using a predefined flow conversion formula, which is specifically described as follows: ； Among them, the For the runoff parameters, the Dt represents the area parameter of the region to be measured, and Dt represents the runoff depth parameter. The meltwater depth parameter and the runoff depth parameter are added to the data to be measured to generate the data set to be measured.

4. The method for assessing the anthropogenic climate impact of runoff in high-altitude areas as described in claim 1, characterized in that, The step of simulating the historical simulation data based on the daily runoff assessment model to output the historical simulated runoff specifically includes: The historical simulation data are then averaged to generate the first weighted dataset. The first weighted dataset is input into the daily runoff assessment model for simulation, thereby outputting the historical simulated runoff; The step of simulating the natural simulation data based on the daily runoff assessment model to output the natural simulated runoff specifically includes: The natural simulation data are then ensembled and averaged to generate a second weighted dataset. The second weighted dataset is input into the daily runoff assessment model for simulation, thereby outputting the natural simulated runoff.

5. The method for assessing the anthropogenic climate impact of runoff in high-altitude areas as described in claim 4, characterized in that, The step of performing a t-test on the difference between the historical simulated runoff and the natural simulated runoff, and determining whether the difference reaches a significant level by using the p-value output after the t-test, specifically includes: Obtain the mean and standard deviation of each runoff parameter in the historical simulated runoff, and record them as the first mean and the first standard deviation; Obtain the mean and standard deviation of each runoff parameter in the natural simulated runoff, and record them as the second mean and the second standard deviation; The p-value is calculated using a set t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, and then outputs the corresponding p-value. Obtain the set significance level value, and look up the matching critical value from the set critical value table based on the significance level value; Determine whether the p value is less than the critical value; If so, the difference between the historical simulated runoff and the natural simulated runoff is determined to be significant. Furthermore, after the step of calculating and outputting the corresponding p-value using a predefined t-value estimation formula based on the first mean, the first standard deviation, the second mean, and the second standard deviation, the method further includes: A probability distribution map is generated based on each of the calculated p values, and the probability value of each of the p values ​​exceeding the set limit value is calculated within the probability distribution map. Determine whether the probability value exceeds a set threshold; If so, proceed with the step of determining whether the runoff in the area to be measured is affected by human-induced global climate change.

6. A system for assessing the anthropogenic climate impact of runoff in high-altitude areas, characterized in that, include: The first acquisition unit is used to acquire a set of test data for the test area, wherein the set of test data is a set of test data arranged in a set time series within the test area. The training unit is used to acquire a set learning model and train the learning model using the test data to output a daily runoff assessment model, which is used to perform runoff inversion and prediction based on the input meteorological variables. The second acquisition unit is used to acquire historical simulation data and natural simulation data from a set database. The historical simulation data is climate simulation data of the area to be tested after human disturbance, and the natural simulation data is climate simulation data of the area to be tested without human disturbance. The first simulation unit is used to simulate the historical simulation data according to the daily runoff assessment model, thereby outputting the historical simulated runoff; The second simulation unit is used to simulate the natural simulation data according to the daily runoff assessment model, thereby outputting the natural simulated runoff; The test unit is used to perform a t-test on the difference between the historical simulated runoff and the natural simulated runoff, and to determine whether the difference reaches the significance level by the p-value output after the t-test. The judgment unit is used to determine if the runoff in the area to be measured is affected by human-induced global climate change. The training unit is used to acquire a predefined learning model and train the model using the test data to output a daily runoff assessment model. This model is used to perform runoff inversion and prediction based on input meteorological variables, specifically including: The dataset to be tested is split according to the set time series to generate a training dataset and a validation dataset. The learning model is trained using the training dataset to output a model to be evaluated. The validation dataset is input into the model to be evaluated for iteration, and the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated are obtained. Determine whether the determination coefficient is greater than a preset threshold; If so, then the model to be evaluated is the daily runoff evaluation model; The step of inputting the validation dataset into the model to be evaluated for iteration and obtaining the judgment coefficients of the runoff simulation prediction curve finally output by the model to be evaluated specifically includes: Acquire set observation data, wherein the set observation data is observation data that is consistent with the data acquisition parameter type in the verification dataset; The validation dataset is input into the model to be evaluated for iteration, thereby outputting model evaluation data; Obtain the set efficiency coefficient calculation formula, and calculate the Nse efficiency coefficient between the model evaluation data and the observation data based on the efficiency coefficient calculation formula; The output Nse efficiency coefficient is denoted as the judgment coefficient.

7. A computer device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements a method for assessing the anthropogenic climate impact of runoff in high-altitude areas as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a method for assessing the anthropogenic climate impact of runoff in high-altitude areas as described in any one of claims 1 to 5.

Images