A method and system for predicting daily runoff interval in alpine region considering the influence of climate change

Abstract

The application discloses a kind of high-cold area daily runoff interval prediction method and system considering the influence of climate change, method includes: collecting hydro-meteorological data, dividing hydrological unit and constructing each unit water system topological relation;Three water source Xinanjiang model is constructed to be subsequently applied in each hydrological unit;Based on degree-day factor method, build snowmelt module embedded into Xinanjiang model, and include basin snow and snowmelt runoff in the calculation of basin precipitation-runoff;Based on hydrological statistics, construct uncertainty module, select the model effective parameter set suitable for basin and drive model output prediction flow set, and then estimate the prediction flow interval of a certain confidence level, as daily runoff interval prediction result.The application effectively depicts the influence of hydro-meteorological uncertainty on runoff change, and makes up the deficiency that traditional hydrological model is poor in high-cold area high water flow prediction, can provide technical support for water resources management and cascade reservoir group scheduling in high-cold area.

Description

Technical Field

[0001] This invention relates to the field of water resources management and planning, and in particular to a method and system for predicting runoff flow. Background Technology

[0002] Runoff forecasting technology is a crucial support for water resource planning and management, as well as flood and drought control. Under the backdrop of climate change, the water cycle is altering, the probability of extreme hydrological events is increasing, and the uncertainty in runoff forecasting is becoming more prominent. High-altitude and cold regions, in particular, possess unique plateau and mountain climates and glacial meltwater runoff characteristics, making them sensitive to climate change. Furthermore, the complex mechanisms of runoff formation limit the accuracy of traditional deterministic hydrological runoff forecasting methods. Therefore, there is an urgent need to develop inter-regional runoff forecasting methods and systems for high-altitude and cold regions that can simultaneously consider snowmelt runoff and the impact of uncertainties. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to propose a method and system for predicting daily runoff intervals in high-altitude and cold regions that takes into account the impact of climate change, effectively characterizing the influence of hydrological and meteorological uncertainties on runoff changes, and making up for the shortcomings of traditional hydrological models in predicting high water flow in high-altitude and cold regions.

[0004] Technical solution: A method for predicting daily runoff intervals in high-altitude and cold regions considering the impact of climate change, comprising the following steps:

[0005] Step S1: Collect flow, precipitation, evaporation and temperature data observed by hydrological and meteorological stations in the study area, and use ArcGIS software to divide hydrological units based on digital elevation information and construct the water system topology of each unit;

[0006] Step S2: Construct the Xin'anjiang River model with three water sources, including modules for evapotranspiration calculation, runoff calculation, division of the three water sources, and confluence calculation. Based on the topological relationship of the water system of each hydrological unit, construct a river flow calculation module to obtain the total outflow calculation results of the basin.

[0007] Step S3: Construct a snowmelt module based on the degree-day factor method and embed it into the Xin'anjiang model. Incorporate watershed snow accumulation and snowmelt runoff into the calculation of watershed precipitation-runoff, making the improved Xin'anjiang model suitable for daily runoff prediction in high-altitude and cold regions.

[0008] Step S4: Construct an uncertainty module based on hydrological statistics, select the effective parameter set of the model applicable to the watershed, drive the improved Xin'anjiang model to output the predicted flow set, estimate the model prediction range at a specified confidence level, and obtain the daily runoff interval prediction results.

[0009] Further, step S1 includes:

[0010] Step S1a: Use the TauDEM plugin in ArcGIS to divide the hydrological units, and select the average flow calculated by Grid NetworkOrder as the threshold to extract the water system in the study area;

[0011] Step S1b: The Thiessen polygon method is used to calculate the corresponding weights of the meteorological station in each hydrological unit. Based on this, a weighted average is performed to obtain the areal precipitation, evaporation and temperature of each hydrological unit as inputs to the hydrological model.

[0012] Step S1c: Analyze the topological relationship of each hydrological unit, and summarize the process of calculating the river flow unit by unit from upstream to downstream based on the topological relationship of each hydrological unit. This will facilitate the subsequent input of the total precipitation runoff calculation results of each hydrological unit to obtain the total discharge of the basin outlet.

[0013] Furthermore, in step S3, the construction of the snow melting module includes:

[0014] Step S3a: Utilize two critical temperatures and To distinguish precipitation patterns, when the temperature < At that time, the precipitation was in the form of snowfall; when the temperature... > At that time, the precipitation form is rain; when the temperature... At that time, the precipitation pattern was a mixture of rain and snow. When performing model calculations, the observed daily average temperature was used as the basis. Determine the amount of snowfall for the day. ;

[0015] Step S3b: Unlike the direct runoff calculation for rainfall, snowfall first becomes part of the watershed snow cover, and the amount of snow cover in the watershed during the time period is calculated accordingly. The calculation formula is:

[0016]

[0017] in, This represents the initial snow accumulation in the watershed, in mm.

[0018] Step S3c: The precipitation amount after deducting snowfall is the rainfall amount. Therefore, the precipitation amount needs to be corrected to obtain the corrected precipitation amount. The corrected formula is as follows, representing the actual rainfall:

[0019]

[0020] Step S3d: Based on the different sources of heat absorbed by the snow, snow melting is divided into high-temperature snow melting, which is caused by heat from the air, and rain-induced snow melting, which is caused by heat carried by rainfall. The snow melting amounts of these two parts are calculated separately and then added together to obtain the total snow melting amount.

[0021] Furthermore, in step S3a, the daily snowfall... The specific calculation formula is as follows:

[0022]

[0023] in, This represents the precipitation amount during the time period, expressed in mm.

[0024] Furthermore, in step S3d, the total snowmelt amount is calculated using the degree-day factor method, and the specific formula is as follows:

[0025]

[0026] in, Total snowmelt amount, in mm; The day factor is expressed in mm / (°C). d); Snow surface temperature, in °C, is the temperature of the snow surface when the total snow melts. Greater than the watershed snow cover At that time, take .

[0027] Further, step S4 includes:

[0028] Step S4a: Assuming the prior distribution of the model parameters is uniform, the Monte Carlo method is used to randomly sample the model parameters, and 10,000 parameter sets are obtained by randomly sampling within the parameter range.

[0029] Step S4b: Determine the likelihood function and set the threshold for the likelihood function value. Input the 10,000 sets of parameters randomly sampled in step S4a into the model, calculate the set of likelihood function values ​​for each set of parameters corresponding to each hydrological station, and filter out the effective parameter sets according to the threshold.

[0030] Step S4c: Drive the improved Xin'anjiang model with the selected effective parameter group and sort the calculation results to obtain the cumulative likelihood distribution of the forecast flow at different times. Select the 5% and 95% fractions as the upper and lower limits of the interval prediction range as the estimation results of the interval prediction.

[0031] Furthermore, in step S4b, the certainty coefficient, which reflects the degree of overlap between the measured flow rate and the simulated forecast flow rate process, is used as the likelihood function. The threshold value of the likelihood function is set to 0.5, and its function expression is as follows:

[0032]

[0033] in, It is the likelihood value of the i-th parameter group; Let V be the variance of the model prediction error for the i-th parameter group; It is the variance of the measured values ​​of the i-th parameter group.

[0034] This invention also provides a daily runoff interval prediction system for high-altitude and cold regions that takes into account the impact of climate change, comprising:

[0035] The data preprocessing module collects flow, precipitation, evaporation and temperature data observed by hydrological and meteorological stations in the study area, and uses ArcGIS software to divide hydrological units based on digital elevation information and construct the water system topology of each unit.

[0036] The hydrological model construction module constructs a three-source Xin'anjiang model, including modules for evapotranspiration calculation, runoff calculation, three-source division and confluence calculation, and constructs a river flow calculation module based on the topological relationship of the water system of each hydrological unit to obtain the total outflow calculation results of the basin.

[0037] The hydrological model optimization module, based on the degree-day factor method, constructs a snowmelt module and embeds it into the Xin'anjiang model. This incorporates watershed snow accumulation and snowmelt runoff into the calculation of watershed precipitation-runoff, making the improved Xin'anjiang model suitable for daily runoff prediction in high-altitude and cold regions.

[0038] The daily runoff interval prediction module constructs an uncertainty module based on hydrological statistics, selects the effective parameter set of the model applicable to the watershed, drives the improved Xin'anjiang model to output the predicted flow set, and estimates the model prediction range at a specified confidence level to obtain the daily runoff interval prediction results.

[0039] The present invention also provides a computer device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, wherein when the programs are executed by the processors, they implement the steps of the method for predicting daily runoff intervals in high-altitude cold regions considering the impact of climate change as described above.

[0040] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting daily runoff intervals in high-altitude and cold regions considering the impact of climate change as described above.

[0041] Beneficial effects: Compared with the prior art, the beneficial effects of the present invention are as follows:

[0042] (1) The present invention constructs a snow melting module, taking into account the unique snow melting runoff characteristics of high-altitude and cold regions, and incorporates the snow melting process into the rainfall runoff calculation process, thereby improving the applicability of the hydrological model for predicting daily runoff in high-altitude and cold regions.

[0043] (2) The present invention constructs an uncertainty module to quantify the comprehensive uncertainty impact of climate change, hydrological and meteorological conditions in the study area and the uncertainty of the model structure itself on the runoff prediction results from the perspective of hydrological statistics, thus enriching the runoff prediction information.

[0044] (3) The present invention couples the snow melting module and the uncertainty module to predict the daily runoff interval, which enriches the theoretical library of runoff prediction methods in high-altitude and cold regions and makes up for the shortcomings of traditional prediction methods in high water flow prediction. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the overall process flow of the method of the present invention;

[0046] Figure 2 This is a schematic diagram illustrating the specific calculation process of the present invention;

[0047] Figure 3 This is a graph showing the prediction results for a 90% confidence interval in an embodiment of the present invention. Detailed Implementation

[0048] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0049] This invention proposes a method for predicting daily runoff intervals in high-altitude and cold regions, taking into account the impact of climate change. Figure 1 and Figure 2 The specific implementation methods of each step included in the method are as follows:

[0050] Step S1: Collect flow, precipitation, evaporation and temperature data from hydrological and meteorological stations in the study area, and use ArcGIS software to divide hydrological units based on digital elevation information and construct the water system topology of each unit.

[0051] The specific steps are as follows:

[0052] Step S1a: Use the TauDEM plugin in ArcGIS to divide the hydrological units, and select the average flow calculated by Grid NetworkOrder as the threshold to extract the water system in the study area;

[0053] Step S1b: The Thiessen polygon method is used to calculate the corresponding weights of the meteorological station in each hydrological unit. Based on this, a weighted average is performed to obtain the areal precipitation, evaporation and temperature of each hydrological unit as inputs for the subsequent step S2 to construct the Xin'anjiang model.

[0054] The Thiessen polygon method was used to divide all meteorological stations into weights for each hydrological unit. For all meteorological stations in each hydrological unit, a weighted average was calculated for precipitation, evaporation, and temperature according to the weights. Meteorological stations within a hydrological unit were always included in the weighted average calculation. Even if a meteorological station was not within a hydrological unit, but was located close to it, it was still included in the weighted average calculation for that hydrological unit if the weight obtained by the Thiessen polygon method was not 0. The weighted average calculation process for precipitation, evaporation, and temperature for each meteorological station was consistent.

[0055] Step S1c: Analyze the topological relationship of each hydrological unit, and summarize the process of calculating the river flow unit by unit from upstream to downstream based on the topological relationship of each hydrological unit. This will facilitate the subsequent input of the total precipitation runoff calculation results of each hydrological unit to obtain the total discharge of the basin outlet.

[0056] Furthermore, in step S1c, the Muskingan method is used to calculate the total discharge from the hydrological unit outlet to the watershed outlet, based on the distance of the river segment from the hydrological unit outlet to the watershed outlet. (km) and flood propagation time (h) Estimate the Muskingan parameters for each hydrological unit. and The range was determined, and the parameters were calibrated and verified. The relevant calculation formulas are as follows:

[0057]

[0058]

[0059] in, For the riverbed gradient, and This represents the numerical value of a cross-section under steady flow conditions, which can be calculated based on measured hydrological data.

[0060] Step S2: Construct the Xin'anjiang River model with three water sources, including modules for evapotranspiration calculation, runoff calculation, division of the three water sources and confluence calculation, and construct a river flow calculation module based on the topological relationship of the water system of each hydrological unit to obtain the total outflow calculation results of the basin.

[0061] The Xin'anjiang model calculates the total precipitation runoff for each hydrological unit. Generally, the total precipitation runoff of the upstream hydrological unit is the outflow. For downstream hydrological units, the outflow comes partly from the unit's total precipitation runoff (calculated by the Xin'anjiang model) and partly from the outflow entering the unit from upstream (calculated from the upstream unit's outflow through the river channel flow). If the two hydrological units are connected in series, for example, the total outflow from hydrological unit A enters hydrological unit B, then the total outflow Th from hydrological unit A will be used. AThe flow rate Th at the outlet of hydrological unit B was calculated using the river flow calculation method. A-B The total precipitation runoff (Th) calculated by the Xin'anjiang model for hydrological unit B is compared with that of hydrological unit B. Bcal The outlet flow Th of hydrological unit B is obtained by superposition. B If the outflows of two hydrological units are in parallel and their outflows are unrelated, they can be calculated simultaneously at the watershed outlet section using the river flow calculation method and superimposed to participate in the calculation of the total outflow of the entire watershed.

[0062] Step S2 specifically includes the following steps:

[0063] Step S2a: Divide the soil into three layers according to the vertical heterogeneity of the soil, and calculate the evapotranspiration using a three-layer evapotranspiration model.

[0064] Step S2b: The full-soil-water-runoff model is adopted, which assumes that at any location in the watershed, before the soil moisture content reaches full (i.e., field capacity), all rainfall replenishes the soil moisture content and does not generate runoff; when the soil is full, all subsequent rainfall generates runoff.

[0065] Step S2c: Using a free-water reservoir structure, the three water sources are divided. First, the total runoff is calculated according to the full-storage runoff model. The water enters the free water reservoir for storage and regulation, and is then divided into surface runoff. underground runoff He Rang Zhong Liu Three parts.

[0066] Step S2d: The calculation of the confluence of the three water sources is divided into three parts: surface runoff confluence, interflow confluence, and groundwater runoff confluence, all of which adopt the linear reservoir algorithm. After the three types of runoff are calculated separately, the confluence calculation of the river network per unit area is then performed.

[0067] Step S2e: After calculating each hydrological unit according to steps S2a~S2d, the total precipitation runoff of each hydrological unit is obtained. This runoff is then substituted into the river flow calculation process constructed in step S1c and used as the river flow calculation module of the Xin'anjiang model to obtain the total discharge calculation result of the basin outlet.

[0068] Step S3: Construct a snowmelt module based on the degree-day factor method and embed it into the Xin'anjiang model. This incorporates watershed snow accumulation and snowmelt runoff into the calculation of watershed precipitation-runoff, making the model suitable for daily runoff prediction in high-altitude and cold regions.

[0069] Specifically, the following steps are included:

[0070] Step S3a: Precipitation can be divided into rainfall and snowfall according to its form. Assuming two critical temperatures... and When the temperature < At that time, the precipitation was in the form of snowfall; when the temperature... > At that time, the precipitation form is rain; when the temperature... At that time, the precipitation pattern was a mixture of rain and snow. When performing model calculations, the observed daily average temperature was used as the basis. The amount of snowfall on that day can be determined. .

[0071] Furthermore, in step S3a, the daily snowfall... The specific calculation formula is as follows:

[0072]

[0073] in, This represents the precipitation amount during the time period, expressed in mm.

[0074] Step S3b: Unlike the direct runoff calculation for rainfall, snowfall first becomes part of the watershed snow cover, and the amount of snow cover in the watershed during the time period is calculated accordingly. The calculation formula is:

[0075]

[0076] in, This represents the initial snow accumulation in the watershed, expressed in mm.

[0077] Step S3c: The precipitation amount after deducting snowfall is the rainfall amount. Therefore, the precipitation amount needs to be corrected to obtain the corrected precipitation amount. The corrected formula is as follows, representing the actual rainfall:

[0078]

[0079] Step S3d: Based on the different sources of heat absorbed by the snow, snow melting can be divided into high-temperature snow melting, which is caused by heat from the air, and rain-induced snow melting, which is caused by heat carried by rainfall. The total snow melting amount is calculated and added together to obtain the total snow melting amount.

[0080] Furthermore, in step S3d, the total snowmelt amount is calculated using the degree-day factor method, and the specific formula is as follows:

[0081]

[0082] in, This is the total snowmelt volume, expressed in mm. This total snowmelt volume is a theoretically calculated value, and is therefore also called the theoretical snowmelt volume. The day factor is expressed in mm / (°C). d); This refers to the snow surface temperature, expressed in °C, typically taken as a constant of 0, based on the theoretical snow melt volume. Greater than the watershed snow cover At that time, take .

[0083] Step S3e: After completing the snowmelt calculation, the snowmelt module is embedded into the Xin'anjiang model. The specific processing method is as follows: when the rainfall is 0, the total snowmelt is taken as the input of the Xin'anjiang model for runoff generation and confluence calculation; when the rainfall is not 0, the rainfall is added to the total snowmelt and runoff generation and confluence calculation is performed.

[0084] Step S4: Construct an uncertainty module based on hydrological statistics, select the effective set of model parameters applicable to the watershed, drive the improved model to output the predicted flow set, estimate the model prediction range at a certain confidence level, and obtain the daily runoff interval prediction results.

[0085] Specifically, the steps include the following:

[0086] Step S4a: Assuming the prior distribution of the model parameters is uniform, the Monte Carlo method is used to randomly sample the model parameters, and 10,000 parameter sets are obtained by randomly sampling within the parameter range.

[0087] Step S4b: Determine the likelihood function and set the threshold for the likelihood function value. Input the 10,000 sets of parameters randomly sampled in step S4a into the model, calculate the set of likelihood function values ​​for each set of parameters corresponding to each hydrological station, and filter out the effective parameter sets according to the threshold.

[0088] Furthermore, in step S4b, the certainty coefficient, which reflects the degree of overlap between the measured flow rate and the simulated forecast flow rate process, is used as the likelihood function. The threshold value of the likelihood function is set to 0.5, and its function expression is as follows:

[0089]

[0090] in, It is the likelihood value of the i-th parameter group; Let V be the variance of the model prediction error for the i-th parameter group; It is the variance of the measured values ​​of the i-th parameter group.

[0091] Step S4c: Drive the improved model with the selected set of effective parameter values ​​and sort the calculation results to obtain the cumulative likelihood distribution of the forecast flow at different times. Select the 5% and 95% fractions as the upper and lower limits of the interval prediction range as the estimation results of the interval prediction.

[0092] It should be noted that the calculation result obtained by the improved model here is the total outlet volume of the entire basin, including the calculation of each hydrological unit and the total outlet volume of the entire basin obtained by the river flow calculation module. The calculation of each hydrological unit is included in the calculation process of the model, as detailed in step S2.

[0093] To demonstrate the effectiveness of the proposed method, the daily runoff of Tangnaihai Station, a hydrological control station at the outlet of the Yellow River source area, was used as an example. Daily runoff data for this station from 2007 to 2016 were extracted from the Yellow River Hydrological Yearbook. Meteorological data from six meteorological stations in the Yellow River source area—Madoi, Dari, Ruoergai, Maqin, Guinan, and Xinghai—from 2007 to 2016 were used, including precipitation, temperature, and evaporation. The period from 2007 to 2014 (8 years) was selected as the calibration period, and 2015-2016 as the validation period. An algorithm program was written in Python to construct the interval prediction model of this invention. Coverage was selected. and interval average bandwidth As evaluation indicators, among them This indicates the percentage of the predicted range that covers the measured value. This represents the average bandwidth of the prediction interval over the entire time series. The relevant calculation formula is as follows:

[0094]

[0095]

[0096]

[0097] in, For time, Indicates the length of time. express Real-time measured values This represents the lower bound of the uncertainty interval. The above formula represents the upper limit of the uncertainty interval, indicating that... The larger the value, The smaller the value, the better the performance of the confidence interval.

[0098] The performance index calculation results of the model of this invention are shown in Table 1 and Figure 3 As shown in Table 1 and Figure 3 As can be seen, most of the measured daily runoff falls within the 90% range of the model in this invention. Therefore, the model in this invention is feasible for predicting the uncertainty range of daily runoff and can quantify the uncertainty of the daily runoff prediction results to a certain extent.

[0099] Table 1. Daily runoff forecast indicators for Tangnaihai Station

[0100]

[0101] Based on the same technical concept as the method, the present invention also provides a daily runoff interval prediction system for high-altitude and cold regions that considers the impact of climate change, comprising:

[0102] The data preprocessing module collects flow, precipitation, evaporation and temperature data observed by hydrological and meteorological stations in the study area, and uses ArcGIS software to divide hydrological units based on digital elevation information and construct the water system topology of each unit.

[0103] The hydrological model construction module constructs a three-source Xin'anjiang model, including modules for evapotranspiration calculation, runoff calculation, three-source division and confluence calculation, and constructs a river flow calculation module based on the topological relationship of the water system of each hydrological unit to obtain the total outflow calculation results of the basin.

[0104] The hydrological model optimization module, based on the degree-day factor method, constructs a snowmelt module and embeds it into the Xin'anjiang model. This incorporates watershed snow accumulation and snowmelt runoff into the calculation of watershed precipitation-runoff, making the improved Xin'anjiang model suitable for daily runoff prediction in high-altitude and cold regions.

[0105] The daily runoff interval prediction module constructs an uncertainty module based on hydrological statistics, selects the effective parameter set of the model applicable to the watershed, drives the improved Xin'anjiang model to output the predicted flow set, and estimates the model prediction range at a specified confidence level to obtain the daily runoff interval prediction results.

[0106] It should be understood that the high-altitude runoff interval prediction system considering the impact of climate change in the embodiments of the present invention can realize all the technical solutions in the above method embodiments. The functions of each functional module can be specifically implemented according to the methods in the above method embodiments. The specific implementation process can be referred to the relevant descriptions in the above embodiments, which will not be repeated here.

[0107] The present invention also provides a computer device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, wherein when the programs are executed by the processors, they implement the steps of the method for predicting daily runoff intervals in high-altitude cold regions considering the impact of climate change as described above.

[0108] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for predicting daily runoff intervals in high-altitude and cold regions considering the impact of climate change as described above.

[0109] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus (systems), computer devices, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0110] This invention is described with reference to a flowchart of a method according to embodiments of the invention. It should be understood that each step in the flowchart and combinations thereof can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the process. Figure 1 A device for a function specified in one or more processes.

[0111] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 The function specified in one or more processes.

[0112] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 Steps of a specified function in one or more processes.

Claims

1. A method for predicting daily runoff intervals in high-altitude and cold regions considering the impact of climate change, characterized in that, Includes the following steps: Step S1: Collect flow, precipitation, evaporation and temperature data observed by hydrological and meteorological stations in the study area, and use ArcGIS software to divide hydrological units based on digital elevation information and construct the water system topology of each unit; Step S2: Construct the Xin'anjiang River model with three water sources, including modules for evapotranspiration calculation, runoff calculation, division of the three water sources and confluence calculation, and construct a river flow calculation module based on the topological relationship of the water system of each hydrological unit to obtain the total outflow calculation results of the basin. Step S3: Construct a snowmelt module based on the degree-day factor method and embed it into the Xin'anjiang model. This incorporates watershed snow cover and snowmelt runoff into the calculation of watershed precipitation-runoff, making the improved Xin'anjiang model suitable for daily runoff prediction in high-altitude and cold regions. The construction of the snowmelt module includes: Step S3a: Utilize two critical temperatures and To distinguish precipitation patterns, the average temperature of the day < At that time, the precipitation was in the form of snowfall; the average temperature of the day was... > At that time, the precipitation was in the form of rain; the average temperature of the day was... At that time, the precipitation pattern was a mixture of rain and snow. When performing model calculations, the observed daily average temperature was used as the basis. Determine the amount of snowfall for the day. ; Step S3b: Unlike the direct runoff calculation for rainfall, snowfall first becomes part of the watershed snow cover, and the amount of snow cover in the watershed during the time period is calculated. The calculation formula is: in, This represents the initial snow accumulation in the watershed, in mm. Step S3c: The precipitation amount after deducting snowfall is the rainfall amount. Therefore, the precipitation amount needs to be corrected to obtain the corrected precipitation amount. The corrected formula is as follows, representing the actual rainfall: Step S3d: Based on the different sources of heat absorbed by the snow, snow melting is divided into high-temperature snow melting, which is caused by heat from the air, and rain-induced snow melting, which is caused by heat carried by rainfall. The snow melting amounts of these two parts are calculated separately and then added together to obtain the total snow melting amount. In step S3a, the daily snowfall amount The specific calculation formula is as follows: in, This represents the precipitation amount during the time period, in mm. In step S3d, the total snowmelt amount is calculated using the degree-day factor method, and the specific formula is as follows: in, Total snowmelt amount, in mm; The day factor is expressed in mm / (°C). d); Snow surface temperature, in °C, is the temperature of the snow surface when the total snow melts. Greater than the watershed snow cover At that time, take ; Step S4: Construct an uncertainty module based on hydrological statistics, select the effective parameter set of the model applicable to the watershed, drive the improved Xin'anjiang model to output the predicted flow set, estimate the model prediction range at a specified confidence level, and obtain the daily runoff interval prediction results.

2. The method according to claim 1, characterized in that, Step S1 includes: Step S1a: Use the TauDEM plugin in ArcGIS to divide the hydrological units, and select the average flow calculated by Grid Network Order as the threshold to extract the water system in the study area; Step S1b: The Thiessen polygon method is used to calculate the corresponding weights of the meteorological station in each hydrological unit. Based on this, a weighted average is performed to obtain the areal precipitation, evaporation and temperature of each hydrological unit as input to the hydrological model. Step S1c: Analyze the topological relationship of each hydrological unit, and summarize the process of calculating the river flow unit by unit from upstream to downstream based on the topological relationship of each hydrological unit. This will facilitate the subsequent input of the total precipitation runoff calculation results of each hydrological unit to obtain the total discharge of the watershed outlet.

3. The method according to claim 1, characterized in that, Step S4 includes: Step S4a: Assuming the prior distribution of the model parameters is uniform, the Monte Carlo method is used to randomly sample the model parameters, and 10,000 parameter sets are obtained by randomly sampling within the parameter range. Step S4b: Determine the likelihood function and set the threshold for the likelihood function value. Input the 10,000 sets of parameters randomly sampled in step S4a into the model, calculate the set of likelihood function values ​​for each set of parameters corresponding to each hydrological station, and filter out the effective parameter sets according to the threshold. Step S4c: Drive the improved Xin'anjiang model with the selected effective parameter group and sort the calculation results to obtain the cumulative likelihood distribution of the forecast flow at different times. Select the 5% and 95% fractions as the upper and lower limits of the interval prediction range as the estimation results of the interval prediction.

4. The method according to claim 3, characterized in that, In step S4b, the certainty coefficient, which reflects the degree of overlap between the measured flow rate and the simulated forecast flow rate process, is used as the likelihood function. The threshold value of the likelihood function is set to 0.5, and its function expression is as follows: in, It is the likelihood value of the i-th parameter group; Let V be the variance of the model prediction error for the i-th parameter group; It is the variance of the measured values ​​of the i-th parameter group.

5. A daily runoff interval prediction system for high-altitude cold regions considering the impact of climate change, used to implement the daily runoff interval prediction method for high-altitude cold regions considering the impact of climate change as described in any one of claims 1-4, characterized in that, include: The data preprocessing module collects flow, precipitation, evaporation and temperature data observed by hydrological and meteorological stations in the study area, and uses ArcGIS software to divide hydrological units based on digital elevation information and construct the water system topology of each unit. The hydrological model construction module constructs a three-source Xin'anjiang model, including modules for evapotranspiration calculation, runoff calculation, three-source division and confluence calculation, and constructs a river flow calculation module based on the topological relationship of the water system of each hydrological unit to obtain the total outflow calculation results of the basin. The hydrological model optimization module, based on the degree-day factor method, constructs a snowmelt module and embeds it into the Xin'anjiang model. This incorporates watershed snow accumulation and snowmelt runoff into the calculation of watershed precipitation-runoff, making the improved Xin'anjiang model suitable for daily runoff prediction in high-altitude and cold regions. The daily runoff interval prediction module constructs an uncertainty module based on hydrological statistics, selects the effective parameter set of the model applicable to the watershed, drives the improved Xin'anjiang model to output the predicted flow set, and estimates the model prediction range at a specified confidence level to obtain the daily runoff interval prediction results.

6. A computer device, characterized in that, include: One or more processors; Memory; And one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, wherein when the programs are executed by the processors, they implement the steps of the method for predicting daily runoff intervals in high-altitude cold regions considering the impact of climate change as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for predicting daily runoff intervals in high-altitude and cold regions that takes into account the impact of climate change, as described in any one of claims 1-4.

Images