Available water volume prediction method and system

Abstract

The present invention provides a method and system for predicting available water volume, relating to the field of artificial intelligence technology. The method includes: fitting a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume in the target basin and the near-surface air temperature based on the available water volume in the target basin during the first time period; determining the preliminary predicted value of the available water volume in the target basin during the second time period corresponding to each global climate model based on the air temperature data of the target basin output by multiple global climate models and the hook-shaped response function; constructing an emergence constraint model according to the air temperature change trend of each global climate model during the third time period and the corresponding preliminary predicted value of the available water volume; and substituting the air temperature change trend of the target basin during the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume in the target basin during the second time period. The present invention predicts the available water volume in the target basin in the future period by fitting the hook-shaped response function and combining the emergence constraint method.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an available water volume prediction method and system. Background Art

[0002] Since the Industrial Revolution, the emission of greenhouse gases such as carbon dioxide in the atmosphere has increased, the energy balance and material circulation process of the earth's land and atmosphere have changed, the global temperature has continued to rise, coral reefs and tropical rainforests have died out on a large scale, extreme disasters such as rainstorms and floods have occurred frequently, and the imbalance and inequality of the social and economic system and the ecological environment have become increasingly severe. In order to cope with the negative impact of the warming rate, it is crucial to strengthen scientific research on the response mechanism of water resources to climate change.

[0003] In recent years, domestic and foreign scholars have estimated the available water on land through various meteorological and hydrological variables such as precipitation, evapotranspiration, runoff, and terrestrial water storage, and found that the available water on land in most parts of the world has a significant downward trend, but some studies have also found that the available water on land in some parts of the world has shown an upward trend. Existing studies use different meteorological and hydrological variables to represent the available water on land, resulting in large differences in research results, making it difficult to form a unified scientific conclusion.

[0004] Existing studies generally assess the evolution trends of global and regional terrestrial water availability in the past and present, but rarely predict the evolution of terrestrial water availability under future climate change scenarios. Therefore, it is difficult to scientifically support integrated water resources management and sustainable social and economic development. Summary of the invention

[0005] The present invention provides a method and system for predicting available water volume, which are used to solve the defect in the prior art that there is a lack of a method for the evolution of available water volume on land under future climate change scenarios, and to realize a method for predicting available water volume in a future river basin.

[0006] The present invention provides a method for predicting available water volume, comprising:

[0007] Based on the available water volume of the target basin in the first time period, a hook-shaped response function is obtained to characterize the relationship between the predicted value of available water volume of the target basin and the near-ground temperature;

[0008] Based on the temperature data of the target watershed output by multiple global climate models and the hook-type response function, determine a preliminary prediction value of available water volume in the target watershed corresponding to each global climate model in the second time period;

[0009] constructing an emergence constraint model according to the temperature change trend of each global climate model in the third time period and the corresponding preliminary prediction value of available water;

[0010] Substitute the temperature change trend of the target river basin in the fourth time period into the emergence constraint model to obtain the predicted available water volume of the target river basin in the second time period;

[0011] Wherein, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0012] According to a method for predicting available water volume provided by the present invention, before the step of fitting the available water volume of the target river basin in the first time period to obtain a hook-shaped response function representing the relationship between the predicted available water volume of the target river basin and the near-surface air temperature, the method further includes:

[0013] Based on the terrestrial water storage data of each grid point of the target river basin in the first time period, obtain the basin-average terrestrial water storage series of the target river basin;

[0014] Take the ratio of the monthly runoff series of the target river basin in the first time period to the basin area of the target river basin as the basin runoff depth series;

[0015] Determine the sum of the basin-average terrestrial water storage series and the basin runoff depth series as the available water volume of the target river basin in the first time period.

[0016] According to a method for predicting available water volume provided by the present invention, before the step of obtaining the basin-average terrestrial water storage series of the target river basin based on the terrestrial water storage data of each grid point of the target river basin in the first time period, the method further includes:

[0017] Based on several gravity satellite data of each grid point of the target river basin in the first time period, invert the terrestrial water storage data of each grid point of the target river basin in the first time period.

[0018] According to a method for predicting available water volume provided by the present invention, before the step of obtaining the basin-average terrestrial water storage series of the target river basin based on the terrestrial water storage data of each grid point of the target river basin in the first time period, the method further includes:

[0019] In the case where there are missing values in several gravity satellite data of each grid point of the target river basin in the first time period, construct and train a random forest model, where the random forest model represents the relationship between each driving factor in the target river basin and the terrestrial water storage;

[0020] Use the random forest model to reconstruct the complete terrestrial water storage data of each grid point of the target river basin in the first time period.

[0021] According to a method for predicting available water volume provided by the present invention, before the step of taking the ratio of the monthly runoff series of the target river basin in the first time period to the basin area of the target river basin as the basin runoff depth series, the method further includes:

[0022] Construct multiple artificial intelligence models, which are used to output monthly runoff data for the first time period according to the input monthly runoff data for partial time periods.

[0023] Determine the artificial intelligence model with the highest Nash efficiency coefficient as the optimal model.

[0024] Based on the optimal model, determine the monthly runoff series of the target basin in the first time period.

[0025] According to a method for predicting available water volume provided by the present invention, the step of determining the preliminary predicted value of the available water volume of the target basin corresponding to each global climate model in the second time period specifically includes:

[0026] Calculate the first multi-year average basin temperature of the target basin in the third time period.

[0027] Calculate the second multi-year average basin temperature of the target basin corresponding to each global climate model in the third time period.

[0028] Calculate the third multi-year average basin temperature of the target basin in the second time period.

[0029] Calculate the sum of the first multi-year average basin temperature and the third multi-year average basin temperature, and then subtract the third multi-year average basin temperature. Project the obtained value into the hook-shaped response function to determine the preliminary predicted value of the available water volume of the target basin in the second time period.

[0030] The present invention also provides an available water volume prediction device, including:

[0031] A fitting module, which is used to fit the available water volume of the target basin in the first time period to obtain a hook-shaped response function characterizing the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature.

[0032] A determination module, which is used to determine the preliminary predicted value of the available water volume of the target basin corresponding to each global climate model in the second time period based on the air temperature data of the target basin output by multiple global climate models and the hook-shaped response function.

[0033] A construction module, which is used to construct an emergence constraint model according to the temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume.

[0034] A prediction module, which is used to substitute the temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume of the target basin in the second time period.

[0035] Among them, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0036] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the available water volume prediction method described in any one of the above is implemented.

[0037] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the available water volume prediction method described in any one of the above is implemented.

[0038] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the available water volume prediction method described in any one of the above is implemented.

[0039] The available water volume prediction method and system provided by the present invention predict the available water volume of the target basin in the future period by fitting the Hook response function and combining the emergence constraint method, providing an important and highly operable reference basis for the global and regional water disaster risk assessment and early warning under climate change scenarios, and providing engineering reference value for coping with future climate disasters, scientifically formulating emission reduction strategies, and comprehensive management of basin water resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 is one of the flow schematic diagrams of the available water volume prediction method provided by the present invention;

[0042] Figure 2 is one of the schematic diagrams of the Hook response function obtained by fitting in the available water volume prediction method provided by the present invention;

[0043] Figure 3 is the schematic diagram of the sliding window in the available water volume prediction method provided by the present invention;

[0044] Figure 4 is the second schematic diagram of the Hook response function obtained by fitting in the available water volume prediction method provided by the present invention;

[0045] Figure 5 is the structural schematic diagram of the available water volume prediction device provided by the present invention;

[0046] Figure 6 It is a schematic structural diagram of the electronic device provided by the present invention. Specific embodiments

[0047] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] The following combines Figures 1 to 4 to introduce the available water volume prediction method of the present invention. As Figure 1 shown, it includes:

[0049] Step 101: Fit a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume in the target basin and the near-surface air temperature based on the available water volume in the target basin during the first time period;

[0050] The target basin is the basin for which the available water volume in the future period needs to be predicted.

[0051] The first time period represents the historical period during which relevant hydrometeorological data of the target basin can be obtained. Therefore, various data related to the basin water volume in the target basin during the first time period can be obtained, and the available water volume in the target basin during the first time period can be determined WA , that is, the available water volume in the target basin during the historical period.

[0052] Since the data of the target basin during the first time period is used to fit the hook-shaped response function, it is therefore preferable to set a longer first time period. In this embodiment, the first time period is the complete time period from 1950 to 2022.

[0053] Furthermore, obtain the average monthly air temperature of the target basin during the first time period.

[0054] On this basis, the binning scaling function can better describe the non-linear relationship between variables and has been widely used in recent years to analyze the response characteristics of extreme precipitation and runoff to global warming. Based on the available water volume in the target basin WA and the average monthly air temperature of the target basin during the first time period, construct a Hook response function based on the binning scaling function.

[0055] Specifically, taking the average monthly temperature of the basin as the sorting criterion, the temperature-basin available water volume combination series in the same month within the first time period is divided into 12 bins with the same sample size. The median of the daily temperature is used to represent the temperature of the bin, and the average value of the basin available water volume in each temperature bin is used to represent the available water volume, finally obtaining 12 groups of temperature-available water volume series.

[0056] Through research, it is found that precipitation and runoff generally show a Hook structure with a "first rising and then falling" trajectory with respect to the near-surface temperature. Therefore, the locally weighted sliding regression smoothing method is used to fit the 12 groups of temperature-available water volume series, thereby drawing the Hook structure of temperature-available water volume in the target basin, as Figure 2 shown.

[0057] Among them, the temperature at the inflection point of the Hook structure is called the peak point temperature, that is, Figure 2 T in pp .

[0058] It can be understood that the fitted Hook response function represents the relationship between the predicted value of the available water volume in the target basin and the near-surface temperature. On this basis, by obtaining the near-surface temperature in the future time period of the target basin, the corresponding available water volume in the future time period can be determined based on the Hook response function.

[0059] Step 102, based on the temperature data of the target basin output by multiple global climate models and the hook-type response function, determine the preliminary predicted value of the available water volume in the target basin corresponding to each global climate model in the second time period;

[0060] For this purpose, the temperature data of the target basin in the future time period is obtained based on multiple global climate models as the near-surface temperature.

[0061] Among them, the second time period represents the future time period to be predicted.

[0062] Optionally, in this embodiment, 35 global climate models are selected, and the available water volume in the target basin under the 35 global climate models in the second time period is obtained as the preliminary predicted value of the available water volume.

[0063] Step 103, according to the temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume, construct an emergent constraint model;

[0064] Furthermore, the emergent constraints technology is a method for reducing the prediction uncertainty of climate models. It is based on the relationship between observational data and model simulations, and by identifying the statistical relationship between certain observable variables in the model ensemble and the future climate response, it constrains the prediction of future climate change.

[0065] Therefore, based on the preliminary predicted value of the available water volume, an emergence constraint model is constructed to obtain a more accurate predicted result of the available water volume.

[0066] Optionally, first calculate the annual average temperature change trend of each global climate model in the third time period, denoted as .

[0067] Among them, the third time period also represents the historical time period. Optionally, the third time period can intersect, partially overlap, or completely overlap with the first time period.

[0068] It can be understood that the third time period is the historical time period for which the output data of the global climate model can be obtained. Specifically, in this embodiment, it is from 1985 to 2014.

[0069] Furthermore, integrate the data of 35 global climate models to obtain 35 paired combinations of the annual average temperature change trend in the historical period and the preliminary predicted value WA fut1 of the available water volume in the future period. The constructed emergence constraint model is specifically:

[0070] ;

[0071] In the formula, a and b represent the parameters of the emergence constraint model.

[0072] Through the above method, a parameter-containing emergence constraint model can be obtained. On this basis, the least squares method is used to solve the parameters a and b of the parameter-containing emergence constraint model.

[0073] Step 104, substitute the temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume of the target basin in the second time period;

[0074] Among them, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0075] Furthermore, based on the reanalysis data of the ERA5-Land dataset, deduce the change trend of the temperature in the third time period and substitute this trend term into the constructed emergence constraint model, specifically as follows:

[0076] ;

[0077] In the formula, represents the predicted result of the change in the available water volume of the basin in the future period after correction, that is, the predicted value of the available water volume of the target basin in the second time period finally obtained; It represents the temperature change trend of the target basin in the fourth time period, obtained based on the ERA5-Land dataset (the fifth-generation land reanalysis dataset of the European Centre for Medium-Range Weather Forecasts).

[0078] Among them, the fourth time period also represents the historical period, that is, through the temperature change trend in the historical period, the prediction results of the available water volume in the future period are corrected.

[0079] Optionally, in this embodiment, the fourth time period is the same as the third time period, and is set to be from 1985 to 2014.

[0080] The present invention predicts the available water volume of the target basin in the future period by fitting the Hook response function and combining the emergence constraint method, providing an important and highly operable reference basis for the global and regional water disaster risk assessment and early warning under climate change scenarios, and providing engineering reference value for coping with future climate disasters, scientifically formulating emission reduction strategies and comprehensive management of basin water resources.

[0081] Before the step of fitting the hook-shaped response function representing the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature based on the available water volume of the target basin in the first time period in the available water volume prediction method of the present invention, the following steps are further included:

[0082] Based on the terrestrial water storage data of each grid point of the target basin in the first time period, obtain the basin-average terrestrial water storage series of the target basin;

[0083] Take the ratio of the monthly runoff series of the target basin in the first time period to the basin area of the target basin as the basin runoff depth series;

[0084] Determine the sum of the basin-average terrestrial water storage series and the basin runoff depth series as the available water volume of the target basin in the first time period.

[0085] Grid the target basin, based on the collected hydrometeorological data of the target basin in the first time period, obtain the terrestrial water storage data of each grid point of the target basin in the first time period, and use the Thiessen polygon method to obtain the basin-average terrestrial water storage series of the target basin based on the grid point data TWS .

[0086] Furthermore, based on the pre-collected hydrometeorological data of the target basin in the first time period, obtain the monthly runoff series of the target basin in the first time period Q .

[0087] Divide the monthly runoff of the basin by the basin area A , and obtain the basin runoff depth series.

[0088] Next, calculate the cumulative values of the basin-average terrestrial water storage series and the monthly runoff depth series of the basin to obtain the available water volume of the target basin in the first time period:

[0089] ;

[0090] In the formula, WA is the available water volume of the target basin, Q is the monthly runoff, A is the basin area, TWS is the terrestrial water storage series.

[0091] In the available water volume prediction method of the present invention, before the step of obtaining the basin-average terrestrial water storage series of the target basin based on the terrestrial water storage data of each grid point of the target basin in the first time period, the following steps are further included:

[0092] Based on several gravity satellite data of each grid point of the target basin in the first time period, the terrestrial water storage data of each grid point of the target basin in the first time period is inversely obtained.

[0093] Collect several gravity satellite data of each grid point of the target basin in the first time period, and inversely obtain the terrestrial water storage data of each grid point of the target basin in the first time period.

[0094] Among them, institutions such as the GFZ German Research Centre for Geosciences, the Jet Propulsion Laboratory (JPL) of the California Institute of Technology in the United States, the Center for Space Research at the University of Texas at Austin in the United States (CSR), and the Goddard Space Flight Center (GSFC) of the National Aeronautics and Space Administration in the United States are responsible for data interpretation and release of monthly-scale global gravity field output data as gravity satellite data.

[0095] Optionally, select one or more of the gravity satellite data to inversely obtain the terrestrial water storage data.

[0096] In this embodiment, the latest sixth-generation (RL06) products of JPL, CSR, and GSFC are simultaneously used. These three data sources have different spatial resolutions and all provide the mascon (mass concentration) monthly-scale equivalent water height (subtracting the 2004 - 2009 average field) based on the solution of the concentrated mass block gravity field, and finally output a long-term TWSA (Terrestrial Water Storage Anomaly) dataset.

[0097] To account for the uncertainties that may be caused by different data products, three sets of GRACE / GRACE-FO (Gravity Recovery And Climate Experiment / GRACE Follow-On) gravity satellite datasets were interpolated to a 0.25°×0.25° spatial grid, and the mean value of each product was taken at each time step. Finally, monthly terrestrial water storage data from 2002 to 2022 was obtained.

[0098] In the available water volume prediction method of the present invention, before the step of obtaining the basin-averaged terrestrial water storage series of the target basin based on the terrestrial water storage data of each grid point in the target basin during the first time period, the following steps are further included:

[0099] In the case where there are missing values in several gravity satellite data of each grid point in the target basin during the first time period, a random forest model is constructed and trained, and the random forest model characterizes the relationship between each driving factor in the target basin and the terrestrial water storage.

[0100] The random forest model is used to reconstruct the complete terrestrial water storage data of each grid point in the target basin during the first time period.

[0101] It can be understood that if the first time period is set from 2002 to 2022, the terrestrial water storage data of the target basin during the first time period can be directly retrieved through the gravity satellite data.

[0102] However, in this embodiment, the first time period is set from 1950 to 2022, that is, there are missing values in the gravity satellite data of each grid point in the target basin during the first time period. Therefore, by constructing a random forest model, the complete terrestrial water storage data during the first time period is reconstructed.

[0103] Specifically, first, the relative humidity of each grid point in the target basin is determined based on the Clausius-Clapeyron thermodynamic equation RH .

[0104] Among them, the Clausius-Clapeyron thermodynamic equation is defined as follows:

[0105] ;

[0106] In the formula, is the saturation vapor pressure at the air temperature T , T 0 is the first integration constant, taking 273.16K; is the second integration constant, taking 611Pa; is the latent heat of vaporization constant, taking ; is the water vapor gas constant, taking ; T is the air temperature and is an input variable of the Clausius-Clapeyron thermodynamic equation.

[0107] Meteorological and hydrological data of each grid point in the target basin are collected in advance to obtain meteorological and hydrological observation samples. The 2-meter air temperature and dew point temperature in the meteorological and hydrological observation samples of each grid point in the target basin are respectively used as input variables and substituted into the Clausius-Clapeyron thermodynamic equation to calculate the relative humidity ( RH ) of each grid point. Specifically: , where T dew is the dew point temperature, and T 2m is the 2m air temperature.

[0108] Furthermore, the specific humidity q is the ratio of the water vapor mass to the total mass of the air mass, and is derived using the ERA5 near-surface air pressure p and dew point temperature. The formula is as follows:

[0109] ;

[0110] On this basis, as Figure 3 shown, for each grid point in the target basin, a spatial sliding window threshold of 5×5 is set, and it slides sequentially within this interval range. The data of the grid points involved in this spatial range are selected as the input of the random forest model, which can improve the robustness of the machine learning model.

[0111] A random forest model is used to construct the relationship model between each driving factor within the above spatial range and the GRACE gravity satellite terrestrial water storage data.

[0112] Among them, the driving factors are data during the period from 2002 to 2022, specifically including:

[0113] The first category: variables obtained or derived from ERA5-Land, including air temperature, relative humidity, snowfall, precipitation, runoff depth, shortwave radiation, and longwave radiation;

[0114] The second category: ENSO (El Nino Southern Oscillation) indices, including sea surface temperature anomaly indices in NINO 1+2 region, NINO 3 region, NINO 3.4 region, and NINO 4 region;

[0115] Category 3: Average series of global climate model output data, including monthly average temperature, monthly relative humidity, monthly precipitation, monthly snowfall, monthly runoff depth, monthly specific humidity, monthly shortwave radiation intensity, and monthly longwave radiation. Among them, the average series of global climate model output data refers to the average of the output results of 35 global climate models, that is, a set of output series is obtained for each variable.

[0116] For each grid point, a lag time of 2 months is considered. That is, for each month, the driving factors of that month and the previous 1 - 2 months are selected as the inputs of the random forest model. The random forest model is constructed using the driving factors and GRACE gravity satellite data during the period from 2002 to 2022, and is used to simulate the terrestrial water storage, expressed as:

[0117] ;

[0118] In the formula, represents the terrestrial water storage simulated by the random forest model at time t, represents the input variables at time t, represents the input variables at time t - 1, represents the input variables at time t - 2; F represents the random forest model.

[0119] Furthermore, the gradient descent method is used to calibrate the parameters of the random forest model for each grid point, and then the random forest model established by driving with the long - series dataset since 1950 is used to reconstruct the long - series terrestrial water storage dataset since 1950, that is, based on the terrestrial water storage data from 2002 to 2022, the terrestrial water storage data of the target basin in the first time period is reconstructed.

[0120] In the available water volume prediction method of the present invention, before the step of taking the ratio of the monthly runoff series of the target basin in the first time period to the basin area of the target basin as the basin runoff depth series, the following steps are also included:

[0121] Construct multiple artificial intelligence models, which are used to output the monthly runoff data in the first time period according to the input monthly runoff data of partial time periods;

[0122] Determine the artificial intelligence model with the highest Nash efficiency coefficient as the optimal model;

[0123] Based on the optimal model, determine the monthly runoff series of the target basin in the first time period.

[0124] Furthermore, since it is also difficult to obtain the complete monthly runoff data of the basin in the first time period, therefore, in combination with the artificial intelligence model, the monthly runoff series of the basin in the first time period is determined based on the available partial - time - period monthly runoff data of the basin.

[0125] Specifically, the Thiessen polygon method is first used to derive the monthly series of basin-averaged ERA5-Land and global climate model output data. The main idea is to convert the grid dataset into a basin-averaged dataset.

[0126] Secondly, multiple artificial intelligence models are used to construct a monthly runoff simulation model. In this embodiment, five artificial intelligence models, namely artificial neural network, support vector machine, long short-term memory model, random forest, and Gaussian linear regression model, are selected.

[0127] The driving factors of each artificial intelligence model include the following two categories:

[0128] The first category is variables obtained or derived from ERA5-Land, including air temperature, relative humidity, snowfall, precipitation, runoff depth, shortwave radiation, and longwave radiation;

[0129] The second category is the average series of global climate model output data, including monthly average air temperature, monthly relative humidity, monthly precipitation, monthly snowfall, monthly runoff depth, monthly specific humidity, monthly shortwave radiation intensity, and monthly longwave radiation.

[0130] Optionally, based on the driving factors and the runoff observation data of the basin hydrological stations, five artificial intelligence models are constructed, and considering a 7-month lag time, the monthly runoff series since the establishment of the hydrological stations is simulated, expressed as:

[0131] ;

[0132] In the formula, represents the monthly runoff simulated by the k-th machine learning model at time t, represents the input variables at time t, represents the input variables at time t-1, represents the input variables at time t-2, represents the input variables at time t-7; F k represents the k th artificial intelligence model, k k = 1, 2, …, 5.

[0133] The gradient descent method is used to calibrate the five artificial intelligence models to obtain the optimal parameters, and then the Nash-Sutcliffe efficiency coefficient (NSE) of the monthly runoff simulated by each artificial intelligence model and the observed monthly runoff is calculated:

[0134] ;

[0135] In the formula, is the simulated runoff sequence; and are the measured runoff series and its mean value respectively; NT is the length of the runoff series. The closer the NSE is to 1, the better the simulation effect indicates.

[0136] Calculate the NSE of 5 artificial intelligence models respectively, and select the artificial intelligence model with the highest NSE for reconstructing the long-term monthly runoff; use the optimal artificial intelligence model calibrated by the long-term dataset since 1950 to reconstruct the long-term monthly runoff dataset since 1950.

[0137] That is, based on the short-term data that can be obtained, reconstruct the monthly runoff series of the target basin in the first time period.

[0138] In the available water volume prediction method of the present invention, the step of determining the preliminary predicted value of the available water volume of the target basin in the second time period corresponding to each global climate model specifically includes:

[0139] Calculate the first multi-year basin average temperature of the target basin in the third time period;

[0140] Calculate the second multi-year basin average temperature of the target basin corresponding to each global climate model in the third time period;

[0141] Calculate the third multi-year basin average temperature of the target basin in the second time period;

[0142] Calculate the sum of the first multi-year basin average temperature and the third multi-year basin average temperature, then subtract the third multi-year basin average temperature, and project the obtained value into the hook-shaped response function to determine the preliminary predicted value of the available water volume of the target basin in the second time period.

[0143] Specifically, use the Thiessen polygon to calculate the first multi-year basin average temperature of the target basin in the third time period, denoted as T obs ; then, for each global climate model, use the Thiessen polygon to calculate the second multi-year basin average temperature of the target basin in the third time period, denoted as T his .

[0144] On this basis, use the Thiessen polygon to calculate the third multi-year average temperature of the target basin in the second time period, that is, in the future period, denoted as T fut .

[0145] As Figure 4 shown, project T obs into the fitted Hook response function to obtain the corresponding available water volume of the target basin in the third time period, denoted as WA obs .

[0146] Then, for each global climate model, further obtain Tobs +(T fut -T his ), the available water volume in the corresponding basin of the Hook response function is used as the preliminary predicted value WA of the available water volume of the target basin in the second time period fut1 .

[0147] In the above way, the preliminary predicted values of the available water volume of the target basin in the second time period can be determined one by one under 35 global climate models.

[0148] It should be noted that the methods for collecting and organizing some of the data in the above text are as follows:

[0149] Determine the target basin. For each grid point of the target basin, monthly-scale meteorological and hydrological observation data from 1950 to 2022 are collected from the fifth-generation atmospheric reanalysis dataset (ERA5-Land) of the European Centre for Medium-Range Weather Forecasts, specifically including 2m air temperature, air pressure, dew point temperature, snowfall, precipitation, runoff depth, shortwave radiation, and longwave radiation.

[0150] Further obtain the time series of 4 ENSO (El Nino Southern Oscillation) indices from 1950 to the present, specifically including the sea surface temperature anomaly index in the NINO 1+2 region, the sea surface temperature anomaly index in the NINO 3 region, the sea surface temperature anomaly index in the NINO 3.4 region, and the sea surface temperature anomaly index in the NINO 4 region.

[0151] Further obtain the monthly average air temperature, monthly relative humidity, monthly precipitation, monthly snowfall, monthly runoff depth, monthly average specific humidity, monthly shortwave radiation intensity, and monthly longwave radiation in the historical period (1950-2014) under each global climate model; for the future period (2015-2100), obtain the monthly average air temperature data of each global climate model under the SSP585 shared socioeconomic pathway, and further obtain the monthly relative humidity, monthly precipitation, monthly snowfall, monthly runoff depth, monthly average specific humidity, monthly shortwave radiation intensity, and monthly longwave radiation during the period 2015-2022 in the future.

[0152] Collect the monthly runoff data of the control hydrological station at the outlet section of the target basin; and interpolate the ERA5-Land and global climate model output data into a 0.25°×0.25° spatial grid.

[0153] The available water volume prediction system provided by the present invention will be described below. The available water volume prediction system described below can be mutually referred to the available water volume prediction method described above.

[0154] As Figure 5As shown in the figure, the available water volume prediction system includes a fitting module 501, a determination module 502, a construction module 503, and a prediction module 504:

[0155] The fitting module is used to fit a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume in the target basin and the near-surface air temperature based on the available water volume in the target basin during the first time period;

[0156] The target basin is the basin for which the available water volume in the future period needs to be predicted.

[0157] The first time period represents the historical period during which relevant hydrometeorological data of the target basin can be obtained. Therefore, various data related to the basin water volume in the target basin during the first time period can be obtained to determine the available water volume in the target basin during the first time period WA , that is, the available water volume in the target basin during the historical period.

[0158] Since the data of the target basin during the first time period is used to fit the Hook response function, it is therefore preferable to set a longer first time period. In this embodiment, the first time period is the complete time cycle from 1950 to 2022.

[0159] Furthermore, the average monthly air temperature of the target basin during the first time period is obtained.

[0160] On this basis, the binning scaling function can better depict the nonlinear relationship between variables and has been widely used in recent years to analyze the response characteristics of extreme precipitation and runoff to global warming. Based on the available water volume in the target basin during the first time period WA , the average monthly air temperature of the basin, a hook-shaped response function is constructed based on the binning scaling function.

[0161] Specifically, taking the average monthly air temperature of the basin as the sorting criterion, the same-month air temperature-available water volume combination series during the first time period is divided into 12 bins with the same sample size. The median of the daily air temperature is taken to represent the temperature of the bin, and the average value of the available water volume in each air temperature bin is used to represent the available water volume. Finally, 12 groups of air temperature-available water volume series are obtained.

[0162] Through research, it is found that precipitation and runoff generally show a Hook structure with a "first rising and then falling" trajectory with respect to the near-surface air temperature. Therefore, the local weighted sliding regression smoothing method is used to fit the 12 groups of air temperature-available water volume series, thereby drawing the Hook structure of air temperature-available water volume in the target basin, as Figure 2 shown.

[0163] Among them, the air temperature at the inflection point of the Hook structure is called the peak point temperature, that is, Figure 2 T in pp .

[0164] It can be understood that the Hook response function obtained by fitting represents the relationship between the predicted value of the available water volume in the target basin and the near-surface air temperature. On this basis, by obtaining the near-surface air temperature in the future period of the target basin, the corresponding available water volume in the future period can be determined based on the Hook response function.

[0165] A determination module, configured to determine a preliminary predicted value of the available water volume in the target basin in the second time period corresponding to each global climate model based on the air temperature data of the target basin output by multiple global climate models and the Hook response function;

[0166] Therefore, based on multiple global climate models, the air temperature data of the target basin in the future period is obtained as the near-surface air temperature.

[0167] Wherein, the second time period represents the future time period to be predicted.

[0168] Optionally, in this embodiment, 35 global climate models are selected, and the available water volume in the target basin under the 35 global climate models in the second time period is obtained as the preliminary predicted value of the available water volume.

[0169] A construction module, configured to construct an emergent constraint model according to the air temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume;

[0170] Furthermore, the Emergent Constraints is a method for reducing the prediction uncertainty of climate models. It is based on the relationship between observational data and model simulations, and by identifying the statistical relationship between certain observable variables in the model ensemble and the future climate response, it constrains the prediction of future climate change.

[0171] Therefore, based on the preliminary predicted value of the available water volume, an emergent constraint model is constructed to obtain a more accurate predicted result of the available water volume.

[0172] Optionally, first calculate the annual average air temperature change trend of each global climate model in the third time period, denoted as .

[0173] Wherein, the third time period also represents the historical time period. It can be understood that the third time period is the historical time period for which the output data of the global climate model can be obtained. In this embodiment, it is specifically from 1985 to 2014.

[0174] Furthermore, the data of 35 global climate models are integrated to obtain 35 paired combinations of the historical period annual average air temperature change trend and the future period WA fut1 . The constructed emergent constraint model is specifically:

[0175] ;

[0176] In the formula, a and b represent the parameters of the emergence constraint model.

[0177] By the above method, an emergence constraint model with parameters can be obtained. On this basis, the least squares method is used to solve the parameters of the emergence constraint model with parameters a and b .

[0178] A prediction module, configured to substitute the temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain a predicted value of the available water volume of the target basin in the second time period;

[0179] Wherein, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0180] Furthermore, based on the reanalysis data of the ERA5-Land dataset, the change trend of temperature in the third time period is derived and substituted into the constructed emergence constraint model, specifically as follows:

[0181] ;

[0182] In the formula, represents the predicted result of the change in the available water volume of the basin in the future period after correction, that is, the predicted value of the available water volume of the target basin in the second time period finally obtained; represents the temperature change trend of the target basin in the fourth time period, obtained based on the ERA5-Land dataset.

[0183] Wherein, the fourth time period also represents the historical period, that is, the predicted result of the available water volume in the future period is corrected by the temperature change trend in the historical period.

[0184] Optionally, in this embodiment, the fourth time period is the same as the third time period, and is set to be from 1985 to 2014.

[0185] The present invention predicts the available water volume of the target basin in the future period by fitting the Hook response function and combining the emergence constraint method, providing an important and highly operable reference basis for the risk assessment and early warning of global and regional water and drought disasters under climate change scenarios, and providing engineering reference value for coping with future climate disasters, scientifically formulating emission reduction strategies, and comprehensive management of basin water resources.

[0186] Figure 6 Illustrates a schematic diagram of the physical structure of an electronic device, such as Figure 6As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communication bus 640. Among them, the processor 610, the communications interface 620, and the memory 630 complete communication with each other through the communication bus 640. The processor 610 may call the logical instructions in the memory 630 to execute the available water volume prediction method, which includes: fitting a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume in the target basin and the near-surface air temperature based on the available water volume in the target basin during the first time period; determining the preliminary predicted value of the available water volume in the target basin during the second time period corresponding to each global climate model based on the air temperature data of the target basin output by multiple global climate models and the hook-shaped response function; constructing an emergence constraint model according to the air temperature change trend in the third time period of each global climate model and the corresponding preliminary predicted value of the available water volume; substituting the air temperature change trend in the fourth time period of the target basin into the emergence constraint model to obtain the predicted value of the available water volume in the target basin during the second time period; wherein, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0187] In addition, when the logical instructions in the above-mentioned memory 630 are implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0188] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the available water volume prediction method provided by each of the above methods. The method includes: fitting a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature based on the available water volume of the target basin in the first time period; determining the preliminary predicted value of the available water volume of the target basin in the second time period corresponding to each global climate model based on the temperature data of the target basin output by multiple global climate models and the hook-shaped response function; constructing an emergence constraint model according to the temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume; substituting the temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume of the target basin in the second time period; wherein, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0189] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the available water volume prediction method provided by each of the above methods. The method includes: fitting a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature based on the available water volume of the target basin in the first time period; determining the preliminary predicted value of the available water volume of the target basin in the second time period corresponding to each global climate model based on the temperature data of the target basin output by multiple global climate models and the hook-shaped response function; constructing an emergence constraint model according to the temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume; substituting the temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume of the target basin in the second time period; wherein, the first time period, the third time period, and the fourth time period are all earlier than the second time period.

[0190] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0191] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0192] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting available water volume, characterized in that Including: Fitting a hook-shaped response function based on the available water volume of the target basin in the first time period to characterize the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature; Based on the air temperature data of the target basin output by multiple global climate models and the hook-shaped response function, determining the preliminary predicted value of the available water volume of the target basin corresponding to each global climate model in the second time period; Constructing an emergence constraint model according to the air temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume; Substituting the air temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume of the target basin in the second time period; Wherein, the first time period, the third time period and the fourth time period are all earlier than the second time period; Before the step of fitting a hook-shaped response function based on the available water volume of the target basin in the first time period to characterize the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature, it further includes: Based on the terrestrial water storage data of each grid point of the target basin in the first time period, obtaining the basin-average terrestrial water storage series of the target basin; Taking the ratio of the monthly runoff series of the target basin in the first time period to the basin area of the target basin as the basin runoff depth series; Determining the sum of the basin-average terrestrial water storage series and the basin runoff depth series as the available water volume of the target basin in the first time period; Before the step of obtaining the basin-average terrestrial water storage series of the target basin based on the terrestrial water storage data of each grid point of the target basin in the first time period, it further includes: Based on several gravity satellite data of each grid point of the target basin in the first time period, inversely obtaining the terrestrial water storage data of each grid point of the target basin in the first time period; Before the step of obtaining the basin-average terrestrial water storage series of the target basin based on the terrestrial water storage data of each grid point of the target basin in the first time period, it further includes: In the case where there are missing data in several gravity satellite data of each grid point of the target basin in the first time period, constructing and training a random forest model, where the random forest model characterizes the relationship between each driving factor in the target basin and the terrestrial water storage; Using the random forest model to reconstruct the complete terrestrial water storage data of each grid point of the target basin in the first time period.

2. The available water volume prediction method according to claim 1, characterized in that Before the step of taking the ratio of the monthly runoff series of the target basin in the first time period to the basin area of the target basin as the basin runoff depth series, it further includes: Constructing multiple artificial intelligence models, where the artificial intelligence models are used to output the monthly runoff data of the first time period according to the input monthly runoff data of partial time periods; Determining the artificial intelligence model with the highest Nash efficiency coefficient as the optimal model; Based on the optimal model, determining the monthly runoff series of the target basin in the first time period.

3. The available water volume prediction method according to claim 1 or 2, characterized in that The step of determining the preliminary predicted value of the available water volume of the target basin corresponding to each global climate model in the second time period specifically includes: Calculating the first multi-year basin-average air temperature of the target basin in the third time period; Calculate the second multi-year basin average temperature of the target basin corresponding to each global climate model in the third time period; Calculate the third multi-year basin average temperature of the target basin in the second time period; Calculate the sum of the first multi-year basin average temperature and the third multi-year basin average temperature, then subtract the third multi-year basin average temperature, project the obtained value into the hook-shaped response function, and determine the preliminary predicted value of the available water volume of the target basin in the second time period.

4. A available water volume prediction system, characterized in that, Include: A fitting module for fitting a hook-shaped response function characterizing the relationship between the predicted value of the available water volume of the target basin and the near-surface air temperature based on the available water volume of the target basin in the first time period; A determination module for determining the preliminary predicted value of the available water volume of the target basin in the second time period corresponding to each global climate model based on the air temperature data of the target basin output by multiple global climate models and the hook-shaped response function; A construction module for constructing an emergence constraint model according to the temperature change trend of each global climate model in the third time period and the corresponding preliminary predicted value of the available water volume; A prediction module for substituting the temperature change trend of the target basin in the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume of the target basin in the second time period; Wherein, the first time period, the third time period and the fourth time period are all earlier than the second time period; The determination model is specifically used to obtain the basin average terrestrial water storage series of the target basin based on the terrestrial water storage data of each grid point of the target basin in the first time period; Take the ratio of the monthly runoff series of the target basin in the first time period to the basin area of the target basin as the basin runoff depth series; Determine the sum of the basin average terrestrial water storage series and the basin runoff depth series as the available water volume of the target basin in the first time period; The determination module is further used to invert the terrestrial water storage data of each grid point of the target basin in the first time period based on a plurality of gravity satellite data of each grid point of the target basin in the first time period; The determination module is further used to construct and train a random forest model in the case where there are missing values in a plurality of gravity satellite data of each grid point of the target basin in the first time period, and the random forest model characterizes the relationship between each driving factor in the target basin and the terrestrial water storage; Use the random forest model to reconstruct the complete terrestrial water storage data of each grid point of the target basin in the first time period.

5. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the available water volume prediction method according to any one of claims 1 to 3.

6. A non-transitory 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 available water volume prediction method according to any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the available water volume prediction method according to any one of claims 1 to 3.

Images