Extreme low water change attribution and low water resource potential assessment method

Abstract

The invention discloses an extreme low water change attribution and low water resource potential evaluation method, which belongs to the technical field of water resource calculation, and comprises the following steps of: obtaining raster data of runoff and driving factors of the runoff in a research area, performing data preprocessing, deducing the frequency distribution of runoff depth of each grid unit by adopting a non-parametric kernel density estimation method, and calculating the runoff depth of each grid unit; and counting the proportion of the grids meeting the guarantee rate runoff depth threshold in each year in the total grid, and further determining a low water typical year under each guarantee rate. A nonlinear regression simulation model of coupling optimization deep learning is adopted to carry out extreme low-water runoff depth regression simulation research under different driving factor combinations, and a key driving factor combination influencing low-water runoff depth change is mined based on a regression simulation precision result; and analyzing the constraint effect of each key driving factor on the extreme low water and a possible threshold value by adopting a constraint line deduction technology, and evaluating the low water resource potential of a typical year of the extreme low water.

Description

Technical Field

[0001] The present invention relates to the technical field of water resource calculation, and in particular to a method for attributing extreme low water changes and evaluating low water resource potential. Background Art

[0002] The attribution of extreme dry runoff changes and the assessment and analysis of dry runoff resource potential play an important role and significance in deeply understanding the causes of extreme events, promoting ecological and environmental protection, and enhancing disaster prevention and mitigation capabilities. In-depth analysis of the key driving factors of extreme dry runoff, scientific assessment of dry runoff resource potential, optimization of water resource allocation, and improvement of water resource utilization efficiency. Extreme dry runoff is an extreme state of river runoff. At present, there are several unresolved issues in the study of extreme dry runoff: First, traditional runoff attribution methods such as statistical analysis, hydrological models, and climate elasticity coefficients have high requirements for the temporal continuity of hydrological sequences, but extreme dry runoff sequences are discontinuous time series. Taking the annual scale as an example, extreme dry years are often separated by several years or even more than ten years. Traditional runoff attribution methods cannot meet the attribution needs of extreme dry runoff changes. Second, due to spatial limitations, traditional runoff change attribution is still based on runoff data from one or several hydrological stations, failing to consider the spatial heterogeneity of runoff changes and their influencing factors, reducing the reliability of attribution results. Third, there is currently a lack of evaluation technology that can evaluate the potential of dry runoff resources in a basin when extreme dry runoff scenarios occur. Therefore, it is urgently necessary to adopt scientific and efficient methods to explore the key driving factors affecting the spatial changes of extreme droughts from the perspective of spatial differentiation, attribute the changes in extreme droughts, and improve the accuracy of extreme drought simulation and drought resource potential assessment. Summary of the invention

[0003] The purpose of the present invention is to solve the problems existing in the prior art and to propose a method for attributing extreme low water changes and evaluating low water resource potential.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: A method for attributing extreme low water changes and assessing low water resource potential comprises the following steps: S1. Obtain raster data of runoff and its driving factors in the study area and perform data preprocessing; S2. Based on the runoff depth grid data corresponding to step S1 in the study area, the frequency distribution of runoff depth of each grid cell is derived using the non-parametric kernel density estimation method, and the proportion of grids that meet the runoff depth threshold of the guarantee rate in each year is counted to the total grids, and then the typical dry year under each guarantee rate is determined; S3. After obtaining the typical dry year corresponding to step S2 of the study area, extract the runoff depth raster data and driving factor raster data corresponding to the typical dry year, combine the fishing net generation technology and the random forest model, and determine the importance of each driving factor to the extreme low water runoff depth; S4. Based on the results of the importance of driving factors, a nonlinear regression simulation model with coupled optimization deep learning is used to carry out regression simulation analysis of extreme low-flow runoff depth under different driving factor combinations. Based on the accuracy of regression simulation, the key driving factor combinations that affect the changes in low-flow runoff depth are identified and mined; S5. Based on the results of attribution analysis, the constraint line inference technique is used to analyze the constraint effects of key driving factors on extreme drought and possible thresholds, and to evaluate the dry water resource potential in typical years of extreme drought.

[0005] Preferably, the preprocessing of the runoff and its driving factor raster data in step S1 includes: Step 11, perform missing value test, outlier test, and white noise processing on each type of data; Step 12, reprojecting and resampling each type of data to solve the problem of inconsistent resolution and non-strict pixel alignment among various types of data.

[0006] Preferably, step S2 comprises: Step 21, firstly, based on the annual runoff depth grid data of the study area over the years, the non-parametric kernel density estimation method is used to deduce the frequency distribution of the annual runoff depth of each grid unit; firstly, the kernel function is used to deduce the probability density function, and then the marginal distribution function of the heavy rain samples of each duration of the grid unit is derived based on the kernel estimated probability density function; assuming that a grid Annual runoff depth sample data is from a univariate continuous population distribution , the overall probability density function is , its kernel density estimation function can be expressed as: In the formula, yes Kernel density estimation function of ; is the kernel function, and ; It is called window width or bandwidth ; Step 22, determining the low-flow runoff depth threshold value under each guarantee rate; based on the frequency distribution result of step 21, given a specific low-flow guarantee rate P, the low-flow runoff depth threshold value RP corresponding to the low-flow guarantee rate P can be deduced; Step 23, determine the typical dry year under each guarantee rate; calculate the proportion of grids that meet the guarantee rate runoff depth threshold in each year to the total grids, and select the year with the largest proportion as the typical year; take 75%, 90%, and 95% three low water guarantee rates as examples: 75% low water runoff depth refers to the runoff depth being less than the 75% guarantee rate runoff depth threshold but greater than the 90% guarantee rate runoff depth threshold; 90% low water runoff depth refers to the runoff depth being less than the 90% guarantee rate runoff depth threshold but greater than the 95% guarantee rate runoff depth threshold; 95% low water runoff depth refers to the runoff depth being less than the 95% guarantee rate runoff depth threshold.

[0007] Preferably, step S3 comprises: Step 31, after obtaining the typical dry year corresponding to step S2 of the study area, extracting the runoff depth raster data and driving factor raster data corresponding to the typical dry year; Step 32, using fishing net construction and sampling technology to collect spatially continuous extreme low-flow runoff depth and its driving factor combination data in the study area in a typical year; Step 33, using the random forest model, identify the importance of each candidate driving factor on the change of extreme low-flow runoff depth in a typical year; the importance calculation formula of the random forest model is: In the formula, Represents variables in random forest Importance score; is the out-of-bag sample data; is the out-of-bag sample data observations; and For the The fitted value corresponding to the observation value is In the tree, the former randomly permutes the variables The latter is before the observation value of the random permutation variable After the observation value of is the discriminant function, that is, when When , the value is 1, otherwise It is also the discriminant function, that is, when When , the value is 1, otherwise Represents the number of trees in the random forest model.

[0008] Preferably, step S4 comprises: Step 41, firstly, a nonlinear regression model is used to perform regression analysis on the extreme low-flow runoff depth under different driving factor combinations, and the residual under each driving factor combination is recorded; Taking three variables as an example, the basic form of the nonlinear regression model is as follows In the formula, Y is the dependent variable of the nonlinear regression model, X1~X3 are the independent variables of the nonlinear regression model, a0 is the constant term of the nonlinear regression model, a1~a3 are the linear term coefficients of the nonlinear regression model, a4~a6 are the cross term coefficients of the nonlinear regression model, and a7~a9 are the square term coefficients of the nonlinear regression model. From the basic form, it can be seen that the nonlinear regression model adopted fully considers the influence of the constant term, linear term, cross term and square term on the regression process. The first 70% of the time series is used as the rate period of the regression model, and the last 30% is used as the validation period of the regression model; the residuals of the regression model in the rate period and validation period are In the formula, and are the residuals of the nonlinear regression model in the rate period and the validation period respectively; and are the simulated value and true value of the nonlinear regression model at the rate period, respectively; and are the simulated value and true value of the nonlinear regression model in the validation period, respectively; Step 42, using an LSTM deep learning model based on differential creative search optimization parameters to simulate the residual of the nonlinear regression model; The LSTM model calculation steps are as follows: In the formula, is the error of the residual simulation result; They are the forget gate, input gate and output gate of the LSTM model respectively; is the memory unit state; Represents the weight of the forget gate; Represents the weight of the input gate; for The weight of is the weight of the output gate; Characterize the bias term of the forget gate; is the bias term of the input gate; is the bias term of the output gate; and yes The offset vector of is the state variable of the model, For recursive input; is the sigmoid activation function; is the hyperbolic tangent activation function; The rate period and validation period of the LSTM deep learning model are consistent with the nonlinear regression model; obtain the simulated residual sequence of the training period and the test period and ; Then, the simulation results of the original nonlinear regression model are corrected using the residuals obtained by simulation, that is, the simulation results of the nonlinear regression simulation model based on coupled optimization deep learning are obtained; In the formula, and The final simulation results of the nonlinear regression simulation model based on coupled optimization deep learning in the rate period and the verification period; Step 43, using Nash efficiency coefficient NSE, R 2 , RMSE evaluation indicators, to evaluate the accuracy of extreme low-flow runoff simulation by the nonlinear regression simulation model based on coupled optimization deep learning under each group of driving factor combinations, and take the driving factor combination with the best accuracy as the key driving factor combination affecting the deep change of low-flow runoff.

[0009] Preferably, the step S5 comprises: Step 51, based on the identified key driving factor combination, the constraint line derivation technology is used to obtain the constraint line of each driving factor in the key driving factor combination on the extreme low-flow runoff depth, revealing the constraint effect of each driving factor on the extreme low-flow runoff depth; the constraint line derivation steps are as follows: A two-dimensional scatter plot was drawn with each driving factor as the limiting variable and the extreme low-flow runoff depth as the response variable; The value range of the limiting variable is divided into several groups, and the 99.9% quantile point of the extreme low-water runoff depth value is selected from each group as the boundary point; Perform optimal fitting of polynomial functions on all boundary points to extract constraint lines and determine their functional equations; Step 52, based on the constraint line result, the best driving factor threshold value that may exist in each type of driving factor; Step 53, based on the identification result of the driving factor threshold, evaluate the low water resource potential that the region may reach in the corresponding typical extreme dry year under the condition of regulating the driving factor; expressed in terms of the average low water runoff depth of the basin, the calculation formula for the low water resource potential is: In the formula, is the low water resource potential of the study area under the conditions of a typical dry year, mm; represents the actual average low-flow runoff depth of the basin in a typical year, mm; is the corresponding constraint line equation; is the optimal driving factor threshold; It indicates the low-flow depth that the basin can reach when the driving factor is infinitely close to the optimal driving factor threshold.

[0010] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention provides a method for attributing extreme low water changes and evaluating low water resource potential. First, a non-parametric kernel density estimation method is used to deduce the frequency distribution of runoff depth of each grid unit, and then the typical low water year under each guarantee rate in the study area is determined. Secondly, after determining the typical year, the importance of each driving factor to the extreme low water runoff depth is determined by combining the fishing net generation technology and the random forest model. Then, a nonlinear regression simulation model of coupled optimization deep learning is used to carry out regression simulation research on extreme low water runoff depth under different driving factor combinations, and based on the regression simulation accuracy results, the key driving factor combination that affects the change of low water runoff depth is mined. Finally, based on the results of attribution analysis, the constraint line deduction technology is used to analyze the constraint effect of each key driving factor on extreme low water and the possible threshold value, and the low water resource potential in the typical year of extreme low water is evaluated.

[0011] 2. This method can effectively solve the problem that traditional runoff attribution methods are not applicable to non-continuous low-flow time series, and realizes the coupling of nonlinear regression model and deep learning theory, which not only effectively eliminates the interference of irrelevant factors on attribution, but also improves the accuracy of attribution. At the same time, based on the results of attribution analysis combined with constraint line derivation technology, it provides a new and effective path for regional assessment of dry water resource potential in extreme dry years. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 The spatial distribution results of the 75% low-flow runoff depth threshold; Figure 3 The spatial distribution results of 90% low-flow runoff depth threshold; Figure 4 The spatial distribution results of 95% low-flow runoff depth threshold; Figure 5 The results of the ratio of the number of low-water grids to the total number of grids that meet each guarantee rate condition in the 1960s; Figure 6 The results of the ratio of the number of low-water grids to the total number of grids that meet each guarantee rate condition in the 1970s; Figure 7 The results of the ratio of the number of low-water grids to the total number of grids that meet each guarantee rate condition in the 1980s; Figure 8 The results of the ratio of the number of low-water grids to the total number of grids that meet each guarantee rate condition in the 1990s; Fig. 9 The results of the ratio of the number of low-water grids to the total number of grids that meet each guarantee rate condition in the 2000s; Fig.10 The results of the ratio of the number of low-water grids to the total number of grids that meet each guarantee rate condition in the 2010s; Fig.11The importance of each driving factor to the extreme low-flow runoff depth; Fig.12 Process diagram of the simulation results of extreme low-flow runoff depth during the validation period using the nonlinear regression simulation model based on coupled optimization deep learning without optimizing the influencing factors; Fig.13 The nonlinear regression simulation model based on coupled optimization deep learning, whose influencing factors were not optimized, was used to evaluate the simulation accuracy of extreme low-flow runoff depth during the validation period; Fig.14 Process diagram of the simulation results of extreme low-flow runoff depth during the validation period using the nonlinear regression simulation model based on coupled optimization deep learning after the influencing factors are optimized; Fig.15 The results of the simulation accuracy evaluation of extreme low-flow runoff depth by the nonlinear regression simulation model based on coupled optimization deep learning after the influencing factors were optimized during the validation period; Fig.16 The constraining effect of rainfall on extreme low-flow runoff depth; Fig.17 The constraining effect of temperature on extreme low-flow runoff depth; Fig.18 The constraining effect of potential evaporation on extreme low-flow runoff depth; Fig.19 The constraining effect of NDVI on extreme low-flow runoff depth; Fig. 20 The constraining effect of SPEI on extreme low-flow runoff depth; Fig.21 Constraining effect of irrigation water on extreme low-flow runoff depth. DETAILED DESCRIPTION

[0013] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0014] Example, see Figures 1 to 21 , a method for attributing extreme low water changes and assessing low water resource potential, comprising the following steps: S1. Obtain raster data of runoff and its driving factors in the study area and perform data preprocessing; S2. Based on the runoff depth grid data corresponding to step S1 in the study area, the frequency distribution of runoff depth of each grid cell is derived using the non-parametric kernel density estimation method, and the proportion of grids that meet the runoff depth threshold of the guarantee rate in each year is counted to the total grids, and then the typical dry year under each guarantee rate is determined; S3. After obtaining the typical dry year corresponding to step S2 of the study area, extract the runoff depth raster data and driving factor raster data corresponding to the typical dry year, combine the fishing net generation technology and the random forest model, and determine the importance of each driving factor to the extreme low water runoff depth; S4. Based on the results of the importance of driving factors, a nonlinear regression simulation model with coupled optimization deep learning is used to carry out regression simulation analysis of extreme low-flow runoff depth under different driving factor combinations. Based on the accuracy of regression simulation, the key driving factor combinations that affect the changes in low-flow runoff depth are identified and mined; S5. Based on the results of attribution analysis, the constraint line inference technique is used to analyze the constraint effects of key driving factors on extreme drought and possible thresholds, and to evaluate the dry water resource potential in typical years of extreme drought.

[0015] In this embodiment, the middle and lower reaches of the Yangtze River are taken as an example. The grid runoff depth data collected the monthly runoff depth grid data of China from 1961 to 2018, with a data resolution of 0.25°. The monthly runoff depth grid data of the middle and lower reaches of the Yangtze River from 1961 to 2018 were obtained by clipping with ACRGIS software, and then the annual runoff depth grid data of the middle and lower reaches of the Yangtze River from 1961 to 2018 were deduced. For the driving factor data, precipitation data, temperature data, potential evapotranspiration data, NDVI data, altitude data, slope data, meteorological drought index SPEI, ecological drought index TVDI, GDP, population density, human activity index (HI), domestic water withdrawal, and agricultural irrigation water withdrawal were collected in the middle and lower reaches of the Yangtze River.

[0016] After obtaining the annual runoff depth raster data of the middle and lower reaches of the Yangtze River from 1961 to 2018, the annual runoff depth data of the middle and lower reaches of the Yangtze River from 1961 to 2018 were collected by using the fishing net construction technology and spatial sampling technology. Then, the frequency distribution of the runoff depth of each grid unit was derived by the non-parametric kernel density estimation method, and the low-water runoff depth thresholds under 75% guarantee rate, 90% guarantee rate and 95% guarantee were derived for the middle and lower reaches of the Yangtze River. The results are as follows Figure 2~Figure 4 shown.

[0017] Based on the low-flow runoff depth threshold results of each grid unit at each guarantee rate, the proportion of grids that meet the low-flow conditions of each guarantee rate in each year can be deduced ( Figures 5 to 10 ). In the present invention, 75% low-water runoff depth refers to a runoff depth that is less than the 75% guaranteed runoff depth threshold and greater than the 90% guaranteed runoff depth threshold; 90% low-water runoff depth refers to a runoff depth that is less than the 90% guaranteed runoff depth threshold and greater than the 95% guaranteed runoff depth threshold; 95% low-water runoff depth refers to a runoff depth that is less than the 95% guaranteed runoff depth threshold. Figures 5 to 10It can be seen that in terms of 75% low-water runoff depth, the number of grids belonging to 75% low-water runoff depth in 1966 accounted for the largest proportion of the total number of grids, which was 40%, and 1966 was determined as the 75% low-water typical year. In terms of 75% low-water runoff depth, the number of grids belonging to 75% low-water runoff depth in 1966 accounted for the largest proportion of the total number of grids, which was 40%, and 1966 was determined as the 75% low-water typical year in the middle and lower reaches of the Yangtze River. In terms of 90% low-water runoff depth, the number of grids belonging to 90% low-water runoff depth in 1978 accounted for the largest proportion of the total number of grids, which was 49%, and 1978 was determined as the 90% low-water typical year in the middle and lower reaches of the Yangtze River. As for the 95% low-flow runoff depth, the number of grids belonging to the 95% low-flow runoff depth in 2011 accounted for the largest proportion of the total number of grids, which was 64%. Therefore, 2011 was determined as the 95% low-flow typical year in the mainstream area of ​​the middle and lower reaches of the Yangtze River. In summary, the regional extreme low-flow typical year determined by the present invention is reasonable and reliable.

[0018] After the typical dry year is determined, the attribution of extreme dry season changes, simulation and evaluation of dry season resource potential can be carried out. The present invention takes the extreme dry season with a 95% guaranteed rate as an example. From the results of the determination of the typical dry year, it can be seen that the typical dry year with a 95% guaranteed rate in the middle and lower reaches of the Yangtze River is 2011. Therefore. The 2011 runoff depth raster data is obtained as the extreme dry season runoff depth data. In addition, the corresponding driving factor data for 2011 are also obtained, including: rainfall data, temperature data, potential evapotranspiration data, NDVI data, altitude data, slope data, meteorological drought index SPEI, ecological drought index TVDI, GDP, population density, human activity index (HI), domestic water consumption, and agricultural irrigation water consumption. On this basis, the random forest model is used to judge the importance of each driving factor to the extreme dry season runoff depth. The results are as follows: Fig.11 As shown. Fig.11 It can be seen that rainfall is the most important factor affecting the 95% extreme low-water runoff depth in the middle and lower reaches of the Yangtze River, followed by temperature and meteorological drought index SPEI. Among the five indicators that characterize the impact of human activities, irrigation water is the most important factor affecting the 95% extreme low-water runoff depth in the middle and lower reaches of the Yangtze River, followed by population density and domestic water use. Ecological drought index TVDI, slope and HI are unimportant influencing factors.

[0019] Then, the nonlinear regression simulation model of coupled optimization deep learning was used to perform probabilistic regression simulation analysis of extreme low-flow runoff depth. The results are as follows: Figure 12 to Figure 15 As shown. Among them, Fig.12 and Fig.13 This is the regression process and accuracy evaluation result of the influencing factors that have not been optimized. When the influencing factors have not been optimized, they are disturbed by unimportant factors, and the performance of the regression model is poor. 2The Nash efficiency coefficient NSE was only 0.7149 and 0.6239 respectively, with an error value of more than ±100 mm. The unimportant influencing factors were eliminated one by one according to their importance from small to large, and regression analysis was performed. The regression accuracy results of each unimportant factor were recorded and compared. The results showed that when the ecological drought index TVDI, slope, human activity index (HI), altitude, GDP, domestic water consumption, and population density PD were eliminated in turn, the model regression result accuracy evaluation index R 2 The Nash efficiency coefficient NSE shows an overall upward trend. After eliminating other influencing factors, the accuracy evaluation index R 2 and Nash efficiency coefficient NSE continue to decrease. Therefore, the optimal combination of influencing factors is selected as rainfall, temperature, potential evapotranspiration, NDVI, meteorological drought index SPEI, and agricultural irrigation water consumption. The regression process and accuracy evaluation results are shown in Fig.14 and Fig.15 After the influencing factors were optimized, the interference of unimportant factors was effectively eliminated, and the performance of the regression model was significantly improved. 2 The Nash efficiency coefficient NSE reached 0.8470 and 0.8433 respectively, and the error value was within ±100mm.

[0020] Finally, the constraint line technology was used to analyze the constraint effects of precipitation, temperature, potential evapotranspiration, NDVI data, meteorological drought index SPEI, and agricultural irrigation water withdrawal on the extreme low-flow runoff depth in the middle and lower reaches of the Yangtze River, to determine the threshold of the most suitable driving factor, and to evaluate the low-water resource potential when the extreme low-flow typical year scenario in the middle and lower reaches of the Yangtze River occurs. The results of the constraint effects of precipitation, temperature, potential evapotranspiration, NDVI data, meteorological drought index SPEI, and agricultural irrigation water withdrawal on the extreme low-flow runoff depth in the middle and lower reaches of the Yangtze River are as follows: Figure 16 to Figure 21 As shown. Figure 16 to Figure 21It can be seen that the constraint capacity of precipitation on the extreme low-flow runoff depth in the middle and lower reaches of the Yangtze River decreases with the increase of precipitation, and there is no optimal rainfall threshold; before the annual average temperature reaches 16℃, the constraint capacity of temperature on the extreme low-flow runoff depth in the middle and lower reaches of the Yangtze River decreases with the increase of temperature, and after the temperature reaches 16℃, the constraint capacity of temperature on the extreme low-flow runoff depth in the middle and lower reaches of the Yangtze River increases with the increase of temperature, and 16℃ is the optimal temperature threshold; the constraint capacity of potential evaporation on the extreme low-flow runoff depth in the middle and lower reaches of the Yangtze River increases with the increase of potential evaporation, and there is no optimal evaporation threshold; the NDVI in the middle and lower reaches of the Yangtze River is mainly distributed in 0.5~ Before NDVI reaches 0.8, the ability of NDVI to constrain the extreme low-water runoff depth in the middle and lower reaches of the Yangtze River decreases with the increase of NDVI. After NDVI reaches 0.8, the ability of NDVI to constrain the extreme low-water runoff depth in the middle and lower reaches of the Yangtze River increases with the increase of NDVI, and 0.8 is the optimal NDVI threshold. The ability of SPEI to constrain the extreme low-water runoff depth in the middle and lower reaches of the Yangtze River decreases with the increase of SPEI value, and there is no optimal SPEI threshold. The ability of irrigation water consumption to constrain the extreme low-water runoff depth in the middle and lower reaches of the Yangtze River increases with the increase of irrigation water consumption, and there is no optimal irrigation water consumption threshold.

[0021] Based on the results of constraint effect and threshold analysis, it can be seen that there are optimal thresholds for temperature and NDVI, which can be adjusted to reach the thresholds to obtain the optimal low-water runoff depth. Among them, temperature is related to carbon emissions, while NDVI is related to the underlying vegetation, which can be adjusted through ecological construction projects. After obtaining the optimal temperature and NDVI thresholds, the optimal thresholds are substituted into the corresponding constraint line equation to obtain the low-water runoff depth under the threshold conditions. The difference between the low-water runoff depth under the threshold conditions and the actual low-water runoff depth in the typical year is the low-water resource potential of the desired dry typical year. The evaluation results of the low-water resource potential of the 95% dry typical year (2011) in the middle and lower reaches of the Yangtze River are shown in Table 1. It can be seen from the table that the low-water runoff depth in the 95% dry typical year (2011) in the middle and lower reaches of the Yangtze River is 1096mm under the optimal temperature threshold conditions, and the corresponding low-water resource potential is 375.52. In the typical year of 95% dry season (2011) in the middle and lower reaches of the Yangtze River, the low-water runoff depth under the optimal NDVI threshold conditions was 1136 mm, and the corresponding low-water resource potential was 415.52.

[0022] The present invention provides a method for attributing extreme low water changes and evaluating low water resource potential. Firstly, a non-parametric kernel density estimation method is used to deduce the frequency distribution of runoff depth of each grid unit, and then the typical low water year under each guarantee rate in the study area is determined. Secondly, after the typical year is determined, the importance of each driving factor to the extreme low water runoff depth is determined by combining fishing net generation technology and random forest model. Then, a nonlinear regression simulation model of coupled optimization deep learning is used to carry out regression simulation research on extreme low water runoff depth under different driving factor combinations. Based on the regression simulation accuracy results, the key driving factor combination that affects the change of low water runoff depth is mined. Finally, based on the result of attribution analysis, the constraint line deduction technology is used to analyze the constraint effect of each key driving factor on extreme low water and the possible threshold value, so as to evaluate the low water resource potential in the typical year of extreme low water.

[0023] This method can effectively solve the problem that traditional runoff attribution methods are not applicable to non-continuous low-flow time series, and realizes the coupling of nonlinear regression model and deep learning theory, which not only effectively eliminates the interference of irrelevant factors on attribution, but also improves the accuracy of attribution. At the same time, based on the results of attribution analysis combined with constraint line derivation technology, it provides a new and effective path for regional assessment of dry water resource potential in extreme dry years.

[0024] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for attributing extreme low water changes and assessing low water resource potential, characterized in that: The following steps are involved: S1. Obtain raster data of runoff and its driving factors in the study area and perform data preprocessing; S2. Based on the runoff depth grid data corresponding to step S1 in the study area, the frequency distribution of runoff depth of each grid cell is derived using the non-parametric kernel density estimation method, and the proportion of grids that meet the runoff depth threshold of the guarantee rate in each year is counted to the total grids, and then the typical dry year under each guarantee rate is determined; S3. After obtaining the typical dry year corresponding to step S2 of the study area, extract the runoff depth raster data and driving factor raster data corresponding to the typical dry year, combine the fishing net generation technology and the random forest model, and determine the importance of each driving factor to the extreme low water runoff depth; S4. Based on the results of the importance of driving factors, a nonlinear regression simulation model with coupled optimization deep learning is used to carry out regression simulation analysis of extreme low-flow runoff depth under different driving factor combinations. Based on the accuracy of regression simulation, the key driving factor combinations that affect the changes in low-flow runoff depth are identified and mined; S5. Based on the results of attribution analysis, the constraint line inference technique is used to analyze the constraint effects of key driving factors on extreme drought and possible thresholds, and to evaluate the dry water resource potential in typical years of extreme drought.

2. The method for attributing extreme low water changes and evaluating low water resource potential according to claim 1, characterized in that: The preprocessing of the runoff and its driving factor raster data in step S1 includes: Step 11, perform missing value test, outlier test, and white noise processing on each type of data; Step 12, reprojecting and resampling each type of data to solve the problem of inconsistent resolution and non-strict pixel alignment among various types of data.

3. The method for attributing extreme low water changes and evaluating low water resource potential according to claim 1, characterized in that: The step S2 comprises: Step 21, firstly, based on the annual runoff depth grid data of the study area over the years, the non-parametric kernel density estimation method is used to deduce the frequency distribution of the annual runoff depth of each grid unit; firstly, the kernel function is used to deduce the probability density function, and then the marginal distribution function of the heavy rain samples of each duration of the grid unit is derived based on the kernel estimated probability density function; assuming that a grid Annual runoff depth sample data is from a univariate continuous population distribution , the overall probability density function is , its kernel density estimation function can be expressed as: In the formula, yes Kernel density estimation function of ; is the kernel function, and ; It is called window width or bandwidth h>0; Step 22, determining the low-flow runoff depth threshold value under each guarantee rate; based on the frequency distribution result of step 21, given a specific low-flow guarantee rate P, the low-flow runoff depth threshold value RP corresponding to the low-flow guarantee rate P can be deduced; Step 23, determine the typical dry year under each guarantee rate; calculate the proportion of grids that meet the guarantee rate runoff depth threshold in each year to the total grids, and select the year with the largest proportion as the typical year; take 75%, 90%, and 95% three low water guarantee rates as examples: 75% low water runoff depth refers to the runoff depth being less than the 75% guarantee rate runoff depth threshold but greater than the 90% guarantee rate runoff depth threshold; 90% low water runoff depth refers to the runoff depth being less than the 90% guarantee rate runoff depth threshold but greater than the 95% guarantee rate runoff depth threshold; 95% low water runoff depth refers to the runoff depth being less than the 95% guarantee rate runoff depth threshold.

4. The method for attributing extreme low water changes and evaluating low water resource potential according to claim 1, characterized in that: The step S3 comprises: Step 31, after obtaining the typical dry year corresponding to step S2 of the study area, extracting the runoff depth raster data and driving factor raster data corresponding to the typical dry year; Step 32, using fishing net construction and sampling technology to collect spatially continuous extreme low-flow runoff depth and its driving factor combination data in the study area in a typical year; Step 33, using the random forest model, identify the importance of each candidate driving factor on the change of extreme low-flow runoff depth in a typical year; the importance calculation formula of the random forest model is: In the formula, Represents variables in random forest Importance score; is the out-of-bag sample data; is the first sample in the out-of-bag data. observations; and For the The fitted value corresponding to the observation value is In the tree, the former randomly permutes the variables The latter is before the observation value of the random permutation variable After the observation value of is the discriminant function, that is, when When , the value is 1, otherwise It is also the discriminant function, that is, when When , the value is 1, otherwise Represents the number of trees in the random forest model.

5. The method for attributing extreme low water changes and evaluating low water resource potential according to claim 1, characterized in that: The step S4 comprises: Step 41, firstly, a nonlinear regression model is used to perform regression analysis on the extreme low-flow runoff depth under different driving factor combinations, and the residual under each driving factor combination is recorded; Taking three variables as an example, the basic form of the nonlinear regression model is as follows In the formula, Y is the dependent variable of the nonlinear regression model, X1~X3 are the independent variables of the nonlinear regression model, a0 is the constant term of the nonlinear regression model, a1~a3 are the linear term coefficients of the nonlinear regression model, a4~a6 are the cross term coefficients of the nonlinear regression model, and a7~a9 are the square term coefficients of the nonlinear regression model; The first 70% of the time series is used as the rate period of the regression model, and the last 30% is used as the validation period of the regression model; the residuals of the regression model in the rate period and validation period are In the formula, and are the residuals of the nonlinear regression model in the rate period and the validation period respectively; and are the simulated value and true value of the nonlinear regression model at the rate period, respectively; and are the simulated value and true value of the nonlinear regression model in the validation period, respectively; Step 42, using an LSTM deep learning model based on differential creative search optimization parameters to simulate the residual of the nonlinear regression model; The LSTM model calculation steps are as follows: In the formula, is the error of the residual simulation result; They are the forget gate, input gate and output gate of the LSTM model respectively; is the memory unit state; Represents the weight of the forget gate; Represents the weight of the input gate; for The weight of is the weight of the output gate; Characterize the bias term of the forget gate; is the bias term of the input gate; is the bias term of the output gate; and yes The offset vector of is the state variable of the model; For recursive input; is the sigmoid activation function; is the hyperbolic tangent activation function; The rate period and validation period of the LSTM deep learning model are consistent with the nonlinear regression model; obtain the simulated residual sequence of the training period and the test period and ; Then, the simulation results of the original nonlinear regression model are corrected using the residuals obtained by simulation, that is, the simulation results of the nonlinear regression simulation model based on coupled optimization deep learning are obtained; In the formula, and The final simulation results of the nonlinear regression simulation model based on coupled optimization deep learning in the rate period and the verification period; Step 43, using Nash efficiency coefficient NSE, R 2 , RMSE evaluation indicators, to evaluate the accuracy of extreme low-flow runoff simulation by the nonlinear regression simulation model based on coupled optimization deep learning under each group of driving factor combinations, and take the driving factor combination with the best accuracy as the key driving factor combination affecting the deep change of low-flow runoff.

6. The method for attributing extreme low water changes and evaluating low water resource potential according to claim 1, characterized in that: The step S5 comprises: Step 51, based on the identified key driving factor combination, the constraint line derivation technology is used to obtain the constraint line of each driving factor in the key driving factor combination on the extreme low-flow runoff depth, revealing the constraint effect of each driving factor on the extreme low-flow runoff depth; the constraint line derivation steps are as follows: A two-dimensional scatter plot was drawn with each driving factor as the limiting variable and the extreme low-flow runoff depth as the response variable; The value range of the limiting variable is divided into several groups, and the 99.9% quantile point of the extreme low-water runoff depth value is selected from each group as the boundary point; Perform optimal fitting of polynomial functions on all boundary points to extract constraint lines and determine their functional equations; Step 52, based on the constraint line result, the best driving factor threshold value that may exist in each type of driving factor; Step 53, based on the identification result of the driving factor threshold, evaluate the low water resource potential that the region may reach in the corresponding typical extreme dry year under the condition of regulating the driving factor; expressed in terms of the average low water runoff depth of the basin, the calculation formula for the low water resource potential is: In the formula, is the low water resource potential of the study area under the conditions of a typical dry year, mm; represents the actual average low-flow runoff depth of the basin in a typical year, mm; is the corresponding constraint line equation; is the optimal driving factor threshold; It indicates the low-flow depth that the basin can reach when the driving factor is infinitely close to the optimal driving factor threshold.

Images