CN-121980938-A - Yangtze river extreme dry water encounter analysis method and system based on CMIP6 climate data and SWAT hydrologic model

CN121980938ACN 121980938 ACN121980938 ACN 121980938ACN-121980938-A

Abstract

The invention provides a method and a system for analyzing extreme dry water encounters of a Yangtze river based on CMIP6 climate data and SWAT hydrologic model, wherein the method comprises the steps of collecting and preprocessing multi-source basic data, covering historical and future CMIP6 climate mode data, a CN05.1 grid point meteorological data set, meteorological hydrologic site observation data, DEM, land utilization and soil data. And adopting spatial interpolation to downscale the CMIP6 data, matching with the CN05.1 resolution, and finishing data deviation correction by a quantile mapping daily deviation correction method. Based on geographic and underlying data, a Yangtze river basin SWAT model and a yellow river basin SWAT model are built in an ArcGIS, and runoff parameter calibration is completed through SWAT-CUP by combining observation and reduction of runoff data. And outputting the history and future simulated runoffs by using the corrected CMIP6 data driving model. And constructing joint distribution by means of Gumbel-Hougarrd Copula function and Bayesian model weighting method, establishing a full-scale encounter matrix, and carrying out extreme withered water encounter analysis of two drainage areas.

Inventors

LIU JIE
ZHOU YANG

Assignees

武汉大学
刘杰

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (7)

1. The invention provides a method for analyzing extreme dry water encounters of Yangtze river based on CMIP6 climate data and SWAT hydrologic model, which is characterized by comprising the following steps: Collecting and preprocessing multisource basic data, including historical and future period CMIP6 climate pattern data, precipitation and air temperature CN05.1 grid observation data sets, meteorological site observation data, digital Elevation (DEM) data, land utilization data, soil data and runoff data after hydrological site restoration; Performing downscaling processing on the historical and future CMIP6 climate mode data by adopting spatial interpolation to ensure that the historical and future CMIP6 climate mode data are consistent with the spatial resolution of the CN05.1 grid observation data, finishing the deviation correction of the downscaling CMIP6 data by combining the CN05.1 dataset by a daily deviation correction (DBC) method based on fractional number mapping, and generating corrected historical and future CMIP6 climate mode data; Inputting observation data of a meteorological site to obtain a historical period river basin simulated runoff, and calibrating parameters closely related to the runoff in the model by combining a SWAT-CUP parameter calibration tool with the runoff data after the reduction of the hydrologic site to form a calibrated SWAT model; And constructing two-dimensional joint distribution under each climate mode by utilizing Gumbel-Hougarrd Copula function, determining weight by using a Bayesian model weighting (BMA) method, obtaining weighted joint distribution of future period, establishing a abundant encounter matrix by combining the historical period (reducing runoff based on the hydrological site), and carrying out the extreme dry encounter analysis of the Yangtze river.
2. The method of claim 1, wherein collecting and preprocessing multisource base data including historical and future period CMIP6 climate pattern data, precipitation and air temperature CN05.1 grid observation data sets, meteorological site observation data, digital Elevation (DEM) data, land use data, soil data and hydrologic site post-restoration runoff data comprises: extracting the historical period and future period CMIP6 climate pattern data, precipitation and air temperature CN05.1 grid observation data set, the DEM data, the land utilization data and the soil data in a flow domain based on ArcGIS software; Filling the depression on the DEM data based on ArcGIS software, and reclassifying the land utilization data and the soil data; The spatial resolutions of the DEM data, the land utilization data and the soil data are kept consistent through spatial interpolation; and projecting the DEM data, the land utilization data and the soil data by using a unified projection coordinate system based on ArcGIS software.
3. The method for analyzing the extreme dry water encounter of Yangtze river based on CMIP6 climate data and SWAT hydrologic model as recited in claim 1, wherein the method for performing downscaling on the historical and future CMIP6 climate pattern data by spatial interpolation to make the historical and future CMIP6 climate pattern data consistent with the CN05.1 grid observation data spatial resolution, performing the downscaling on the downscaling CMIP6 data by combining the CN05.1 dataset by a Daily Bias Correction (DBC) method based on fractional mapping, and generating corrected historical and future CMIP6 climate pattern data comprises the following steps: Interpolating the historical period and future period CMIP6 climate pattern data into a 0.25 ° ×0.25° regular grid by bilinear interpolation to make it consistent with the spatial resolution of the precipitation and air temperature CN05.1 grid-dotted observation dataset; The system deviation of the data is corrected by adopting a DBC method and the CMIP6 climate pattern data of the historical period and the future period after the downscaling, the system deviation is assumed to have the same simulation error in each quantile of the climate variable of the historical period and the future period, and the occurrence frequency and the magnitude of the daily precipitation series are corrected sequentially by combining the LOCI (Local INTENSITY SCALING) method and the DT (Daily Translation) method. Firstly, correcting the occurrence probability of precipitation based on a LOCI method, taking 0.1mm as an occurrence threshold of actually measured precipitation, determining the precipitation threshold of each month of a simulation series according to the occurrence frequency of actually measured daily precipitation in different months, judging that the precipitation threshold is rainy when the daily precipitation is higher than the threshold, otherwise, judging that the precipitation is rainless, enabling the occurrence frequencies of actually measured precipitation and the simulation series to be consistent, then calculating the systematic deviation of the frequency distribution functions of actually measured daily precipitation (air temperature) and the history simulation series in each month, and deducing correction coefficients corresponding to 0%, 1%, and 100% quantile, and finally using the coefficients for correcting simulated long-series weather data according to the following specific calculation formula: Wherein, the And Respectively correcting the daily precipitation and the air temperature series of the mth month; And Respectively setting a daily precipitation and air temperature series in the mth month before the history period correction; ( ) And ( ) Corresponding to the same quantile; And ( And ) Respectively measuring and simulating series of cumulative distribution functions of precipitation (air temperature) in a history period; And Is the inverse of the cumulative distribution function of the measured series of historic daily precipitation and air temperature.
4. The method for analyzing extreme dry water encounters of Yangtze river based on CMIP6 climate data and SWAT hydrologic model as claimed in claim 1, wherein based on said DEM data, said land utilization and said soil data, establishing said Yangtze river basin and yellow river basin SWAT model in ArcGIS software, inputting said meteorological site observation data to obtain historical period basin simulated runoffs, calibrating parameters closely related to runoffs in model by means of SWAT-CUP parameter calibration tool in combination with said hydrologic site reduced runoff data, forming calibrated SWAT model comprising: Identifying a river network by the DEM data, completing construction of a Hydrological Response Unit (HRU) based on the land utilization data and the soil data, inputting the meteorological site observation data, and operating a SWAT model to obtain historical period river basin simulation runoff data; For the problem of SWAT model multi-site calibration, parameters are partitioned more specifically according to the river basin sites, each hydrological site is the outlet of the parameter partition, the multi-site calibration is split into single site calibration, the principle of calibrating upstream and downstream is followed, and a SWAT-CUP parameter calibration tool is used for completing the parameter calibration work of the model.
5. The method for analyzing the extreme dry encounter of the Yangtze river based on the CMIP6 climate data and the SWAT hydrologic model according to claim 1, wherein the method for analyzing the extreme dry encounter of the Yangtze river is characterized in that the corrected SWAT model with the CMIP6 data driving rate outputs historical and future period analog runoff data, a two-dimensional joint distribution under each climate mode is constructed by utilizing a Gumbel-Hougarrd Copula function, weights are determined by a Bayesian model weighting (BMA) method, the future period weighted joint distribution is obtained, and a full dry encounter matrix is built by combining the historical period (reducing runoff based on the hydrologic site) joint distribution, so that the analysis for the extreme dry encounter of the Yangtze river is developed, and the method comprises the following steps: The key point of the Copula function is that the used Copula function is Gumbel-Hougarrd Copula function, the parameter theta is solved, the expression of the Gumbel-Hougarrd Copula function and the relation between the parameter theta and Kendall rank correlation coefficient tau are as follows: Wherein u and v are edge distribution functions of two variables; The BMA method is used for solving the weight of the two-dimensional joint distribution of the historical period and the future period of each climate mode, and then the weighted average two-dimensional joint distribution is taken as a research object, and the concrete principle is as follows: Let B be the weighted output result, the measured data d= [ X, Y, O ], wherein X, Y is the hydrologic site restored runoff data, O is the empirical frequency calculated by the hydrologic site restored runoff data, f= [ f 1 ,f 1 ,…,f K ] is the set of calculated frequencies obtained by inputting X, Y into K two-dimensional distribution functions, and the empirical frequency O has the following calculation formula: Wherein m (i) is the number of combined observations satisfying the condition x is not more than x i and y is not more than y i in the combined observation sample, and N is the total number of observations; the calculation formula for calculating the frequency f k is as follows: Wherein, the The calculated frequency for the kth Copula function. The probability density function of B can be obtained from the Bayes full probability formula as follows: Wherein, the The posterior probability of the calculated frequency f k of the kth two-dimensional distribution model after the actual measurement data D is given, and the posterior probability represents the adaptation degree of the calculated frequency f k to the empirical frequency O, namely the weight w k calculated by BMA; weighting the posterior distribution of the result B under the conditions of the calculated frequency f k and the measured data D of the k two-dimensional joint distribution function; when the calculated frequency f and the empirical frequency O conform to the normal distribution, the BMA weighted frequency can be derived as: Wherein, the Mean value f k and variance E is a desired function value, w k is the weight calculated by BMA; Weighted average is carried out on Copula functions established by K modes, and a weighted two-dimensional joint distribution function is obtained: Wherein, the Weights assigned to each model after using the BMA method; Kendall rank correlation coefficient as Gumbel-Hougarrd Copula function in kth mode; for the two-dimensional distribution function after the BMA weighting, And For its edge distribution function. And obtaining two-dimensional joint distribution of the historical period according to the runoff data after the hydrologic station is restored, establishing a matrix of the historical period and the future period in the withered meeting, and analyzing the extreme withered meeting of the yellow river of the Yangtze river.
6. A Yangtze river extreme withered water encounter analysis system based on CMIP6 climate data and a SWAT hydrological model, comprising: The acquisition module is used for acquiring and preprocessing multisource basic data, including historical and future period CMIP6 climate pattern data, precipitation and air temperature CN05.1 grid observation data sets, meteorological site observation data, digital Elevation (DEM) data, land utilization data, soil data and runoff data after hydrological site reduction; The correction module is used for performing downscaling processing on the historical and future CMIP6 climate mode data by adopting spatial interpolation to ensure that the historical and future CMIP6 climate mode data are consistent with the spatial resolution of the CN05.1 grid observation data, finishing deviation correction of the downscaling CMIP6 data by combining the CN05.1 data set based on a daily deviation correction (DBC) method of fractional number mapping, and generating corrected historical and future period CMIP6 climate mode data; The modeling module is used for constructing the Yangtze river basin and yellow river basin SWAT model in ArcGIS software based on the DEM data, the land utilization and the soil data, inputting the meteorological site observation data to obtain historical period basin simulated runoffs, and calibrating parameters closely related to the runoffs in the model by means of a SWAT-CUP parameter calibration tool and combining the runoff data after the hydrological site reduction to form a calibrated SWAT model; The analysis module is used for outputting historical and future period simulated runoff data by using the SWAT model after the corrected CMIP6 data driving rate, constructing two-dimensional joint distribution under each climate mode by utilizing Gumbel-Hougarrd Copula function, determining weights by a Bayesian model weighting (BMA) method, obtaining future period weighted joint distribution, establishing a abundant encounter matrix by combining the historical period (reducing runoff based on the hydrological site), and carrying out the extreme dry encounter analysis of the Yangtze river.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the program, implements the method of analyzing yellow-Yangtze river extreme dry encounter based on CMIP6 climate data and SWAT hydrologic model as claimed in any one of claims 1 to 5.

Description

Yangtze river extreme dry water encounter analysis method and system based on CMIP6 climate data and SWAT hydrologic model Technical Field The invention relates to the technical field of water resource prediction and cross-river basin water resource management, in particular to a method and a system for analyzing extreme dry water encounter of a Yangtze river based on CMIP6 climate data and SWAT hydrologic model. Background Ensuring a sufficient source of water to meet the ever-increasing demands of population growth and economic development is a significant challenge facing humans in this century. Among the numerous interventions that provide water resource safety, cross-basin water diversion is a common engineering approach to directly supplementing water resources to reduce water resource shortage areas. Undoubtedly, the water stability of the water source area directly relates to the smooth achievement of the water regulating task and the basic guarantee of the water resource safety of the water receiving area. The inter-government climate change specialization committee (IPCC) was pointed out in the sixth evaluation report (AR 6) that global surface temperatures increased by 0.99 ℃ in 2001-2020 compared to 1850-1900 and that the temperature rise was expected to be no less than 1.5 ℃ in 2021-2040. The climate model shows that in a warmed climate background, extreme rainfall events become more common, have a great influence on runoff, and the frequency and intensity of extreme events such as drought and waterlogging are increased. The interaction of temperature rise and extreme water inflow causes uncertain influence on irrigation, power generation, water supply and reservoir ecology, and brings great risk to cross-river basin water diversion engineering. Sustainability of a cross-basin water transfer project is closely related to changes of water resources, and future operation of the project is very important to consider future trends of runoffs in a water taking area and a receiving area. Future global warming may further threaten the availability of water resources, increasing the uncertainty of water demand across the Yangtze river-yellow river basin. In addition, extreme weather hydrographic events tend to be frequent and concurrent in varying environments. Composite extreme events spanning multiple spatial scales become more frequent, and the difficulty of guaranteeing the water resource safety of large-scale waterbasins increases. For cross-basin water diversion projects, the hydrographic drought connection between the water source water intake and the receiving area is not negligible, and the damage caused by the compound extreme event is larger than that caused by the independent event. Under the background, the multi-watershed water encountering law identification, extreme withered underwater water resource safety guarantee, cross-watershed multi-line multi-water source joint allocation and the like are focuses and research fronts of attention at home and abroad, and are important practical problems to be solved urgently The withered water encounter is essentially one of the hydrologic abundant, and Copula functions are widely used in hydrology, including joint frequency analysis of precipitation, drought, flooding, and other extreme events. The Copula function is simpler in the aspect of calculating continuous probability, can flexibly construct hydrologic variable joint distribution with edge distribution being random distribution, can analyze probability contour maps, and can reflect combination states more intuitively. Meanwhile, the Copula function has no limitation on the edge distribution type, can describe nonlinear and asymmetric correlation among variables, has great flexibility and adaptability, and becomes an important research tool in the field of multivariate hydrologic analysis and calculation. Development of future abundant encounter research is highly dependent on reliable runoff prediction results, and students at home and abroad have developed a great deal of work around runoff prediction under climate change, and form a mature research thought of 'future climate situation-hydrologic simulation-runoff prediction'. Accurate climate prediction is therefore important for studying climate change and its effects. Global climate pattern (GCMs) is an important basis for studying past climate mechanisms and future climate change. In recent years, the world climate research program (WRCP) has become a central element of climate change assessment by coordinating GCMs the coupling pattern comparison programs of past, present and future climate design and distribution. Currently, CMIP has evolved to a sixth stage (CMIP 6), where the climate pattern of CMIP6 has been improved and enhanced in terms of resolution and physical parameterization scheme, as compared to the previous stages. But due to systematic deviations of the modes and the coarser spatial resolution. The downscaling and cor