CN115983511A - Rainfall estimation method and system based on improved statistical downscaling method - Google Patents
Rainfall estimation method and system based on improved statistical downscaling method Download PDFInfo
- Publication number
- CN115983511A CN115983511A CN202310285989.4A CN202310285989A CN115983511A CN 115983511 A CN115983511 A CN 115983511A CN 202310285989 A CN202310285989 A CN 202310285989A CN 115983511 A CN115983511 A CN 115983511A
- Authority
- CN
- China
- Prior art keywords
- precipitation
- data
- gcms
- rainfall
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a precipitation estimation method and system based on an improved statistical downscaling method, which comprises the steps of collecting basic data of a daily scale in a preset range and preprocessing the basic data; acquiring preprocessed basic data, calculating precipitation parameters of observation precipitation data and simultaneous GCMs precipitation data of all ground rainfall stations, and acquiring an average value of the precipitation parameters; calling a CLIGEN model to generate scaled-down GCMs precipitation data based on the averaged precipitation parameters; calculating a spatial correlation coefficient of the observation precipitation data; evaluating the accuracy of the GCMs downscaling precipitation data, if the accuracy meets the requirement, replacing the GCMs precipitation data in the same period with the GCMs precipitation data in the future scene, and performing downscaling treatment to obtain the downscaling precipitation characteristics in the future period. The method can perform downscaling processing on the GCMs precipitation data under the situations of different shared socioeconomic paths and typical concentration paths in the future, predict precipitation in the future period, and has high prediction precision.
Description
Technical Field
The invention relates to a precipitation data estimation method, in particular to a precipitation estimation method and system based on an improved statistical downscaling method.
Background
Global Climate Modes (GCMs) are considered to be the most important and feasible method for researching the influence of climate change at present, can provide important data input for researching the influence of climate change on watershed water circulation, and have important significance for making strategies for coping with climate risk. But limited by the complexity and computational load of global climate systems, GCMs usually provide grid-scale climate change information with a spatial resolution of several hundred kilometers, while hydrologic models usually require watershed or site-specific scale climate data, resulting in mismatch between input data and hydrologic model spatial scale.
Therefore, spatial downscaling of the GCMs data is required. Both dynamic and statistical downscaling methods are used to make up for the gap between the GCMs output and the hydrological model data requirements. The statistical downscaling method can correct the system error of the global climate mode, and is widely used without considering the influence of the boundary condition on the prediction result. As an effective downscaling method, the random weather generator can reproduce a daily sequence with similar statistical characteristics, but most weather generators can only simulate single-site weather data which is irrelevant in space, and the accuracy of the simulation of the extreme flow of the basin and the output of the hydrological model is affected.
Therefore, the inventor provides a multi-site statistics downscaling method, the method performs downscaling processing of time and space scales at the same time through steps of space downscaling, time downscaling, spatial correlation reconstruction and the like, converts data of a coarse grid daily scale into data of a fine grid daily scale, and has the problem that processing precision of daily scale rainfall trend information of different sites in the same grid is insufficient, namely the problem of secondary grid nonuniformity exists during processing. Therefore, further research and innovation is needed.
Disclosure of Invention
The purpose of the invention is as follows: a precipitation estimation method based on an improved statistical downscaling method is provided, and a system for realizing the method is further provided to solve the problems in the prior art.
According to one aspect of the invention, a precipitation estimation method based on an improved statistical downscaling method is provided, and the method comprises the following steps:
s1, acquiring basic data of a daily scale within a preset range, wherein the basic data comprises observed precipitation data of a ground rainfall station and at least two groups of GCMs precipitation data; preprocessing the basic data to obtain preprocessed observation rainfall data and preprocessed GCMs rainfall data;
s2, acquiring the preprocessed basic data, calculating precipitation parameters of the observation precipitation data and the simultaneous GCMs precipitation data of each ground rainfall station, and acquiring an average value of the precipitation parameters; calling a CLIGEN model to generate downscaled GCMs precipitation data based on the averaged precipitation parameters;
s3, calculating spatial correlation coefficients of the observation precipitation data of all the stations, and performing spatial correlation reconstruction on the reduced-scale GCMs precipitation data by using a two-stage shuffle algorithm; reconstructing the annual variation trend of downscaling data of the GCMs in the same period according to the annual variation trend of the observed rainfall data;
and S4, evaluating the accuracy of the GCMs downscaling precipitation data, replacing the GCMs precipitation data in the same period with the GCMs precipitation data in the future scene if the accuracy meets the preset requirement, performing downscaling treatment to obtain downscaling precipitation characteristics in the future period, and drawing related images.
According to an aspect of the application, the step S2 is further:
s21, calculating precipitation parameters of the observation precipitation data and the GCMs precipitation data of the same period of each ground rainfall station, wherein the precipitation parameters comprise a monthly daily precipitation average value, a monthly daily precipitation standard deviation, a monthly daily precipitation skewness coefficient, a monthly daily precipitation-precipitation transition probability and a monthly daily non-precipitation transition probability;
s22, determining four GCM grid center points closest to the ground rainfall station positions according to an angular distance calculation formula; respectively calculating the change rate of the CLIGEN precipitation parameters in each period, and calculating the CLIGEN precipitation parameters in each period at the position of the ground rainfall station by using a bilinear interpolation method;
and S23, performing set averaging on the CLIGEN precipitation parameters of the GCMs, and calling a CLIGEN model to generate reduced GCMs precipitation data based on the averaged precipitation parameters.
According to one aspect of the application, the angular distance calculation formula is: d = cos -1 （cosθ d cosθ s （cosβ d cosβ s +sinθ d sinβ d ）+sinθ d sinθ s ）；θ d Is the target rainfall station latitude;β d a target rainfall station longitude;θ s the latitude of the central point of the GCM data grid is obtained;β s longitude of the center point of the GCM data grid;
the process of calculating the change rate of the CLIGEN precipitation parameter in each period is as follows: CP (CP) m =（CP m-p /CP m-c ）×CP o-c ；
CP m A CLIGEN parameter that is calculated future time period GCM precipitation data;CP m-p a CLIGEN parameter for future period GCM precipitation data;CP m-c a CLIGEN parameter which is historical GCM precipitation data;CP o-c the CLIGEN parameter of the precipitation data was observed for historical periods.
According to an aspect of the application, the step S1 is further:
s11, acquiring GCMs precipitation data, wherein the GCMs precipitation data adopt historical precipitation data and future precipitation data of a preset number of CMIP6 modes; the historical precipitation data is historical test data, and the future precipitation data is SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5 which share social and economic paths and typical concentration path combination scenario data; the historical period GCMs precipitation data preparation needs to be intercepted based on the observation precipitation data time period so that the historical experimental precipitation data of the GCMs are consistent with the observation precipitation data time period;
s12, carrying out statistical interpolation on each GCMs precipitation data to form a grid with preset precision by adopting a bilinear interpolation method;
s13, calculating and adopting a correlation coefficient, a standard root mean square error, a relative standard deviation and a TSS index to evaluate the precision of the precipitation data of the GCMs, and if the TSS index of a certain GCM is larger than a threshold value, using the precipitation data as input data of precipitation estimation and importing the input data into a database;
and S14, performing deviation correction on the GCMs precipitation data with the accuracy reaching the threshold value by adopting an equidistant cumulative distribution function method.
According to an aspect of the application, the step S3 is further:
step S31, the basic data of the daily scale is arranged according to the month, and 12 precipitation data matrixes with the size of n multiplied by k are constructedX]Wherein the data matrix of the observed precipitation is recorded as [ [ alpha ] ]X obs ]The CLIGEN precipitation data matrix is recorded as [, ]X GEN ]N is the total number of days per month in the study period,kthe number of the rainfall stations;
step S32, setting the CLIGEN rainfall data matrixX GEN ]In each row of data, sorting according to size to obtain rank of each dataiVan der Warden score matrix [ S ] based on rank order computation matrix]Wherein S = Φ -1 （i/m+1），SA van der waals score;Φ -1 is the inverse function of the normal distribution;irank for each data in each column;mthe total number of data in each row;
step S33Respectively calculating and observing rainfall data matrixX obs ]The spearman rank correlation coefficient to obtaink×kMatrix of (2)C S-obs ]And judging the matrix [ 2 ]C S-obs ]Whether the matrix is positive or not, and if so, obtaining the matrix [ 2 ] by using the Cholesky decomposition methodC S-obs ]Upper triangular matrix of (2)R]If not, the matrix [ 2 ]C S-obs ]Decomposition into eigenvectorsVAnd a characteristic valueDSubstituting the negative value in the characteristic vector with positive number, and using the corrected characteristic value to make an angular matrixD’Production matrix [ 2 ]C S-obs-m ]And finally will matrix [ 2 ]C S-obs-m ]The matrix is obtained by using Cholesky decomposition method after standardization treatmentC S-obs ]The upper triangular matrix of (2)R]；
Step S34, reconstructing the Van der Waarden score matrix [ 2 ]S * ]The calculation formula is as follows: [S * ]=[S][R](ii) a Calculating the Van der Waals score matrixS * ]In order of (1), the CLIGEN precipitation data matrix [ 2 ]X GEN ]According to the van der Waals score matrixS * ]Is adjusted to obtain a reconstructed matrix [ 2 ]X GEN-R ]；
Step S35, calculate the matrix [ [ alpha ] ]X GEN-R ]The Pearson correlation coefficient matrix of (2)C P-GEN-R ]Establishing a Pearson correlation coefficient matrixC P-GEN-R ]And matrixC S-obs ]The linear relationship of (1): [C S-obs ]=a×[C P-GEN-R ]+b；
Step S36, calculating Pearson correlation coefficient matrix of the observed precipitation dataC P-obs ]Using Pearson correlation coefficient matrix [ 2 ]C P-obs ]Will be the formula [, ]C S-obs ]=a×[C P-GEN-R ]2 in [ + b ]C P-GEN-R ]Performing substitution to obtain a new matrixC S-obs ’ ]；
Step S37, use the matrix [ [ alpha ] ]C S-obs ’ ]Substitution of [ 2 ] in step S35C S-obs ]And repeating the steps S35 to S36 until the precision meets the requirement.
According to one aspect of the application, the method further comprises the following precipitation tendency recovery process:
step S38, calculating correlation coefficients among all stations based on rainfall station observation rainfall sequences, selecting one rainfall station as a control station, and using the rainfall station with the largest average correlation coefficient with other rainfall stations as the control station;
step S39, rearranging the row data according to the precipitation generation structure of the control station in rainy days, randomly arranging the precipitation-free days of the row, and adjusting the row data in other rows; and calculating the annual precipitation order of the observed precipitation data, and adjusting the historical test data of the GCMs at the same site and the same period according to the annual precipitation order of the observed precipitation data to ensure that the annual precipitation trend is unchanged.
According to an aspect of the application, the step S4 is further:
s41, calculating CLIGEN parameters, space correlation coefficients and average absolute errors, standard root mean square errors and decision coefficients of watershed precipitation of the observed precipitation data and historical test data of the GCMs in the same period; evaluating the precision of a downscaling method based on a CLIGEN model and a two-stage shuffle algorithm;
s42, acquiring or calculating precipitation characteristics, wherein the precipitation characteristics comprise average daily, monthly and annual precipitation of each station and extreme precipitation indexes, and calculating the average annual precipitation of a basin in a future period;
and S43, drawing a comparison histogram, a scatter diagram, a box diagram and a frequency curve based on the rainfall characteristics.
According to an aspect of the present application, the step S39 further includes:
s39a, acquiring an observed rainfall sequence in a preset period of the control station, extracting daily rainfall data, dividing preset intervals according to a maximum value and a minimum value, and setting a marker value for each interval; constructing a mapping relation between daily rainfall data and the marker; converting the observation precipitation sequence of each rainfall station into a marker numerical sequence to form a marker numerical sequence set;
s39b, acquiring historical test data of the GCMs in the same period, sequencing the precipitation data according to the annual precipitation order of the observation precipitation data, constructing daily precipitation and constructing an identifier numerical sequence set;
and S39c, calculating the sum of the identifier numerical sequence of the observation precipitation sequence of each station and the Euclidean distance of precipitation data acquired by the GCMs one by one, judging whether the sum is smaller than a preset value, and if the sum is smaller than the preset value, considering that the adjusted annual precipitation trend accords with the expectation.
According to an aspect of the present application, the step S22 further includes:
s22a, reading the position parameter of each ground rainfall station and the position parameter of each GCM grid central point, determining four GCM grid central points closest to each ground rainfall station by adopting an angular distance calculation formula, calculating the repetition times of each GCM grid central point, and constructing a weight coefficient matrix of a GCM grid;
s22b, acquiring GCM precipitation data of a future period and a historical period and CLIGEN precipitation parameters of observation precipitation data, and calculating the change rate of the CLIGEN precipitation parameters of each period; judging whether the difference value of the rainfall parameter change of each ground rainfall station in each grid exceeds a threshold value one by one, and if so, marking the grid as a first downscaling grid;
and S22c, judging whether abnormal points exist in the precipitation parameters of each ground rainfall station in each grid, and if the abnormal points exist, calculating the CLIGEN precipitation parameters of each period at the position of each ground rainfall station by using a bilinear interpolation method.
According to another aspect of the invention, a precipitation estimation system based on an improved statistical downscaling method is provided, and comprises:
at least one processor; and
a memory communicatively coupled to at least one of the processors; wherein the content of the first and second substances,
the memory stores instructions executable by the processor to implement a method of precipitation prediction based on an improved statistical downscaling method according to any one of the preceding claims.
Has the advantages that: according to the invention, by calculating the distance between the rainfall station and the adjacent grid and carrying out interpolation processing, the problem that the rainfall stations in the same GCM output grid can adopt the same change rate in the prior art is solved, and the respective change rate is calculated for the rainfall stations in each output grid, so that the rainfall can be accurately predicted. The application solves the technical problem existing in the prior art all the time.
Drawings
FIG. 1 is an overall flow chart of the present invention.
Fig. 2 is a flow chart of step S2 of the present invention.
FIG. 3 is a scatter plot of the spatial correlation coefficient of the downscaled data versus the spatial correlation coefficient of the observed precipitation data according to the present invention.
FIG. 4 is a box-type plot comparing rainfall prediction data and observation data of the watershed precipitation in the future period of the invention.
FIG. 5 is a schematic diagram of the method before reconstruction of the trend of the river basin year rainfall data in the historical period.
FIG. 6 is a schematic diagram of the trend reconstruction of the river basin year rainfall data in the historical period.
FIG. 7 is a schematic diagram of the method for reconstructing the trend of river basin year precipitation data in the future period.
FIG. 8 is a schematic diagram of the reconstructed trend of river basin year precipitation data in the future period.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that the present invention may be practiced without one or more of these specific details. In other instances, well-known features have not been described in order to avoid obscuring the invention.
Embodiment 1, as shown in fig. 1, a method for predicting precipitation based on an improved statistical downscaling method is provided, which includes the following steps:
s1, acquiring basic data of a daily scale within a preset range, wherein the basic data comprises observation precipitation data of a ground rainfall station and at least two groups of GCMs precipitation data; preprocessing the basic data to obtain preprocessed observation rainfall data and preprocessed GCMs rainfall data;
s2, acquiring the preprocessed basic data, calculating precipitation parameters of the observation precipitation data and the simultaneous GCMs precipitation data of each ground rainfall station, and acquiring an average value of the precipitation parameters; calling a CLIGEN model to generate scaled-down GCMs precipitation data based on the averaged precipitation parameters;
s3, calculating spatial correlation coefficients of the observed precipitation data of each station, and performing spatial correlation reconstruction on the reduced-scale GCMs precipitation data by using a two-stage shuffle algorithm; reconstructing the annual variation trend of downscaling data of the GCMs in the same period according to the annual variation trend of the observed rainfall data;
and S4, evaluating the accuracy of the GCMs downscaling precipitation data, replacing the GCMs precipitation data in the same period with the GCMs precipitation data in the future scene if the accuracy meets the preset requirement, performing downscaling treatment to obtain downscaling precipitation characteristics in the future period, and drawing related images.
Embodiment 2, as shown in fig. 2, the step S2 further comprises:
s21, calculating precipitation parameters of the observed precipitation data and the simultaneous GCMs precipitation data of each ground rainfall station, wherein the precipitation parameters comprise a monthly daily precipitation mean value, a monthly daily precipitation standard deviation, a monthly skewness coefficient of daily precipitation, a monthly daily precipitation-precipitation transition probability and a monthly daily non-precipitation transition probability;
s22, determining four GCM grid center points closest to the ground rainfall station positions according to an angular distance calculation formula; respectively calculating the change rate of the CLIGEN rainfall parameters of each period, and calculating the CLIGEN rainfall parameters of each period at the position of the ground rainfall station by using a bilinear interpolation method;
and S23, performing set averaging on the CLIGEN precipitation parameters of the GCMs, and calling a CLIGEN model to generate reduced GCMs precipitation data based on the averaged precipitation parameters.
Practice ofExample 3, the angular distance calculation formula is: d = cos -1 （cosθ d cosθ s （cosβ d cosβ s +sinθ d sinβ d ）+sinθ d sinθ s ）；θ d Is the target rainfall station latitude;β d is the target rainfall station longitude;θ s the latitude of the central point of the GCM data grid is obtained;β s longitude of the center point of the GCM data grid;
the process of calculating the change rate of the CLIGEN precipitation parameter in each period comprises the following steps: CP (CP) m =（CP m-p /CP m-c ）×CP o-c ；
CP m A CLIGEN parameter that is calculated future time period GCM precipitation data;CP m-p a CLIGEN parameter for future period GCM precipitation data;CP m-c CLIGEN parameter of GCM precipitation data in historical period;CP o-c the CLIGEN parameter of the precipitation data was observed for historical periods.
In the aspect of data selection of future rainfall data estimation, the scheme uses CMIP6 global climate mode data, and compared with the previous climate mode data, the data has more perfect climate change scenes and higher simulation precision, and the precision of the rainfall estimation result can be effectively improved; in the case of downscaling model selection, its application is limited because CLIGEN is a single-site weather generator and there is no mature method to calculate the CLIGEN parameters for future periods of precipitation. The method solves the problem of poor spatial correlation of each station by using a two-stage Shuffles algorithm, and further provides a GLIGEN parameter generation method for precipitation in the future period, so that the application range of a CLIGEN model is greatly improved; in the aspect of annual precipitation tendency, the random weather generator randomly generates precipitation according to probability, so that the rainfall area after the precipitation is difficult to ensure to be the same as the original rainfall sequence, and the identification of the hydrological process of the drainage basin can be influenced.
Aiming at the problem, the invention provides a method for reconstructing the annual precipitation tendency of the downscaling data based on the rank of the original sequence. The spatial correlation and annual precipitation trend reconstruction method provided by the invention can be used as a post-processing technology to be coupled with any single-site scaling-down method, so that the application range of other single-point scaling-down methods is enriched, and a possible way is provided for processing other meteorological data.
The method solves the defects that the existing single-site statistical downscaling method neglects spatial correlation and cannot accurately simulate the annual precipitation trend, and solves the problem of time-space mismatching between climate model output and hydrological model data requirements, and improves the universality of the single-site statistical downscaling method and the accuracy of estimated precipitation data.
Embodiment 4, the step S39 further includes:
s39a, acquiring an observed rainfall sequence in a preset period of the control station, extracting daily rainfall data, dividing preset intervals according to a maximum value and a minimum value, and setting a marker value for each interval; constructing a mapping relation between daily rainfall data and the marker; converting the observation precipitation sequence of each rainfall station into a marker numerical sequence to form a marker numerical sequence set;
s39b, acquiring historical test data of the GCMs in the same period, sequencing the precipitation data according to the annual precipitation order of the observation precipitation data, constructing daily precipitation and constructing an identifier numerical sequence set;
and S39c, calculating the sum of the identifier numerical sequence of the observation precipitation sequence of each station and the Euclidean distance of precipitation data acquired by the GCMs one by one, judging whether the sum is smaller than a preset value, and if the sum is smaller than the preset value, considering that the adjusted annual precipitation trend accords with the expectation.
In the prior art, the future trend is not corrected, and the data trend of the rainfall generated by the GCM is possibly opposite to the actual situation or has a large difference, so that the forecast cannot be better performed.
Embodiment 5, the step S22 further includes:
s22a, reading the position parameter of each ground rainfall station and the position parameter of each GCM grid central point, determining four GCM grid central points closest to each ground rainfall station by adopting an angular distance calculation formula, calculating the repetition times of each GCM grid central point, and constructing a weight coefficient matrix of a GCM grid;
s22b, acquiring GCM precipitation data of a future period and a historical period and CLIGEN precipitation parameters of observation precipitation data, and calculating the change rate of the CLIGEN precipitation parameters of each period; judging whether the difference value of the rainfall parameter change of each ground rainfall station in each grid exceeds a threshold value one by one, and if so, marking the grid as a first downscaling grid;
and S22c, judging whether abnormal points exist in the precipitation parameters of each ground rainfall station in each grid, and if the abnormal points exist, calculating the CLIGEN precipitation parameters of each period at the position of each ground rainfall station by using a bilinear interpolation method.
In order to solve the problem that the accuracy is insufficient in space and future rainfall characteristics cannot be truly reflected through downscaling of monthly scale data to daily scale data in the prior art, the method for downscaling the daily scale rainfall data to the daily scale in space is provided. In the application, the daily scale precipitation data are downscaled to the level of the daily scale rainfall site, the variation trends of different rainfall sites in the same grid can be the same or different, and the real physical process can be reflected more accurately.
Example 6, the procedure of the invention is as follows:
step 1, collecting rainfall data observed by a ground rainfall station and various GCMs (general purpose computing machines) rainfall data in a preset range in a rainfall data preparation module, carrying out quality inspection and screening on the observed rainfall data, determining a used time range, further carrying out interpolation processing on the various GCMs rainfall data, and importing the rainfall data into a database;
step 2, evaluating the precision of the GCMs precipitation data in a GCMs data screening and processing module, screening the GCMs precipitation data, and correcting the deviation of the selected climate mode precipitation data;
step 3, calculating CLIGEN precipitation parameters of the observation precipitation data and the contemporaneous GCMs precipitation data in a GCMs data downscaling module, further, collecting and averaging the CLIGEN precipitation parameters of all GCMs, and calling a CLIGEN model to generate downscaled GCMs precipitation data based on the CLIGEN precipitation parameters of the collection average;
step 4, calculating a spatial correlation coefficient of the observed precipitation data in a GCMs (generalized likelihood monitoring) downscaling data spatial correlation reconstruction module, performing spatial correlation reconstruction on the GCMs downscaling data by using a two-stage shuffle algorithm, and further reconstructing the annual variation trend of the downscaling data of the GCMs in the same period according to the annual precipitation variation trend of the original precipitation data;
step 5, calculating an average Absolute Error (MAE), a Standard Root Mean Square Error (SRMSE) and a decision Coefficient (decision coeffient) in a GCMs data downscaling precision evaluation module, and evaluating the precision of the GCMs downscaling data;
step 6, replacing the 3 kinds of contemporaneous GCMs rainfall data in the step by GCMs rainfall data under the situations of four shared socioeconomic paths and typical concentration paths in the future in a rainfall data estimation module, repeating the step 3 to the step 4, carrying out scale reduction treatment on the rainfall data in the future, and further reconstructing the estimated annual change trend of the rainfall data;
and 7, analyzing the precipitation characteristics in the future period in a future precipitation analysis module, drawing related images, and importing the estimated future precipitation data into a database.
Example 7, the procedure of step 1 was further as follows:
step 11, 20 CMIP6 mode historical and future period rainfall data are adopted in the GCMs data, further, the historical period data are historical test data, and the future period rainfall data are SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5 four kinds of shared socioeconomic path and typical concentration path combination scenario data;
step 12, intercepting the historical period GCMs precipitation data preparation based on the observation precipitation data period to ensure that the historical period GCMs precipitation test precipitation data and the observation precipitation data period have consistency;
and step 13, the interpolation processing is to adopt a bilinear interpolation method to carry out statistical interpolation on all GCMs precipitation data to form a grid with the angle of 0.1 degrees multiplied by 0.1 degrees.
Example 8, the procedure of step 2 was further as follows:
step 2-1, evaluating the accuracy of the GCMs precipitation data by using a standardized Taylor diagram, further, adopting evaluation criteria comprising correlation coefficients, standard root mean square errors and relative standard deviations by the standardized Taylor diagram, and introducingTSS(Taylor Sill Score) the accuracy of the GCMs data was evaluated comprehensively,TSScloser to 1 indicates higher accuracy.
Step 2-2, the judgment standard of screening GCMs precipitation data isTSSValue, ifTSSIf the value is larger than 0.5, the mode is used as input data of rainfall estimation and is imported into a database;
step 2-3, the deviation correction method is an equidistant cumulative distribution function method (EDCDFm);
further, the cumulative distribution function of the precipitation data adopts a cumulative distribution function of mixed two-parameter gamma distribution, and the formula is as follows:G（x）=（1-P）·H（x）+P·F（x) (ii) a In the formula (I), the compound is shown in the specification,Pis the percentage of rainy months;H(x) Is a jump function, 0 in the absence of rain and 1 in the presence of rain;F(x) A Cumulative Distribution Function (CDF) of precipitation data is fitted for the two-parameter gamma distribution. Wherein the two-parameter gamma distribution fits a probability density function of the precipitation data.
Example 9, the procedure of step 3 was further as follows:
step 3-1, the CLIGEN precipitation parameters comprise a monthly daily precipitation mean value (Pw), a monthly precipitation standard deviation (Sd), a skewness coefficient (Sview) of the monthly precipitation, and a monthly daily precipitation transition probability: precipitation-precipitation probability (P (W | W)), and precipitation-free probability (P (W | D)), the calculation formula is as follows:
（P（W|W））=N ww /（N wd +N ww ）；
（P（W|D））=N dw /（N dw +N dd ）
in the formula (I), the compound is shown in the specification,N ww days for precipitation-precipitation;N wd days of precipitation-no precipitation;N dw days of no precipitation-precipitation;N dd days without precipitation-days without precipitation.
And 3-2, calculating the future period rainfall parameters based on a Delta method and a bilinear interpolation method (a Delta-BI method), wherein the Delta method does not consider the secondary network heterogeneity of the GCM, and rainfall stations in the same GCM output grid adopt the same change rate when the Delta method is used for calculating the future rainfall. Further, the Delta-BI method comprises the following calculation process: firstly, determining four GCM grid central points closest to the positions of rainfall stations according to an angular distance formula; secondly, calculating and respectively calculating the change rate of the CLIGEN rainfall parameters in the future period by using a Delta method; thirdly, calculating the CLIGEN rainfall parameters of the rainfall station position in the future period by using a bilinear interpolation method; finally, the CLIGEN precipitation parameters of all GCMs are averaged together. The calculation formula is as follows:
the angular distance calculation formula: d = cos -1 （cosθ d cosθ s （cosβ d cosβ s +sinθ d sinβ d ）+sinθ d sinθ s ）；θ d A target rainfall station latitude;β d a target rainfall station longitude;θ s the latitude of the central point of the GCM data grid is obtained;β s longitude is the center point of the GCM data grid;
process for calculating the rate of change of CLIGEN precipitation parameters for each periodComprises the following steps: CP (CP) m =（CP m-p /CP m-c ）×CP o-c ；
CP m A CLIGEN parameter that is calculated future time period GCM precipitation data;CP m-p CLIGEN parameter for future period GCM precipitation data;CP m-c CLIGEN parameter of GCM precipitation data in historical period;CP o-c the CLIGEN parameter of the precipitation data was observed for historical periods.
Example 10, the procedure of step 4 is further as follows:
step 4-1, daily precipitation data are arranged according to the month to obtain 12 precipitation data with the size of 12n×kMatrix of (2)X]Wherein the observed data are expressed as [ 2 ]X obs ]The CLIGEN data is expressed as [, ]X GEN ]In whichnRepresenting the total number of days per month during the study period,kthe number of the rainfall stations;
step 4-2, mixingX GEN ]In each row of data, sorting according to size to obtain rank of each dataiFurther, a matrix [ S ] of Van der Waerden scales (van der Waerden orders) based on rank calculation matrix]The calculation formula is as follows: s = Φ -1 (i/m + 1) wherein S is a van der Waals score; phi (phi) of -1 Is an inverse function of the normal distribution; i is the rank of each data in each column; m is the total number of data in each column.
Step 4-3, calculating the observation data matrix respectivelyX obs ]The Spearman rank correlation coefficient (Spearman rank correlation coefficient) of (1) to obtaink×kMatrix of (2)C S-obs ]；
Step 4-4, judging [ [ 2 ] ]C S-obs ]If the matrix is positive, the matrix is obtained by using Cholesky decomposition methodC S-obs ]The upper triangular matrix of (2)R]If not, further converting the value of [ 2 ]C S-obs ]Decomposition into eigenvectorsVAnd a characteristic valueDSubstituting the negative values in the eigenvector with very small positive numbers, and further, using the corrected eigenvalues to make an angular matrixD’Production [ 2 ]C S-obs-m ]Finally, the termC S-obs-m ]The matrix is obtained by using Cholesky decomposition method after standardization treatmentC S-obs ]The upper triangular matrix of (2)R]The normalized formula is:C S-obs =C S-obs-m /（diag（C S-obs-m ）diag（C S-obs-m ）’） 1/2 (ii) a In the formula (I), the compound is shown in the specification,C S-obs the Spearman rank correlation coefficient matrix for the observed precipitation after normalization processing and feature vector correction.
Step 4-5, obtaining the matrix by using Cholesky decomposition methodC S-obs ]Upper triangular matrix of [ R ]]；
Step 4-6, reconstructing the van der Waals score matrix [ 2 ]S * ]The calculation formula is as follows: [S * ]=[S][R]；
Step 4-7, calculating [ solution ]S * ]Further, the matrix [ 2 ]X GEN ]According to [ 2 ]S * ]Is adjusted to obtain a reconstructed matrix [ 2 ]X GEN-R ]；
Step 4-8, calculating the matrixX GEN-R ]The Pearson correlation coefficient [ c ], [C P-GEN-R ]Creation ofC P-GEN-R ]And 2C S-obs ]The linear relationship of (1): [C S-obs ]=a×[C P-GEN-R ]+b；
Step 4-9, calculating Pearson correlation coefficient of observed precipitation dataC P-obs ]Use ofC P-obs ]Subjecting the product of step 4 to step 8 to the formula [, ]C P-GEN-R ]Performing substitution to obtain a new matrixC S-obs ’ ]；
Step 4 to 10, useC S-obs ’ ]Replacing [ 2 ] in step 4-5C S-obs ]And repeating the steps 4-5 to 4-9 until the precision meets the requirement.
Example 11, further includes a trend recovery process, which is specifically as follows:
4-11, selecting one rainfall station as a control station, wherein the selection standard is that correlation coefficients among all stations are calculated based on rainfall sequence observed by the rainfall station, and the rainfall station with the largest average correlation coefficient with other rainfall stations is used as the control station;
4-12, rearranging the row of data according to the precipitation generation structure of the control station in rainy days, randomly arranging the precipitation-free days of the row, and further adjusting the row of data in other rows;
and 4-13, calculating the annual precipitation order of the observed precipitation data, and further adjusting the historical test data of the GCMs at the same site and the same period according to the annual precipitation order of the observed precipitation data so as to ensure that the annual precipitation trend is unchanged.
Example 12, the procedure of step 5 was further as follows:
step 5-1, calculating CLIGEN parameters, space correlation coefficients and average absolute errors, standard root mean square errors and decision coefficients of watershed precipitation of observed precipitation data and historical test data of GCMs in the same period;
and 5-2, evaluating the precision of the downscaling method based on the CLIGEN model and the two-stage shuffle algorithm.
Example 13, the procedure of step 6 was further as follows:
step 6-1, the future four shared socioeconomic routes and typical concentration route scenes comprise SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5;
6-2, repeating the steps 3-1 to 4-12, and estimating precipitation data of different scenes in a future period;
and 6-3, calculating the annual precipitation order of the original GCMs in the future period, and further adjusting the annual precipitation data of the GCMs of the station after the scale reduction according to the annual precipitation order of the original GCMs so as to ensure that the annual precipitation trend is unchanged.
Example 14, the procedure of step 7 was further as follows:
and 7-1, the precipitation characteristics comprise average daily, monthly and annual precipitation of each site, extreme precipitation indexes and average annual precipitation of the basin in the future period.
And 7-2, drawing a comparison histogram, a scatter diagram, a box type diagram and a frequency curve based on the calculation result of the step 7-1.
Taking a certain river basin as an example, the water resource of the river basin is relatively scarce, and the spatial and temporal distribution of the river basin is uneven. The ecological environment and socioeconomic development of the watershed are very sensitive to climatic changes, especially to precipitation changes. The watershed is not influenced by resource development at present, and the hydrological process is mainly controlled by natural factors. In this embodiment, the daily precipitation data observed at 10 sites in a certain watershed and the 20 GCMs precipitation data of CMIP6 are used.
The method mainly comprises the following steps:
s1, preparing precipitation data: data on observed daily precipitation at 10 sites in a watershed from 1976 to 2014, historical experimental precipitation data on 20 GCMs of CMIP6 from 1976 to 2014, and 7 combined scenario precipitation data on 20 GCMs of CMIP6 from 2015 to 2100 were collected, all interpolated on a 0.1 ° × 0.1 ° grid.
And S2, respectively evaluating the precision of 20 GCMs precipitation data by using a standardized Taylor diagram, screening the GCMs precipitation data, carrying out deviation correction on the screened GCMs precipitation data by using an equidistant cumulative distribution function method (EDCDFm method), and further calculating the annual precipitation statistical characteristics of GCMs averaged by a multimode set for comparing the precision improvement conditions before and after the deviation correction. The data are specifically as follows: the mean of the observed data was 378.56, the standard deviation was 84.64;
before offset correction GCMs data: mean 502.65, standard deviation 23.82, mean absolute error: the relative deviation of 131.40 is 0.45 and the root mean square error is 156.24.
Pre-offset-bound GCMs: the mean was 375.46, the standard deviation was 82.05, the mean absolute error was 1.51, the relative deviation was 0.005, and the root mean square error was 103. And S3, calculating CLIGEN precipitation parameters of the observation precipitation data and the GCMs precipitation data, further generating downscaling contemporaneous GCMs precipitation data based on the precipitation parameters, performing spatial correlation reconstruction on the GCMs precipitation data by using a two-stage shuffle algorithm, further reconstructing the annual precipitation variation trend of the historical GCMs precipitation data according to the annual precipitation variation of the observation precipitation data, and reconstructing the annual precipitation variation trend of the future GCMs precipitation data according to the annual precipitation variation of the original GCMs precipitation data.
As shown in fig. 3, fig. 3 is an embodiment of the present invention: and (4) a scatter diagram of the spatial correlation coefficient of the downscaling data and the spatial correlation coefficient of the observation rainfall data.
And S4, calculating the average daily, monthly and annual precipitation of each site in the future period and an extreme precipitation index, calculating the average annual precipitation of the drainage basin in the future period, and analyzing precipitation characteristics in the future period.
Fig. 4 is a box-type graph comparing rainfall amount of the forecast data and the observation data of the rainfall in the watershed of the future period, as shown in fig. 4. It can be seen from fig. 4 that the effect of the present application is superior to that of the prior art.
In the embodiment, aiming at the condition that a plurality of days of gas generators at present use a single station as a research object, the downscaling data of each station lacks spatial correlation, and the extreme value simulation precision of hydrological elements in a flow domain is poor. The method is based on the CLIGEN random day generator and a two-stage Shuffle algorithm, firstly, a method for generating CLIGEN precipitation parameters in the future period is provided, and the application of the CLIGEN random day generator in precipitation estimation is expanded; secondly, the spatial correlation of precipitation of each station is reconstructed by using a two-stage Shuffle algorithm, the method can also be used as a post-processing technology to be combined with other statistical downscaling methods, the conversion from a single-station downscaling method to a multi-station downscaling method is realized, and the method has a wide application prospect; finally, an annual precipitation trend reconstruction method based on the original precipitation sequence order is provided, the problem that annual precipitation data generated by a statistical downscaling method cannot guarantee the trend is solved, and a method system in the statistical downscaling field is expanded. In a specific embodiment, it is found that the new downscaling method can be adopted to obtain a conclusion that the method is not completely consistent with the single-site downscaling method, that is, although the single-site downscaling method has higher simulation precision in the aspect of annual precipitation mean value, the range of the generated watershed precipitation is too small because the spatial correlation of each site is not considered, and the simulation of extreme precipitation is not good.
As shown in fig. 5 to 8, in the embodiment of the present invention: the trend of the river basin year rainfall forecast data in the historical period and the future period is reconstructed. The Euclidean distances between measured annual precipitation and duration periods MME (multimode average), BCC-CSM2-MR and CMCC-ESM2 are 228, 322 and 310 respectively, so that reconstruction can be carried out according to the trend of MME in the future.
The above embodiments relate mainly to geographical information, rain station observation and precipitation products of 20 global climate modes in terms of information utilization. It should be noted that the method has strong expansibility, and other single-site downscaling methods can all adopt the method. As noted above, while the present invention has been shown and described with reference to certain preferred embodiments, it is not to be construed as limited thereto. Various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. The rainfall estimation method based on the improved statistical downscaling method is characterized by comprising the following steps of:
s1, acquiring basic data of a daily scale within a preset range, wherein the basic data comprises observed precipitation data of a ground rainfall station and at least two groups of GCMs precipitation data; preprocessing the basic data to obtain preprocessed observation rainfall data and preprocessed GCMs rainfall data;
s2, acquiring the preprocessed basic data, calculating precipitation parameters of the observation precipitation data and the simultaneous GCMs precipitation data of each ground rainfall station, and acquiring an average value of the precipitation parameters; calling a CLIGEN model to generate downscaled GCMs precipitation data based on the averaged precipitation parameters;
s3, calculating spatial correlation coefficients of the rainfall data observed by each rainfall station, and performing spatial correlation reconstruction on the reduced-scale GCMs (generalized computer systems) rainfall data by using a two-stage shuffle algorithm; reconstructing the annual variation trend of downscaling data of the GCMs in the same period according to the annual precipitation variation trend of the observed precipitation data;
and S4, evaluating the accuracy of the GCMs downscaling precipitation data, replacing the GCMs precipitation data in the same period with the GCMs precipitation data in the future scene if the accuracy meets the preset requirement, performing downscaling treatment to obtain downscaling precipitation characteristics in the future period, and drawing related images.
2. The method for estimating precipitation based on the improved statistical downscaling method according to claim 1, wherein the step S2 further comprises:
s21, calculating precipitation parameters of the observed precipitation data and the simultaneous GCMs precipitation data of each ground rainfall station, wherein the precipitation parameters comprise a monthly daily precipitation mean value, a monthly daily precipitation standard deviation, a monthly skewness coefficient of daily precipitation, a monthly daily precipitation-precipitation transition probability and a monthly daily non-precipitation transition probability;
s22, determining four GCM grid central points closest to the ground rainfall station according to an angular distance calculation formula; respectively calculating the change rate of the CLIGEN rainfall parameters of each period, and calculating the CLIGEN rainfall parameters of each period at the position of the ground rainfall station by using a bilinear interpolation method;
and S23, performing collective averaging on the CLIGEN precipitation parameters of the GCMs, and calling a CLIGEN model to generate reduced GCMs precipitation data based on the averaged precipitation parameters.
3. The improved statistical downscaling method based precipitation estimation method according to claim 2, wherein the angular distance calculation formula is: d = cos -1 （cosθ d cosθ s （cosβ d cosβ s +sinθ d sinβ d ）+sinθ d sinθ s ）；θ d A target rainfall station latitude;β d is the target rainfall station longitude;θ s the latitude of the central point of the GCM data grid is obtained;β s longitude of the center point of the GCM data grid;
the process of calculating the change rate of the CLIGEN precipitation parameter in each period is as follows: CP (CP) m =（CP m-p /CP m-c ）×CP o-c ；
CP m A CLIGEN parameter that is calculated future time period GCM precipitation data;CP m-p CLIGEN parameter for future period GCM precipitation data;CP m-c CLIGEN parameter of GCM precipitation data in historical period;CP o-c the CLIGEN parameter of the precipitation data was observed for historical periods.
4. The method for predicting precipitation based on the improved statistical downscaling method according to any one of claims 1 to 3, wherein the step S1 further comprises:
s11, acquiring GCMs precipitation data, wherein the GCMs precipitation data adopt historical precipitation data and future precipitation data of a preset number of CMIP6 modes; the historical precipitation data is historical test data, and the future precipitation data is SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5 which share social and economic paths and typical concentration path combination scenario data; the historical period GCMs precipitation data preparation needs to be intercepted based on the observation precipitation data time period so that the historical experimental precipitation data of the GCMs are consistent with the observation precipitation data time period;
s12, carrying out statistical interpolation on each GCMs precipitation data to form a grid with preset precision by adopting a bilinear interpolation method;
s13, calculating and adopting a correlation coefficient, a standard root mean square error, a relative standard deviation and a TSS index to evaluate the precision of the precipitation data of the GCMs, and if the TSS index of a certain GCM is larger than a threshold value, using the precipitation data as input data of precipitation estimation and importing the input data into a database;
and S14, performing deviation correction on the GCMs precipitation data with the accuracy reaching the threshold value by adopting an equidistant cumulative distribution function method.
5. The method for estimating precipitation according to claim 4, wherein the step S3 is further to:
step S31, the basic data of the daily scale is divided intoArranging the moon, constructing 12 precipitation data matrixes with the size of n multiplied by kX]Wherein the data matrix of the observed precipitation is recorded as [ [ alpha ] ]X obs ]The CLIGEN precipitation data matrix is recorded as [, ]X GEN ]N is the total number of days per month in the study period,kthe number of the rainfall stations;
step S32, making the CLIGEN rainfall data matrix [ 2 ]X GEN ]In each row of data, sorting according to size to obtain rank of each dataiVan der Warden score matrix [ S ] based on rank order computation matrix]Wherein S = Φ -1 (i/m + 1), S is Van der Waals score; phi -1 Is the inverse function of the normal distribution; i is the rank of each data in each column; m is the total number of data in each row;
step S33, calculating the matrix of the data of the observed precipitationX obs ]Is obtained by the spearman rank correlation coefficient ofk×kMatrix of (2)C S-obs ]And judging the matrix [ 2 ]C S-obs ]Whether the matrix is a positive definite matrix or not, and if so, obtaining the matrix by using the Cholesky decomposition methodC S-obs ]The upper triangular matrix of (2)R]If not, the matrix is further setC S-obs ]Decomposition into eigenvectorsVAnd a characteristic valueDSubstituting the negative value in the characteristic vector with positive number, and using the corrected characteristic value to make an angular matrixD’Generating matrix [ 2 ]C S-obs-m ]Finally, the matrix [ 2 ]C S-obs-m ]The matrix is obtained by using Cholesky decomposition method after standardization treatmentC S-obs ]Upper triangular matrix of (2)R]；
Step S34, reconstructing the Van der Waarden score matrix [ 2 ]S * ]The calculation formula is as follows: [S * ]=[S] [R](ii) a Calculating the Van der Waals score matrixS * ]In order of (1), the CLIGEN precipitation data matrix [ 2 ]X GEN ]According to the van der Waals score matrixS * ]Is adjusted to obtain a reconstructed matrix [ 2 ]X GEN-R ]；
Step S35, calculating the matrix [ [ 2 ] ]X GEN-R ]The Pearson correlation coefficient matrix of (2)C P-GEN-R ]Establishing a Pearson correlation coefficient matrixC P-GEN-R ]And matrixC S-obs ]The linear relationship of (c): [C S-obs ]=a×[C P-GEN-R ]+b；
Step S36, calculating Pearson correlation coefficient matrix of the observed precipitation dataC P-obs ]Using Pearson correlation coefficient matrix [ ]C P-obs ]Will be the formula [, ]C S-obs ]=a×[C P-GEN-R ]+ b inC P-GEN-R ]Performing substitution to obtain a new matrixC S-obs ’ ]；
Step S37, use the matrix [ [ alpha ] ]C S-obs ’ ]Substitution of the matrix in step S35C S-obs ]And repeating the steps S35 to S36 until the precision meets the requirement.
6. The method for predicting precipitation based on the improved statistical downscaling method according to claim 5, wherein the step S3 further comprises a precipitation trend recovery process:
step S38, calculating correlation coefficients among all stations based on rainfall station observation rainfall sequences, selecting one rainfall station as a control station, and using the rainfall station with the largest average correlation coefficient with other rainfall stations as the control station;
s39, rearranging the row data according to a precipitation generation structure of the control station in rainy days, randomly arranging the precipitation-free days of the row, and adjusting the row of data in other rows; and calculating the annual precipitation order of the observed precipitation data, and adjusting the historical test data of the GCMs at the same site and the same period according to the annual precipitation order of the observed precipitation data so as to ensure that the annual precipitation trend is unchanged.
7. The method for predicting precipitation based on the improved statistical downscaling method according to claim 6, wherein the step S4 further comprises:
s41, calculating CLIGEN parameters, space correlation coefficients and average absolute errors, standard root mean square errors and decision coefficients of watershed precipitation of the observed precipitation data and historical test data of the GCMs in the same period; evaluating the precision of a downscaling method based on a CLIGEN model and a two-stage shuffle algorithm;
s42, acquiring or calculating precipitation characteristics, wherein the precipitation characteristics comprise average daily, monthly and annual precipitation of each station and extreme precipitation indexes, and calculating the average annual precipitation of a basin in a future period;
and S43, drawing a comparison histogram, a scatter diagram, a box diagram and a frequency curve based on the rainfall characteristics.
8. The method for predicting precipitation based on the improved statistical downscaling method of claim 7, wherein the step S39 further comprises:
s39a, acquiring an observed rainfall sequence in a preset period of the control station, extracting daily rainfall data, dividing preset intervals according to a maximum value and a minimum value, and setting a marker value for each interval; constructing a mapping relation between daily rainfall data and the marker; converting the observation precipitation sequence of each rainfall station into a marker numerical sequence to form a marker numerical sequence set;
s39b, acquiring historical test data of the GCMs in the same period, sequencing the precipitation data according to the annual precipitation order of the observation precipitation data, constructing daily precipitation and constructing an identifier numerical sequence set;
and S39c, calculating the sum of the identifier numerical sequence of the observation precipitation sequence of each station and the Euclidean distance of precipitation data acquired by the GCMs one by one, judging whether the sum is smaller than a preset value, and if the sum is smaller than the preset value, considering that the adjusted annual precipitation trend accords with the expectation.
9. The method for predicting precipitation based on the improved statistical downscaling method according to claim 8, wherein the step S22 further comprises:
s22a, reading the position parameter of each ground rainfall station and the position parameter of each GCM grid central point, determining four GCM grid central points closest to each ground rainfall station by adopting an angular distance calculation formula, calculating the repetition times of each GCM grid central point, and constructing a weight coefficient matrix of a GCM grid;
s22b, acquiring GCM precipitation data of a future period and a historical period and CLIGEN precipitation parameters of observation precipitation data, and calculating the change rate of the CLIGEN precipitation parameters of each period; judging whether the difference value of the rainfall parameter change of each ground rainfall station in each grid exceeds a threshold value one by one, and if so, marking the grid as a first downscaling grid;
and S22c, judging whether abnormal points exist in the precipitation parameters of each ground rainfall station in each grid, and if the abnormal points exist, calculating the CLIGEN precipitation parameters of each period at the position of each ground rainfall station by using a bilinear interpolation method.
10. Precipitation prediction system based on improved statistics downscaling method, characterized by comprising:
at least one processor; and
a memory communicatively coupled to at least one of the processors; wherein, the first and the second end of the pipe are connected with each other,
the memory stores instructions executable by the processor to implement the method of precipitation prediction based on an improved statistical downscaling method of any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310285989.4A CN115983511B (en) | 2023-03-22 | 2023-03-22 | Precipitation prediction method and system based on improved statistical downscaling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310285989.4A CN115983511B (en) | 2023-03-22 | 2023-03-22 | Precipitation prediction method and system based on improved statistical downscaling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115983511A true CN115983511A (en) | 2023-04-18 |
| CN115983511B CN115983511B (en) | 2023-05-23 |
Family
ID=85970588
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310285989.4A Active CN115983511B (en) | 2023-03-22 | 2023-03-22 | Precipitation prediction method and system based on improved statistical downscaling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115983511B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116449460A (en) * | 2023-04-23 | 2023-07-18 | 水利部交通运输部国家能源局南京水利科学研究院 | Regional month precipitation prediction method and system based on convolution UNet and transfer learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1744185A1 (en) * | 2005-07-13 | 2007-01-17 | CIMA-Università degli Studi di Genova | Method for estimating a variable geophysical field particularly for forecasting atmospheric precipitation |
| CN110597873A (en) * | 2019-08-23 | 2019-12-20 | 北京师范大学 | Precipitation data estimation method, precipitation data estimation device, precipitation data estimation equipment and storage medium |
| CN110874630A (en) * | 2019-10-23 | 2020-03-10 | 中国人民解放军空军研究院战场环境研究所 | Deep learning-based numerical model product downscaling refinement method |
| CN114677059A (en) * | 2022-05-26 | 2022-06-28 | 水利部交通运输部国家能源局南京水利科学研究院 | Method and system for comprehensively evaluating precision of inversion precipitation product by integrating time-space indexes |
| CN115357847A (en) * | 2022-08-28 | 2022-11-18 | 水利部交通运输部国家能源局南京水利科学研究院 | Day scale star-ground precipitation fusion method based on error decomposition |
-
2023
- 2023-03-22 CN CN202310285989.4A patent/CN115983511B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1744185A1 (en) * | 2005-07-13 | 2007-01-17 | CIMA-Università degli Studi di Genova | Method for estimating a variable geophysical field particularly for forecasting atmospheric precipitation |
| CN110597873A (en) * | 2019-08-23 | 2019-12-20 | 北京师范大学 | Precipitation data estimation method, precipitation data estimation device, precipitation data estimation equipment and storage medium |
| CN110874630A (en) * | 2019-10-23 | 2020-03-10 | 中国人民解放军空军研究院战场环境研究所 | Deep learning-based numerical model product downscaling refinement method |
| CN114677059A (en) * | 2022-05-26 | 2022-06-28 | 水利部交通运输部国家能源局南京水利科学研究院 | Method and system for comprehensively evaluating precision of inversion precipitation product by integrating time-space indexes |
| CN115357847A (en) * | 2022-08-28 | 2022-11-18 | 水利部交通运输部国家能源局南京水利科学研究院 | Day scale star-ground precipitation fusion method based on error decomposition |
Non-Patent Citations (5)
| Title |
|---|
| JAMIL A.A. ANACHE等: "Land use and climate change impacts on runoff and soil erosion at the hillslope scale in the Brazilian Cerrado", 《SCIENCE OF THE TOTAL ENVIRONMENT》 * |
| XIAOPING SHI等: "Multisite and multivariate GCM downscaling using a distribution-free shuffle procedure for correlation reconstruction", 《CLIMATE RESEARCH》 * |
| YUQUAN QU等: "Inter-comparison of several soil moisture downscaling methods over the Qinghai-Tibet Plateau, China", 《JOURNAL OF HYDROLOGY》 * |
| 殷水清: "土壤侵蚀研究中降雨过程随机模拟综述", 《地理科学进展》 * |
| 胡庆芳等: "环境变化对流域水文水资源的影响评估及不确定性研究进展", 《气候变化研究进展》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116449460A (en) * | 2023-04-23 | 2023-07-18 | 水利部交通运输部国家能源局南京水利科学研究院 | Regional month precipitation prediction method and system based on convolution UNet and transfer learning |
| CN116449460B (en) * | 2023-04-23 | 2023-10-10 | 水利部交通运输部国家能源局南京水利科学研究院 | Regional month precipitation prediction method and system based on convolution UNet and transfer learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115983511B (en) | 2023-05-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nipen et al. | Adopting citizen observations in operational weather prediction | |
| CN108761574B (en) | Rainfall estimation method based on multi-source information fusion | |
| Cheung et al. | Application of macroecological theory to predict effects of climate change on global fisheries potential | |
| CN111027175A (en) | Method for evaluating social and economic influences of flood based on coupling model integrated simulation | |
| CN113032733B (en) | Remote sensing rainfall error correction method and system based on nonlinear classification regression analysis | |
| CN109359411B (en) | Estimation method for vegetation coverage of marsh wetland under influence of climate change | |
| CN114254802B (en) | Prediction method for vegetation coverage space-time change under climate change drive | |
| Marchesoni-Acland et al. | Intra-day solar irradiation forecast using RLS filters and satellite images | |
| CN112861072A (en) | Satellite-ground multi-source rainfall self-adaptive dynamic fusion method | |
| CN115983511A (en) | Rainfall estimation method and system based on improved statistical downscaling method | |
| Otache et al. | ARMA modelling of Benue River flow dynamics: comparative study of PAR model | |
| Chen et al. | Trust me, my neighbors say it's raining outside: ensuring data trustworthiness for crowdsourced weather stations | |
| CN110119556B (en) | Space-time evolution analysis method for regional water source conservation function | |
| Hou et al. | Improving the forecast accuracy of ECMWF 2-m air temperature using a historical dataset | |
| Li et al. | A nonstationary runoff frequency analysis for future climate change and its uncertainties | |
| Wang et al. | Spatial Variation of Extreme Rainfall Observed From Two Century‐Long Datasets | |
| Monjo et al. | Estimation of future extreme rainfall in Barcelona (Spain) under monofractal hypothesis | |
| CN117094516B (en) | Urban group month living water demand prediction method based on fixed effect model | |
| Kreuwel et al. | Forecasting day-ahead 1-minute irradiance variability from numerical weather predictions | |
| Boyko et al. | Post-processing climate projections of precipitation for the Po river basin: will Italy's North become water-constrained? | |
| CN116151015A (en) | Meteorological forecast downscaling algorithm | |
| CN116341841A (en) | Runoff forecast error correction method, apparatus, device, medium and program product | |
| CN115936238A (en) | Method, system, equipment and medium for predicting middle and long term output of global wind power | |
| CN109886497B (en) | Ground air temperature interpolation method based on latitude improved inverse distance weighting method | |
| Huang et al. | Hybrid Intra-hour Solar PV Power Forecasting using Statistical and Skycam-based Methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |