CN114910981A - Quantitative evaluation method and system for rainfall forecast overlap and newly added information in adjacent forecast periods - Google Patents
Quantitative evaluation method and system for rainfall forecast overlap and newly added information in adjacent forecast periods Download PDFInfo
- Publication number
- CN114910981A CN114910981A CN202210675212.4A CN202210675212A CN114910981A CN 114910981 A CN114910981 A CN 114910981A CN 202210675212 A CN202210675212 A CN 202210675212A CN 114910981 A CN114910981 A CN 114910981A
- Authority
- CN
- China
- Prior art keywords
- forecast
- precipitation
- data
- adjacent
- rainfall
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- 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
- Environmental & Geological Engineering (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of precipitation forecast, and provides a quantitative evaluation method and a quantitative evaluation system for precipitation forecast overlap and new information of adjacent forecast periods, wherein the quantitative evaluation method comprises the following steps: acquiring rainfall forecast data of adjacent forecast periods to be evaluated and observation rainfall data corresponding to the rainfall forecast data as original samples; determining certainty coefficients of the observed precipitation data and precipitation forecast data of adjacent forecast periods; according to the deterministic coefficients of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods, the deterministic coefficients of repeated interpretation of the precipitation forecasts of the adjacent forecast periods and the deterministic coefficients of the independent interpretation of the precipitation forecasts of the target forecast period are calculated on the basis of the set operation and are used as the quantization results of the overlapping and newly added information of the precipitation forecasts of the adjacent forecast periods; and calculating the quantitative result reference distribution of the rainfall forecast overlap and the newly added information of the adjacent forecast periods, and performing unilateral significance test to obtain a significance result for evaluating the rainfall forecast overlap and the newly added information of the adjacent forecast periods.
Description
Technical Field
The invention relates to the technical field of precipitation forecast, in particular to a quantitative evaluation method and system for precipitation forecast overlap and new information in adjacent forecast periods.
Background
The accurate seasonal rainfall forecast has important value and wide application prospect in the fields of disaster prevention and reduction, water resource planning and management and the like of natural disasters such as flooding, drought and the like. The major climate centers around the world began developing Global Climate Models (GCMs) that produced valuable Global forecasts of precipitation under different scenarios by characterizing various key climate-related physical processes. In different scenarios, the forecast provides information with overlapping intersections and portions of variance. Wherein, the forecast period has a great influence on the generated forecast information.
At present, it is proposed to compare precipitation forecast and observed precipitation data subjected to spatial interpolation processing in prediction areas in different prediction time periods, calculate data difference indexes, and use the data difference indexes for precipitation forecast evaluation according to all the data difference indexes. However, in practical applications, it is difficult to determine whether effective information is provided by rainfall forecasts in adjacent forecast periods, whether overlapping or newly added information exists between adjacent forecast periods, how much overlapping or newly added information is, whether results are significant, and the like, and there is a certain limitation.
Disclosure of Invention
The invention provides a quantitative evaluation method and system for rainfall forecast overlap and newly added information in adjacent forecast periods, aiming at overcoming the defects that whether effective information is provided by rainfall forecasts in adjacent forecast periods or not is difficult to judge, the amount of overlapped or newly added information cannot be quantized, and certain limitation exists in the existing rainfall forecast evaluation method.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a quantitative evaluation method for rainfall forecast overlap and new information of adjacent forecast periods comprises the following steps:
and S1, acquiring precipitation forecast data of adjacent forecast periods to be evaluated and corresponding observed precipitation data as original samples.
And S2, determining the certainty coefficient of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods according to the regression equation of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the regression equation of the union set of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the precipitation forecast data of the adjacent forecast periods.
And S3, calculating a certainty coefficient repeatedly explained by the rainfall forecast of the adjacent forecast periods and a certainty coefficient separately explained by the rainfall forecast of the target forecast period based on collective operation according to the certainty coefficient of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods, and using the certainty coefficient as the quantization results of the superposition of the rainfall forecast of the adjacent forecast periods and the newly added information.
And S4, calculating the quantized result reference distribution of the adjacent forecast precipitation forecast overlap and the newly added information, and performing single-side significance test according to the quantized result reference distribution and the original sample to obtain a significance result for evaluating the adjacent forecast precipitation forecast overlap and the newly added information.
Furthermore, the invention also provides a quantitative evaluation system of the rainfall forecast overlap and the newly added information of the adjacent forecast periods, and the quantitative evaluation method of the rainfall forecast overlap and the newly added information of the adjacent forecast periods is applied. Which comprises the following steps:
and the observation rainfall query module is used for querying and acquiring observation rainfall data of corresponding time and area according to the input rainfall forecast data of the adjacent forecast periods to be evaluated.
And the certainty coefficient calculation module is used for constructing a regression equation of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods, constructing a regression equation of a union set of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods, and calculating the certainty coefficient of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods through the regression equation.
And the quantification module is used for calculating the certainty coefficient repeatedly explained by the rainfall forecast in the adjacent forecast periods based on collective operation according to the certainty coefficient of the observed rainfall data and the rainfall forecast data in the adjacent forecast periods, calculating the certainty coefficient independently explained by the rainfall forecast in the target forecast period, and outputting the calculated certainty coefficient as the quantification result of the superposition of the rainfall forecast in the adjacent forecast periods and the newly added information.
And the significance evaluation module is used for calculating the quantized result reference distribution of the adjacent forecast period rainfall forecast overlap and the newly added information, performing single-side significance test according to the quantized result reference distribution and the original sample, and outputting a significance result for evaluating the adjacent forecast period rainfall forecast overlap and the newly added information.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that: the method combines set operation and linear regression deterministic coefficient, further carries out significance test on the distribution of the deterministic coefficient, simply and effectively judges whether the rainfall forecast of adjacent forecast periods provides effective information or not, quantifies the overlapped or newly increased information amount in the rainfall forecast of the adjacent forecast periods, and effectively distinguishes overlapped and newly increased components in the observed rainfall information provided by the rainfall forecast of the adjacent forecast periods, thereby providing reference for the business use of the rainfall forecast.
Drawings
Fig. 1 is a flowchart of a quantitative evaluation method for adjacent forecast precipitation forecast overlap and new information according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of the difference and overlap of precipitation forecasts in adjacent forecast periods.
Fig. 3 is a spatial distribution diagram of observed precipitation information provided by forecast precipitation with forecast age of 0 months.
Fig. 4 is a spatial distribution diagram of observed precipitation information provided by forecasted precipitation with a forecast period of 1 month.
Fig. 5 is a spatial distribution diagram of observed precipitation information provided by forecast precipitation with forecast periods of 0 and 1 month.
Fig. 6 is a spatial distribution diagram of overlapping components in observed precipitation information provided by forecasted precipitation for 0 and 1 month forecasted periods.
Fig. 7 is a spatial distribution plot of new components in observed precipitation information provided for a forecast period of 0 months versus a forecast period of 1 month.
Fig. 8 is a spatial distribution diagram of significance test results of forecast precipitation overlap information with forecast periods of 0 month and 1 month.
Fig. 9 is a spatial distribution diagram of the significance test result of the forecast rainfall increase information with a forecast period of 0 month relative to a forecast period of 1 month.
Fig. 10 is an architecture diagram of a system for quantitatively evaluating the overlap and addition of rainfall forecasts in adjacent forecast periods according to an embodiment of the present invention.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
The present embodiment provides a quantitative evaluation method for adjacent forecast precipitation forecast overlap and new information, as shown in fig. 1, which is a flowchart of the quantitative evaluation method for adjacent forecast precipitation forecast overlap and new information in the present embodiment.
The quantitative evaluation method for rainfall forecast overlap and new information in adjacent forecast periods provided by the embodiment comprises the following steps of:
and S1, acquiring precipitation forecast data of adjacent forecast periods to be evaluated and corresponding observed precipitation data as original samples.
And S2, determining the certainty coefficient of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods according to the regression equation of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the regression equation of the union set of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the precipitation forecast data of the adjacent forecast periods.
And S3, calculating the deterministic coefficient of repeated interpretation of the rainfall forecast of the adjacent forecast period and the deterministic coefficient of the interpretation of the rainfall forecast of the single target forecast period based on the collective operation according to the deterministic coefficient of the observed rainfall data and the rainfall forecast data of the adjacent forecast period, and taking the deterministic coefficients of the repeated interpretation of the rainfall forecast of the adjacent forecast period and the deterministic coefficient of the interpretation of the rainfall forecast of the single target forecast period as the quantized results of the overlapping and newly added information of the rainfall forecast of the adjacent forecast period.
And S4, calculating quantized result reference distribution of the adjacent forecast period rainfall forecast overlap and the newly added information, and performing single-side significance test according to the quantized result reference distribution and the original sample to obtain a significance result for evaluating the adjacent forecast period rainfall forecast overlap and the newly added information.
In the embodiment, the deterministic coefficients of the set operation and the linear regression are combined, overlapped and newly added components in information provided by rainfall forecast in adjacent forecast periods are simply and effectively distinguished, and the quantized and significance-evaluated results are output, so that reference is provided for service use of rainfall forecast products.
In an alternative embodiment, the step of S2, the step of obtaining a regression equation based on the union of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods, includes: respectively establishing observation rainfall o k Forecast precipitation f equivalent to forecast period of 0 month 0 Regression equation of, observing precipitation o k Forecast precipitation f with forecast period of 1 month 1 And observing the precipitation o k Forecast precipitation union with adjacent forecast period (f) 0 ∪f 1 ) Further, the certainty coefficient determined by the above three regression equations is obtained.
The method comprises the steps of establishing a regression equation of observed precipitation data and precipitation forecast data with a forecast period of 0 month, and determining a certainty coefficient R of the observed precipitation data and the precipitation forecast data with the forecast period of 0 month by comparing the sum of squares of residuals and the total variance of observed precipitation 2 (o～f 0 ) (ii) a The expression is as follows:
a regression equation of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month is established, and the certainty coefficient R of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month is determined by comparing the sum of squares of the residuals and the total variance of the observed precipitation 2 (o～f 1 ) (ii) a The expression is as follows:
constructing a union set of observed precipitation data and precipitation forecast data of adjacent forecast periods (f) 0 ∪f 1 ) And determining a union of the observed precipitation data and the precipitation forecast data of adjacent forecast periods by comparing the sum of squared residuals and the total variance of the observed precipitation (f) 0 ∪f 1 ) Coefficient of certaintyR 2 (o～(f 0 ∪f 1 ) ); the expression is as follows:
in the formula o k Represents observed precipitation data of the k year, f 0,k Precipitation forecast data representing 0 month forecast of the k-th year, f 0 Representing precipitation forecast data with forecast period of 0 month in the original sample; f. of 1,k Precipitation forecast data representing a forecast period of 1 month in the k-th year, f 1 Representing precipitation forecast data with forecast period of 1 month in the original sample; alpha is alpha 1 、α 2 、α 3 Respectively, the intercept terms, beta, of the corresponding linear regression models 1 、β 2 、β 3,1 And beta 3,2 Respectively, the slope term, epsilon, of the corresponding linear regression model 1,k 、ε 2,k 、ε 3,k Respectively are residual error terms of corresponding linear regression models; where K is 1,2, K is the total years of the original sample.
Wherein the content of the first and second substances,
to observe the average precipitation over many years:
further, in step S3, the specific steps of calculating the certainty factor repeatedly explained by the precipitation forecasts in the adjacent forecast periods and the certainty factor separately explained by the precipitation forecast in the target forecast period according to the certainty factors of the observed precipitation data and the precipitation forecast data in the adjacent forecast periods include:
and S3.1, calculating the variance repeatedly explained by the rainfall forecast in the adjacent forecast period based on set operation as the certainty coefficient of the certainty coefficient according to the certainty coefficient of the observed rainfall data and the rainfall forecast data in the adjacent forecast period. The expression is as follows:
R 2 (o～(f 0 ∩f 1 ))＝R 2 (o～f 0 )+R 2 (o～f 1 )-R 2 (o～(f 0 ∪f 1 ))。
and S3.2, calculating the variance of the individual precipitation forecast interpretations in the target forecast period based on collective operation according to the deterministic coefficients of the observed precipitation data and the precipitation forecast data in the adjacent forecast periods as deterministic coefficients of the observed precipitation data and the precipitation forecast data in the adjacent forecast periods. The expression is as follows:
R 2 (o～(f 0 /f 1 ))＝R 2 (o～(f 0 ∪f 1 ))-R 2 (o～f 1 )。
in the embodiment, the deterministic coefficient repeatedly explained by the adjacent forecast period rainfall forecast and the deterministic coefficient separately explained by the target forecast period rainfall forecast are used as the quantitative results of the adjacent forecast period rainfall forecast overlap and the newly added information, and the significance results of the adjacent forecast period rainfall forecast overlap and the newly added information are further obtained through significance test, so that the rainfall forecast product can be effectively evaluated.
In an optional embodiment, in step S4, the original sample is processed by using a bootstrap method to obtain a quantized result reference distribution of the adjacent forecast precipitation forecast overlap and the new information.
The bootstrap method uses an Empirical Distribution Function (EDF) as an estimator of a Cumulative Distribution Function (CDF).
In this embodiment, the step of processing the original sample by using the bootstrap method includes:
s4.1, randomly scrambling the rainfall forecast data of the adjacent forecast periods in the original sample, and repeatedly executing the steps S2 and S3 to a preset iteration number n to obtain a deterministic coefficient set R containing n elements of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods ′2 (o～(f 0 ∩f 1 ) And R) ′2 (o～(f 0 /f 1 ))。
R i ′2 (o～(f 0 ∩f 1 ))∈R ′2 (o～(f 0 ∩f 1 ) Representing the adjacent forecast precipitation forecast obtained at the i-th iterationThe repeatedly explained variances, i ═ 1, 2. N variances R i ′ 2 (o～(f 0 ∩f 1 ) A set of deterministic coefficients R forming a repeated interpretation of the precipitation forecasts in adjacent forecast periods ′2 (o～(f 0 ∩f 1 ))。
R i ′2 (o～(f 0 /f 1 ))∈R ′2 (o～(f 0 /f 1 ) I-1, 2.., n) represents the variance of the individual target forecast precipitation interpretations obtained at the i-th iteration. N variances R i ′ 2 (o～(f 0 /f 1 ) A set of deterministic coefficients R forming a single interpretation of the forecast of precipitation for the target forecast period ′2 (o～(f 0 /f 1 ))。
In the present embodiment, a deterministic coefficient set R is formed ′2 (o～(f 0 ∩f 1 ) And R) ′2 (o～(f 0 /f 1 ) Reference distribution as a result of quantization.
And S4.2, performing single-side significance test according to the quantization result reference distribution and the original sample.
Specifically, an empirical distribution function is used as an estimator of the total cumulative distribution function, and an empirical distribution function value of an original sample (namely, precipitation forecasts of adjacent forecast periods to be evaluated) is calculated; the expression is as follows:
wherein I (·) is an indicator function; r i ′ 2 Reference distribution, R, representing the quantified result of the overlap of precipitation forecasts and new information of adjacent forecast periods 2 And the certainty coefficient of the observed precipitation data and the precipitation forecast data of the adjacent forecast period is represented.
For a preset significance level index alpha, setting a significance threshold value as (1-alpha) multiplied by 100% of a quantization result reference distribution, and then performing single-side significance test on an empirical distribution function value of an original sample and the quantization result reference distribution by adopting a single-side test method:
when the empirical distribution function value of the original sample
If so, evaluating that the rainfall forecast overlap and newly added information of the adjacent forecast periods are obvious;
otherwise, judging that the rainfall forecast overlap and the newly added information of the adjacent forecast periods are not significant.
In one embodiment, the significance level index α is 0.1.
And outputting significance results of the rainfall forecast overlap and the newly added information of the adjacent forecast periods to obtain quantitative evaluation results of the rainfall forecast overlap and the newly added information of the adjacent forecast periods.
In the embodiment, the set operation and the linear regression certainty coefficient are combined, and a bootstrap method and a single-side significance test are further combined, so that overlapped and newly added components in observed rainfall information provided by rainfall forecasts in adjacent forecast periods are simply and effectively distinguished, and reference is provided for the service use of the rainfall forecasts.
Further, as shown in fig. 2, the difference and overlap of the precipitation forecasts in adjacent forecast periods are shown by the wien diagram. The quantitative evaluation method provided by the embodiment can distinguish the overlapping component in the observed precipitation information provided by the precipitation forecast with the forecast period of 0 month and the forecast period of 1 month from the newly added component in the observed precipitation information provided by the precipitation forecast with the forecast period of 1 month.
Example 2
In this embodiment, a quantitative evaluation method for rainfall forecast overlap and new information in adjacent forecast periods is applied to perform a test in the embodiment 1.
And S1, acquiring precipitation forecast data of adjacent forecast periods to be evaluated and corresponding observed precipitation data as original samples.
In this embodiment, a National Oceanic and Atmospheric Administration (NOAA) Climate Prediction Center (CPC) global daily scale Unified rainfall meter Database (URD) is used as a terrestrial reference to observe precipitation, a National Center for Environmental Prediction (NCEP) second generation Model System (CFSv 2) is used as Forecast precipitation data, and both the observed precipitation and Forecast precipitation spatial resolutions are 1 ° × 1 ° in North American Multi-Model environment (NMME) experiments. To avoid the strong influence of initialization, the present embodiment focuses on seasonal precipitation data, and the target season is summer (June-July-August, JJA).
And S2, determining the certainty coefficient of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods according to the regression equation of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the regression equation of the union set of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the precipitation forecast data of the adjacent forecast periods.
In this embodiment, 3 regression equations are established for the precipitation data for each grid:
1) observing the regression equation of the precipitation and the forecast precipitation with forecast period of 0 month;
2) observing the regression equation of the precipitation and the forecast precipitation with the forecast period of 1 month;
3) and observing the regression equation of the union set of the rainfall and forecast rainfall with forecast period of 0 month and 1 month.
Specifically, a regression equation of the observed precipitation data and the precipitation forecast data with the forecast period of 0 month is constructed, and the certainty coefficient R of the observed precipitation data and the certainty coefficient R of the precipitation forecast data with the forecast period of 0 month are determined by comparing the sum of squares of the residual errors and the total variance of the observed precipitation 2 (o～f 0 )。
As shown in fig. 3, the spatial distribution of the observed precipitation information provided for the forecast precipitation with forecast period of 0 month corresponds to the certainty factor R calculated in step S2 2 (o～f 0 ). As can be seen, in the central part of North America, North south America, east south America, south America, east Australia, southeast Asia, etc., the forecast precipitation period of 0 month is predicted to provide the viewAnd more rainfall information is measured.
A regression equation of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month is established, and the certainty coefficient R of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month is determined by comparing the sum of squares of the residuals and the total variance of the observed precipitation 2 (o～f 1 )。
As shown in fig. 4, the spatial distribution of the observed precipitation information provided for the forecast precipitation with a forecast period of 1 month corresponds to the certainty factor R calculated in step S2 2 (o～f 1 ). As can be seen from the figure, in regions such as the north south america, the south east america, the east australia, and the south east asia, the observation rainfall information provided by the forecast rainfall for 1 month in forecast period is more.
Comparing fig. 3 and 4, in central north america, eastern south america, southern africa, western australia, etc., more information is provided by forecasts with an expectation of 0 months than by forecasts with an expectation of 1 month; in some regions, such as eastern australia and southeast asia, the information provided by adjacent forecast periods is of comparable size and similar spatial distribution, and there may be some overlap.
Constructing a union set of observed precipitation data and precipitation forecast data of adjacent forecast periods (f) 0 ∪f 1 ) And determining a union of observed precipitation data and precipitation forecast data for adjacent forecast periods by comparing the sum of squared residuals and the total variance of the observed precipitation (f) 0 ∪f 1 ) Coefficient of certainty R 2 (o～(f 0 ∪f 1 ))。
As shown in FIG. 5, the spatial distribution of the observed precipitation information provided for forecast precipitation with forecast periods of 0 and 1 month, respectively, corresponds to R calculated in S2 2 (o～(f 0 ∪f 1 )). Overall, when the union of both is used as an interpretation variable, the most information can be interpreted.
S3, calculating a certainty coefficient repeatedly explained by the rainfall forecast of the adjacent forecast periods and a certainty coefficient separately explained by the rainfall forecast of the target forecast period based on collective operation according to the certainty coefficient of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods, and using the certainty coefficient as the quantization results of the superposition of the rainfall forecast of the adjacent forecast periods and the newly added information
In order to quantify the overlapping and adding information of the rainfall forecasts in the adjacent forecast periods, the variance repeatedly explained by the rainfall forecasts in the adjacent forecast periods and the variance explained by the rainfall forecasts in the target forecast period are obtained by using a set operation method. Namely, R obtained by the collective operation in step S3 in embodiment 1 2 (o～(f 0 ∩f 1 ))、R 2 (o～(f 0 /f 1 ))。
As shown in fig. 6, the spatial distribution of overlapping components in observed precipitation information provided for forecasted precipitation for 0 and 1 month forecasted periods. As can be seen, the regions where the information provided by the adjacent forecast overlaps are distributed mainly in the south north america, the south east america, the east australia and the south east asia.
As shown in fig. 7, the spatial distribution of the new components in the observed precipitation information is provided for a forecast period of 0 months relative to a forecast period of 1 month. As can be seen from the figure, the newly added information is distributed more sparsely, such as in the central part of North America, in the south of south America, in the south of Africa, etc.
And S4, calculating the quantized result reference distribution of the adjacent forecast precipitation forecast overlap and the newly added information, and performing single-side significance test according to the quantized result reference distribution and the original sample to obtain a significance result for evaluating the adjacent forecast precipitation forecast overlap and the newly added information.
This embodiment applies to two variances R determined according to the deterministic coefficient 2 (o～(f 0 ∩f 1 ) And R) 2 (o～(f 0 /f 1 ) Respectively, were tested for significance. As shown in fig. 8 and 9, fig. 8 is a spatial distribution diagram of significance test results of forecast precipitation overlap information with forecast periods of 0 month and 1 month, and fig. 9 is a spatial distribution diagram of significance test results of forecast precipitation newly increase information with forecast periods of 0 month and 1 month. As can be seen from the figure, the quantitative results of the overlapping and adding information in the observed precipitation information provided by the forecast precipitation with forecast periods of 0 month and 1 month are significant in the above-mentioned regions。
The experimental results show that the quantitative evaluation method for the adjacent forecast period rainfall forecast overlap and the newly added information can effectively quantify the overlap and the difference between the information provided by the adjacent forecast period rainfall forecast, can evaluate the significance of the information, and can provide reference for forecast service.
Example 3
The embodiment provides a quantitative evaluation system for the adjacent forecast rainfall forecast overlap and the newly added information, which is applied to the quantitative evaluation method for the adjacent forecast rainfall forecast overlap and the newly added information provided in the embodiment 1.
Fig. 10 is an architecture diagram of a system for quantitatively evaluating the overlap of rainfall forecasts and addition information in adjacent forecast periods according to this embodiment.
In the quantitative evaluation system for rainfall forecast overlap and new information in adjacent forecast periods provided by this embodiment, the system includes:
and the observation rainfall query module 100 is used for querying and acquiring observation rainfall data of corresponding time and area according to the input rainfall forecast data of the adjacent forecast periods to be evaluated.
And the certainty coefficient calculation module 200 is used for constructing a regression equation of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods, constructing a regression equation of a union set of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods, and calculating the certainty coefficient of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods through the regression equation.
And the quantification module 300 is configured to calculate, based on collective operation, a certainty coefficient repeatedly explained by the precipitation forecasts in the adjacent forecast periods and a certainty coefficient separately explained by the precipitation forecast in the target forecast period according to the certainty coefficient of the observed precipitation data and the precipitation forecast data in the adjacent forecast periods, and output the calculated certainty coefficient as a quantification result of the overlap and addition information of the precipitation forecasts in the adjacent forecast periods.
The significance evaluation module 400 is used for calculating quantized result reference distribution of the adjacent forecast precipitation forecast overlap and the newly added information, performing single-side significance test according to the quantized result reference distribution and the original sample, and outputting a significance result for evaluating the adjacent forecast precipitation forecast overlap and the newly added information.
In an alternative embodiment, the deterministic coefficient calculation module 200 includes a first calculation unit 210, a second calculation unit 220, and a third calculation unit 230.
The first calculating unit 210 is configured to construct a regression equation between the observed precipitation data and the precipitation forecast data with the forecast period of 0 month, and determine a certainty coefficient between the observed precipitation data and the precipitation forecast data with the forecast period of 0 month by comparing the sum of squares of residuals and the total variance of the observed precipitation.
The second calculating unit 220 is configured to construct a regression equation between the observed precipitation data and the precipitation forecast data with the forecast period of 1 month, and determine a certainty coefficient between the observed precipitation data and the precipitation forecast data with the forecast period of 1 month by comparing the sum of squares of the residuals and the total variance of the observed precipitation.
The third calculating unit 230 is configured to construct a regression equation of a union set of the observed precipitation data and precipitation forecast data with forecast periods of 0 month and 1 month, and determine a certainty coefficient of a union set of the observed precipitation data and precipitation forecast data of an adjacent forecast period by comparing a sum of squares of residuals and a total variance of the observed precipitation.
In an alternative embodiment, the significance evaluation module 400 includes a reference distribution analysis unit 410, a sample distribution analysis unit 420, and a significance verification unit 430.
The reference distribution analysis unit 410 is configured to randomly scramble precipitation forecast data of adjacent forecast periods in the original samples, sequentially input the scrambled original samples to the deterministic coefficient calculation module 200 and the quantization module 300, and repeatedly execute the above steps to a preset iteration number n to obtain a deterministic coefficient set repeatedly explained by the precipitation forecast of the adjacent forecast periods, which includes n elements, and a deterministic coefficient set repeatedly explained by the precipitation forecast of a target forecast period, which includes n elements, as a quantized result reference distribution.
The sample distribution analyzing unit 420 is configured to calculate an empirical distribution function value of the original sample by using an empirical distribution function as an estimator of the total cumulative distribution function.
The significance checking unit 430 is configured to take (1- α) × 100% of the quantization result reference distribution as a significance threshold according to the quantization result reference distribution output by the reference distribution analyzing unit 410, where α is a preset significance level index; performing one-sided significance test according to the empirical distribution function value of the original sample output by the sample distribution analyzing unit 420, and outputting an evaluation result determined to be significant if the empirical distribution function value of the original sample is greater than 1-alpha; otherwise, outputting an evaluation result which is judged to be not significant.
In a specific implementation process, the rainfall forecast data of adjacent forecast periods to be evaluated are input into the quantitative evaluation system, wherein the observed rainfall query module 100 queries and obtains the observed rainfall data of corresponding time and area, and matches the rainfall forecast data of adjacent forecast periods to be evaluated one by one according to time domain and grid, and then transmits the data to the certainty coefficient calculation module 200.
In the deterministic coefficient calculation module 200, the first calculation unit 210 inputs the observed precipitation data and the precipitation forecast data with the forecast period of 0 month into the regression equation established by the calculation unit, and outputs the deterministic coefficients of the observed precipitation data and the precipitation forecast data with the forecast period of 0 month. The second calculating unit 220 inputs the observed precipitation data and the precipitation forecast data with the forecast period of 1 month into the regression equation constructed by the second calculating unit, and outputs the certainty coefficients of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month. The third calculating unit 230 inputs the observed precipitation data and the precipitation forecast data with forecast periods of 0 month and 1 month into a regression equation established by the third calculating unit, and outputs certainty coefficients of the observed precipitation data and the precipitation forecast data with forecast periods of 0 month and 1 month. The deterministic coefficient calculation module 200 transmits the deterministic coefficients output by the first, second and third calculation units 210, 220 and 230, respectively, to the quantization module 300.
The quantization module 300 calculates the deterministic coefficient repeatedly explained by the adjacent forecast precipitation forecast based on the collective operation according to the input deterministic coefficient, calculates the deterministic coefficient separately explained by the target forecast precipitation forecast, and outputs the calculated deterministic coefficient to the significance evaluation module 400 as the quantization result of the overlapping and new information of the adjacent forecast precipitation forecast.
In the significance evaluation module 400, the reference distribution analysis unit 410 scrambles precipitation forecast data of adjacent forecast periods to be evaluated and corresponding observed precipitation data thereof, then sequentially inputs the data into the deterministic coefficient calculation module 200 and the quantization module 300, and repeatedly executes the above steps to a preset iteration number n to obtain a deterministic coefficient set repeatedly explained by the adjacent forecast periods containing n elements and a deterministic coefficient set separately explained by the target forecast period containing n elements as quantized result reference distribution.
The sample distribution analyzing unit 420 calculates an empirical distribution function value of the rainfall forecast of the adjacent forecast period to be evaluated by using an empirical distribution function as an estimator of a total cumulative distribution function according to the quantization result of the adjacent forecast period rainfall forecast overlap and the new addition information output by the quantization module 300 and the quantization result reference distribution obtained by the reference distribution analyzing unit 410.
The significance checking unit 430 takes (1- α) × 100% of the quantized result reference distribution as a significance threshold according to the quantized result reference distribution output by the reference distribution analyzing unit 410, wherein α is a preset significance level index; performing one-side significance test according to the empirical distribution function value of the original sample output by the sample distribution analyzing unit 420, and outputting an evaluation result determined to be significant if the empirical distribution function value of the original sample is greater than 1- α; otherwise, outputting an evaluation result which is judged to be not significant.
And outputting significance results of the rainfall forecast overlap and the newly added information of the adjacent forecast periods to obtain quantitative evaluation results of the rainfall forecast overlap and the newly added information of the adjacent forecast periods.
The same or similar reference numerals correspond to the same or similar parts;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A quantitative evaluation method for rainfall forecast overlap and new information of adjacent forecast periods is characterized by comprising the following steps:
s1, acquiring precipitation forecast data of adjacent forecast periods to be evaluated and observation precipitation data corresponding to the precipitation forecast data as original samples;
s2, determining certainty coefficients of the observed precipitation data and the precipitation forecast data of the adjacent forecast periods according to the regression equation of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the regression equation of the union set of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods and the precipitation forecast data of the adjacent forecast periods;
s3, calculating a certainty coefficient repeatedly explained by the rainfall forecast of the adjacent forecast periods and a certainty coefficient separately explained by the rainfall forecast of the target forecast period based on collective operation according to the certainty coefficient of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods, and using the certainty coefficient as the quantization results of the superposition of the rainfall forecast of the adjacent forecast periods and the newly added information;
and S4, calculating the quantized result reference distribution of the adjacent forecast precipitation forecast overlap and the newly added information, and performing single-side significance test according to the quantized result reference distribution and the original sample to obtain a significance result for evaluating the adjacent forecast precipitation forecast overlap and the newly added information.
2. The quantitative evaluation method of claim 1, wherein the step of S2 is a step of using a regression equation based on the combination of the observed precipitation data, the precipitation forecast data of the adjacent forecast periods, and the precipitation forecast data of the adjacent forecast periods to calculate the regression equation, and comprises:
constructing observation precipitation data and precipitation forecast number with forecast period of 0 monthAccording to the regression equation, and by comparing the sum of the squares of the residual errors and the total variance of the observed rainfall, the certainty coefficient R of the observed rainfall data and the rainfall forecast data with the forecast period of 0 month is determined 2 (o～f 0 ) (ii) a The expression is as follows:
a regression equation of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month is established, and the certainty coefficient R of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month is determined by comparing the sum of squares of the residuals and the total variance of the observed precipitation 2 (o～f 1 ) (ii) a The expression is as follows:
constructing a union set of observed precipitation data and precipitation forecast data of adjacent forecast periods (f) 0 ∪f 1 ) And determining a union of observed precipitation data and precipitation forecast data for adjacent forecast periods by comparing the sum of squared residuals and the total variance of the observed precipitation (f) 0 ∪f 1 ) Coefficient of certainty R 2 (o～(f 0 ∪f 1 ) ); the expression is as follows:
in the formula o k Represents observed precipitation data of the k year, f 0,k Precipitation forecast data representing 0 month forecast of the k-th year, f 0 Representing precipitation forecast data with forecast period of 0 month in the original sample; f. of 1,k Precipitation forecast data representing a forecast period of 1 month in the k-th year, f 1 Representing precipitation forecast data with forecast period of 1 month in the original sample; alpha is alpha 1 、α 2 、α 3 Respectively, intercept of corresponding linear regression modelsTerm, beta 1 、β 2 、β 3,1 And beta 3,2 Respectively, the slope term, epsilon, of the corresponding linear regression model 1,k 、ε 2,k 、ε 3,k Respectively residual error terms of the corresponding linear regression models; where K is 1,2, K is the total years of the original sample.
3. The quantitative evaluation method according to claim 2, wherein the step of S3 includes the following specific steps:
calculating the variance repeatedly explained by the rainfall forecast in the adjacent forecast periods based on set operation as the certainty coefficient of the certainty coefficient according to the certainty coefficient of the observed rainfall data and the rainfall forecast data in the adjacent forecast periods; the expression is as follows:
R 2 (o～(f 0 ∩f 1 ))＝R 2 (o～f 0 )+R 2 (o～f 1 )-R 2 (o～(f 0 ∪f 1 ))；
calculating the variance of the individual precipitation forecast interpretation in the target forecast period based on collective operation as the certainty coefficient of the observation precipitation data and the precipitation forecast data of the adjacent forecast periods; the expression is as follows:
R 2 (o～(f 0 /f 1 ))＝R 2 (o～(f 0 ∪f 1 ))-R 2 (o～f 1 )。
4. the quantitative evaluation method according to any one of claims 1 to 3, wherein in the step S4, the original sample is processed by using a bootstrap method to obtain a quantitative result reference distribution of adjacent forecast rainfall forecast overlap and new information.
5. The quantitative estimation method of claim 4, wherein the step of processing the raw sample by using the bootstrap method comprises:
randomly scrambling precipitation forecast data of adjacent forecast periods in the original sample, and repeatedly executing the steps S2 and S3 to a preset iteration number n to obtain a solution containing n precipitation forecast dataDeterministic coefficient set R of observed precipitation data of elements and precipitation forecast data of adjacent forecast periods ′2 (o～(f 0 ∩f 1 ) And R) ′2 (o～(f 0 /f 1 ))；
Wherein the content of the first and second substances,
representing a set of variances repeatedly interpreted by adjacent forecast precipitation forecasts obtained at the ith iteration,
representing a set of variances, i ═ 1, 2.., n, interpreted by the target forecast precipitation forecast alone, obtained at the i-th iteration;
set of deterministic coefficients R ′2 (o～(f 0 ∩f 1 ) And R) ′2 (o～(f 0 /f 1 ) Reference distribution as a result of quantization.
6. The quantitative evaluation method of claim 4, wherein in the step of S4, the step of performing a one-sided significance test on the original sample according to the quantitative result reference distribution comprises:
calculating an empirical distribution function value of an original sample by using an empirical distribution function as an estimator of a total cumulative distribution function; the expression is as follows:
wherein I (·) is an indicator function;
indicating the overlap and novelty of precipitation forecasts in adjacent forecast periodsReference distribution of quantization results of information increase, R 2 A certainty factor representing the observed precipitation data and precipitation forecast data for adjacent forecast periods;
for a preset significance level index alpha, setting a significance threshold value as (1-alpha) multiplied by 100% of a quantization result reference distribution, and then performing single-side significance test on an empirical distribution function value of an original sample and the quantization result reference distribution by adopting a single-side test method:
when the empirical distribution function value of the original sample
If so, evaluating that the rainfall forecast overlap and newly added information of the adjacent forecast periods are obvious; otherwise, judging that the rainfall forecast overlap and the newly increased information of the adjacent forecast periods are not significant.
7. A quantitative evaluation system for rainfall forecast overlap and new information of adjacent forecast periods is applied to the quantitative evaluation method of any one of claims 1 to 6, and is characterized by comprising the following steps of:
the observation rainfall query module is used for querying and acquiring observation rainfall data of corresponding time and area according to input rainfall forecast data of adjacent forecast periods to be evaluated;
the deterministic coefficient calculation module is used for constructing a regression equation of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods, constructing a regression equation of a union set of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods, and calculating the deterministic coefficient of the observed rainfall data and the rainfall forecast data of the adjacent forecast periods through the regression equation;
the quantitative module is used for calculating the deterministic coefficient repeatedly explained by the rainfall forecast of the adjacent forecast period based on collective operation according to the deterministic coefficient of the observed rainfall data and the rainfall forecast data of the adjacent forecast period, calculating the deterministic coefficient separately explained by the rainfall forecast of the target forecast period, and outputting the calculated deterministic coefficient as the quantitative result of the overlapping and newly added information of the rainfall forecast of the adjacent forecast period;
and the significance evaluation module is used for calculating the quantized result reference distribution of the adjacent forecast period rainfall forecast overlap and the newly added information, performing single-side significance test according to the quantized result reference distribution and the original sample, and outputting a significance result for evaluating the adjacent forecast period rainfall forecast overlap and the newly added information.
8. The quantitative evaluation system of claim 7, wherein the deterministic coefficient computation module comprises a first computation unit, a second computation unit, and a third computation unit; wherein:
the first calculation unit is used for constructing a regression equation of the observed precipitation data and the precipitation forecast data with the forecast period of 0 month, and determining the certainty coefficient of the observed precipitation data and the precipitation forecast data with the forecast period of 0 month by comparing the sum of squares of the residual errors and the total variance of the observed precipitation;
the second calculation unit is used for constructing a regression equation of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month, and determining the certainty coefficient of the observed precipitation data and the precipitation forecast data with the forecast period of 1 month by comparing the sum of squares of the residual errors and the total variance of the observed precipitation;
and the third calculation unit is used for constructing a regression equation of the observed precipitation data and the precipitation forecast data union set with forecast periods of 0 month and 1 month, and determining the certainty coefficient of the observed precipitation data and the precipitation forecast data union set of the adjacent forecast periods by comparing the sum of squares of the residual errors and the total variance of the observed precipitation.
9. The quantitative evaluation system of claim 7, wherein the significance evaluation module comprises a reference distribution analysis unit, a sample distribution analysis unit and a significance check unit; wherein:
the reference distribution analysis unit is used for randomly scrambling precipitation forecast data of adjacent forecast periods in the original samples, inputting the scrambled original samples into the deterministic coefficient calculation module and the quantization module in sequence, and repeatedly executing the steps to a preset iteration number n to obtain a deterministic coefficient set containing n elements and repeatedly explained by the precipitation forecast of the adjacent forecast periods and a deterministic coefficient set containing n elements and independently explained by the precipitation forecast of a target forecast period as quantized result reference distribution;
the sample distribution analysis unit is used for adopting an empirical distribution function as an estimator of a total cumulative distribution function and calculating an empirical distribution function value of an original sample;
the significance checking unit is used for taking (1-alpha) multiplied by 100% of the quantized result reference distribution as a significance threshold according to the quantized result reference distribution output by the reference distribution analysis unit, wherein alpha is a preset significance level index; performing one-side significance test according to the empirical distribution function value of the original sample output by the sample distribution analysis unit, and outputting an evaluation result judged to be significant if the empirical distribution function value of the original sample is greater than 1-alpha; otherwise, outputting an evaluation result which is judged to be not significant.
10. A quantitative evaluation system for rainfall forecast overlap and newly added information in adjacent forecast periods is characterized by comprising a memory and a processor, wherein the memory stores a computer program; the processor, when executing the computer program, performs the steps of the quantitative evaluation method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210675212.4A CN114910981B (en) | 2022-06-15 | 2022-06-15 | Quantitative evaluation method and system for rainfall forecast overlap and newly added information in adjacent forecast periods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210675212.4A CN114910981B (en) | 2022-06-15 | 2022-06-15 | Quantitative evaluation method and system for rainfall forecast overlap and newly added information in adjacent forecast periods |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114910981A true CN114910981A (en) | 2022-08-16 |
| CN114910981B CN114910981B (en) | 2023-04-11 |
Family
ID=82770422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210675212.4A Active CN114910981B (en) | 2022-06-15 | 2022-06-15 | Quantitative evaluation method and system for rainfall forecast overlap and newly added information in adjacent forecast periods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114910981B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170176640A1 (en) * | 2014-03-28 | 2017-06-22 | Northeastern University | System for Multivariate Climate Change Forecasting With Uncertainty Quantification |
| US20170261646A1 (en) * | 2016-03-08 | 2017-09-14 | China Institute Of Water Resources And Hydropower Research | Self-correcting multi-model numerical rainfall ensemble forecasting method |
| CN110058328A (en) * | 2019-01-30 | 2019-07-26 | 沈阳区域气候中心 | Summer Precipitation in Northeast China multi-mode combines NO emissions reduction prediction technique |
| CN111898093A (en) * | 2020-07-01 | 2020-11-06 | 武汉大学 | Multi-dimensional post-processing method, device and system suitable for weather ensemble forecasting |
| CN112884209A (en) * | 2021-01-29 | 2021-06-01 | 河海大学 | Weather method and mathematical statistics method-based medium and long-term rainfall forecasting method |
| CN113743022A (en) * | 2021-09-26 | 2021-12-03 | 华北电力大学 | High-precision climate change data storage and visualization method |
| CN113917568A (en) * | 2021-10-13 | 2022-01-11 | 中山大学 | Rainfall forecast capacity and remote correlation effect correlation method based on certainty coefficient |
| WO2022011728A1 (en) * | 2020-07-16 | 2022-01-20 | 中山大学 | Spatial auto-correlation clustering method for remote correlation mode |
-
2022
- 2022-06-15 CN CN202210675212.4A patent/CN114910981B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170176640A1 (en) * | 2014-03-28 | 2017-06-22 | Northeastern University | System for Multivariate Climate Change Forecasting With Uncertainty Quantification |
| US20170261646A1 (en) * | 2016-03-08 | 2017-09-14 | China Institute Of Water Resources And Hydropower Research | Self-correcting multi-model numerical rainfall ensemble forecasting method |
| CN110058328A (en) * | 2019-01-30 | 2019-07-26 | 沈阳区域气候中心 | Summer Precipitation in Northeast China multi-mode combines NO emissions reduction prediction technique |
| CN111898093A (en) * | 2020-07-01 | 2020-11-06 | 武汉大学 | Multi-dimensional post-processing method, device and system suitable for weather ensemble forecasting |
| WO2022011728A1 (en) * | 2020-07-16 | 2022-01-20 | 中山大学 | Spatial auto-correlation clustering method for remote correlation mode |
| CN112884209A (en) * | 2021-01-29 | 2021-06-01 | 河海大学 | Weather method and mathematical statistics method-based medium and long-term rainfall forecasting method |
| CN113743022A (en) * | 2021-09-26 | 2021-12-03 | 华北电力大学 | High-precision climate change data storage and visualization method |
| CN113917568A (en) * | 2021-10-13 | 2022-01-11 | 中山大学 | Rainfall forecast capacity and remote correlation effect correlation method based on certainty coefficient |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114910981B (en) | 2023-04-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gudmundsson et al. | The Global Streamflow Indices and Metadata Archive (GSIM)–Part 2: Quality control, time-series indices and homogeneity assessment | |
| Loucks et al. | System sensitivity and uncertainty analysis | |
| Funk et al. | The centennial trends greater horn of Africa precipitation dataset | |
| Han et al. | Drought forecasting based on the remote sensing data using ARIMA models | |
| Johnson et al. | Measurement of GCM skill in predicting variables relevant for hydroclimatological assessments | |
| Beecham et al. | Statistical downscaling of multi-site daily rainfall in a South Australian catchment using a Generalized Linear Model. | |
| Li et al. | Multi-site multivariate downscaling of global climate model outputs: an integrated framework combining quantile mapping, stochastic weather generator and Empirical Copula approaches | |
| D’Ambrosio et al. | Characterising the hydrological regime of an ungauged temporary river system: a case study | |
| Hennemuth et al. | Statistical methods for the analysis of simulated and observed climate data: applied in projects and institutions dealing with climate change impact and adaptation | |
| Yeh et al. | Stochastic model for drought forecasting in the Southern Taiwan Basin | |
| Zafarani et al. | Applicability of different ground-motion prediction models for northern Iran | |
| Huang et al. | Predictive performance of ensemble hydroclimatic forecasts: Verification metrics, diagnostic plots and forecast attributes | |
| Gebremichael et al. | Error uncertainty analysis of GPCP monthly rainfall products: A data-based simulation study | |
| Pipunic et al. | An evaluation and regional error modeling methodology for near‐real‐time satellite rainfall data over Australia | |
| Du et al. | Multivariate return period‐based ground motion selection for improved hazard consistency over a vector of intensity measures | |
| Gilleland et al. | A new approach to testing forecast predictive accuracy | |
| Morawietz et al. | Systematic evaluation of autoregressive error models as post-processors for a probabilistic streamflow forecast system | |
| Beuchat et al. | A robust framework for probabilistic precipitations downscaling from an ensemble of climate predictions applied to Switzerland | |
| Wang et al. | A pathway analysis method for quantifying the contributions of precipitation and potential evapotranspiration anomalies to soil moisture drought | |
| Liu et al. | A conceptual prediction model for seasonal drought processes using atmospheric and oceanic standardized anomalies: application to regional drought processes in China | |
| Yamaguchi et al. | Observation‐based evaluation of ensemble reliability | |
| Yousaf et al. | Development of a new hybrid ensemble method for accurate characterization of future drought using multiple global climate models | |
| Dormann | Response to Comment on" Methods to account for spatial autocorrelation in the analysis of species distributional data: a review" | |
| Tedla et al. | Evaluation of factors affecting the quality of citizen science rainfall data in Akaki catchment, Addis Ababa, Ethiopia | |
| Li et al. | A nonstationary runoff frequency analysis for future climate change and its uncertainties |
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 |