CN108171007B - One kind being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value - Google Patents
One kind being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value Download PDFInfo
- Publication number
- CN108171007B CN108171007B CN201810036068.3A CN201810036068A CN108171007B CN 108171007 B CN108171007 B CN 108171007B CN 201810036068 A CN201810036068 A CN 201810036068A CN 108171007 B CN108171007 B CN 108171007B
- Authority
- CN
- China
- Prior art keywords
- forecast
- data
- runoff
- variable
- sequence
- 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.)
- Expired - Fee Related
Links
- 238000012545 processing Methods 0.000 title claims abstract description 37
- 230000007774 longterm Effects 0.000 title claims abstract description 19
- 238000013277 forecasting method Methods 0.000 title claims abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 79
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims abstract description 24
- 238000001556 precipitation Methods 0.000 claims description 30
- 238000010219 correlation analysis Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 11
- 241001269238 Data Species 0.000 claims description 8
- 238000005259 measurement Methods 0.000 claims description 7
- 230000000052 comparative effect Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 9
- 238000007619 statistical method Methods 0.000 abstract description 4
- 238000003860 storage Methods 0.000 description 18
- 230000000875 corresponding effect Effects 0.000 description 16
- 230000008901 benefit Effects 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 239000007787 solid Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000012252 genetic analysis Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/14—Rainfall or precipitation gauges
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Environmental & Geological Engineering (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Hydrology & Water Resources (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses one kind to be based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value, is related to hydrographic data process field.The method: forecast area and forecasting period are determined；It is forecast using forecast area of the numerical value set forecasting procedure to forecasting period, obtains average runoff forecast data conclusion α, be denoted as Q_Model_Forecast；The forecast area of forecasting period is forecast using extra large temperature distant correlation technique, average runoff forecast data conclusion β is obtained, is denoted as Q_SST_Forecast；The anomaly value Ano for calculating Q_SST_Forecast, when | Ano | >=50%, use Q_SST_Forecast to replace Q_Model_Forecast as final forecast conclusion；When | Ano | < 50% uses Q_Model_Forecast as final forecast conclusion.The present invention can be realized the moon scale Runoff Forecast that leading time is 1 year, in trend prediction and prediction of extremum, the more currently used numerical value set forecasting procedure of the method for the invention or statistical method precision with higher.
Description
Technical field
The present invention relates to hydrographic data process fields, more particularly to one kind to be based on the modified middle length of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value
Phase Runoff Forecast method.
Background technique
Accurate Medium-and Long-Term Runoff Forecasting can provide basin future water resources situation in longer leading time, to have
Effect instructs the scheduling of the engineerings such as reservoir to practice, and watershed water resource integrated configuration and utilization have important economic benefit and society
Benefit.It is existing to be divided into four classes from big classification about Medium-and Long-Term Runoff Forecasting: physics genetic analysis method, Statistics Method,
Intelligent method and Numerical Prediction Method.
Physics genetic analysis method mainly analyzes phenomena such as atmospheric circulation, solar activity, planetary position, earth rotation to pre-
The Influencing Mechanism of object is reported, thus the method that the factor realizes Runoff Forecast according to weather report.
Mathematical statistics rule be find the correlativity of predictor and Forecasting Object according to a large amount of historical summaries, and
According to the method that this correlativity realizes Medium-long Term Prediction, predictor is single-factor or multiple-factor.
Intelligent method is then according to intelligence such as fuzzy mathematics, gray system, artificial neural network, wavelet analysis, chaos analysis
Energy analysis method, constructs the nonlinear model of predictor and Forecasting Object, thus the method for realizing Medium-long Term Prediction.
Existing Numerical Prediction Method is that more rapidly a kind of forecasting procedure and one kind of future development become current development
Gesture models the meteorological model characteristics of motion using math equation, realizes the explanation and simulation of correlated phenomena, and comes accordingly
The following runoff number is predicted.Exist currently, the advanced degree of existing Numerical Forecast Technology has become one country of measurement
The important symbol of art technology strength.However, because meteorological model system has stronger uncertain and won ton passivity, therefore,
Math equation based on reductionism design is difficult to portray the movement of the relevant weather hydrology completely, therefore, based on meteorological model system
The Medium-and Long-Term Runoff Forecasting result reliability that existing Numerical Prediction Method obtains is low, and numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM is met the tendency of to solve this problem
And it gives birth to.Numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM constructs DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM sample using the method for initial value disturbance, and it is flat to do weighting to DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM sample
Obtain final forecast conclusion.Due to model prediction error approximation Normal Distribution, the weighted average of this great amount of samples, energy
Enough random errors and part system error for effectively overcoming each forecast sample are higher for the trend prediction forecast skill of runoff.
Although existing conventional needle centering long-period runoff numerical value set forecasting procedure can effectively reflect tendency information, averagely up and down to many years
Runoff Forecast precision it is higher, but due to this method have " peak load shifting " effect, that is, be easy extreme high level is dragged down, will be extremely low
Value is drawn high, and this method is caused to be difficult to effectively forecast extreme event.
Summary of the invention
The purpose of the present invention is to provide one kind to be based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value,
Being difficult to solve the problems, such as that existing numerical value set forecasting procedure obtains effectively forecasts extreme event.
To achieve the goals above, of the present invention to be based on the modified Medium-and Long-Term Runoff Forecasting side of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value
Method, which comprises
S1 determines forecast area and forecasting period；
S2 is forecast using forecast area of the numerical value set forecasting procedure to forecasting period, obtains average Runoff Forecast
Data conclusion α, is denoted as Q_Model_Forecast；
S3 forecasts the forecast area of forecasting period using the extra large distant correlation technique of temperature, obtains average Runoff Forecast number
According to conclusion β, it is denoted as Q_SST_Forecast；
The anomaly value of Q_SST_Forecast is calculated according to formula (1):
Ano is anomaly value,For the same period long-time average annual value of forecasting period；
S4, forecast conclusion of the forecast area in forecasting period
When | Ano | >=50%, then there is extreme case in the runoff of the forecasting period obtained based on the extra large distant correlation technique of temperature
Possibility is high, and Q_SST_Forecast is used to replace Q_Model_Forecast as final forecast conclusion；
As | Ano | < 50%, what extreme case occurred in the runoff of the forecasting period obtained based on the extra large distant correlation technique of temperature can
Energy property is low, uses Q_Model_Forecast as final forecast conclusion.
Preferably, S1, specifically:
Forecast demand is received, and extracts section and forecasting period to be forecast in the forecast demand, it will section be forecast
Upper basin as forecast basin.
Preferably, S2, specifically:
S21 obtains the precipitation rate forecast data of publication in nearest 5 days from global climate forecast system；
S22 parses the precipitation speed data for obtaining global grid formula to the precipitation rate forecast data, to mutually in the same time
Grid-net type precipitation speed data take arithmetic mean to get each grid by 6h mean precipitation rate；
S23, calculate obtain each grid by 6h accumulative rainfall amount；
S24 obtains forecast area by face mean rainfall P for 24 hours according to formula (2)；
PiIt is the 6h accumulative rainfall amount of i-th of mesh point in forecast area, n is the quantity of forecast area mesh point；
According to forecast area by face mean rainfall for 24 hours, estimated prediction region each nature specified in forecasting period
The moon accumulative rainfall amount forecast data of the moon；
S25, if the moon accumulative rainfall amount forecast data P for the calendar month t that forecast basin is specified in forecasting periodt, obtain
Calendar month t and P in historytThe immediate multiple actual measurement precipitation, that is, corresponding run-off of each actual measurement precipitation, will be each
The arithmetic mean of instantaneous value of run-off, in the Runoff Forecast opinion of calendar month t, repeats S25 as forecast basin, until obtaining all fingers
Obtained all Runoff Forecast opinions are formed average runoff forecast data conclusion α, are denoted as by the Runoff Forecast opinion for determining calendar month
Q_Model_Forecast。
Preferably, S3, specifically:
S31 obtains the warm data in global grid type sea；
S32, since being obtained from 1981, the section to be forecast of forecast area specifies each year calendar of moon x in forecasting period
History average diameter flow data, is denoted as sequence one；
The temperature sequence that different grids in the warm data in global grid type sea shift to an earlier date each year in m month is obtained, sequence two is denoted as,
m≤12；
Sequence one and sequence two are subjected to correlation analysis, obtain with the maximum sea position of one relative coefficient of sequence with
Month in advance；Sequence a will be denoted as with the highest sequence two of one relative coefficient of sequence；
S33, if i-th _ r row, jth _ r column shift to an earlier date the warm data sequence in sea of leadMonth in the warm data in global grid type sea
For sequence a, the comparative diagram of sequence a and sequence one are drawn, it is corresponding that the curve of sequence a is in the sea temperature data A at time least significant end
Diameter flow data be section to be forecast in forecasting period specify moon x runoff to be forecast；
S34, selection and the immediate multiple sea temperature data points of the warm data A in sea from extra large warm data history data, and obtain
Historical traffic data corresponding to each warm data point in sea, then by the arithmetic average of obtained multiple historical traffic datas,
The runoff forecast data that is averaged specifies the runoff to be forecast of moon x as section to be forecast in forecasting period, multiple by what is obtained
Section of the maximum value and minimum value of historical traffic data as average runoff forecast data, by average runoff forecast data and its
Section is denoted as Q_SST_Forecast as average runoff forecast data conclusion β, complete to using the sea distant correlation technique of temperature to pre-
The forecast area for section of giving the correct time is forecast, is obtained.
It is highly preferred that in S32, correlation analysis, specifically: the correlation of variable X and variable Y is obtained using formula (3)
Coefficient rXY, formula (3) are as follows:
If rXYIt is positive, then variable X and variable Y correlation；
If rXYIt is negative, then variable X and variable Y negative correlation；
If rXYBe 0, then variable X and variable Y be independent variable；
If rXYIt is 1, then variable X and the linear correlativity of variable Y；
N is the sequence length of variable X and variable Y；WithFor the arithmetic average of variable X and variable Y；
Variable X is sequence one, and variable Y is sequence two or variable X is sequence two, and variable Y is sequence one.
It is highly preferred that maximal correlation property coefficient is greater than 0.6.
The beneficial effects of the present invention are:
The method of the invention combines numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM with the distant related extremum prediction of extra large temperature, makes full use of set of values
Advantage of the forecast in trend prediction, and the distant related advantage in prediction of extremum of sea temperature are closed, can be realized leading time is 1 year
Moon scale Runoff Forecast, in trend prediction and prediction of extremum, the more currently used numerical value set of the method for the invention is pre-
Reporting method or statistical method precision with higher.
Detailed description of the invention
Fig. 1 is the frame process signal based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value
Figure；
Fig. 2 is Hai Wen and runoff correlation analysis flow diagram；
Fig. 3 is based on distant relevant Runoff Forecast schematic diagram；
Fig. 4 is forecast area boundary schematic diagram；
Fig. 5 be June runoff at preceding 1 year April, 317 row, 9 column position sea temperature data compare schematic diagram；
Fig. 6 be July runoff at 131 row of preceding 1 year September, 155 column position sea temperature data compare schematic diagram；
Fig. 7, which is August runoff, compares schematic diagram with sea temperature data at preceding 1 year March, 30 row, 13 column position；
Fig. 8, which is September runoff, compares schematic diagram with sea temperature data at preceding 2 months 1 year 94 row, 4 column positions；
Fig. 9 is tradition forecast and this method value of forecasting contrast schematic diagram.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing, to the present invention into
Row is further described.It should be appreciated that the specific embodiments described herein are only used to explain the present invention, it is not used to
Limit the present invention.
Referring to Fig.1, about herein described based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value
It is described in detail:
(1) forecast area and forecasting period are determined
Forecast demand is received, and extracts section and forecasting period to be forecast in the forecast demand, it will section be forecast
Upper basin as forecast basin.
The above basin perimeter of section determines zone boundary (longitude and latitude array) that the step is primarily used to extract areal rainfall
(i.e. rainfall tiling to the depth on drainage area) forecast result, which is the data of grid form, on basin side
Point in boundary is weighted and averaged to obtain areal rainfall.In addition, also demand should determine forecasting period according to weather report, such as forecast certain section
The run-off in June, 2017.
(2) numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM:
1) the above basin perimeter precipitation numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM of section: downloading global climate forecast system (CFS, Climate
Forecast System) precipitation rate forecast data (file is started with prate, daily publication 4 when time).Global climate
The downloading network address of forecast system are as follows: http://www.ftp.ncep.noaa.gov/data/nccf/com/cfs/prod/
cfs/。
The forecast file of the present embodiment downloading and the forecast documenting time nearest 5 days, there is 4 files daily, utilizes
Wgrib.exe tool parses the file of the grib2 format of downloading, obtain global grid formula precipitation speed data (by
6h, 1 ° × 1 ° of resolution ratio), Dos environment is issued orders format are as follows:
wgrib2.exeprate.01.2017060800.daily.grb2-match"PRATE"-csvCFSV2_
20170608_00_1_PRATE_6_china.txt
Wherein 20170608 after prate.01 represent the data publication time, and 00 of its back when representing data publication
Secondary, CFSV2_20170608_00_1_PRATE_6_china.txt represents the filename (can arbitrarily name) saved, has recorded
Each grid by 6h mean precipitation rate, unit mm/s.By 6h mean precipitation rate obtain after, can multiplied by 3600*6 seconds,
Obtain each grid by 6h accumulative rainfall amount.Forecast district is obtained according to the following formula by face mean rainfall for 24 hours.
Wherein, PiIt is the 6h accumulative rainfall amount of i-th of mesh point in forecast area, n is the quantity of forecast area mesh point.
Acquisition research area after accumulative rainfall amount forecast data, can further synthesize a moon accumulative rainfall amount forecast data for 24 hours.
2) according to nearest 5 days (at daily 4 time), totally 20 forecast samples specify the precipitation forecast in month to believe in research area
Breath obtains the precipitation forecast that research area specifies month using the method for arithmetic mean.Obtain the month and Precipitation forecast in history
Immediate 5 actual measurements precipitation is measured, corresponding run-off is obtained according to the corresponding time, and be averaged to 5 diameter flow datas
Runoff Forecast opinion Q_Model_Forecast as specified month.
(3) based on the distant related extreme value amendment of sea temperature
1) the warm data in lattice pattern sea are obtained
The s from network address ht: //www.metoffice.gov.uk/hadobs/hadisst/data/download.html
Place obtains the warm data set in global grid type sea (file name similar to HadISST1_SST_2017.txt.gz, wherein 2017 represent
Time has recorded average warm data in sea of each month in file).In order to reduce influence of the climate change to method validity, palpus
Download later data file in 1980.
2) correlation of analysis sea warm data and diameter flow data
By the average diameter flow data of the forecast moon since 1981 and the different grids in the whole world shift to an earlier date different months (12 months with
It is interior) temperature sequence in each year does correlation analysis, and the calculation formula of correlation analysis is formula (2).
Wherein: rXYFor relative coefficient, if it is canonical variable X and variable Y correlation, the variable if being negative
X and variable Y negative correlation represent variable X if it is 0 and variable Y are independent variable, represented if if it is 1 variable X and
Variable Y is linear relationship；N is the sequence length of variable X and variable Y；WithFor the arithmetic mean of variable X and variable Y
Value.
Section diameter flow data (many years sequence) correlation analysis in Global Sst data (many years sequence) and forecast month
The result is that obtain with the highest sea position of Inflow Sequence related coefficient with month, specific algorithm process are as shown in Figure 2 in advance.
3) it generates and is based on distant relevant Runoff Forecast
By Hai Wen and runoff correlation analysis, i-th _ r row in Global Sst raster data, jth _ r column is obtained and have shifted to an earlier date
The Inflow Sequence correlation highest of the sea of leadMonth warm data and forecast month.Such as the 20th row, 16 column, March (in advance 1
A month) the warm data in sea and the diameter flow data related coefficient highest in certain section April.The absolute value of this related coefficient often may be used
To reach 0.6 or more, Runoff Forecast, such as Fig. 2 can be made by history set of metadata of similar data.
In Fig. 3, solid broken line is runoff process (such as certain section 1981 to 2016 that is averaged certain same period many years section month
The runoff process of year September part), the dotted line of lower section is the extra large Wen Xu of some highest sea position of corresponding correlation
Column.It should be noted that the time of lower section sea warm broken line and flow broken line is not consistent, but it have passed through the knot after translation
Fruit, because being to have taken leadMonth month forward when doing correlation analysis.Therefore, the warm broken line in sea is more relative to flow broken line
A number (point in figure in open circles) is gone out.Since the corresponding data on flows of the warm data in sea in open circles does not occur also, because
This can be used the sea temperature data to give a forecast future traffic data.Prediction technique is historical data Comparison Method, specially selection history
It is upper with the immediate warm data points in 5 seas of the point, sea temperature data point in Fig. 3 where hollow triangle, according to 5 selected seas
Warm data point finds out corresponding historical traffic data.Finally, doing arithmetic mean to 5 historical traffic datas, future is obtained
Certain month forecast conclusion Q_SST_Forecast, and with the maximum value and minimum value of 5 historical traffic datas, as certain following moon
Average diameter flows out existing possibility section.
4) logarithm runoff DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM carries out extreme value amendment
In step (2), the average Runoff Forecast conclusion in forecast some month of section is obtained using numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM
Q_Model_Forecast.Meanwhile in step (3), forecast some month of section is obtained using the extra large distant relevant method of temperature
Average Runoff Forecast conclusion Q_SST_Forecast.Two values are using distinct methods for the same section, same
The forecast conclusion in month.From introduction above it is found that numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM is calibrated to trend prediction, but has the effect of " peak load shifting "
It answers, it is difficult to effectively forecast extreme case, and the statistical methods such as distant correlation, it is calibrated to the assurance of extreme case.Therefore, using Hai Wen
Distant correlation forecast data logarithm DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM conclusion carries out extreme value amendment, to improve overall forecast precision.Specific steps are as follows: first
First, the anomaly situation of Q_SST_Forecast is judged, such as formula (3).
Wherein: Ano is anomaly value,For the same period long-time average annual value for forecasting month.
Then, judge | Ano | whether it is higher than 50%, thinks to think based on the sea distant relevant method of temperature if being higher than 50%
The runoff in month is forecast it is possible that extreme case, then use Q_SST_Forecast to replace Q_Model_Forecast as most
Whole forecast conclusion.Otherwise, use Q_Model_Forecast as final forecast conclusion.
Long-term runoff DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM method can effectively reflect tendency information in tradition, pre- to many years average upper and lower runoff
It reports precision higher, but since this method has " peak load shifting " effect, that is, is easy to drag down extreme high level, and extreme low value is drawn
Height causes this method to be difficult to effectively forecast that (the extreme high level of runoff or extreme low value, extreme event of the present invention refer to extreme event
50%) anomaly absolute value is more than.With computer technology and big data technology grow up based on long in distant relevant runoff
Phase forecasting procedure, due to it can be found that various predictors and Forecasting Object correlativity, and correlativity evaluation for pole
End event is more sensitive, therefore forecast extreme event that can be more accurate.It is pre- for trend in order to give full play to DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM
The advantage that extreme event is forecast in the advantage of report and distant related forecast, it is of the present invention to be based on numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value
The conclusion for the extreme value Information revision DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM that modified Medium-and Long-Term Runoff Forecasting method is forecast using distant related forecasting procedure.
Embodiment
The step of according to described in summary of the invention, select Danjiangkou Reservoir June in flood season in 2017, July, August, in September
For Long Term Runoff Forecasting, this patent is described in detail and embodiment explanation.The forecast manufacture time is set to May 5 in 2017
Day.
(1) forecast area and forecasting period are determined
The selected Danjiangkou Reservoir storage above basin of section is forecast area, and forecasting period is in June, 2017, July, August
And September, prediction variable are monthly average storage, unit m3/ s. forecast area boundary such as Fig. 5).Area above category in Danjiangkou in figure
In Hanjiang River upstream, across Shaanxi, Henan, Hubei San Sheng.Solid dot represents CFS data and is located inside Hanjiang River in Fig. 4
Lattice point and the lattice point that can nearby characterize Han River precipitation event.
(2) numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM
Download the precipitation rate forecast data of global climate forecast system (CFS, Climate Forecast System)
(file is started with prate, is issued daily secondary at 4), downloading network address is http://www.ftp.ncep.noaa.gov/
Data/nccf/com/cfs/prod/cfs/, the forecast file of nearest 5 days of downloading, there is 4 files daily.20 texts are downloaded altogether
Part is as shown in the table:
Table 1CFS raw data file download list
It is parsed using file of the wgrib.exe tool to the grib2 format of downloading, obtains the precipitation of global grid formula
Speed data (by 6h, 1 ° × 1 ° of resolution ratio), concrete operations are successively to run such as to issue orders under Dos environment:
Table 2CFS original data processing order
Text file after parsing have recorded each grid by 6h mean precipitation rate, unit mm/s.Each grid
Numerical value multiplied by 3600*6 seconds, obtain each grid by 6h accumulative rainfall amount.According to the lattice point in Fig. 4, obtained using formula (1)
Forecast district by face mean rainfall for 24 hours, and by daily rainfall statistics be in June, 2017, July, August and September moon accumulative rainfall amount, tool
Body numerical value such as following table.
The forecast of 3 accumulative rainfall amount of table
| Forecast area | Time | Moon accumulative rainfall amount (mm) |
| The above Hanjiang River in Danjiangkou | In June, 2017 | 112 |
| The above Hanjiang River in Danjiangkou | In July, 2017 | 160 |
| The above Hanjiang River in Danjiangkou | In August, 2017 | 135 |
| The above Hanjiang River in Danjiangkou | In September, 2017 | 113 |
Record and same period precipitation are averagely put in storage according to -2016 years 1981 Danjiangkou Reservoir June, July, August, September
Live state information, such as table 4,
4 Danjiangkou Reservoir of table is put in storage table corresponding with basin precipitation
Referring to table 4, obtains and dropped in history in June, 2017, July, August, immediate 5 actual measurements of September Precipitation forecast amount
Water obtains corresponding run-off according to the corresponding time, and takes the average runoff as specified month pre- 5 diameter flow datas
It reports shown in opinionaire 5.
Table 5 is the average runoff forecast data that forecast area specifies month in forecasting period
| Forecasting Object | Time | Runoff (m3/s) |
| Danjiangkou storage | In June, 2017 | 1752.2 |
| Danjiangkou storage | In July, 2017 | 2496 |
| Danjiangkou storage | In August, 2017 | 2001 |
| Danjiangkou storage | In September, 2017 | 1799.8 |
(3) based on the distant related extreme value amendment of sea temperature
1) the warm data in global grid type sea are obtained
From network address https: //www.metoffice.gov.uk/hadobs/hadisst/data/download.html
Place obtains the warm data set in global grid type sea (file name similar to HadISST1_SST_2017.txt.gz, wherein 2017 represent
Time has recorded average warm data in sea of each month in file).In order to reduce influence of the climate change to method validity, palpus
Download later data file in 1980.The format of original document is .gz file, and listed files is as shown in table 6:
The original document for the warm data in global grid type sea that 6 format of table is .gz
| Serial number | Compressed file title |
| 1 | HadISST1_SST_1961-1990.txt.gz |
| 2 | HadISST1_SST_1991-2003.txt.gz |
| 3 | HadISST1_SST_2004.txt.gz |
| 4 | HadISST1_SST_2005.txt.gz |
| 5 | HadISST1_SST_2006.txt.gz |
| 6 | HadISST1_SST_2007.txt.gz |
| 7 | HadISST1_SST_2008.txt.gz |
| 8 | HadISST1_SST_2009.txt.gz |
| 9 | HadISST1_SST_2010.txt.gz |
| 10 | HadISST1_SST_2011.txt.gz |
| 11 | HadISST1_SST_2012.txt.gz |
| 12 | HadISST1_SST_2013.txt.gz |
| 13 | HadISST1_SST_2014.txt.gz |
| 14 | HadISST1_SST_2015.txt.gz |
| 15 | HadISST1_SST_2016.txt.gz |
The .gz ending of original document each in table 6, which is deleted, can obtain corresponding text file.
2) correlation of analysis sea warm data and diameter flow data
By the average diameter flow data of the forecast moon since 1981 and the different grids in the whole world shift to an earlier date different months (12 months with
It is interior) temperature sequence in each year does correlation analysis (calculation formula such as following formula).
Wherein: rXYFor relative coefficient, if it is canonical variable X and variable Y correlation, the variable if being negative
X and variable Y negative correlation represent variable X if it is 0 and variable Y are independent variable, represented if if it is 1 variable X and
Variable Y is linear relationship；N is the sequence length of variable X and variable Y；WithFor the arithmetic mean of variable X and variable Y
Value.
Section diameter flow data (many years sequence) correlation analysis in Global Sst data (many years sequence) and forecast month
The result is that obtain with the highest sea position of Inflow Sequence related coefficient with month, specific algorithm process are as shown in Figure 2 in advance.This
The diameter flow data that step is used is as shown in table 7:
7 Danjiangkou Reservoir monthly average of table storage
Analysis by above step it is available with 6,7,8, the sea position in some highest month of September part correlation
(ranks number of locating text file), shown in specific the following table 8:
Table 8 and 6,7,8, the sea position in some highest month of September part relative coefficient and related coefficient
| Runoff month | Extra large temperature month | Extra large potential temperature sets i_r (row) | Extra large potential temperature sets j_r (column) | Related coefficient |
| 6 | Preceding in April, 1 | 317 | 9 | 0.54 |
| 7 | Preceding part of in September, 1 | 131 | 155 | -0.65 |
| 8 | Preceding in March, 1 | 30 | 13 | 0.61 |
| 9 | Preceding 1 year 2 month | 94 | 4 | 0.69 |
3) it generates and is based on distant relevant Runoff Forecast
By previous step Hai Wen and runoff correlation analysis, i-th _ r row in Global Sst raster data, jth _ r have been obtained
Column shift to an earlier date the Inflow Sequence correlation highest of the sea warm data and forecast month of leadMonth, according to the forecast side described before
Method, draws Hai Wen and runoff comparative diagram, production 6,7,8, the forecast of September part, as shown in Fig. 5, Fig. 6, Fig. 7 and Fig. 8.
In Fig. 5, Fig. 6, Fig. 7 and Fig. 8, solid broken line is averaged for certain same period many years section month, and (such as certain is disconnected for runoff process
The runoff process of 1981 to the parts of in September, 2016 in face), the dotted line of lower section is some highest sea of corresponding correlation
The warm sequence in the sea of position.It should be noted that the time of lower section sea warm broken line and flow broken line is not consistent, but pass through
It is after translation as a result, because being taken leadMonth month forward when doing correlation analysis.Therefore, the warm broken line in sea is opposite
A number is had more in flow broken line, exactly this number can predict the following runoff.Prediction technique is historical data
Comparison Method, specially choose in history with the point (the last one extra large warm spot) immediate warm data point in 5 seas, according to selected
The warm data point in 5 seas finds out corresponding historical traffic data.Arithmetic mean is done to 5 historical traffic datas, obtains future
Certain month forecast conclusion Q_SST_Forecast, and with the maximum value and minimum value of 5 historical traffic datas, as certain following moon
Average diameter flows out existing possibility section.
Therefore, using the distant relevant forecasting procedure of sea temperature is based on, on May 5th, 2017 in June, 2017, July, August, 9
The average storage of the moon does Medium-long Term Prediction, and obtaining forecast result is respectively 1000m3/s、1000m3/s、900m3/s、3200m3/ s,
Corresponding forecast anomaly is -20%, -57%, -58% and+52%.As it can be seen that the forecast knot based on the extra large distant correlation technique of temperature
By thinking that July, August and September will appear extreme event.
4) logarithm runoff DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM carries out extreme value amendment
Based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value according to this patent, using being based on
The extreme value in July, August and September that Hai Wenyao correlation technique is forecast, is modified DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM conclusion, obtains finally forecasting knot
Fruit is 1752.2m3/s、1000m3/s、900m3/s、3200m3/s.Forecast result of the forecast result relative to conventional method, greatly
The prediction ability for extreme event is improved greatly.The forecasting procedure carries out in the storage forecast of Danjiangkou flood season in 2017 for the first time
Scientific dispatch has been carried out to reservoir using the forecast conclusion using, successful predicting flood season water situation, relevant departments, has been ensured
Route normal operation in 2017, economic and social benefit are significant.
Summarize: the present embodiment is with the middle length of Danjiangkou Reservoir 2017 major flood season June, July, August and autumn flood phase September
For phase storage forecast, illustrate the advantage of herein described method.It is counted according to live state information, Danjiangkou Reservoir in June, 2017,
July, August, the average storage of September are respectively 1511m3/s、805m3/s、721m3/ s and 3700m3/s.June, July, August, September
Many years be averagely put in storage respectively 1242m3/s、2332m3/s、2127m3/ s and 2110m3/s.Thus in June, 2017,7 are obtained
The moon, August, September average storage anomaly be+20%, -65%, -65% and+75%.Using traditional DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM method,
On May 5th, 2017 does Medium-long Term Prediction to the average storage of in June, 2017, July, August, September, and obtaining forecast result is respectively
1752.2m3/s、2496m3/s、2001m3/ s and 1799.8m3/s.Using the distant relevant forecasting procedure of sea temperature is based on, in 2017
May 5 did Medium-long Term Prediction to the average storage of in June, 2017, July, August, September, and obtaining forecast result is respectively 1000m3/
s、1000m3/s、900m3/s、3200m3/ s, corresponding forecast anomaly are -20%, -57%, -58% and+52%.As it can be seen that
Forecast conclusion based on the extra large distant correlation technique of temperature thinks that July, August and September will appear extreme event.Therefore, institute according to the present invention
State 8 utilized based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value based on the distant correlation technique forecast of extra large temperature
The extreme value of the moon, September And October, are modified DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM conclusion, and obtaining final forecast result is 1752.2m3/s、1000m3/
s、900m3/s、3200m3/s.Forecast result of the forecast result relative to conventional method, substantially increases for extreme event
Prediction ability.The forecasting procedure is applied in the storage forecast of Danjiangkou flood season in 2017 for the first time, and successful predicting flood season comes
Regimen condition, relevant departments have carried out scientific dispatch to reservoir using the forecast conclusion, have ensured Route 2017
The normal operation in year, economic and social benefit are significant.Tradition forecasts Fig. 9 visible with the detailed comparisons of the context of methods value of forecasting.
By using above-mentioned technical proposal disclosed by the invention, following beneficial effect: the method for the invention has been obtained
Numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM is combined with the distant related extremum prediction of extra large temperature, makes full use of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM excellent in trend prediction
Gesture, and the distant related advantage in prediction of extremum of sea temperature, can be realized the moon scale Runoff Forecast that leading time is 1 year, are becoming
In gesture prediction and prediction of extremum, the more currently used numerical value set forecasting procedure of the method for the invention or statistical method tool
There is higher precision.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
Depending on protection scope of the present invention.
Claims (4)
1. one kind is based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value, which is characterized in that the method packet
It includes:
S1 determines forecast area and forecasting period；
S2 is forecast using forecast area of the numerical value set forecasting procedure to forecasting period, obtains average runoff forecast data
Conclusion α, is denoted as Q_Model_Forecast；
S3 forecasts the forecast area of forecasting period using the extra large distant correlation technique of temperature, obtains average runoff forecast data knot
By β, it is denoted as Q_SST_Forecast；
The anomaly value of Q_SST_Forecast is calculated according to formula (1):
Ano is anomaly value,For the same period long-time average annual value of forecasting period；
S4, forecast conclusion of the forecast area in forecasting period
When | Ano | >=50%, then there is the possibility of extreme case in the runoff of the forecasting period obtained based on the extra large distant correlation technique of temperature
Property it is high, use Q_SST_Forecast to replace Q_Model_Forecast as finally forecasting conclusion；
As | Ano | there is a possibility that extreme case in the runoff of < 50%, the forecasting period obtained based on the extra large distant correlation technique of temperature
It is low, use Q_Model_Forecast as final forecast conclusion；
S2, specifically:
S21 obtains the precipitation rate forecast data of publication in nearest 5 days from global climate forecast system；
S22 parses the precipitation speed data for obtaining global grid formula to the precipitation rate forecast data, to lattice mutually in the same time
Net formula precipitation speed data takes arithmetic mean to get each grid by 6h mean precipitation rate；
S23, calculate obtain each grid by 6h accumulative rainfall amount；
S24 obtains forecast area by face mean rainfall P for 24 hours according to formula (2)；
PiIt is the 6h accumulative rainfall amount of i-th of mesh point in forecast area, n is the quantity of forecast area mesh point；
According to forecast area by face mean rainfall for 24 hours, estimated prediction region each calendar month specified in forecasting period
Month accumulative rainfall amount forecast data；
S25, if the moon accumulative rainfall amount forecast data P for the calendar month t that forecast basin is specified in forecasting periodt, obtain in history
Calendar month t and PtThe immediate multiple actual measurement precipitation, that is, corresponding run-off of each actual measurement precipitation, by each run-off
Arithmetic mean of instantaneous value as forecast basin in the Runoff Forecast opinion of calendar month t, S25 is repeated, until obtain all specified natures
Obtained all Runoff Forecast opinions are formed average runoff forecast data conclusion α, are denoted as Q_ by the Runoff Forecast opinion of the moon
Model_Forecast；
S3, specifically:
S31 obtains the warm data in global grid type sea；
S32, since being obtained from 1981, the section to be forecast of forecast area specifies each year history of moon x flat in forecasting period
Equal diameter flow data, is denoted as sequence one；
The temperature sequence that different grids in the warm data in global grid type sea shift to an earlier date each year in m month is obtained, is denoted as sequence two, m≤
12；
Sequence one and sequence two are subjected to correlation analysis, obtain with the maximum sea position of one relative coefficient of sequence in advance
Month；Sequence a will be denoted as with the highest sequence two of one relative coefficient of sequence；
S33, if the warm data sequence in sea that i-th _ r row, jth _ r column shift to an earlier date leadMonth in the warm data in global grid type sea is sequence
A is arranged, the comparative diagram of sequence a and sequence one are drawn, the curve of sequence a is in the corresponding diameter of the warm data A in the sea at time least significant end
Flow data is the runoff to be forecast that section to be forecast specifies moon x in forecasting period；
S34, selection and the immediate multiple sea temperature data points of the warm data A in sea from extra large warm data history data, and obtain each
Historical traffic data corresponding to extra large temperature data point is then by the arithmetic average of obtained multiple historical traffic datas, i.e., flat
Equal Runoff Forecast data specify the runoff to be forecast of moon x, the multiple history that will be obtained as section to be forecast in forecasting period
Section of the maximum value and minimum value of data on flows as average runoff forecast data, by average runoff forecast data and its section
As average runoff forecast data conclusion β, it is denoted as Q_SST_Forecast, is completed using the sea distant correlation technique of temperature to forecasting period
Forecast area forecast.
2. being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value according to claim 1, feature exists
In, S1, specifically:
Forecast demand is received, and extracts section and forecasting period to be forecast in the forecast demand, by the upper of section to be forecast
Basin is as forecast basin.
3. being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value according to claim 1, feature exists
In, in S32, correlation analysis, specifically: the relative coefficient r of variable X and variable Y is obtained using formula (3)XY, formula (3)
Are as follows:
Xi∈ X, X are Inflow Sequence；Yi∈ Y, Y are the warm sequence in sea；
If rXYIt is positive, then variable X and variable Y correlation；
If rXYIt is negative, then variable X and variable Y negative correlation；
If rXYBe 0, then variable X and variable Y be independent variable；
If rXYIt is 1, then variable X and the linear correlativity of variable Y；
N is the sequence length of variable X and variable Y；WithFor the arithmetic average of variable X and variable Y；
Variable X is sequence one, and variable Y is sequence two or variable X is sequence two, and variable Y is sequence one.
4. being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value according to claim 1, feature exists
In maximal correlation property coefficient is greater than 0.6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810036068.3A CN108171007B (en) | 2018-01-15 | 2018-01-15 | One kind being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810036068.3A CN108171007B (en) | 2018-01-15 | 2018-01-15 | One kind being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108171007A CN108171007A (en) | 2018-06-15 |
| CN108171007B true CN108171007B (en) | 2019-02-05 |
Family
ID=62514413
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810036068.3A Expired - Fee Related CN108171007B (en) | 2018-01-15 | 2018-01-15 | One kind being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108171007B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636032B (en) * | 2018-12-11 | 2021-08-24 | 中国科学院深圳先进技术研究院 | Precipitation forecast method, system, terminal and storage medium based on multi-mode integration |
| CN109816167A (en) * | 2019-01-18 | 2019-05-28 | 昆仑(重庆)河湖生态研究院(有限合伙) | Runoff Forecast method and Runoff Forecast device |
| CN109902115B (en) * | 2019-01-28 | 2022-03-04 | 中山大学 | Method for programmatically extracting region and watershed data oriented to raster data |
| CN109884734A (en) * | 2019-02-18 | 2019-06-14 | 哈尔滨工程大学 | A kind of ocean temperature Similarity Method for Forecasting based on similar disparity |
| CN110133756A (en) * | 2019-04-26 | 2019-08-16 | 长威信息科技发展股份有限公司 | A kind of weather station precipitation amount correction method |
| CN110135652B (en) * | 2019-05-23 | 2023-04-07 | 吉林大学 | Long-term flood season runoff prediction method |
| CN111338005B (en) * | 2020-03-02 | 2021-07-30 | 中国人民解放军国防科技大学 | Prediction method for generation frequency of tropical cyclone in northwest Pacific ocean on monthly scale |
| CN113254878B (en) * | 2021-05-19 | 2022-05-17 | 中国电建集团昆明勘测设计研究院有限公司 | Method for judging water temperature structure of reservoir in hydraulic and hydroelectric engineering |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7136756B1 (en) * | 2004-11-02 | 2006-11-14 | Vieux And Associates, Inc. | Method for determining runoff |
| CN102867106A (en) * | 2012-08-14 | 2013-01-09 | 贵州乌江水电开发有限责任公司 | Method and system for predicting short-term running water |
| CN107292098A (en) * | 2017-06-15 | 2017-10-24 | 河海大学 | Medium-and Long-Term Runoff Forecasting method based on early stage meteorological factor and data mining technology |
| CN107423857A (en) * | 2017-07-31 | 2017-12-01 | 长江水利委员会水文局 | A kind of long-term water multiple target joint probability Forecasting Methodology in region |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104091074B (en) * | 2014-07-12 | 2017-10-10 | 长安大学 | A kind of MEDIUM OR LONG RANGE HYDROLOGIC FORECAST METHOD based on empirical mode decomposition |
| CN106169024A (en) * | 2016-06-29 | 2016-11-30 | 中国南方电网有限责任公司 | The southern trough influence research method to Medium-and Long-Term Runoff Forecasting |
| CN107341318B (en) * | 2017-07-17 | 2020-07-24 | 河海大学 | Simulation method of full-river-based monthly runoff time displacement two-dimensional matrix |
-
2018
- 2018-01-15 CN CN201810036068.3A patent/CN108171007B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7136756B1 (en) * | 2004-11-02 | 2006-11-14 | Vieux And Associates, Inc. | Method for determining runoff |
| CN102867106A (en) * | 2012-08-14 | 2013-01-09 | 贵州乌江水电开发有限责任公司 | Method and system for predicting short-term running water |
| CN107292098A (en) * | 2017-06-15 | 2017-10-24 | 河海大学 | Medium-and Long-Term Runoff Forecasting method based on early stage meteorological factor and data mining technology |
| CN107423857A (en) * | 2017-07-31 | 2017-12-01 | 长江水利委员会水文局 | A kind of long-term water multiple target joint probability Forecasting Methodology in region |
Non-Patent Citations (4)
| Title |
|---|
| 变化环境下气象水文预报研究进展;雷晓辉等;《水利学报》;20180131;第49卷(第1期);第606-612页 |
| 基于ELman神经网络的锦屏一级水电站年平均径流结合预报研究;贺志尧等;《水电能源科学》;20171031;第35卷(第10期);第25-28页 |
| 洮儿河镇西站径流长期预报研究;许士国等;《水文》;20071031;第27卷(第5期);第86-89页 |
| 结合大气环流和遥相关信息的集合径流预报方法及其应用;杨龙等;《清华大学学报 （自然科学版）》;20131231;第53卷(第5期);第9-14页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108171007A (en) | 2018-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108171007B (en) | One kind being based on the modified Medium-and Long-Term Runoff Forecasting method of numerical value DATA PROCESSING IN ENSEMBLE PREDICTION SYSTEM extreme value | |
| Armal et al. | Trends in extreme rainfall frequency in the contiguous United States: Attribution to climate change and climate variability modes | |
| Mwangi et al. | Forecasting droughts in East Africa | |
| Siuta et al. | WRF hub-height wind forecast sensitivity to PBL scheme, grid length, and initial condition choice in complex terrain | |
| Komma et al. | Ensemble prediction of floods–catchment non-linearity and forecast probabilities | |
| CN107292098A (en) | Medium-and Long-Term Runoff Forecasting method based on early stage meteorological factor and data mining technology | |
| Schulze | Impacts of global climate change in a hydrologically vulnerable region: challenges to South African hydrologists | |
| Shi et al. | Temporal and spatial variations of potential evaporation and the driving mechanism over Tibet during 1961–2001 | |
| CN102495937A (en) | Prediction method based on time sequence | |
| El Alaoui El Fels et al. | Rainfall regionalization and variability of extreme precipitation using artificial neural networks: a case study from western central Morocco | |
| Diodato et al. | Storminess and environmental changes in the Mediterranean Central Area | |
| CN116756481B (en) | Watershed future hydrologic drought probability analysis method based on time-varying gain model and Copula | |
| Pichard et al. | Hydro-climatology of the Lower Rhône Valley: historical flood reconstruction (AD 1300–2000) based on documentary and instrumental sources | |
| Gaur et al. | Gridded extreme precipitation intensity–duration–frequency estimates for the Canadian landmass | |
| Bajamgnigni Gbambie et al. | Added value of alternative information in interpolated precipitation datasets for hydrology | |
| Kormos et al. | 31 years of hourly spatially distributed air temperature, humidity, and precipitation amount and phase from Reynolds Critical Zone Observatory | |
| Modi et al. | Investigating the role of snow water equivalent on streamflow predictability during drought | |
| Tiancheng et al. | Application of Improved Naive Bayesian‐CNN Classification Algorithm in Sandstorm Prediction in Inner Mongolia | |
| Araki et al. | Evaluation of GLDAS soil moisture seasonality in arid climates | |
| Suprit et al. | Simulating the daily discharge of the Mandovi River, west coast of India | |
| Zhang et al. | Simulation of daily precipitation from CMIP5 in the Qinghai-Tibet Plateau | |
| Branković et al. | Downscaling of ECMWF ensemble forecasts for cases of severe weather: Ensemble statistics and cluster analysis | |
| Bastin et al. | Inter Annual Variability of wind speed in India | |
| Boisserie et al. | Calibrated forecasts of extreme windstorms using the extreme forecast index (EFI) and shift of tails (SOT) | |
| Meher | Rainfall and runoff estimation using hydrological models and ANN techniques |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190205 |