CN106709181B - A kind of hydrological distribution model rating method based on multiple programming and agent approach - Google Patents
A kind of hydrological distribution model rating method based on multiple programming and agent approach Download PDFInfo
- Publication number
- CN106709181B CN106709181B CN201611206541.5A CN201611206541A CN106709181B CN 106709181 B CN106709181 B CN 106709181B CN 201611206541 A CN201611206541 A CN 201611206541A CN 106709181 B CN106709181 B CN 106709181B
- Authority
- CN
- China
- Prior art keywords
- model
- hydrological
- population
- peak value
- calibration
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of hydrological distribution model rating method based on multiple programming and agent approach, it is intended to improve the calibration efficiency and simulation precision of Watershed Hydrologic Models.The described method includes: 1) Parallel Programming Environment based on MPICH building, coupling open source hydrological distribution model source program and multi-objective optimization algorithm；2) calibration is carried out to hydrological model for diurnal courses sequence and peak value sequence respectively, and establishes corresponding model framework；3) linear reservoir for merging above-mentioned model using agent approach, forms the runoff simulation of recombination.The beneficial effects of the present invention are: parallel multiple target operation can effectively improve the calibration efficiency of hydrological distribution model, slip is fixed time；Modular approach can make full use of peak value model and the respective advantage of diurnal courses model, be remarkably improved simulation precision.
Description
Technical field
The present invention relates to hydrological model technical fields, relate in particular to a kind of distribution based on multiple programming and agent approach
Formula hydrological model rating method.
Background technique
With the aggravation of urban sprawl and climate change, the generation of the natural calamities such as flood arid is more and more frequent, seriously
National economy and people life property safety are threaten, accurately mould is carried out to flood and drought process using Watershed Hydrologic Models
It fits and forecasts with unerring accuracy to alleviation the condition of a disaster important in inhibiting.
Hydro-Model Parameter Calibration Technology calibration is the important step of hydrological simulation, and the calibration of model parameter directly influences hydrological model
Prediction performance.It is numerous that hydrological distribution model is related to parameter, and calibration is complicated, adjusts ginseng not only inefficiency manually, expends a large amount of
Time and efforts, and be difficult to find that optimal parameter group, bring very big uncertainty to model result, significantly limit
The application of hydrological distribution model.Therefore, the automatic calibration technology of parameter is appealed to, the calibration of hydrological distribution model will be significantly improved
Efficiency and simulation precision.But as the calculating process of hydrological distribution model is increasingly sophisticated fine, hydrological simulation is more and more smart
While true, it is increasingly longer also to lead to time of single simulation, even if the time of consuming is still very using automatic calibration technology
It is long.
Practical Project shows to be usually present multiple conflicting objective functions in hydrologic process, traditional based on monocular
Target prioritization scheme only only accounts for the feature of hydrologic process in a certain respect, cannot sufficiently reflect the different dynamic of Hydrology
Scholarship and moral conduct is to cause simulation precision insufficient.
Summary of the invention
To solve the deficiencies in the prior art, the purpose of the present invention is to provide a kind of point based on multiple programming and agent approach
Cloth hydrological model rating method, to solve hydrological distribution model, time-consuming during calibration, low efficiency, and precision is not high
Problem.
To realize above-mentioned target, a kind of hydrological distribution model rating method based on multiple programming and agent approach is special
Sign is, includes the following steps:
1) the hydrometeorological data that hydrological model needs are collected, meteorological data are interpolated into the resolution ratio of model needs, really
Cover half type waits for calibration parameter；
2) threshold value for determining super sizing technique is ground using super sizing technique (Peaks Over Threshold methods) selection
The peak value sequence of diurnal courses during studying carefully.
3) MPI based on MPICH (Message-Passing Interface) Parallel Programming Environment, by multiple-objection optimization
Algorithm ε-NSGA II and hydrological model source program are coupled, and are made with the meteorological data that step 1) is collected, and actual measurement hydrographic data
For input, with the assorted efficiency of receiving for receiving assorted efficiency factor NSE/D and relative deviation Bias/D and peak value sequence of diurnal courses sequence
Coefficient NSE/P and relative deviation Bias/P is as objective function, calibration hydrological model；
4) target Watershed Hydrologic Models are established according to calibration result, linear reservoir is integrated using agent approach, forms recombination
Inflow Sequence.
In above-mentioned technical proposal, step 2) specifically:
(1) threshold value T is set, and the setting of threshold value need to guarantee that the peak value finally chosen meets independence requirement；
(2) runoff peak value is chosen for more than the run-off of the threshold value.
The independence requirement are as follows:
(1) natural logrithm that the number of days interval between two neighboring peak value must be greater than drainage area (sq. mi) adds 5；
Intermediate flow between (2) two adjacent peaks must be less than 75% of smaller value in the two peak-datas.
Multi-objective optimization algorithm ε-NSGA II and hydrological model source program are coupled described in step 3), specific side
Method is as follows:
(1) relevant parameter of II algorithm of ε-NSGA: initial population number N, evolutionary generation gens, crossover probability, variation is set
Probability etc., while the related number and range to calibration parameter of VIC model is set；
(2) MPI is initialized, the process number cSize for including in MPI process group and process of this process in communication set are obtained
Number pRank；
(3) Population Size is randomly generated by main program master and is the parent population P0 (i.e. N number of parameter group) of N, and passed through
Processor active task is distributed to cSize subprogram (being separately operable hydrological model according to parameter group) by MPI_Bcast function；
(4) according to calculated result, the fitness function for calculating each individual in parent population P0 (receives assorted efficiency factor and relatively
Deviation), it carries out quick non-dominated ranking and crowding calculates, generate the non-dominant archive solution of A0 ε；
(5) the injection mechanism (Injection Scheme) for using 25%, the non-dominant archive solution of A0 ε in (4) are formed
25% individual in new parent population Pt, remaining is 75% by being randomly generated, Population Size Nt；
(6) progeny population Qt, Population Size Nt are generated by selection, intersection, mutation operator；By progeny population Qt and father
Merge for Pt and forms new population Rt=Pt ∪ Qt, Population Size 2Nt；
(7) processor active task is distributed to by cSize subprogram, and root by MPI_Bcast function by main program master
The calculating that quick non-dominated ranking and crowding are carried out according to operation result according to the non-dominated ranking of individual each in group and is gathered around
It squeezes distance selection and generates elite parent population Nt, while generating the non-dominant archive solution of At ε；
(8) increased the random sample that generates by the Population Size of archive population A t according to the 25% of (5) injection mechanism and formed
Population Size is the parent population Pt+1 of 4At.
(9) circulation step (6)-(8), evolutionary generation or other termination conditions until reaching setting.
It is that diurnal courses and peak value runoff select different parameter groups, structure according to the optimum results of II multi-objective Algorithm of ε-NSGA
Build the model/D model specifically for daily stream flow and the model/P model for peak analogue.
It is integrated the output result of model/D model and model/P model to obtain final calibration as follows
As a result QC:
Wherein mmodel/D(Q) and mmodel/PIt (Q) is the distribution coefficient of model/D and model/P model, Qmodel/A(t) and
Qmodel/P(t) be respectively model/D and model/P model linear reservoir.QO, maxDiurnal courses data are periodically surveyed most for rate
Big value, T are the threshold value of super sizing technique, and E (T) is the expectation for the peak value that all super sizing techniques are chosen.QoFor measuring runoff.
By using above-mentioned technological means, the invention has the benefit that
(1) parallel automatic rating method solve hydrological distribution model during calibration low efficiency, time-consuming asks
Topic, is remarkably improved model efficiency, in the same time, the higher calibration result of accuracy can be obtained.
(2) multi-objective optimization algorithm significantly reduces the uncertainty of model parameter, according to multiple target solution, ginseng
Number determination process is more objective, avoids the subjectivity artificially judged.
Further aspect of the present invention is clearly stated that in the description of following specific embodiments.
Detailed description of the invention
Fig. 1 is the flow diagram of example of the present invention.
Fig. 2 is the pseudocode of multiple programming.
Fig. 3 is the solution NSE/D and NSE/P of multi-objective Algorithm, is determined for parameter.
Fig. 4 is diurnal courses single goal calibration result.
Fig. 5 is peak value runoff single goal calibration result.
Fig. 6 is the multiple target calibration result recombinated using agent approach.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and detailed description.For protrusion advantage of the invention, adopt
The variable infiltration capacity model VIC for using University of Washington to propose is embodied as case.
As shown in Figure 1, a kind of hydrological distribution model rating method based on multiple programming and agent approach of the invention, packet
Include following steps:
It extracts and locates using GIS data and Law of DEM Data of the arcgis to target basin
Reason obtains data, the sizing grids such as basin DEM, flow direction and profile and takes 5km × 5km.
Vegetation parameter, soil parameters and the meteorological Parameter File that preparation model needs, file resolution is 5km × 5km.
Vegetation pattern uses the global 1km soil cover classification of University of Maryland's development, is divided into 11 classes.Vegetation parameter packet
Include vegetation structure impedance, minimum stomata impedance, the leaf area index of vegetation, zero-plane displacement, albedo, roughness and root zone
Distribution proportion etc. in each sublevel soil.
Soil data mainly includes base flow exponent, satisfies according to Chinese 1: 100 ten thousand pedological map cutting, soil parameters
With hydraulic conductivity, soil moisture diffusion parameter, soil antecedent water content, net lattice control elevation, soil horizon thickness, average soil
Earth temperature, soil pressure in bubbles, soil content, soil total body density, soil bulk density, critical soil moisture content ratio,
Wilting point soil moisture content ratio, exposed soil surface roughness, accumulated snow rugosity, year-round average temperature, residual moisture content etc..Above-mentioned parameter
Data can be extracted by arcgis to directly determine, and be directly used in model, generally without modification.
Variable infiltration capacity curve parameter, the non-linear base flow maximum value factor, base flow maximum flow rate, base flow hair in soil file
Raw maximum soil water storage capacity factor measure and thickness of soil are determined by parameter calibration.
Meteorological file includes five daily rain amount total amount, highest day temperature, minimum day temperature, wind speed and relative humidity variables.
The peak value sequence of diurnal courses during studying is chosen using super sizing technique, the threshold value of super sizing technique is flat for many years according to basin
Equal run-off determines that value is twice average annual flow, while ensuring that peak-data is greater than 30 in the research phase.
Multi-objective genetic algorithm ε-NSGA II and VIC model source code are coupled based on MPICH, concrete thought
For (as shown in Fig. 2 pseudocode):
(1) relevant parameter of II algorithm of ε-NSGA: initial population number N, evolutionary generation gens, crossover probability, variation is set
Probability, while the related number and range to calibration parameter of VIC model is set；
(2) MPI is initialized, the process number cSize for including in MPI process group and process of this process in communication set are obtained
Number pRank；
(3) Population Size is randomly generated by main program master and is the parent population P0 (i.e. N number of parameter group) of N, and passed through
Processor active task is distributed to cSize subprogram (being separately operable hydrological model according to parameter group) by MPI_Bcast function；
(4) according to calculated result, the fitness function for calculating each individual in parent population P0 (receives assorted efficiency factor and relatively
Deviation), it carries out quick non-dominated ranking and crowding calculates, generate the non-dominant archive solution of A0 ε；
(5) the injection mechanism (Injection Scheme) for using 25%, the non-dominant archive solution of A0 ε in (4) are formed
25% individual in new parent population Pt, remaining is 75% by being randomly generated, Population Size Nt；
(6) progeny population Qt, Population Size Nt are generated by selection, intersection, mutation operator；By progeny population Qt and father
Merge for Pt and forms new population Rt=Pt ∪ Qt, Population Size 2Nt；
(7) processor active task is distributed to by cSize subprogram, and root by MPI_Bcast function by main program master
The calculating that quick non-dominated ranking and crowding are carried out according to operation result according to the non-dominated ranking of individual each in group and is gathered around
It squeezes distance selection and generates elite parent population Nt, while generating the non-dominant archive solution of At ε；
(8) increased the random sample that generates by the Population Size of archive population A t according to the 25% of (5) injection mechanism and formed
Population Size is the parent population Pt+1 of 4At.
(9) circulation step (6)-(8), evolutionary generation or other termination conditions until reaching setting.
Above-mentioned meteorology, soil, vegetation data and measuring runoff data are inputted, according to following objective function come calibration hydrology mould
Type:
N and M is the quantity of diurnal courses and peak value runoff, Q in formulaobsAnd QsimRespectively survey daily flow and simulating sun stream
Amount, QObs, PAnd QSim, PRespectively actual measurement and simulated peak.
The solution of multi-objective optimization algorithm according to Fig.3, is that diurnal courses and peak value runoff select different parameters
Group constructs the VIC/D model specifically for daily stream flow and the VIC/P model for peak analogue.
The output result of VIC/D model and VIC/P model is integrated as follows to obtain final calibration result
QC:
Wherein mVIC/D(Q) and mVIC/PIt (Q) is the distribution coefficient of VIC/D and VIC/P model, QVIC/D(t) and QVIC/P(t) divide
Not Wei VIC/D and VIC/P model output.QO, maxFor the maximum value for surveying diurnal courses data, T is the threshold value of super sizing technique, E
It (T) is the expectation for all threshold values chosen.
Result using above method calibration is as Figure 4-Figure 6, it can be seen that using agent approach integration Inflow Sequence without
By high runoff or low runoff plot and measuring runoff deviation is little.
It is described above that only example of the invention is implemented, it is not intended to restrict the invention, the research for this field
For person, the invention may be variously modified and varied.All to limit in range in claim of the invention, that is done is any
Modification, equivalent replacement, improvement etc., should all be within protection scope of the present invention.
Claims (5)
1. a kind of hydrological distribution model rating method based on multiple programming and agent approach, which is characterized in that including walking as follows
It is rapid:
1) the hydrometeorological data that hydrological model needs are collected, meteorological data is interpolated into the resolution ratio of model needs, determines mould
Type waits for calibration parameter；
2) threshold value for determining super sizing technique chooses the peak value sequence of diurnal courses during studying using super sizing technique；
3) based on the Parallel Programming Environment of MPICH, multi-objective optimization algorithm ε-NSGA II and hydrological model source program are subjected to coupling
It closes, input is used as with the meteorological data that step 1) is collected, and actual measurement hydrographic data, with the assorted efficiency factor of receiving of diurnal courses sequence
NSE/D and relative deviation Bias/D and peak value sequence receive assorted efficiency factor NSE/P and relative deviation Bias/P as target
Function, calibration hydrological model；
4) target Watershed Hydrologic Models are established according to calibration result, linear reservoir is integrated using agent approach, form the runoff of recombination
Sequence；
Multi-objective optimization algorithm ε-NSGA II and hydrological model source program are coupled described in step 3), specific method is such as
Under:
(1) relevant parameter of II algorithm of ε-NSGA: initial population number N, evolutionary generation gens, crossover probability, mutation probability is set,
The related number and range to calibration parameter of VIC model is set simultaneously；
(2) MPI is initialized, the process number of the process number cSize for including in MPI process group and this process in communication set is obtained
pRank；
(3) the parent population P0 that Population Size is N is randomly generated by main program master, and will be transported by MPI_Bcast function
Calculation task distributes to cSize subprogram, and subprogram is separately operable hydrological model according to parameter group；
(4) according to calculated result, the fitness function of each individual in parent population P0 is calculated, that is, receives assorted efficiency factor and relatively inclined
Difference, carries out quick non-dominated ranking and crowding calculates, and generates the non-dominant archive solution of A0 ε；
(5) 25% injection mechanism is used, the non-dominant archive solution of A0 ε in (4) is formed 25% in new parent population Pt
Individual, remaining is 75% by being randomly generated, Population Size Nt；
(6) progeny population Qt, Population Size Nt are generated by selection, intersection, mutation operator；By progeny population Qt and parent Pt
Merge and forms new population Rt=Pt ∪ Qt, Population Size 2Nt；
(7) processor active task is distributed to by cSize subprogram by MPI_Bcast function by main program master, and according to fortune
Calculate result and carry out the calculating of quick non-dominated ranking and crowding, according to the non-dominated ranking of individual each in group and it is crowded away from
Elite parent population Nt is generated from selection, while generating the non-dominant archive solution of At ε；
(8) increased the random sample that generates by the Population Size of archive population A t according to the 25% of (5) injection mechanism and formed population
Size is the parent population Pt+1 of 4At；
(9) circulation step (6)-(8), evolutionary generation or other termination conditions until reaching setting.
2. the hydrological distribution model rating method according to claim 1 based on multiple programming and agent approach, feature
It is, the super sizing technique used in step 2), detailed process are as follows:
(1) threshold value T is set, and the setting of threshold value need to guarantee that the peak value finally chosen meets independence requirement；
(2) runoff peak value is chosen for more than the run-off of the threshold value.
3. the hydrological distribution model rating method according to claim 2 based on multiple programming and agent approach, feature
It is, the independence requires to specifically include:
(1) natural logrithm that the number of days interval between two neighboring peak value must be greater than drainage area adds 5；
Intermediate flow between (2) two adjacent peaks must be less than 75% of smaller value in the two peak-datas.
4. the hydrological distribution model rating method according to claim 1 based on multiple programming and agent approach, feature
Be, target Watershed Hydrologic Models constructed by step 4) include specifically for diurnal courses series model model/D model and
For the model/P model for the peak analogue that super sizing technique is chosen, two group models are respectively configured not according to multiple target solution
Same parameter group.
5. the hydrological distribution model rating method according to claim 4 based on multiple programming and agent approach, feature
It is, the specific algorithm of agent approach described in step 4) is:
Wherein mmodel/DAnd mmodel/PFor the distribution coefficient of model/D and model/P model, Qmodel/D(t) and Qmodel/P(t) respectively
For the linear reservoir of model/D and model/P model, Qo,maxThe maximum value of diurnal courses data is periodically surveyed for rate, T is overdetermination amount
The threshold value of method, E (T) are the expectation for the peak value that all super sizing techniques are chosen, QoFor measuring runoff.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611206541.5A CN106709181B (en) | 2016-12-23 | 2016-12-23 | A kind of hydrological distribution model rating method based on multiple programming and agent approach |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611206541.5A CN106709181B (en) | 2016-12-23 | 2016-12-23 | A kind of hydrological distribution model rating method based on multiple programming and agent approach |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106709181A CN106709181A (en) | 2017-05-24 |
| CN106709181B true CN106709181B (en) | 2019-09-06 |
Family
ID=58903111
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611206541.5A Active CN106709181B (en) | 2016-12-23 | 2016-12-23 | A kind of hydrological distribution model rating method based on multiple programming and agent approach |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106709181B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109086479A (en) * | 2018-06-20 | 2018-12-25 | 浙江大学 | A kind of hydrological distribution model parameter multiple target rating method based on parallel computation |
| CN109086245B (en) * | 2018-06-29 | 2020-09-04 | 浙江大学 | VIC parameter multi-target method based on substitution model |
| CN110288147A (en) * | 2019-06-21 | 2019-09-27 | 北京师范大学 | Watershed Hydrologic Models determination method for parameter and device |
| CN111176865B (en) * | 2019-12-31 | 2023-04-07 | 中国科学院计算机网络信息中心 | Peer-to-peer mode parallel processing method and framework based on optimization algorithm |
| CN112715322B (en) * | 2020-12-22 | 2022-08-09 | 广东省科学院广州地理研究所 | Method and device for obtaining agricultural irrigation water |
| CN117057253B (en) * | 2023-09-28 | 2023-12-08 | 中国水利水电科学研究院 | Hydrological model parameter calibration method based on space discrete shuffling complex evolutionary algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106625A (en) * | 2013-03-08 | 2013-05-15 | 珠江水利委员会珠江水利科学研究院 | Reservoir, sluice and pump cluster combined saltwater tide control and scheduling method |
| CN104318077A (en) * | 2014-10-09 | 2015-01-28 | 水利部交通运输部国家能源局南京水利科学研究院 | Quantitative analysis method for river runoff change caused by climate change and human activity |
| CN104408318A (en) * | 2014-12-03 | 2015-03-11 | 中国水利水电科学研究院 | Seasonal design flow computing method for river-type water functional areas |
| CN105868534A (en) * | 2016-03-24 | 2016-08-17 | 大连理工大学 | Multi-objective optimization sampling based hydrologic model uncertainty analysis method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9772609B2 (en) * | 2013-05-24 | 2017-09-26 | Optirtc, Inc. | Optimized hydromodification management with active stormwater controls |
-
2016
- 2016-12-23 CN CN201611206541.5A patent/CN106709181B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106625A (en) * | 2013-03-08 | 2013-05-15 | 珠江水利委员会珠江水利科学研究院 | Reservoir, sluice and pump cluster combined saltwater tide control and scheduling method |
| CN104318077A (en) * | 2014-10-09 | 2015-01-28 | 水利部交通运输部国家能源局南京水利科学研究院 | Quantitative analysis method for river runoff change caused by climate change and human activity |
| CN104408318A (en) * | 2014-12-03 | 2015-03-11 | 中国水利水电科学研究院 | Seasonal design flow computing method for river-type water functional areas |
| CN105868534A (en) * | 2016-03-24 | 2016-08-17 | 大连理工大学 | Multi-objective optimization sampling based hydrologic model uncertainty analysis method |
Non-Patent Citations (2)
| Title |
|---|
| 《An optimization based sampling approach for multiple metrics uncertainty analysis using generalized likelihood uncertainty estimation》;Rurui Zhou等;《Journal of Hydrology》;20160613;第274-286页 |
| 《基于NSGA-Ⅱ的水文模型参数多目标优化研究》;黄晓敏等;《人民长江》;20120131;第16-21页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106709181A (en) | 2017-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106709181B (en) | A kind of hydrological distribution model rating method based on multiple programming and agent approach | |
| Alatorre et al. | Soil erosion and sediment delivery in a mountain catchment under scenarios of land use change using a spatially distributed numerical model | |
| Batelaan et al. | GIS-based recharge estimation by coupling surface–subsurface water balances | |
| Abdulla et al. | Application of a macroscale hydrologic model to estimate the water balance of the Arkansas‐Red River Basin | |
| CN102354348A (en) | Watershed scale soil moisture remote sensing data assimilation method | |
| CN109086479A (en) | A kind of hydrological distribution model parameter multiple target rating method based on parallel computation | |
| Chea et al. | Flow simulation in an ungauged catchment of Tonle Sap Lake Basin in Cambodia: Application of the HEC-HMS model | |
| Ibrakhimov | Spatial and temporal dynamics of groundwater table and salinity in Khorezm (Aral Sea Basin), Uzbekistan | |
| Ma et al. | Effect of land use/cover changes on runoff in the Min River watershed | |
| Watson et al. | Distributive rainfall–runoff modelling to understand runoff-to-baseflow proportioning and its impact on the determination of reserve requirements of the Verlorenvlei estuarine lake, west coast, South Africa | |
| Sintondji et al. | Modelling the water balance of Ouémé catchment at the Savè outlet in Benin: contribution to the sustainable water resource management. | |
| Jahanshahi et al. | Simulation of the catchments hydrological processes in arid, semi-arid and semi-humid areas. | |
| Samady | Continuous hydrologic modeling for analyzing the effects of drought on the Lower Colorado River in Texas | |
| Dumedah et al. | Integration of an evolutionary algorithm into the ensemble Kalman filter and the particle filter for hydrologic data assimilation | |
| CN117494419A (en) | Multi-model coupling drainage basin soil erosion remote sensing monitoring method | |
| Vafakhah et al. | Spatial variations of runoff generation at watershed scale | |
| CN109388873A (en) | Garden landscape planing method based on mountainous region mima type microrelief temperature | |
| Xie et al. | Assessing the risks of soil erosion and small reservoir siltation in a tropical river basin in Mali using the SWAT model under limited data condition | |
| Ashaolu et al. | Effect of Land Use/Land Cover Change on Groundwater Recharge in Osun Drainage Basin, Nigeria | |
| Parak | Flood estimation for roads, bridges and dams. | |
| Gusma et al. | Soil erosion rate and hazard level at the Sianjo-anjo Reservoir watershed in Indonesia | |
| Sabale et al. | Comparative study between water yield and consumptive use: a case study of khatav taluka | |
| Park et al. | Comparison of watershed streamflow using projected MIROC3. 2 Hires GCM data and observed weather data for 2000-2009 under SWAT simulation | |
| Deus | Streamflow forecasts due precipitation water in a tropical large watershed at Brazil for flood Early warning, based on SWAT model | |
| Togbevi et al. | Assessing the effect of land use and land cover changes on water balance in the Ouriyori basin (Benin, West Africa) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |