Afsane Farpour; Hosein KhozeymeNezhad
Abstract
Today, the use of hydrological models is mainly necessary to simulate changes in water source and flow (runoff and evaporation). Proper modeling of hydrological processes requires the determination of model parameters. In calibration processes, the values of the model parameters are estimated so ...
Read More
Today, the use of hydrological models is mainly necessary to simulate changes in water source and flow (runoff and evaporation). Proper modeling of hydrological processes requires the determination of model parameters. In calibration processes, the values of the model parameters are estimated so that the model can simulate a natural system well. It is generally impossible to estimate the parameters of such models directly due to the large number of parameters and it is necessary to estimate them with the help of optimization tools (model calibration). In the present study, the parameters of the daily Hymod rainfall-runoff model (a simple conceptual rainfall-runoff model) were calibrated using the Whale algorithm (WOA), which is derived from the way whale food is searched. The evaluation of the mentioned calibration method was performed using daily precipitation, evapotranspiration and transpiration data for 5 years and its validation was performed in 5 years in the Leaf River Basin of the United States. The simulated and observed flow rates were compared using correlation coefficient (R2), root mean square error (RMSE) and Nash-Sutcliffe coefficient (NS). The values of error measurement criteria were 0.91, 1.2 and 0.8 for the calibration period and 0.91, 2.5 and 0.83 for the validation period, respectively. Also, the parameters calculated using the whale algorithm, the maximum moisture storage in the area of 216.95 mm, spatial variation of soil moisture storage 0.38, the distribution factor between the two moisture tanks 0.98, the shelf life in the laminar tank 0.08 days and Shelf life in fast flow tank is 0.47 days. Examination of error values showed that the Whale Optimization Algorithm has high efficiency in calibrating rainfall-runoff models.
alireza memari; Karim Solaimani; mahmood habibnejad; Ali Golkarian
Abstract
AbstractDifferent models have been designed to simulate incoming floods into urban areas from rural areas, one of the models is the EPASWMM. Mashhad urban watershed with 8 inde-pendent subbasins and area of 2,472 hectares, located in south of Mashhad city . Study area is the residential area of Mashhad ...
Read More
AbstractDifferent models have been designed to simulate incoming floods into urban areas from rural areas, one of the models is the EPASWMM. Mashhad urban watershed with 8 inde-pendent subbasins and area of 2,472 hectares, located in south of Mashhad city . Study area is the residential area of Mashhad Paydari township (Area= 95.6 ha) that located below the M2, M3 and M4 subbasins. Due to high peak flood events in M2, M3 and M4 subbasins, the capacity of urban transmission channels in the Paydari township is’nt suf-ficient and after every rain, a lot of runoff enters the residential areas. The purpose of this research is evaluate the efficiency of EPASWMM model in simulating runoff. The study area was divided into 12 hydrological units (OUT1 to OUT12) and after determine of input parameters of the model, the runoff gathering network in the form of nodes-channels was introduced into the model. Four rainfall events were used to evaluate and calibrate the model with (NS), (RMSE) and (R2) coefficients. In the calibration section of the model, the value of the (NS) coefficient for discharge in the first and third events was 0.94 and 0.92 respectively and for runoff height, 0.9 and 0.86, respectively and indicates simulation accuracy of EPASWMM model in both parameters. The results of the model evaluation section for the second and fourth events shows that the NS and R2 coefficients for three parameters of discharge, height and runoff were from 0.78 to 0.9 and 0.73 to 0.89 and indicates a large correlation between observational and simulated data. Flood reduction scenarios in hydrological units have been investigated according to the simulation results.
Saeed Nabipay Lashkarian; Mahmood Arabkhedri; Samad Shadfar
Abstract
Accelerated soil erosion and the production and transfer of large amounts of sediment are considered as one of the most important issues in many parts of Iran. Having enough knowledge on soil erosion and sediment production as well as identification of critical areas are necessary in order to reduce ...
Read More
Accelerated soil erosion and the production and transfer of large amounts of sediment are considered as one of the most important issues in many parts of Iran. Having enough knowledge on soil erosion and sediment production as well as identification of critical areas are necessary in order to reduce the adverse effects of this problem through soil and water conservation measures. Due to lack of gauge stations in the outlets’ of small watersheds, empirical models are considered as the proper tools for estimating soil erosion and sedimentation. One of these methods which was developed based on Meditranean sediment data, is the Factorial Scoring Model (FSM). Towards this, the main objective of this research is to compare estimated vs. observed sediment yields in some selected regions of Iran. For this purpose, 58 small chatchments equipped by reservoirs in their outlets were selected in Semnan, Central, Isfahan, Chaharmahal and Bakhtiari, West Azarbaijan, Lorestan, Fars, Golestan and Khorasan Razavi Provinces. The comparison of the results showed that all estimates of the original model (before calibration) are several times bigger than observed values obtained by surveying (0.44 to 459.44 t.km-2.y-1). In order to correct the model coefficients for Iran, calibration was performed in two scenarios; once using all data and in the second secenario after allocating data into three geographic regions, which in both cases resulted in significant improvement of estimates. The Nash‐Sutcliffe Efficiency coefficient for all catchments was 0.39, and for Central, Zagros and Northeast regions were 51.1, 0.66 and 0.78, respectively. Based on obtained findings, the application of original FSM must be avoided in study areas. Instead, we strongly recommend using calibrated models based on observational values for different geographic regions.
Rahim Kazemi; Jahangir Porhemmat
Abstract
Understanding the base flow can be useful in river flow analysis, rainfall-runoff modeling, calibration of models, water resource management in low flow conditions and determination of the amount of water storage in watershed. In this research, 22 stations were selected with the appropriate data and ...
Read More
Understanding the base flow can be useful in river flow analysis, rainfall-runoff modeling, calibration of models, water resource management in low flow conditions and determination of the amount of water storage in watershed. In this research, 22 stations were selected with the appropriate data and common period of the years of 1982-2012. The trend of flow changes during the months of the year was determined for all hydrometric stations in the study area and the driest month was determined. Then, the calibration of six recursive digital algorithms was performed using the long-term data of the driest month of the year and after obtaining optimal parameters of the models, the base flow separation for the whole period was performed. The performance evaluation of the models was done using root mean square error. The results showed that the major part of the river flow in the study area was related to the base flow and the minimum, maximum and average annual base flow index for the whole period was equal to 0.48, 0.62 and 0.56, respectively, representing more than 50% of ground water contribution to stream flow of the studied watersheds. Results of the evaluation of the models using the root mean square error showed that the mean error in the research area for all the methods ranged from 0.025 to 0.044. The minimum was related to Lynie and Holick, and the maximum related to the One-parameter digital filter. Over all, conclusion of the results of the calibration process and investing the correlation between calculated and measured data showed that there was a correlation with a coefficient of explanation of more than 80%. Calibration method with dry season data in the absence of tracer-based methods is suggested as the most suitable method for calibrating digital separation filters in the study area.
Aboalhasan Fathbabadi; Mahnaz Kohneshin; Ali Heshmatpour; Masome Farasati
Abstract
During last decades hydrological models were extesively used in rainfall-runoff modeling. These models contain some constant parameters that must be optimized through appropraite mthods. In addation to model structur, the efficieny of hydrological models depend on these optimized parameters. In this ...
Read More
During last decades hydrological models were extesively used in rainfall-runoff modeling. These models contain some constant parameters that must be optimized through appropraite mthods. In addation to model structur, the efficieny of hydrological models depend on these optimized parameters. In this study, the efficiency of three automatic optimizing algorithms including Dynamically Dimensioned Search (DDS), Shuffled Complex Evolution and Genetic algorithms in calibration lumped hydrological model HyMod in Ghorchay Ramian Catchment were investigated. For these mehods, convergence speed and variability of final optimized values were investigated. Results showed that Genetic algorithm converged faster than two other methods. Following, DDS algorithm converged faster than Shuffled Complex Evolution algorithm. Shuffled Complex Evolution and Genetic algorithms took shorter and longer time per each epock, respectively. Highest and the least variability of final results were obtained for DDS and Shuffled Complex Evolution algorithms, respectively. With respect to final results variability, Shuffled Complex Evolution algorithm was more satable and had better performance than other methods. Using analyse variance and comper means in Shuffled Complex Evolution algorithms for complexes less than 12, the model performance was increased as the number of complexe increased. As alpha value increased, the model performance decreased and model had the best performance at the value of 0.58. Conversely, model performance was increase as beta values increasd and the best perfromnce was obtained for beta equal to 1. For Genetic algorithm, the best performance was obtained when the value of values crossover, mutation and chromosome number was equal to 0.2 and 0.3 and 16, respectively.
nasim arman; Ali Shahbazi; mohammad faraji; somaieh dehdari
Abstract
Water harvesting and surface runoff control systems are the important components of urban planning and development and ignoring these issues is likely to raise crisis. In order to decrease the urban flood damages, the urban runoff is needed to be evaluated correctly. Today some of models are developed ...
Read More
Water harvesting and surface runoff control systems are the important components of urban planning and development and ignoring these issues is likely to raise crisis. In order to decrease the urban flood damages, the urban runoff is needed to be evaluated correctly. Today some of models are developed for urban runoff simulation. One of the most important models in evaluating and management of urban runoff is SWMM. The aim of this study is to evaluate SWMM efficiency on urban runoff simulation in Izeh urban basin. To define design rain, concentration time was computed while considering the duration of the cloudburst as equal to this time. Also three performance indexes of Nash-Sutklif, errors sum of squares and Bias were used in order to model calibration and validation. Moreover, areas of high susceptibility were determined for two, five, 10, 20 and 50 years of return periods. Later, it was found that the principle reason of inundation is the lack of sufficient capacity of water ways. In some points, even with sufficient capacity, inundation occurs, confirmed by model. In these cases the causes stem from the improper design and construction of bridges which has lessen the size of water ways and caused junk clogging. Three rainfall events were recorded on March 13, 2017, March 28, 2017 and April 6, 2017 which were considered in order to calibrate and evaluate the model performance. Along with that, the discharge, depth and velocity of water at the outlet were considered as well. The results of the SWMM application gave indication of a good matchup between discharge, depth and the velocity of runoff for observed and estimated data. In this case, this model could be utilized to well predict the inundation hazard, design and the estimation of the cost and volume of drainage systems, management of watershed and prioritization of region to address flooding issues.
Hoshang Behravan
Abstract
In mountainous watersheds, snow melt can have a significant impact on the water balance and at certain times of the year it could have the most important contribution to runoff. In many parts of the world snow acts as a natural reservoir that can play an important role for water supply. Specifically, ...
Read More
In mountainous watersheds, snow melt can have a significant impact on the water balance and at certain times of the year it could have the most important contribution to runoff. In many parts of the world snow acts as a natural reservoir that can play an important role for water supply. Specifically, high-altitude basins in semi-arid regions with a significant snow component can have a large capability to balance and distribute scarce water resources. In this study the spatially distributed process-oriented hydrological model J2000g was used for the 431 km² large Latyan catchment in Iran. The target was to derive spatially distributed estimates of the quantity and timing of hydrological balance terms and state variables like rainfall, runoff, snow water equivalent (SWE) and snow melt. The model uses the distribution concept of Hydrological Response Units (HRU) to take the spatial variability in the basin into account. The comparison of the separate SWE models resulted in values between 0.28-0.68 for NSE and values between 0.53-0.83 for R². For the catchment models the comparison of the simulated runoff with measured data showed NSE values between 0.78 and 0.82. By these values it can be stated that the hydrological dynamics and the snow processes of the sub-basins within the Latyan catchment could be simulated sufficiently well with J2000g. It can be summarized that the single modules and in particular the snow components of J2000g along with the HRU distribution approach can be considered as suitable for the given project objectives i.e. the assessment of the hydrological dynamics of the Latyan catchment. Hereby, the model can be used to elaborate important hydrological information for a sustainable management of the water resources.
Payam Ebrahimi; Jamileh Salimi Kochi; Mohsen Mohseni Saravi
Abstract
SWAT is a continuous, physically based and distributed hydrologic model which all several hydrological processes like discharge, sediment yield nitrogen are simulated for each one of it. The purpose of this study is to test the efficiency of the Soil and Water Assessment Test (SWAT) and its applicability ...
Read More
SWAT is a continuous, physically based and distributed hydrologic model which all several hydrological processes like discharge, sediment yield nitrogen are simulated for each one of it. The purpose of this study is to test the efficiency of the Soil and Water Assessment Test (SWAT) and its applicability as a flow simulator, and using SWAT-Cup software and the SUFI2 algorithm as a means to calibrate and validate Neka Watershed in Mazandaran Province. Four indices were used to assess the goodness of calibration, viz., P-factor, R-factor, R2 and Nash-Sutcliffe (NS). Runoff data (1995-2004) of four hydrometery stations were used for calibration and (2005-2009) for validation of this watershed. The results of these values for flows at four stations for calibration process in Ablo, Pain zarandin, Karkhane siman and Sefid chah were 0.85, 0.78, 0.78, 0.89 for P-factor; 2.55, 2.03, 1.71, 2.43 for R-factor; 0.76, 0.62, 0.69, 0.71 for Nash-Sutcliffe and 0.71, 0.82, 0.76, 0.63 for R2. The results of validation were 0.87, 0.88, 0.72, 0.72 for P-factor; 3.61, 2.24, 3.56, 1.78 for R-factor; 0.74, 0.66, 0.58, 0.64 for Nash-Sutcliffe (NS) and 0.81, 0.68, 0.73, 0.61 for R2 respectively. In general, the results showed that SWAT could be a proper tool for simulating the flow rate values of the Neka Watershed.
Marzieh Hajimohammadi; Abolfazl Azizian; Bagher Ghermezcheshmeh
Abstract
Knowledge of climate variabilities and their behavior in future periods and their effects in various fields has great importance especially in strategic and macro planning in water resources. This study aims to evaluate the effect of climate change on hydrological condition of the Kan Watershed. For ...
Read More
Knowledge of climate variabilities and their behavior in future periods and their effects in various fields has great importance especially in strategic and macro planning in water resources. This study aims to evaluate the effect of climate change on hydrological condition of the Kan Watershed. For this purpose the HadCM3model under the A2 emission scenario and also statistical downscaling model (SDSM) were applied for temperature and rain variables simulation. Then, SWAT model was used for monthly runoff simulation and SUFI-2 algorithm was used in SWAT-CUP software pack for calibrating and uncertainly analyzing. The performance of SDSM model was evaluated base on MBE and NRMSE parameters, the result indicated that temperature variable was simulated more accurate than of precipitation. The result of the predicting temperature in period (2011-2040) compare with the base period (1961-2001) showed the maximum and minimum temperature will increase by 1.3 and 0.8 °C, respectively. Also, the rainfall will decrease by 3-4 percent for all of selected stations. The most rainfall reduction will be for spring. While in some months of winter an increase of precipitation was predicted. The result of calibration and validation of SWAT model agreed well with the observed data, so that Nash-Sutcliff efficiency coefficient, as objective function, was 0.82 and 0.71, in calibration period (1983-1991) and validation period (1992-1996) respectively. Finally, results of runoff prediction showed an increase in winter and a decrease in other seasons based on climate scenarios. Overall, according to obtained results runoff will decrease by seven percent for future period.
Rahim Kazemi; Amir Safari; Amir Karam; Jahangir Porhemat
Abstract
Estimation of Base Flow Index (BFI), has always been one of the most important issues in hydrology. Base flow separation process, often is done using daily stream flow data. Lack of full coverage of daily data for the whole country, may lead to some errors for estimating the base flow and its index. ...
Read More
Estimation of Base Flow Index (BFI), has always been one of the most important issues in hydrology. Base flow separation process, often is done using daily stream flow data. Lack of full coverage of daily data for the whole country, may lead to some errors for estimating the base flow and its index. In this research, base flow index in some catchments of Karkhe basin based on daily and monthly stream flow data and using six algorithms were extracted. Three based on recursive digital filter (One-parameter, two parameter and Lynie & Hollick algorithms), and the others on simple smoothing including, Local minimum, Fixed interval and sliding interval methods. Results were analyzed using different statistical methods such as standard deviation, mean absolute error, relative error and other descriptive methods. The results showed that minimum relative error of monthly data compared to daily data was related to two parameter algorithm with 2.45% and the maximum was related to Lynie & Hollick algorithm with 84.19 percent .In overall conclusion, two parameter algorithms because of low relative error, minimum Mean Absolute and Root Mean Square Error was recognized as a suitable method for the extraction of base flow using monthly data in the absence of appropriate data daily.
Akbar Hassani; Hosseinali Bahrami; Aliakbar Noroozi; Shahin Oustan
Abstract
Reflectance spectroscopy can be used to study agricultural and environmental aspect of soil that are sensitive to soil organic and inorganic compounds. Despite the extensive studies in the field of visible-near infrared reflectance spectroscopy, there are rare researches in gypseous and calcareous soils. ...
Read More
Reflectance spectroscopy can be used to study agricultural and environmental aspect of soil that are sensitive to soil organic and inorganic compounds. Despite the extensive studies in the field of visible-near infrared reflectance spectroscopy, there are rare researches in gypseous and calcareous soils. The objective of this study was to obtain a model that can predict chemical properties of gypseous soils via reflectance spectroscopy methods. Soil samples were collected from 102 locations in five different provinces in 0-30 cm of depth. Some chemical properties of soils, such amount of gypsum, equivalent calcite, cation exchange capacity, pH, EC, exchangeable calcium, magnesium, sodium and potassium and amount of silt, clay and sand were measured by standard methods in the laboratory. Air-dried soil samples were scanned at one nm resolution from 350 to 2500 nm, and calibrations between properties and reflectance spectra were developed using cross-validation under Partial Least Squares Regression (PLSR) and Boosted Regression Trees (BRT). Raw reflectance and first derivative reflectance data were used separately and combined for all samples in the data set. Data were additionally divided into two random subsets of 70 and 30 percent of the full data, which were each used for calibration and validation. Strongest correlations were obtained with gypsum, equivalent calcium carbonate, cation exchange capacity, exchangeable Ca and Mg, organic matter, sand and clay contents. Overall, BRT provided better predictions when under cross-validation. However, PLSR and BRT results were comparable in terms of prediction accuracy when using separate data sets for calibration and validation. In conclusion, VNIR spectroscopy was variably successful in estimating soil properties and showed its potential for substituting laboratory analyses or providing inexpensive co-variable data in environmental studies.
Foad Naserabadi; Abazar Esmali Ouri; Hossein Akbari; Rokhsareh Rostamian
Abstract
In hydrologic models, a sensitivity analysis could be performed to identify important parameters and reduce their numbers in an easier and faster calibration process. This study describes an application of a useful method for sensitivity analysis of Soil and Water Assessment Tool (SWAT) model. This method ...
Read More
In hydrologic models, a sensitivity analysis could be performed to identify important parameters and reduce their numbers in an easier and faster calibration process. This study describes an application of a useful method for sensitivity analysis of Soil and Water Assessment Tool (SWAT) model. This method determines parameters with the greatest impact on model results and gives a sensitivity ratio and an input-output relationship graph for each parameter. In this study, the SWAT model is used for a monthly rainfall-runoff simulation in Ghareh Su watershed with an area of 4062 km2. Calibration of the model were performed manually, using the recorded weather data from 2003 to 2008 and the validation was performed, using the recorded data from 2009 and 2010. The sensitivity analysis has been performed using One Factor At a Time (OAT) method to evaluate and demonstrate the influences of the model parameters on four major components of water balance, including surface runoff, lateral flow, groundwater and evapotranspiration. The results showed that the parameters of initial SCS runoff curve number for moisture condition Π (CN2), soil available water capacity (SOL_AWC), soil bulk density (SOL_BD), saturated hydraulic conductivity (SOL_K), maximum canopy storage (CANMX), soil evaporation compensation factor (ESCO) and minimum melt rate for snow during the year (SMFMN) have the greatest influence on remarked parts of water balance.
Azade Jabary; Majid Hosseini; Amir Khosrojerdi
Abstract
The present research is looking for determining and evaluation of water balance in Sanjabi Watershed located in Kermanshah Province due to a variety of problems raised by the population increase. For this purpose, the SWAT model parameters have been calibrated by applying required information layers ...
Read More
The present research is looking for determining and evaluation of water balance in Sanjabi Watershed located in Kermanshah Province due to a variety of problems raised by the population increase. For this purpose, the SWAT model parameters have been calibrated by applying required information layers in a period extending from 1995 to 2004 and their sensitivity were analyzed by SUFI2 program. The flow parameters’ sensitivity was computed by P-Value and t-State in SUFI2. The most sensitive parameters were calibrated and thereafter model was validated for a period extending from 2005 to 2007. The model evaluation results showed that the value of NS (Nash Sutcliff), R2 (Coefficient of determination), Mean Absolute Relative Error (MARE) and Percent Difference (D) for calibration period were 0.78, 0.78, 0.2 and -0.2, respectively and for validation period were 0.74, 0.86, 0.3 and 0.01, respectively. Results showed the desired accuracy of SWAT for runoff simulation. Discharges for calibration and validation periods were generally under and overestimated, respectively. The water balance values are calculated as the amount of water available in the soil. So that the water balance resulted as negative in years 2005 (normal year), 2006 (wet year) and 2007 (dry year). Nevertheless, in these three years, autumn and winter monthly water balances were positive while in spring and summer were negative. The results also showed that most water losses were due to evapotranspiration in this watershed.
Hossein Saadati; Shaban Ali Gholami; Foroud Sharifi
Volume 1, Issue 2 , July 2009, , Pages 87-98
Abstract
Lack of hydrometric data is the main issue for research and executive plans. In the present study in order to solve this issue, the capability of SWAT model has been studied in terms of daily discharge simulation in un-gauged watersheds. This model can simulate six hydrological and biological processes ...
Read More
Lack of hydrometric data is the main issue for research and executive plans. In the present study in order to solve this issue, the capability of SWAT model has been studied in terms of daily discharge simulation in un-gauged watersheds. This model can simulate six hydrological and biological processes such as daily discharge, daily sediment, yearly and monthly water balance, water quality, agricultural products and estimation of vegetative cover for different animal grazing management systems. Simulation has been done for both gauged and un-gauged basins in Kasilian catchment for two time periods as calibration and validation of model. Calibration has been done using data derived from 4 years recording data (1987-1981) while validation has been done using 6 years data (1981-1986). Results show that, ABF, CN2 and REVAPC parameters are most sensitive and also Z and T-Laps parameters are low sensitive among other parameters. The model simulated runoff reasonable with r2=0.69, PBIAS=-2.5, and NS=0.72, whereas, the validation of model for the same data showed that r2=0.57, PBIAS=4.3, and NS=-0.54, which indicates that the performance this model is not quite valid for low stream flows because of sensitivity of hydrological parameters. The new version of SWAT model, modified during 2000 to 2005, has a capability in using GIS, thus it can be used in the same study area.
