Yousef Hasheminejhad; Mehdi Homaee; Ali Akbar Noroozi
Abstract
Introduction: Soil salinization is increasing across developing world countries and agricultural production is decreasing as a result of this stress. Climate change could adversely affect soil salinization trend through the decrease in rainfall and increased evapotranspiration in arid regions. Policy ...
Read More
Introduction: Soil salinization is increasing across developing world countries and agricultural production is decreasing as a result of this stress. Climate change could adversely affect soil salinization trend through the decrease in rainfall and increased evapotranspiration in arid regions. Policy and decision makers require continuous and quantitative monitoring of soil salinity to adapt with the adverse effects of climate change and increasing need for food. Indices derived from near surface or satellite based sensors are increasingly applied for monitoring of soil salinity so a considerable number of these indices are introduced already for soil salinity monitoring. Different regression methods have been already used for modeling and verification of developed models amongst them multiple linear regression (including stepwise, forward selection and backward elimination) and partial least square regression are the most important methods.
Materials and Methods: To evaluate different approaches for modeling soil salinity against remotely sensed data, an area of about 50000 ha was selected in Sabzevar- Davarzan plain during 2013 and 2014 years. The locations of sampling points were determined using Latin Hypercube Sampling (LHS) strategy. Sampling density was 97 points for 2013 and 25 points for 2014. All points were sampled down to 90 cm depth in 30 cm increments. Totally 366 soil samples were analyzed in the laboratory for electrical conductivity of saturated extract. Electromagnetic induction device (EM38) was also used to measure bulk soil electrical conductivity for the sampling points at the first year and sampling points and 8 points around it at the second year. Totally 97 and 225 EM measurements were also recorded for first and second years respectively. Mean measured soil EC data were calibrated against the EM measurements. Finding the fair correlations, the EM and EC data could be converted to each other. 23 spectral indices derived from Landsat 8 images in the sampling dates along with DEM were used as independent variables. Multiple Linear Regression (MLR) and Partial Least Square Regression (PLSR) methods were evaluated for their fitness in predicting soil salinity from independent variables in different calibration and verification datasets.
Results and Discussion: Different multiple linear regression approaches using the first year data for training and second year data for testing the models and vice versa were evaluated which produced determination coefficients of about 22 to 88 percent in the training dataset but this regression did not reach to 29 percent in the test dataset. Due to the multiple co-linearity amongst the independent variables the multiple linear regression methods were not applicable to all variables. Excluding the co-linear variables, log- transforming and randomizing them into train and test datasets improved the determination coefficient of model and its validation at an acceptable level. Application of partial least square regression using the original and log- transformed data of first and second years as train and test datasets and vice versa introduced determination coefficients of about 39 to 85 percent in the training dataset but were not able to predict in the test dataset. Random dividing of all data into train and test datasets considerably increased the determination coefficient in the verification dataset. Repeating the randomization showed that the approach has the required consistency for predicting the coefficients of variables.
Conclusions: Wide range of independent variable could be used for predicting soil salinity from remotely sensed data and indices. On the other hand the independent variables generally show multi-colinearity amongst themselves. Correlation matrix, variance inflation factor and tolerance indices could be used to identify multi-colinearity. Removing or scaling the variable with high colinearity could improve the regression. Different data transformation methods including log- transformation could also significantly improve the strength of regression. In this research EM data showed more significant correlations with spectral indices in comparison with laboratorial measured EC data. As the EM38 device measures the reflectance in special range of spectrum this higher correlation could be expected. Such models should be calibrated and verified against ground truth data. Generally a part of data set is used for calibrating (making the model) and the remained for verifying (testing the model). Random dividing of the total data of 2 years into calibration (2/3 of data) and verification (1/3 of data) could significantly improve the regression in the verification data set. This procedure increases the range of variability for data used for calibration and verification and prevents outlier predictions.
Hossein Kheirfam; Mehdi Homaee; Seyed Hamidreza Sadeghi; Behrouz Zarei Darki
Abstract
Introduction: Land degradation and soil losses are common and universal problems which is a pose threat to food security, ecosystem health and consequently sustainable development and human well-being. Meanwhile, improving the chemical and physical properties of biological soil crusts is an effective ...
Read More
Introduction: Land degradation and soil losses are common and universal problems which is a pose threat to food security, ecosystem health and consequently sustainable development and human well-being. Meanwhile, improving the chemical and physical properties of biological soil crusts is an effective factor in soil loss controlling. Also, the chemical properties specially soil nitrogen are the important factors for soil quality determination. To this end, various strategies on techniques of amendments have been implemented to improve soil properties and quality. Although the application of most strategieshave been verified to soil quality,but their application in real conditions is restricted due to detrimental environmental effects, instability, cost and time-consuming and less accessibility. Recently, biological soil crusts enrichment based on soil microorganism inoculation and stimulation has been raised as a biological and useful strategy in soil conservation sciences. Accordingly, the present study aimed to investigate the role of individual and combined inoculation of bacteria and stimulant nutrient material into small-scale plots on soil nitrogen variation as one of the important soil chemical component.
Material and Methods: The study soil was collected from the erosion-prone and poor biological crust of a sub-watershed from Chalusrood watershed located in Mazandaran Province. The soil sampling was carried out from the upper of the soil surface using a 5cm-diameter coring polyvinyl chloride. The sampled soils were air-dried and sieved by a 2 mm-sized mesh. The Nutrient Agar and Tryptic Soy Agar general were used to bacteria isolation. The identification of isolated bacteria was carried out based on available protocols. Effective nitrogen-fixing bacteria were selected and then purified by selective Azotobacter Agar, Modified II and DSMZ1media. The purified bacteria proliferated by LB Broth medium and then inoculated into soil small sized-plots simultaneously with stimulant nutrient material throught spraying technique. The study was conducted at plot scale with 0.5×0.05×0.5 m dimensions and the plots filled by study soil based on standard protocols. The soil samples were taken at once the 7-8 days from surface of soil plots and the amounts of soil nitrogen were measured by using Kjeldahl method. As well as, experiment period was planned about 60 days. The one-way ANOVA and Tukey HSD test were subjected to statistically analyses.
Results and discussion: The results indicated that the Azotobacter sp. and Bacillussubtilis strain were selected as the most appropriate bacteria to be applied for nitrogen fixing in soil. Also, the results showed that the average total organic nitrogen in control plots ranged from 0.082 to 0.136%, which implies the soil limitation of total nitrogen. However, the measured total organic nitrogen in the bacteria, stimulant nutrient, and combined inoculation plots varied from 0.11 to 0.241%, 0.117 to 0.204%, and 0.124 to 0.374%, respectively. These results demonstrated the positive role of inoculated treatments on fixing nitrogen in the soil. Therefore, the population of Azotobacter sp., the Bacillussubtilis strain, was considerably increased after the inoculation process, and this led to converted and fixed atmospheric nitrogen (N2) into utilizable nitrogen (NH4 or NO3) in soil by using the enzyme nitrogenase as a catalyst. The statistical analyses and evaluation results were indicative of a significant (p
Yousef Hasheminejhad; Mahdi Homaee; Ali Akbar Noroozi
Abstract
Introduction: Monitoring and management of saline soils depends on exact and updatable measurements of soil electrical conductivity. Large scale direct measurements are not only expensive but also time consuming. Therefore application of near ground surface sensors could be considered as acceptable time- ...
Read More
Introduction: Monitoring and management of saline soils depends on exact and updatable measurements of soil electrical conductivity. Large scale direct measurements are not only expensive but also time consuming. Therefore application of near ground surface sensors could be considered as acceptable time- and cost-saving methods with high accuracy in soil salinity detection. . One of these relatively innovative methods is electromagnetic induction technique. Apparent soil electrical conductivity measurement by electromagnetic induction technique is affected by several key properties of soils including soil moisture and clay content.
Materials and Methods: Soil salinity and apparent soil electrical conductivity data of two years of 50000 ha area in Sabzevar- Davarzan plain were used to evaluate the sensitivity of electromagnetic induction to soil moisture and clay content. Locations of the sampling points were determined by the Latin Hypercube Sampling strategy, based on 100 sampling points were selected for the first year and 25 sampling points for the second year. Regarding to difficulties in finding and sampling the points 97 sampling points were found in the area for the first year out of which 82 points were sampled down to 90 cm depth in 30 cm intervals and all of them were measured with electromagnetic induction device at horizontal orientation. The first year data were used for training the model which included 82 points measurement of bulk conductivity and laboratory determination of electrical conductivity of saturated extract, soil texture and moisture content in soil samples. On the other hand, the second year data which were used for testing the model integrated by 25 sampling points and 9 bulk conductivity measurements around each point. Electrical conductivity of saturated extract was just measured as the only parameter in the laboratory for the second year samples.
Results and Discussion: Results of the first year showed a significant correlation between electrical conductivity and apparent conductivity with a regression coefficient of 0.78. Although multiple linear regression by inclusion of soil moisture and clay content as independent variables improved the regression coefficient to 0.80 but the effect of clay content was not significant in this multiple model. Sensitivity analysis by maintaining one variable at its average value and changing the second variable also showed greater sensitivity of the model to soil moisture in comparison with soil clay content. Generally under estimation of soil moisture and over estimation of soil clay content produced about 63 to 65 percent Mean Bias Error (MBE) while over estimation of soil moisture and under estimation of soil clay content produced about 35- 37 percent of MBE. So the model is quite sensitive to both parameters and they cannot be estimated in the field by feeling and the other field methods. Simple linear regression model between ECe and EMh was tested on the second year because the errors in estimating soil moisture could be imposed a significant error on estimating soil salinity. Once the model was tested for estimation of soil salinity in the central point based on EMh reading at the center and then it was tested for estimation of soil salinity based on the average EMh of 9 points in each location. Results showed that the correlation between soil salinity and apparent soil electrical conductivity could be improved to 0.98 using the average of 9 measurements instead of 1 measurement.
Conclusion: Based on the results the electromagnetic induction device is sensitive to soil moisture. Although its sensitivity to clay content is less than the sensitivity to moisture content, but the total model error as a result of over estimating soil moisture is about equal to its error resulted from under estimating clay content and vice versa. So the field and feeling methods could not be implemented as inputs for the multiple regression models but these methods have enough accuracy to divide soil samples into two groups of dry and wet soils or sandy or clayey soils, on the other hand measurements of these parameters imposes more cost and time to soil salinity surveys. Results also showed that the repeated EM measurements around each sampling point could improve the strength of the regression. Therefore regarding to the sensitivity of the technique to soil moisture three methods are suggested to improve accuracy of calibration: a)- measurement and calibration under the same moisture conditions; b)- field approximation of soil moisture and dividing soil samples into two groups of dry and moist soils and deriving two different groups of calibration equations.
M. Sarai Tabrizi; H. Babazadeh; mahdi homaee; F. Kaveh Kaveh; M. Parsinejad
Abstract
Introduction: Several mathematical models are being used for assessing the plant response to the salinity of the root zone. The salinity of the soil and water resources is a major challenge for agricultural sector in Iran. Several mathematical models have been developed for plant responses to the salinity ...
Read More
Introduction: Several mathematical models are being used for assessing the plant response to the salinity of the root zone. The salinity of the soil and water resources is a major challenge for agricultural sector in Iran. Several mathematical models have been developed for plant responses to the salinity stress. However, these models are often applicable in particular conditions. The objectives of this study were to evaluate the threshold value of Basil yield reduction, modeling Basil response to salinity and to evaluate the effectiveness of available mathematical models for the yield estimation of the Basil .
Materials and Methods: The extensive experiments were conducted with 13 natural saline water treatments including 1.2, 1.8, 2, 2.2, 2.5, 2.8, 3, 3.5, 4, 5, 6, 8, and 10 dSm-1. Water salinity treatments were prepared by mixing Shoor River water with fresh water. In order to quantify the salinity effect on Basil yield, seven mathematical models including Maas and Hoffman (1977), van Genuchten and Hoffman (1984), Dirksen and Augustijn (1988), and Homaee et al., (2002) were used. One of the relatively recent methods for soil water content measurements is theta probes instrument. Theta probes instrument consists of four probes with 60 mm long and 3 mm diameter, a water proof container (probe structure), and a cable that links input and output signals to the data logger display. The advantages that have been attributed to this method are high precision and direct and rapid measurements in the field and greenhouse. The range of measurements is not limited like tensiometer and is from saturation to wilting point. In this study, Theta probes instrument was calibrated by weighing method for exact irrigation scheduling. Relative transpiration was calculated using daily soil water content changes. A coarse sand layer with 2 centimeters thick was used to decrease evaporation from the surface soil of the pots. Quantity comparison of the used models was done by calculating statistical indices such as maximum error (ME), normalized root mean square error (nRMSE), modeling efficiency (EF), and coefficient of residual mass (CRM). At the end of the experiment, dry matter yield at the different treatments was measured and relative yield was calculated by dividing dry matter yield of treatments on dry matter yield at no stress treatment (control treatment). Leaching requirement in experimental treatments was calculated by Ayarset al., (2012) equation.
Results and Discussion: The results indicated that Basil threshold value based on soil salinity was 2.25
dSm-1 with the yield reduction of 7.2% per dSm-1. The mathematical model of van Genuchten and Hoffman (1984) had a higher precision than other models in simulating Basil yield reduction function based on saturated soil extract salinity. The overall observations revealed that van Genuchten and Hoffman (1984), Steppuhnet al., (2005) and Homaeeet al., (2002) models were accurate for simulating Basil root water uptake and yield response to saturated soil extract salinity. Considering the presented results, it seems that among math-empirical models for salinity stress conditions, model of van Genuchten and Hoffman (1984) is more accurate than Maas and Hoffman (1977), Dirksen and Augustijn (1988) and Homaeeet al., (2002a) models. The works of Green et al., (2006) and Skaggs et al., (2006) came to the same conclusion. Our work indicated that mostly statistical models have lower precision than math-empirical models. Steppuhn et al., (2005a) reported that statistical models had the higher accuracy than math-empirical model of Maas and Hoffman (1977) and among statistical models, the modified Weibull model had the best fit on measured data which is in good agreement with the results of this study.
Conclusion: The goals of this research were to evaluate Basil response to saturated soil extract salinity, to estimate threshold value of Basil crop coefficients, to obtain yield reduction gradient, and also to investigate efficiency of available math-empirical models in estimating reduction functions. The results of this study indicated that the Basil threshold value obtained based on saturated soil extract salinity was 2.25 dSm-1 and the gradient of yield reduction was 7.2% per dSm-1 according to Maas and Hoffman (1977) linear fitting. The reached general conclusion was that among the math-empirical reduction functions, the model of van Genuchten and Hoffman (1984) had the highest accuracy when compared to the models of Maas and Hoffman (1977), Dirksen and Augustijn (1988) and Homaee et al., (2002a). Therefore, it is recommended to use the van Genuchten and Hoffman (1984), Steppuhn et al., (2005), and Homaee et al., (2002) models respectively, instead of the other models in this research.
E. Babaeian; M. Homaee; R. Rahnemaie
Abstract
Phytoextraction is a remediation technology for contaminated soils with lead (Pb). The application of chelating agents can be resulted in high efficiency in this method. In current study, the effect of synthetic and natural chelates applicationon efficiency of lead phytoextraction from soil by carrot ...
Read More
Phytoextraction is a remediation technology for contaminated soils with lead (Pb). The application of chelating agents can be resulted in high efficiency in this method. In current study, the effect of synthetic and natural chelates applicationon efficiency of lead phytoextraction from soil by carrot was investigated. The experiment factors were 1) six levels of Pb (0, 100, 200, 300, 500 and 800 mg Pb kg-1 soil, added as Pb(NO3 )2, 2) chelates (EDTA, NTA and oxalic acid, and 3) chelate concentration (0, 2.5, 5 and 10 mmol kg-1 soil). The results indicated that EDTA effectively increased the Pb content in soil solution. At the highest applied rate (10 mmol EDTA kg-1), it resulted in 463-fold increase in extractable Pb, compared to the control treatment. Pb content in the shoot and taproot increased with the chelates application rates.The highest Pb content in the shoot (342.2±13.9 mg kg-1) and root (310 ±15.5 mg kg-1) occurred in 10 mmol kg-1 EDTA when Pb level was 800 mg kg-1. Pbphytoextraction potential increased with increasing thechelate and Pb concentration. Maximum Pb extraction from soil (1208±26.6 g ha-1 yr-1) during growth season occurred in 10 mmol kg-1 EDTA, when soil Pb level was 800 mg kg-1. It may be concluded that carrot can take up high amount of Pb and concentrate it in its roots and shoots. Thus, it can be introduced as a lead accumulator to phytoextractPb from contaminated soils.
V.R. Jalali; M. Homaee
Abstract
Abstract
Saturated hydraulic conductivity (Ks) is needed for many studies related to water and solute transport, but often cannot be measured because of practical and/or cost-related reasons. Nonparametric approaches are being used in various fields to estimate continuous variables. One type of the ...
Read More
Abstract
Saturated hydraulic conductivity (Ks) is needed for many studies related to water and solute transport, but often cannot be measured because of practical and/or cost-related reasons. Nonparametric approaches are being used in various fields to estimate continuous variables. One type of the nonparametric lazy learning algorithms, a k-nearest neighbor (k-NN) algorithm, was introduced and tested to estimate saturated hydraulic conductivity (Ks) from other soil properties including soil textural fractions, EC, pH, SP, OC, TNV, ρs and ρb. A number of 10 nearest neighbors, based on Cross Validation technique were selected to perform saturated hydraulic conductivity prediction from 151 soil sample attributes. The nonparametric k-NN technique performed mostly equally well, in terms of Pearson correlation coefficient (r=0.801), modeling efficiency (EF=0.65), root-mean-squared errors (RMSE=71.15) maximum error (ME=120.47), coefficient of determination (CD=1.32) and coefficient of residual mass (CRM=-0.046) statistics. It can be concluded that the k-NN technique is a competitive alternative to other techniques such as pedotransfer functions (PTFs) to estimate saturated hydraulic conductivity.
Keywords: k-nearest neighbor (k-NN), Modeling, Saturated hydraulic conductivity
M. Davari; M. Homaee
Abstract
Abstract
Soil Contamination by heavy metals is yet one of the most important environmental concerns. Among heavy metals, Nickel and Cadmium have dangerous influences on human, animals and plants. The objective of this study was to derive a new model for simultaneous phytoextraction of Ni and Cd from ...
Read More
Abstract
Soil Contamination by heavy metals is yet one of the most important environmental concerns. Among heavy metals, Nickel and Cadmium have dangerous influences on human, animals and plants. The objective of this study was to derive a new model for simultaneous phytoextraction of Ni and Cd from contaminated soils. Consequently, a macroscopic model was derived by combining yield reduction functions and relative concentrations of Ni and Cd in plant tissues. To verify the derived model, a clay loam soil was simultaneously contaminated with different concentrations of Ni and Cd. The Ornamental Kale seeds were then seeded in these packed contaminated soils in three replicates. Plants were harvested after full development. The Ni and Cd contents of soil samples and plant materials were extracted by 4M HNO3 oxidation and wet oxidation methods, respectively. The Ni and Cd concentrations were measured by Atomic Absorption Spectrometer (Shimadzu, AA 670-G) and Inductively Coupled Plasma Optical Emission Spectrometry (Varian Vista-PRO). The results indicated that relative yield of Ornamental Kale in the contaminated soils with both Ni and Cd was reduced more than the soil polluted with separate Cd or Ni. The results also indicated that at any given soil Cd concentration, the Ni content of Ornamental Kale increases with increasing soil Ni concentration. Meanwhile, with increasing soil Cd, the Ni content in Ornamental Kale was decreased. Further, at any given Cd content, the amount of Cd in Ornamental Kale was increased by increasing Ni concentration in soil. The results further indicated that the proposed model can well predict Ni phytoextraction from soils contaminated with both Ni and Cd. However, this model could only provide an overall estimate for Cd phytoextraction. It was further concluded that Ornamental Kale due to its high biomass production and high tolerance to Ni and Cd concentrations can be used to remediate low to moderate combined Ni -Cd contaminated soils.
Keywords: Cadmium, Multiplicative theory, Nickel, Ornamental kale, Phytoextraction
A. Farrokhian Firouzi; M. Homaee; E. Klumpp; R. Kasteel; M. Sattari
Abstract
Abstract
Microbial contaminants transport to groundwater is a serious environmental problem that can result in large outbreaks of waterborne diseases. Some of bacteria can travel from vadose zone and cause contamination of groundwater resources. Thus, an accurate prediction of transport and fate of ...
Read More
Abstract
Microbial contaminants transport to groundwater is a serious environmental problem that can result in large outbreaks of waterborne diseases. Some of bacteria can travel from vadose zone and cause contamination of groundwater resources. Thus, an accurate prediction of transport and fate of pathogenic microorganisms in unsaturated soil is needed to protect groundwater resources. The main objectives of this research were quantitative study of bacterial transport and deposition under unsaturated conditions in calcareous soils. A series of column leaching experiment with well-controlled suction and flow rate was conducted. Breakthrough curves (BTCs) of Pseudomonas fluorescens and Cl were measured. After the leaching experiment the bacteria was measured in difference layers of the soil columns. The HYDRUS-1D kinetic attachment-detachment model (AD) was used to evaluate the transport and deposition of bacteria in soil columns. The breakthrough curves in soils were described well by attachment-detachment model. Whereas the model fit underestimate the amount of bacteria retention in the soil columns. The detachment rate was less than 0.001 of the attachment rate, indicating irreversible attachment of bacteria. Most of the cells were retained close to the soil column inlet, and the rate of deposition decreased with depth. Microbial reduction rate for the soil was 10.18-13.34 log m-1. High reduction rate of bacteria was attributed to soil calcium carbonate that has favorable attachment site for bacteria.
Keywords: Bacteria transport, Pseudomonas fluorescens, Calcareous soil, Bacteria Attachment, Unsaturated flow, Bacteria Detachment
E. Babaeian; M. Homaee
Abstract
Abstract
Enhancing phytoextraction with aminopolycarboxylic acids (APCAs) associated with fast growing and metal tolerant plants species has been proposed for the clean-up of heavy metal contaminated soils. The objectives of this study were to assess the efficiency of EDTA and NTA for desorbing Pb from ...
Read More
Abstract
Enhancing phytoextraction with aminopolycarboxylic acids (APCAs) associated with fast growing and metal tolerant plants species has been proposed for the clean-up of heavy metal contaminated soils. The objectives of this study were to assess the efficiency of EDTA and NTA for desorbing Pb from soil and to compare their effects for enhancing of Pb extraction with Land Cress (Barbara verna). The experimental factors were including 0, 100 and 800 mg Pb kg-1 soil, EDTA and NTA (0, 5 and 10 mmol kg-1 soil). The results indicated that EDTA was much more efficient for enhancing root to shoot Pb translocation. In 800 mg Pb kg-1 soil, as a result of 10 mmol EDTA kg-1 soil, a value of 1075 mg Pb kg-1 DW shoot was obtained. The soils treated with EDTA showed higher values of soluble Pb concentration than NTA and no chelate. Also, MLPI was higher (0.87) in presence of 5 mm EDTA kg-1 concentration. In high concentrations of Pb and APCAs, both EDTA and NTA caused acute symptoms on leaves which showed wilting, necrotic areas and curling of borders. Finally, Land Cress due to high lead resistance can be introduced as a Pb hyperaccumulator to chelate-induced phytoextraction technology.
Keywords: Soil Contamination, Phytoextraction, Land Cress (Barbara Verna), Lead, Aminopolycarboxylic Acids
Sh. Ghorbani; M. Homaee; M.H. Mahdian
Abstract
Abstract
Infiltration is a significant process which controls the fate of water in the hydrologic cycle. The direct measurement of infiltration is time consuming, expensive and often impractical because of the large spatial and temporal variability. Artificial Neural Networks (ANNs) are used as an ...
Read More
Abstract
Infiltration is a significant process which controls the fate of water in the hydrologic cycle. The direct measurement of infiltration is time consuming, expensive and often impractical because of the large spatial and temporal variability. Artificial Neural Networks (ANNs) are used as an indirect method to predict the hydrological processes. The objective of this study was to develop and verify some ANNs to predict the infiltration process. For this purpose, 123 double ring infiltration data were collected from different sites of Iran. The parameters of some infiltration models were then obtained; using sum squares error optimization method. Basic soil properties of the two upper pedogenic layers such as initial water content, bulk density, particle-size distributions, organic carbon, gravel content, CaCO3 percent and soil water contents at field capacity and permanent wilting point were obtained for each sampling point. The feedforward multilayer perceptron was used for predicting the infiltration parameters. Two ANNs types were developed to estimate infiltration parameters. The developed ANNs were categorized into two groups; type 1 and type 2 ANNs. For developing type 1 ANNs, the basic soil properties of the first upper soil horizon were used as inputs, hierarchically. While for developing type 2 ANNs the basic soil properties of the two upper soil horizons were used as inputs, using principal component analysis technique. Evaluation results of these two types ANNs showed the better performance of type 1 ANNs in predicting the infiltration parameters. Therefore, this type of ANNs was used for predicting the cumulative infiltration. The reliability test indicated that the developed ANNs for Philip model have the best performance to predict cumulative infiltration with a mean RMSE of 6.644 cm. The developed ANNs for Horton, Kostiakov-Lewis and Kostiakov have the next best ranks, respectively.
Keywords: Multilayer Perceptron, Artificial Neural Networks, Infiltration Models, Soil Infiltration
P. Fathi; Y. Mohammadi; M. Homaee
Abstract
Abstract
Prediction of input flow into water resources is regarded as one of the most important issues in optimum planning and management in producing electro-water energy and optimum allocation of water into different consumption sources. Different parameters affect on input discharge into dams. Climate ...
Read More
Abstract
Prediction of input flow into water resources is regarded as one of the most important issues in optimum planning and management in producing electro-water energy and optimum allocation of water into different consumption sources. Different parameters affect on input discharge into dams. Climate variables including temperature and rainfall have the most effect on input runoff rate to water resource in dry and semi-dry regions like Iran. A suitable monthly runoff-rainfall model is a strong tool to consider the climate changes effect on accessibility of water to produce electro-water energy. The investigations have shown that the relation between runoff rate and effective variables is non-linear and complicated. Artificial Neural Networks due to their unique properties have a tremendous capability in non-linear relations simulation. Artificial Neural Networks establish a great change in analyzing dynamic systems behavior in different water-science engineering. In this paper it has been attempted to design static network to recover the non-linear relations between dependant and independent variables, so that the intelligent discharge estimation of average monthly input to Vahdat dam can be done by its help. In addition, by designing and extension of dynamic neural network model based on times series performance, the amount of the monthly input discharge to the dam was predicted. Considering the capability of Artificial Neural Networks, these networks were used for modeling the rivers monthly discharge non-linear time series. Analysis of time series having two major goals; random mechanism understanding or modeling and future series value prediction was done base on previous ones. Also, the performance of the designed models was evaluated by comparing results of the static and dynamic neural network. The results of the investigation showed that there is a good conformity between the predicted values given by combined neural network and observed data. Furthermore, the results showed that the time series dynamic neural network model predict the monthly discharge more accurate than static model.
Keywords: monthly average discharge, Artificial Neural Networks, time series
