multivariate regression

The geomorphic characteristics of the watersheds are interrelated and the temporal and spatial scale in the form of season and sub-basins affect the concentration of suspended sediment. One of the objectives of this study was to investigate the relationship between suspended sediment concentration and watershed characteristics of Kan River using principal components regression and to recognize the effect of seasons and sub-basins on sediment concentration. The concentration of suspended sediment during four rainfall-runoff events in three seasons and in sub-basins was measured and calculated. The sixteen physiographic and land use characteristics were determined in the sub-basins and the main factors were identified and the scores of each factor for each feature were calculated using principal component analysis (PCA). The results of variance analysis showed that the concentration of suspended sediment was significant in terms of time scale and spring had the highest rate of sedimentation. Redundancy analysis and canonical analysis on the properties that participate in the first factor (PC1) showed the characteristics of the percentage of erodible formation, relatively erodible formation, and percentage of free construction activity, respectively. Road (slope leveling) and stream length are the most essential attributes of sub-basins in the production and concentration of suspended sediment in the study area.

Solar radiation reaching the earth's surface is one of the most important variables used in solar energy projects, hydrological modeling, evaporation and transpiration, meteorology and agriculture. In this research, the functionality of principal component analysis and linear multivariate regression was investigated in predicting the amount of solar radiation in Yazd station between 2010 to 2014 and 2015 to 2021. Variables of average temperature, minimum temperature, maximum temperature, hours of sunshine, relative humidity and solar radiation are received from the Meteorological Organization on a daily basis and variables of extraterrestrial solar radiation, relative distance from the earth to the sun, solar inclination angle and maximum hours of sunshine are calculated with the existing relations and was chosen as input of Principal component analysis method. The correlation of these parameters with the amount of solar radiation shows the presence of a significant positive relationship at the level of 1% between the amount of solar radiation and extraterrestrial radiation, the ratio of sunny hours, the average temperature and the relative distance from the earth to the sun, and with the solar inclination angle at the level of 5% and was observed a significant negative correlation at the level of 1% with relative humidity values. The results of the analysis of the main components showed that the first two components can be selected as the main components according to the eigenvalues of the parameters and the percentage of variance. The correlation coefficient of the first and second components with solar radiation was 0.893 and -0.168, respectively, and as a result, the first component has a higher correlation with the amount of solar radiation than the second component. The rotated value of loading in the first component shows that the minimum temperature, maximum temperature, average temperature and relative humidity are more correlated with solar radiation and the correlation of other factors (extraterrestrial radiation, solar inclination angle and relative distance from the earth to the sun and the ratio of sunny hours) is low in this component. Investigating the relationship between solar radiation and the values of the main components obtained from PC1 and PC2 using multivariate linear regression showed that the effect of meteorological parameters on solar radiation is significant (p < 0.001) and its R2 is was determined as 0.89%. The evaluation criteria of PCR and MLR methods show the ability of PCA method to estimate solar radiation.

Awareness of erosion and its spatial changes is a guide for managers and planners in the important matter of water and soil protection in watersheds. In the present study water flow and sediment data of 40 hydrometric stations in the Middle Alborz region and in the statistical period between the wateAwareness of erosion and its spatial changes is a guide for managers and planners in the important matter of water and soil protection in watersheds. In the present study, with the aim of presenting a map of soil water erosion, water flow and sediment data of 40 hydrometric stations in the Middle Alborz region and in the statistical period between the water years of 1980-1981 to 2020-2021 were used. First, the most important factors affecting erosion and sedimentation in the studied sub-basins including 30 hydrological, physiographic, geomorphological, geological and soil characteristics, climate, land use and vegetation as independent variables and two sediment characteristics; including total yield and specific deposition as dependent variables were identified. Using factor analysis and principal component analysis methods, the most important characteristics of effective basins in sedimentation were identified. In the next stage, the study area basins are divided into six homogeneous groups based on selected characteristics and cluster analysis and multivariate regression stepwise method was used to model between specific sedimentation and basin characteristics. According to the selected regression models, it is determined that the amount of special sediment in the Middle Alborz region to five factors of agricultural land area (rainfed, irrigated and orchards), the area of sub-basins, the total area of erosion-sensitive and Quaternary structures, the average annual discharge and the form factor of the basin depend on the fact that these five factors control 92% of the sediment production changes in the selected sub-basins. Then, considering the sediment transfer coefficient of the sub-basins, the amount of soil erosion was determined and the soil erosion map was drawn for the Middle Alborz region. The results showed that the average sedimentation rate and the average weight of soil erosion in the study area were 3.84 and 13.56 tons per hectare per year, respectively. In general, the factors affecting erosion and sedimentation of the Middle Alborz region can be divided into three groups: land use change due to human activities, geology and physiography.r years 1980-1981 to 2020-2021 were used. First, the most important factors include 30 hydrological, physiographic, geomorphological, geological and soil characteristics climate land use and vegetation as independent variables and two sediment characteristics. The most important characteristics of effective basins in sedimentation were identified. In the next stage, the study area basins are divided into six homogeneous groups based on selected characteristics and cluster analysis and using multivariate regression by stepwise method to model between specific sedimentation and basin characteristics. According to the selected regression models, it is determined that the amount of special sediment in the Middle Alborz region to five factors of agricultural land area (rainfed, irrigated and orchards), the area of sub-basins, the total area of erosion-sensitive and Quaternary structures, The average annual discharge and the form factor of the basin depend on the fact that these five factors control 92% of the sediment production changes in the selected sub-basins. Then, considering the sediment transfer coefficient of the sub-basins, the amount of soil erosion was determined and the soil erosion map was drawn for the Middle Alborz region. The results showed that the average sedimentation rate and the average weight of soil erosion in the study area were 3.84 and 13.56 tons per hectare per year, respectively. In general, the factors affecting erosion and sedimentation of the Middle Alborz region can be divided into three groups: land use change due to human activities, geology and physiography.

Density current is one of the most important factors in the sedimentation process of dams. Increased sediment will reduce dam storage capacity and makes significant challenges for relevant engineers. Therefore, understanding the dynamics of density fluids and related sediment patterns is very efficient for dam reservoir management.

The purpose of this study was to create an intelligent model with appropriate adaptation to laboratory data so that, it can be used in future designs with different variables. Therefore, in this study, the percentage of reduction of density salt current head under the influence of trapezoidal permeable obstacles (aggregates with a diameter of 1 cm), taking into account variables such as discharge, slope, concentration and height of obstacles in laboratory.

Based on the results, the density salt current head was modeled using the artificial neural network feed-forward method and the classical multivariate regression method, and the performance of these two methods was compared. The results showed that the intelligent feed neural network intelligent method in modeling the percentage reduction of density salt current head is significantly superior to the multivariate regression method so that the training, calibration and test regression values ​​ are 0.99, 0.98 and 0.98 were obtained for neural network and 0.92, 0.91 and 0.91 for multivariate regression, respectively.

The performance of the artificial neural network is much better than the multivariate regression method.

This study was conducted to investigate the main factors affecting groundwater level fluctuations in the Najafabad plain. For this analysis, the 10-year period from 2004 to 2014 was considered and changes in temperature, precipitation, in river discharge between the beginning and end of the study area, groundwater withdrawal and in water flow in irrigation and drainage networks, were analyzed. In order to establish the relationship between each of the mentioned components with the changes in groundwater level in the study area, two general classifications were considered to determine the groundwater unit hydrograph. In the first classification, a groundwater unit hydrograph was prepared for all observation wells in the study area. In the second category, this hydrograph was prepared for observation wells around the river. Then, using multivariate regression analysis in SAS software, the best relationship between these components with changes in groundwater level was determined. Results showed that if the changes of groundwater unit hydrograph level are considered in the first group, changes in groundwater withdrawal with a share of 36.19%, had the maximum impact on changes in groundwater level. After that, the changes in the river flow with a share of 28.60%, had the greatest share in groundwater fluctuations. On the other hand, when the groundwater unit hydrograph is considered for the second group, changes in flow in the river with a share of 34.64% will have the maximum effect on changes in groundwater level and the share of groundwater withdrawal will reach to 28.53.

At the beginning of the conversion of precipitation into runoff, some of the precipitation is absorbed by the soil or penetrates deep into the earth and the rest becomes surface runoff. By determining the runoff and erosion simultaneous threshold can help to watersheds proper management. In this study, in order to determine the runoff and erosion simultaneous threshold using erosion different components in different land uses of Gachsaran Formation, a part of the Kuhe Gach watershed of the Izeh city with an area of 1202 hectares was selected. In this study, the relationship between the runoff and erosion simultaneous threshold and erosion different components such as soil permeability and the amount of sediment and runoff in different land uses of Gachsaran Formation was done using multivariate regression. Then, sampling of erosion different components at 6 points with three replicates and at different rainfall intensities of 0.75, 1 and 1.25 mm in min, in 3 land uses of the range, residential area and agricultural lands with the help of the device rain simulator was performed. SPSS and EXCEL software were used for statistical analysis. The results showed that in estimating the runoff and erosion threshold in all three-land uses, the amount of sediment and then the degree of soil permeability had the most positive and negative effects and the amount of runoff had no role in modeling. The positive effect of the amount of sediment in general on the runoff and erosion threshold in five cases and its negative effect in three cases and the positive effect of soil permeability in modeling in general on the runoff and erosion threshold in three cases and affecting its negative is three cases.

Delineation of gully erosion susceptible areas by using statistical models, as well as optimum usage of existing data and information with the least time and cost and more precision, is important. The main objective of this study is to determine the areas accuracy to gully erosion and susceptibility mapping by using data mining of the bivariate Dempster-Shafer, linear multivariate statistical methods and their integration in Semirom watershed, southern Isfahan province. First, the geographical location of a total of 156 randomly gullies were mapped using preliminary reports, satellite imagery interpretation and field survey. In the next step, 14 conditioning parameters of the gullies in the study area were selected including the topographic, geomorphometric, environmental, and hydrologic parameters using the regional environmental characteristics and the multicollinearity test for modeling. Then, the Dempster-Shafer statistical, linear regression, and ensembled methods were developed using 70% of the identified gullies and 14 effective parameters as dependent and independent variables, respectively. The remaining 30% of the gully distribution dataset were used for validation. The results of the multivariate regression model showed that land use, slope and distance to drainage network parameters have the most significant relation to gully occurrence. The gully erosion susceptibility maps were prepared by individual and ensemble methods and they were divided to 5 classes of very low to very high rate. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to validate gully erosion susceptibly maps. The verification results showed that the AUC of ensemble method (0.948) is higher than Dempster-Shafer (0.924) and Multivariate regression (0.864) methods. Also, the the seed core area index (SCAI) value of the ensembled model from very low to very high susceptible classes have a decreasing trend that indicating a proper separation of susceptible classes by this model.

Evaluation of irrigation systems consists of analyzing the measurable performance of the system in operation conditions, is a management tool that allows the system users to make the best use of the irrigation systems. Since various regions in Iran have arid and semi-arid climates, obtaining the actual wind drift and evaporation losses and also making optimized use of water towards sustainable water resources management has a great importance. Since estimation of evaporation and wind losses that has been presented in different climatic conditions has different results, and it is not possible to present a reference model, so the purpose of this study is to investigate the effective climatic parameters on evaporation and wind losses of the Center Pivot system and to present an optimal model using multivariate regression in the study area.

This research has been carried out in the city of Pars-Abad in Ardabil province. The study was conducted in Part of the Moghan Agro-Industrial Complex, at wind speeds of 0-3 and 3-6 meters per second with three replications. Experiments were performed on two Center Pivot Irrigation System in sections 5 and 6 of Moghan Agro-Industrial Complex (1-4-6 and 1-8-6), which included two design treatments. The average altitude of this area is 32 meters above the sea.

Results indicated that the efficiencies of the system at wind speeds of 0-3, 3-6 meters per second were 89.47, 84.47 percent, respectively, and wind speed, temperature, relative humidity, nozzle diameter, and vapor pressure deficit (saturation deficit) were among the factors that influenced wind drift and evaporation losses. Base on the results, the wind factor had the most and vapor pressure deficit had the least effect on wind drift and evaporation losses. Also, the equation of wind drift and evaporation losses was also obtained with the atmospheric conditions of the region, which had the best fit with the measured values. Results shown that the difference between results of the model and observed data was not significant in 1% level of confidence. Moreover, the measured and the modeled percentages obtained for wind drift and evaporation losses varied from 7 to 18 and from 10 to 18 percent, respectively.

The results show a direct relationship between the evaporation and wind losses with wind speed in the region. So that at wind speeds of less than 3 m/s the parameters of Coefficient of Uniformity, Distribution Uniformity and potential and Application Efficiency of the low quarter are in the appropriate range, and at medium and high wind speeds, evaluation parameters are not in the desired range, it is indicating the importance of center pivot system design in windy regions.

This research has been carried out in the city of Pars-Abad in Ardabil province. The study was conducted in Part of the Moghan Agro-Industrial Complex Province at wind speeds of 0-3 and 3-6 meters per second with three replications. For the experiment, two rows of cans at an angle of 3 ͦ were positioned radially along the center of the centripivot system 10 m apart. The sprayer used in this study was F33AS 11/64˝ with 26 m of spraying diameter and operating pressure of the sprinklers varied from about 2 to 2.5 bars.

Results indicated that the efficiencies of the system at wind speeds of 0-3, 3-6 meters per second were 89.47, 84.47 percent, respectively, and wind speed, temperature, relative humidity, nozzle diameter, and vapor pressure deficit (saturation deficit) were among the factors that influenced wind drift and evaporation losses.

The yield of Wheat is influenced by various factors including climate, land management practices and soil properties. It is important to investigate the relations between soil physical and chemical properties and the yield of wheat. In this study, the physical and chemical properties of the soil top layer (0-30 cm) and the yield of Wheat were determined in 34 wheat fields of Nazarabad region in Alborz province, during wheat crop year of 2018. Principal Component Analysis (PCA) was used to select the effective parameters on wheat yield and multivariate linear regression was applied to analyze the relationship between wheat yield and soil properties. Finally four regression equations were presented. Principal Component Analysis (PCA), stepwise linear regression, the highest correlation method and all the measured properties were used respectively to prepare four equations to estimate the Wheat yield. The evaluation of each equation was performed by Geometric Mean Error Ratio (GMER), Geometric Standard Deviation of the Error Ratio (GSDER), Normalized Root Mean Squared Error (NRMSE), and determination coefficient (R2). The results showed that wheat yield varies from 2750 to 10500 kg/ha in the region and the proposed equation using clay, Pava, Cu, porosity, calcium carbonate equivalent and pH with GMER, GSDER, NRMSE and R2 values of 0.99, 1.19, 0.17, and 0.64 respectively, is the most appropriate equation for determining of wheat yield. Therefore, in regions with climatic conditions and soil physical and chemical properties similar to the study area, the yield of wheat can be estimated with an acceptable level, based on the proposed equation.

Snow and snow melt runoff have a major role to play in supplying water. In this study the impact of climate change on runoff caused by snowmelt was studied in Sarab Seyed Ali watershed in north of Lorestan province. In this study, using snow cover level extracted from Landsat 5 and 7 satellite images and climatic data predicted by general circulation model using Lars-WG software snowmelt runoff under three scenarios A1B, A2 and B2 in the year 2010 to 2013 was simulated using the SRM model. The simulation results were evaluated using two statistic including coefficient of determination and volume difference. The value of 0.87 for the coefficient of explanation and -3.62 for the volume difference indicates the accuracy of the model. Then, using these three scenarios, the climate parameters for 2020 to 2060 were predicted through general circulation model, then using the predicted climate parameters, the rate of snow cover was estimated through multivariate regression equations and was into to SRM model. the simulation was done for 2020 to 2060. The results showed that the average runoff in November to May for the three scenarios was 4.3, 4.4 and 4.2 m3/s respectively, while the long-time mean runoff from 1980 to 2013 was 6.07 m3. However, this decrease in future runoff volume can be justified by the reduction of snow cover.

This research was used from climate information such as average humidity, average wind speed, soil temperature and salinity of different soil depths (5, 10, 20, 30, 50, and 100 centimeters) in 10 years (2008-2017). The purpose of this study was to investigate the relationship between climate parameters and soil salinity fluctuations. For this purpose, rice soil salinity data were used which were previously measured. In both seasons of rice cultivation (conducted in one of the nurseries of Alborz province), salinity fluctuations at different depths of soil were recorded over a period of five to seven days, and soil salinity required for the purposes of this study was obtained, and then with the required meteorological data, the research was done. Also, Hurst test to find enough variance of the data to analyze the time series trend, and Mann-Kendal, spearman, and multivariate regression tests were applied for data analysis. The results showed that all parameters of the study based on the Hurst coefficient bigger than 0.5, had a significant variance for time series analysis. The results of Mann-Kendal and spearman tests showed that in surface soil layers (5 and 10 cm) after 10 years period, the soil salinity had a descending trend. However, both tests had an increasing trend in deeper layers (20, 30, 50 and 100 cm) and at depths of 50 and 100 cm observed the significantly increasing trend at a 5% level. Result of the multivariate regression test illustrated that the temperature in depth of 5 cm and average relative humidity (each 16%) and average speed of 60% had the most and the least effect, respectively on the soil salinity in the mentioned soil depth. Also, in four depths of six depths, the temperature in depth of 5 cm was the most effect on soil salinity of different soil depths. As a general result, in areas where the conditions of this present study, cultivation of plants with surface roots is possible because salinity fluctuations in the surface layers are low and it does not pose a threat to the plant under cultivation but crop cultivation with deep roots is limited because salinity fluctuations is an increasing trend that should be addressed so that soil salinity does not have a detrimental effect on plant yield.

Optimized management and utilization of any system requires knowing the working conditions of the system, and optimized utilization becomes possible by evaluating the way the system works. Since various regions in Iran have arid and semi-arid climates, obtaining the actual wind drift and evaporation losses and also making optimized use of water has a great importannce. The present research intended to evaluate a classic semi-permanent sprinkler, to measure wind drift and evaporation losses, and to introduce an optimized model by using multivariate regression for the study region. The study was conducted in Part of the Moghan agro-industrial complex in Ardabil province at wind speeds of 0-3 and 3-6 and higher than 6 meters per second with three replications. Results in ADF 25º nozzle sprinkler model indicated that the efficiencies of the system at wind speeds of 0-3, 3-6, and higher than 6 meters per second were 82, 66, and 43 percent, respectively, and wind speed, temperature, relative humidity, nozzle diameter, and vapor pressure deficit (saturation deficit) were among the factors that influenced wind drift and evaporation losses. The wind factor had the most and vapor pressure deficit the least effect on wind drift and evaporation losses. Furthermore, the equation of the best fit line to the measured values for wind drift and evaporation losses under the prevailing atmospheric conditions in the region was as follows and the difference between results of the model and observed data was not significant in 1% level of confidence. Moreover, the measured and the modeled percentages obtained for wind drift and evaporation losses varied from 6 to 34 and from 11to 35 percent, respectively.

Food security depends on production of basic agricultural crops such as wheat. In rain-fed wheat cultivation, climatic parameters such as temperature and rainfall have much effect on the wheat production. In this research, multi-year meteorological parameters and rain-fed wheat production data of Dezful, Shooshtar, Andimeshk and Masjed-Soleiman counties were used to determine the climatic model production at north of Khouzestan province. To estimate the amount of the effect of different meteorological factors on wheat yield, the linear regression methods at the local and regional scales were used. The findings showed that for estimation of wheat yield, local models need less data model than the regional ones. However, in comparison the local models had more accuracy than the regional model. Also, sensitivity analysis showed that, the mean of maximum temperature and that of precipitation of winter months were most sensitive parameters on wheat yield. In addition, wheat yield increases by the increase in the rain fall. But, rainwater productivity has had wide fluctuation between 0.07 to 0.39 kg/m3 in various years in the region. This shows that more attention should be paid to other factors such as farming factors for increasing rainwater productivity.

Discharge estimation in watersheds with limited statistical data is of interest of researchers especially in developing countries. In many cases, recorded discharges are not available or insufficient in terms of quality and quantity that lead to problems for water resources management plans. Therefore, methods to predict the river discharges in ungauged or with limited recorded data have considerable importance. Various methods including statistical models, time series and expert models have been developed for discharge estimation. The decision tree model is one of expert models that model the nonlinear behavior of data using simple rules production. The objective of this study is evaluation of multivariate regression and decision tree model (M5) to estimate monthly discharge in ungauged watersheds in Golestan Province.
In this study, the Golestan province that includes various watersheds with different characteristics was considered. The physiographic characteristics of studied watershed were extracted and rainfall and temperature climatic parameters were estimated in monthly time scale in GIS environment for the period 1981-2011. The climatic and physiographic parameters were considered as input to multivariate regression and decision trees M5 models and root mean square error (RMSE) and correlation coefficient (R) applied for models evaluation.
According to the results of multiple regression and decision tree models, discharge estimations in wet months were more accurate than dry months. In application of decision tree model the best prediction belonging to March with R=0.93 and RMSE=1.002 while worst prediction was for August with R=0.571 and RMSE=0.635. Moreover, multivariate regression model led to best results in March with R=0.723 and RMSE=2.043 and low accurate prediction in August with R=0.322 and RMSE=1.979. The results of decision tree model were better than multivariate regression model in all months based in calculations.
The results of this study showed that the discharge estimation using multivariate regression and decision tree M5 models in wet months are applicable but in dry months predictions are not accurate because of high rainfall variations, storm patterns and error in rainfall zoning and interpolation. The results indicated that the decision tree model had more accurate results than multivariate regression model considering higher precision and lower error. The decision tree model had higher correlation coefficient with respect to model evaluation criteria.

By using regionalization methods information from gauged sites transform to desired site. Up until now a variety of regionalization approaches have been proposed. In every site it is necessary to evaluate these methods and select the best method. It is of interest to understand how spatial weighted least square regression method based on depth function flood quantiles (SWLSR) compare with multivariate regression (MR) and Physiographical space-based kriging (PSK) methods. In each iteration desired station regarded as ungauged site then using genetic algorithm depth functions weights were optimized, finally (regarding) by taking account similarity between desired site and others sites flood quantiles corresponding to different return periods were estimated. By means of a leave-one-out cross-validation procedure, the performance of SWLSR was compared to MR and PSK methods for prediction of 10, 50 and 100 yr for 26 gauging station in the Southern Alborz. . The Result showed SWLSR approach yielded lower root-mean-square estimation errors and higher Nush Sutcliffe criteria thaneither the MR or the PSK approaches. PSK method estimated foold discharge in ungaged basin better than MR. In depth based approach Nush Sutcliffe criteria values for flood quintiles (Nash–Sutcliffe efficiency values for 10,50 and 100 yr floods were 0.64, 0.65 and 0.65 respectivly) three corresponding to different return periods were similar.In this method relative error to area in small catchment were biger than those obtained in big catchment.

Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.Investigating lowflow characteristics is of vital importance for optimal water resources management and sustainable use of water resources especially in arid and semi arid regions. In this study 26 hydromerty stations in the Sefidroud basin were selected and 7-day annual minimum series were computed as low flow index. Physiographic characteristics of the basin were also gathered in GIS software. Perason type 3 (PIII) and Generalized logistic distributions (GLOG) were identified as the best regional distributions for the west and east homogeneous regions respectively based on the goodness-of-fit test. Low flow regional regression (LFRR) was then developed for each homogeneous regions. According to the established models, drainage area, land use characteristics, average basin slope and basin circularity were identified as important factors in low flows features.

Phosphorus is important as an essential element in the production of agricultural products. On the other hand, its ability to induce essential micronutrient deficiency and its negative effects on the environment, have attracted more attention to this element. The knowledge of phosphorus availability conditions in the soil and consequently the accurate management of fertilizer consumption has a key role in the environmental protection. The degree of phosphorus absorption in the soil depends on the environmental factors, soil characteristics and compositions, and phosphorus fertilizer which have been used. The amount of available phosphorus in the soil has relationship with some of the physical and chemical properties of the soil. Since, the soil characteristics are important factors in the reaction of phosphorus in the soil, the present study aimed to investigate and determine the most important soil characteristics affecting the availability of phosphorus using regression and artificial neural network techniques, in the soils of Sistan plain.
Soil sampling was done in 1.5×1.5 km intervals, from 0-30 cm depth, and 200 soil samples were taken. The amounts of available phosphorus and the soil properties including the percentages of clay , organic matter, calcium carbonate and the amount of pH were measured. Then, stepwise multivariate linear regression analysis was performed to determine linear relation between available phosphorus and the soil properties. In order to model and validate the regression model, respectively 80 and 20% of data were selected and entered into SPSS software. To train the neural network, multilayer perceptron (MLP) network was used by MATLAB 7.6 package. In this type of network, 70% of data is selected for training, 15% for validation and 15% for testing the model. Levenberg-Marquardt algorithm and hyperbolic tangent (as a transfer function) were used to train the network. The numbers of neurons in the hidden layer were calculated based on the trial and error method and finally the best structure was selected according to the highest R2 and the lowest RMSE value. Moreover, quantifying the importance of variables in the neural network was done through employing connection weight approach. In this method, the connection weights of input-hidden and hidden-output neurons were used to indicate the significance of variables.
The values of the coefficient of variation for the soil properties were in the range of 5.66 for pH (the lowest) and 69.90 for available phosphorus (the highest). The high variation of the available phosphorus could be due to the different amounts of phosphorus fertilizers consumption and their diverse rate of conversion to less soluble forms. The validation results of regression and neural network methods showed that the latter technique was more accurate compared with the multivariate linear regression method, in the estimation of available phosphorus, as multi-layer perceptron neural network with 4-6-1 layout predicts nearly 90% of available phosphorus variability using soil properties (percentage of clay, organic matter, calcium carbonate and the amount of pH); however, the obtained regression equation could explain only 43% of phosphorus variances. The reasons for this could be: 1) considering nonlinear relations between the variables in the artificial neural network method, and 2) less sensitivity of this method to the existence of error in input data, comparing with the regression method. The values of R2 and RMSE were 0.43 and 11.23, respectively for training the multivariate linear regression method and 0.91 and 4.28, respectively for training the artificial neural network method. From the investigated soil properties in the current study, the percentage of organic matter and clay were entered in the regression model, and the values of standardized regression coefficient (beta) showed that the first variable is more important to explain the available phosphorus variability. The results of quantifying the importance of variables by the connection weight method showed that pH had the greatest contribution in the variability of phosphorus in the study area. In the other words, the high values of pH were the most important limiting factor for the availability of phosphorus in Sistan soils.
Considering nonlinear and complicated relations between variables, the artificial neural network model is an effective tool to assess the effect of soil properties on the availability of phosphorus in the study area. The results of quantifying the importance of variables by using the connection weight method showed that pH had the greatest contribution in the variability of phosphorus in the study area. In fact, the existence of lime in the soils of the study area, arid climate and lack of precipitation have resulted in the accumulation of basic cations in the soil and consequently increased pH values. Furthermore, the observed average values of pH that are more than 8.5 demonstrated the risk of soil sodicity in the study area. Thus, the management of this area by cultivating tolerant plants could be resulted in increasing organic matter content, which along with using chemical amendments such as sulfur will decrease pH values and increase the availability of phosphorus in Sistan plain. Applying such practices and through it modifying soil characteristics, decreasing the consumption of phosphate fertilizers and preventing their hazardous environmental effects would be expected in long run.

Weather features and their variations have an important role in the yield of agricultural products, especially in rain-fed conditions. The main metrological variables that affected yields consist of precipitation, temperature, soil moisture and solar radiation. Also, drought is one of the major constraints to production, especially the mid-season drought which occurs during the podand seed formation stages and the terminal drought which occurs during the pod filling stage. The results of investigating the relation between drought indices such as Standardized Precipitation Index (SPI), Palmer Drought Severity Index (PDSI), Crop Moisture index (CMI) and Z index with crop yields indicated the capability of these indices to estimate variations in crop yields. The objective of this study in the first step is investigation of relations among wheat and barley crop yields with climatic variables and SPI and RDI drought indices based on Principle Component Analysis (PCA) method at Bojnourd, Mashhad and Birjand stations. In addition, by selecting the prominent variables via PCA method, the best models of estimating each crop’s yield based on multivariate regression methods at selected stations were determined. In this study, the relationship between yields of rain-fed wheat and barley with weather variables consisting of minimum, mean and maximum temperature, precipitation, evapotranspiration and drought indices including SPI and RDI were investigated and modeled at Bojnourd, Mashhad and Birjand stations. For this purpose, the values of each variable were calculated for 34 time scales of 1, 2, 3, 4, 6, and 9 months and wet periods (nine 1-month periods, eight 2- month periods, seven 3- month periods, six 4- month periods, two 6- month periods, one wet period (5 or 7-month) and one 9-month period). After that, the main influencing variables were chosen among investigated time periods for each variable by using the method of principal component analysis (PCA). In continuation, the selected variables via PCA technique were used in the multivariate regression methods to create the best model of predicting wheat and barley yields based on each mentioned variable and combination of them. The performance of the established model was evaluated based on Ideal Point Error (IPE) criteria and the best predicting model of wheat and barley was selected for each region. The results showed that applying PCA technique as a powerful statistical tool leads to decrease of the error and inflation of constructed models. This is done by reducing the volume of data and selecting influencing variables. Based on the PCA results by choosing only four components the 90 percent and greater than variation of crop yields are estimated and the first component includes time periods of spring and winter months. Investigation of the results of the best model at the given stations based on IPE criteria show that the constructed models based on variables of SPI index have more accuracy for predicting yields of wheat and barley at station of Bojnourd, at Mashhad station the created models based on a combination of variables and at Birjand station a model based on a combination of variables and a created model according to RDI variables was used that has more accuracy for predicting yields of wheat and barley, respectively. Comparing the estimated and actual values of wheat and barley yields indicate that the correlation coefficients of the models when applied to estimate the yield of wheat and barley at Bojnourd station resulted in 68 and 69 percent, at Mashhad station 89 and 86 percent and at Birjand station 66 and 74 percent, respectively.The performance evaluation graph shown in Fig. 1 can be used to illustrate model performance and to diagnose model bias. According to the results, a relation between crop yields and combination of metrological variables and drought indices is more positive and stronger than only metrological variables combination. The results showed that the variables of temperature, precipitation and evapotranspiration are to be considered. Also, the evaluation model indicated that the RDI index is more suitable for predicting rain-fed wheat and barley yields.

Prediction of crop yield is one of the managerial and planning tools in agriculture. Meteorological variables and drought indices play an essential role in yield prediction. In this study, the yield of four crops, namely wheat, barley, chickpea and watermelon was predicted in Mashhad and Birjand regions based on meteorological variables and drought indices using three regression models of Ridge, Enter and Step Wise. For this purpose, ten meteorological parameters and seven drought indices were used in three structures as input multivariate regression models. Results showed that among the meteorological variables, the two variables: number of rainy days and maximum air temperature had higher correlation with crops under study. Similarly, among drought indices, Nguyen index showed such higher correlation. It was found that Ridge multivariate model including weather parameters and meteorological drought indices could be recommended instead of two structures of meteorological variables and drought indices for yield prediction. This recommendation was based on multiple regression analysis and higher determination coefficient (R2) and a lower normal root mean squared error (NRMSE). Also, determination coefficients for all four products in both regions were from 0.91 to 0.99 and NRMSE from 0.09 to 0.48. Results also showed that the most prediction accuracy was for wheat with R2 of 0.996 and NRMSE of 0.09 in Mashhad and for chickpea with R2 of 0.999 and NRMSE of 0.09 in Birjand. Similarly,the least accurate prediction belonged to watermelon having R2 of 0.92 and NRMSE of 0.48 in Birjand and R2 of 0.96 and NRMSE of 0.27 in Mashhad. Overally, due to high R2 and low NRMSE, Ridge regression model is recommended for both regions.

Investigation of soil hydraulic properties like permanent wilting point (PWP) and field capacity (F. C) are very important for studding and modeling the soil water and solute transport in soil in which their spatial and temporal variability led to development of indirect methods in prediction of these soil characteristics. Therefore; in the present study in order to evaluate the amount of water at F. C and P. W. P، 63 samples have been taken from 15 pedons in Fashand region. The particle size distribution have been determined by hydrometric method، bulk density by volumetric method (using undisturbed clods)، saturation percentage by weight and percentage of water at F. C and P. W. P by using pressure plate apparatus. We applied the artificial neural network (ANN)، multivariate regression (MR) methods and used several pedotransfer functions (PTFs) to predict the F. C and P. W. P parameters، using the easily measurable characteristics of clay، silt and sand percentage، bulk density and water at saturation percentage. The results showed that the ANN method give the best results followed by MR method and finally the PTFs. Regarding the PTFs، the classic are showed better results relative to parametric and point PTFs. In conclusion، the results of this study showed that، training is very important in increasing the model accuracy of one region.