multiple linear regression

As the global demand for water resources increases, the reduction in water loss, including Evapotranspiration, becomes more obvious. Although many models have been developed to predict evapotranspiration, no universally accepted model for all climate regions has been established. Several soft computational models have been created to circumvent the constraints of empirical models and accurately predict ET. Soft computing models typically require less data and are applicable across various climatic zones. This study aimed to analyze how well two Random Forest models and Multiple Linear Regression could predict ETo in the Ardabil plain region. Meteorological data from the Iranian Meteorological Organization were used to calculate the reference evapotranspiration from 2014 to 2016. In constructing the model, data from 4 meteorological stations were combined to generate a random time series, while the fifth station was reserved for evaluating the models. The assessment metrics used comprised RMSE, R2, and NSE. The RF model achieved higher accuracy with R2, NSE, and RMSE values of 0.74, 0.743, and 8.20 mm, respectively, compared to the MLR model. The results demonstrate that random forest models are reliable tools for forecasting ETo with minimal climate data. In general, using the results of this study and other similar research, we conclude that RF and MLR models simulate potential evapotranspiration with acceptable accuracy but are sensitive to the number of input parameters.

Life on Earth is influenced by precipitation. Precipitation is one of the most significant factors that affect the hydrological cycle. Considering that precipitation is non-linear, complex, and can be changed according to spatial and temporal, estimating the amount of this important atmospheric factor in each month or year for each region and watershed is particularly important in managing and optimizing water resources. Various optimization models and algorithms have been proposed for modeling hydrological systems in recent decades. These algorithms have significantly reduced errors and increased accuracy. Still, since hydrological systems rely on random events, none of the methods can be completely and accurately selected as a superior model for modeling and estimating. The honey badger algorithm is an innovative algorithm that requires a few iterations to achieve an optimal solution, and this increases the speed of reaching the desired results. In current study investigates the performance of three models, including multiple linear regression (MLR), artificial neural network (ANN), and hybrid artificial neural network with honey badger optimization algorithm (HBA-ANN) for modeling the temporal and spatial precipitation in East Azarbaijan province. The best-developed model was selected by evaluation criteria such as R, RMSE, NRMSE, MBE, and NSE, the best model is selected.

The MLR model is one of the methods to analyze and investigate several variables. In this method, the model has one dependent variable and several independent variables, so that a linear equation is generated between the independent variables called X1, X2, ..., Xn and the dependent variable Y. ANN is a black box model of neural networks in the human brain. One of the most used methods is the BP method, which includes two stages. In the first stage, which is entitled feed-forward, the error value is calculated, after comparing output and objective values. In the second stage, which is labeled the back-propagation, the error value calculated in the previous step is corrected. The mentioned two stages continue until the output of the model approaches the desired output. The HBA is a new algorithm that simulates the honey-seeking behavior of a creature called the honey badger. The HBA includes two stages. In the first phase, the locations of this creature are calculated, and in the second phase, the exact distance between the HBA and the prey (dj) is calculated based on the honey intensity (S) and the honey smell intensity (Ij), as well as its new and optimal location for the prey Xnew. In the HBA-ANN model, the HBA algorithm is used to determine the most optimal output value in the ANN and increase performance in modeling. Therefore, the developed hybrid model can have the characteristics of both ANN and HBA methods.

In this study, in the first stage, the temporal modeling, and in the second stage, the spatial modeling of the monthly precipitation of 18 stations in East Azarbaijan province during the period of 1996-2022 using MLP, ANN, and HBA-ANN models has been paid. For temporal modeling of precipitation, one and two-month precipitation delay steps of the stations were used as input parameters. The first 70% of the dataset was selected for the training phase and the last 30% of the dataset was selected for the testing phase. Based on the results obtained from evaluation criteria and graphic diagrams, it can be concluded that the HBA-ANN model indicated significant accuracy compared to other models in the temporal modeling of precipitation. Also, by comparing the results of the stations in the HBA-ANN model, the Heris station with R =0.94, RMSE=2.25, NSE=0.79, NRMSE=0.04, and MBE=1.06 in the testing stage performed better compared with other stations. For spatial modeling of precipitation, the geographic coordinates of the stations, which include longitude, latitude, and altitude, are used as input parameters, and average monthly precipitation is used as the output parameter. From eighteen stations, 70% of the stations were selected for the training phase and 30% of the stations were selected for the testing phase. Based on the results obtained from R=0.95, RMSE=1.03, NSE =0.92, NRMSE = 0.03, and MBE = -0.81 and graphical diagrams, it can be concluded that the HBA-ANN model revealed significant accuracy compared to other models in spatial modeling of precipitation.

Precipitation is one of the most important factors that significantly change the hydrological cycle. Therefore, modeling and estimating this parameter is vital. In this study, the performance of multiple linear regression (MLR), artificial neural network (ANN), and hybrid ANN using honey badger algorithm (HBA-ANN) models were used for the spatial and temporal modeling of precipitation in East Azarbaijan province. For spatial modeling, the time delay steps of one and two months of station precipitation were selected as input parameters. Also, for temporal modeling, the longitude, latitude, and altitude parameters were used. The mentioned models were evaluated by R, RMSE, NSE, NRMSE, and MBE assessment criteria. According to the results of temporal modeling, the HBA-ANN model for all stations, especially Heris station with R equal to 0.94, RMSE equal to 2.25, NSE equal to 0.79, NRMSE equal to 0.04, and MBEequal to 1.06 is selected as the superior model. Also, based on the results obtained from spatial modeling, the HBA-ANN model with R equal to 0.95, RMSE equal to 1.03, NSE equal to 0.92, NRMSE equal to 0.03, and MBEequal to -0.81 was selected as the best model. The MLR and ANN models, respectively, presented a relatively poor performance compared to the developed hybrid model.

Artificial intelligence models as powerful methods in modeling nonlinear complex problems, have a significant ability and this has been proven in numerous articles. Artificial intelligence has been used in various issues, including engineering, medicine, etc. The success of this method in comparison with analytical and numerical methods, their easiness, speed and accuracy caused to open their place among researchers as much as possible. Today, Considering that one of the challenges of human life is the issues related to water resources management, so in this study, an attempt has been made to investigate the performance of artificial intelligence and regression models in the cases of water resources. Various researches have been done in the case of modeling and parametric analysis of water resources. However, in this study, artificial intelligence (Learning Machine) models were used to simulate the qualitative and quantitative parameters of water. The models used in this study are: Self-Adapting Extreme Learning Machine (SAELM), Least Square Support Vector Machine (LSSVR), Adaptive Neuro Fuzzy Inference System (ANFIS) and Multiple Linear Regression (MLR) model which was used to predict changes in hydrogeological parameters. Today, due to the growing global population, one of the most important challenges is access to safe drinking water. In our country, Iran, due to its location in the semi-arid region and low rainfall, this danger is felt more than ever. One of the serious issues is the salinity leakage into groundwater resources. In this study, an attempt has been made to simulate the leakage of salinity dynamic flow into the freshwater resources of the coastal aquifer, using artificial intelligence and statistical models. At the end, the simulation results and the accuracy of the models are given. The study area in this work, is Mighan Wetland and Mighan aquifer in Markazi province. Annual rainfall occurs in small amounts in this area. According to the statistical results provided by synoptic and rain gauge stations in the region, the maximum and minimum rainfall values range from 461 mm in the northeast to 208 mm in the center of Arak plain. The hydraulic outlet of the aquifer to the Mighan plain is located in the center of the plain. The water entering the Mighan plain and leaves the system due to evaporation from the water table. Observatory wells were used to sampling this lake due to its saline water. The wells were located in an area called Vismeh near the lake.

In this study, qualitative and quantitative parameters: water salinity, total dissolved solids (TDS), chlorine ion (cl), sampling time (t), electrical conductivity (EC), Salinity and groundwater level (GWL) were simulated. In this work, Adaptive Neuro Fuzzy Inference System (ANFIS), Least square support vector machine (LSSVM), Self Adaptive Extreme learning machine (SAELM) and Multiple linear regression (MLR) models were used for simulation. In this study, data from 173 months of sampling were used. 70% of the sample size was used for training and 30% for testing models.

Simulation was performed using artificial intelligence models and regression model. The simulation results showed higher accuracy of artificial intelligence models. After simulation and obtaining the results, then the uncertainty analysis was performed by Wilson Score method without continuity correction. In this method, the prediction error (ei), the mean prediction error (Mean) and the standard deviation of the error is (Se). If the mean error value of a model in predicting the target variable is positive, it means that the performance of the model is Over Estimated. Also, if the average value of the model error is negative, the performance of the model is Under Estimated. Moreover, the results of Uncertainty Analysis with a significance of 5% were obtained. and finally we briefly write the subsequent performance Over Estimated (OS) and Under Estimated (US).

The results showed that different models were successful in predicting water parameters. In order to comprehensively evaluate the accuracy of the models in the simulation, the performance of the models was measured by five approaches. The proposed approaches were: 1) Evaluation of prediction by accuracy chart, 2) Performance evaluation by mathematical indices, 3) Performance evaluation, by Uncertainty Analysis by Wilson Score method without continuity correction, 4) Accuracy evaluation by error distribution charts and 5) Performance evaluation by discrepancy rate (DR) charts. Finally, all the results are given at the end of each section, respectively.Approach 1- According to the prediction accuracy charts, 16 charts were drawn and the most accurate models of which are depicted in Figures 4 to 7. After modeling, the results showed that the most accurate models in simulating groundwater parameters were SAELM model in GWL simulation. According to the results, SAELM model in GWL and EC simulation, LSSVM in TDS simulation and MLR in Salinity simulation were the superior midel, Respectively.Approach 2- According to the performance measurement indices, finally the results showed that SAELM model was the best model in simulating parameters (EC) and (GWL). The LSSVM model was also the most accurate model in modeling (TDS). MLR model was the best model in (Salinity) parameter simulation.Approach 3- Uncertainty analysis was performed based on Wilson score method. The performance of the models in the simulation showed that the performance of the SAELM model was determined as Under estimated and other superior models in simulation had Over estimated performance.Approach 4- Based on the error distribution diagrams, the best accuracy was assigned to SAELM and MLR models.Approach 5- Based on the discrepancy ratio, SAELM and MLR models were estimated to be the most accurate models in the simulation.

Water infiltration to soil play an important role in irrigation management, storage moisture in soil especially in dry and semi-dry area and increasing agronomy yield. Understanding water infiltration to soil is of important in designing and applying water conservation methods, flood and runoff control and soil erosion management. Additionally, accurately measuring water infiltration to soil in different times is more important for predicting water storage in root zone, irrigation designing and planning and agronomy management. In other hand, water infiltration to soil is a base for precision agriculture, therefore, producing accurate maps for water infiltration to soil play an important role in land management and applying precision agriculture. Modeling soil water infiltration at the field scale with ruler of calcareous, saline and sodic conditions is important for a better understanding of infiltration processes in these soils and future of infiltration modeling. The present study aimed for estimating water infiltration to soil at different times using soil spatial prediction functions and spatial estimators.

in present study, 72 soil samples were collected using a random sampling method in Marvdasht plain, Fars Province. In selected points, soil bulk density, sand silt, clay, pH, electrical conductivity, calcium carbonate, solution sodium, solution calcium and magnesium and organic carbon contents. For measuring water infiltration to soil, the double ring method were used. Multiple linear regression (MLR), artificial neural network (ANN) and spatial estimators were used for deriving soil spatial prediction functions models between water infiltration to soil in different times including 5, 10, 20, 45, 90, 150, 210 and 270 min. In this study, the readily available soil properties and auxiliary variables such as remote sensing and topography data were used in soil spatial prediction functions.

The results of evaluating regression and artificial neural networks models based on the mean error (ME), coefficient of determination (R2) and root mean square error (RMSE) criteria in testing phase were showed that the developed artificial neural network models in present study performed better than multiple linear regression models in water infiltration to soil prediction at different times. Moreover, the results showed that the combined estimators (artificial neural network - Kriging) performed better than ordinary kriging model for estimating water infiltration to soil.

In totally, the results of this study showed that the applying soil spatial prediction functions (using auxiliary variables such as remote sensing data and topography data with the readily available soil properties) had a great potential to predict spatial estimation of water infiltration to soil at most considered times.

Cation exchange capacity (CEC) is one of the most important soil chemical properties that plays an important role in soil fertility. However, standard laboratory methods for measuring CEC are difficult, costly, and time-consuming. The aim of this study was to use new methods such as 1) Pedotransfer Functions (PTF) based on the basic soil properties using Multiple Linear Regression (MLR), 2) soil spectroscopy (Vis–NIR, 400 – 2500 nm) using Partial Least Squares Regression (PLSR), and Support Vector Regression (SVR), for estimating soil CEC. Also, from the regression coefficient analysis, key wavelengths were introduced. For this purpose, CEC was measured using the sodium acetate method for 72 soil samples collected, and spectral reflection of soil samples was determined using spectroscopy. These methods are made from a calibration set (70% of data) and evaluated with a validation set (30% of data). The results showed that Vis - NIR method performed better than PTF. In this study, wavelength ranges around 566, 854, 1354, 1418, 1906, 2071, 2203, 2319, and 2341 nm were investigated as the key wavelengths for estimation of CEC. Furthermore, the results of prediction models showed that SVR has a better performance than PLSR. This study proved that Vis-NIR is a promising method for soil CEC estimation.

Quantitative and qualitative analysis of water resources has become one of the most widely used topics in water resources research today. In this research, data mining, artificial intelligence, mathematical techniques have been used to simulate water behavior and estimate its parameters changes. The models used to estimate hydrogeological parameters are Self-adaptive Extreme learning machine (SAELM), Least square support vector machine (LSSVM), Adaptive Neuro-Fuzzy Inference System (ANFIS) and Multiple linear regression (MLR) models. Also, to evaluate the performance of these models, the accuracy of the models was assessed in the form of 5 approaches. The results showed that the SAELM model was the best model based on the simulation and correlation diagrams. Based on accuracy evaluation indices, the SAELM model with RMSE, MAPE and, R indices equal to 0.1545, 0.0070, and 0.9979, respectively, had the highest accuracy in hydrogeological parameters prediction. Based on Uncertainty Analysis by the Wilson Score method, the performance of the top model (SAELM) was estimated to be underestimated. Also, based on the error ratio diagrams, the most accurate results were related to the SAELM model. Finally, the SAELM model was assigned the lowest error rate using the error distribution diagrams.

Soil water Infiltration plays an important role in the water cycle of nature. However, since the direct measurement of soil water infiltration is laborious, time-consuming and expensive. Therefore, in this study was conducted to investigate the possibility of estimating the coefficients of water infiltration equations such as Kastiakov, Kastiakov-Lewis, Horton and USA Soil Conservation Service (SCS) using readily available soil properties and to evaluate the validity of these functions outside of their derivation regions. Therefore, soil physical properties and cumulative infiltration were measured in two different regions (T1 and T2). Parametric functions were derived using T1 location measurements and validation was performed using independent T2 measurements. Cumulative infiltration was measured using double rings with three replications at 78 points. Parametric functions were created using multiple linear regression. The highest accuracy of parametric functions in the derivation stage was related to SCS equations with the coefficient of explanation (R2) equal to 0.66 and the lowest accuracy with the value of R2 equal to 0.04 was related to the power of Horton equation. In the validation stage, the accuracy of the functions showed a large decrease. R2 changes range from 0.00 to 0.26, normalized root mean square error (nRMSE) range from 1.76 to 80.93 and geometric mean error ratio (GMER) range from 0.94 to 1.73 Was obtained. Therefore, in this study, the use of parametric functions to estimate the coefficients of infiltration equations outside of their derivation regions was not efficient.

Aggregate stability of soils informs about their relative strengths against erosive forces and mechanical disruption. In this research, a hybrid Genetic Algorithm-Artificial Neural Network method was used to select the best subset of features affecting the mean weight diameter (MWD. In addition, the ability of ANNs and multiple linear regression (MLR) for quantifying the relationship between the MWD index and some soil properties was assessed. After the modeling process, the importance of the selected features in relation to spatial variability of aggregate stability was investigated. In order to prepare a suitable data set; MWD index and some soil features were measured in collected soils from 90 sampling points. Feature selection results showed that six soil features including clay, sand, organic matter, calcium carbonate, electrical conductivity, and sodium adsorption ratio had the greatest effect on the aggregates stability of the studied soils. According to the MWD modeling results, the obtained values of coefficient of determination (R2), mean absolute error percentage (MAEP), and root mean square error (RMSE) for the ANN model performance were 0.94, 21.39, and 0.07% respectively. These findings indicated that the developed ANN model was able to predict the complex and nonlinear relationships between the MWD index and the soil properties selected by the algorithm. Based on the sensitivity analysis results, calcium carbonate equivalent, sand particles, and organic matter were identified as key factors in estimating aggregate stability. Overall, this study provides a robust framework for the prediction of aggregate stability and identifying the most determinant parameters influencing it in arid and semi-arid soils that could be applied to other regions with similar challenges.

Rainfed cultivation is one of the best ways to rich the sustainable development with optimal use of green water. About 75 percent of the world's cultivated areas are rainfed, which forms important part of the international economy. For a long time rainfed cultivation has been a common method in low-rainfall areas of Iran and mentioned method is one of the main ways of producing crops yet. Rainfall is an important factor in rainfed agriculture, so its limitations and non-conformity with planting season can reduce the production efficiency. Wheat growing period consists of three seasons of the year. Rainfall in spring, fall and winter is effective on wheat growth. Cereal grains, especially wheat, are the most important food source in the world. After cereals, legumes are considered to be the main human food source and have significant nutritional and agricultural importance. In addition to being high in quality and valuable, suitable and complementary protein for cereals in the nutritional pattern, lentils are also among the legumes that stabilize the air nitrogen in the soil, which lends itself to crop rotation. Thus, intermittent cultivation of rainfed wheat with rainfed lentil is an important factor in stabilizing production in the developing countries. Given that Iran is in arid and semi-arid regions and its population is growing, it is important to plan, and evaluate the past in order to predict the future. Zanjan province is the fifth country with 6.9% of total grains production. In this study, the effect of precipitation in water requirement of these two important products and their water stress has been studied in the climatic conditions of Zanjan plain. Zanjan province is located at the longitude of 47º 10' to 50 º 5' East and the latitude of 35º 25'37º10' North. The area of Zanjan province is about 39369 square kilometers. According to the De Marten climate classification, the tropical, temperate and subtropical climatic zones can be identified at the applied level. Average rainfall over the past decade has been reported as 301 mm. For conducting this study, statistical data including meteorological and crop yield data 2004-2014 obtained from the Meteorological and Agriculture-Jahad Organization, respectively. Zanjan synoptic station with the latitude and longitude of 48º 31' East and 36º 39' North, respectively, is located 1663 meters above the sea level. In the last 50 year, the annual precipitation trend has been decreasing and this trend is affecting the water availability of rainfed plants in the region and can decrease yield of crops. Although in the long term the average temperature trend is slightly constant, the yield has also declined with decreasing rainfall. Regarding the importance of wheat and lentil in arid lands of Zanjan plain, these two main rainfed crops were investigated in this study by determining the regression relationship between yield and evapotranspiration with precipitation and water stress coefficient. The results showed that the average evapotranspiration values of wheat and lentil for Zanjan plain during the study period were 398 and 262 mm and the mean temporal stress coefficients were 36 and 33%, respectively. Regression analysis showed the highest correlation between actual evapotranspiration and yield. The highest correlation was observed between temporal stress coefficient and yield of rainfed wheat. But, in lentils this trend was different and the correlation between water stress coefficient (Ks) and yield was more than the precipitation and lentil yield. For 25% decrease in precipitation, wheat yield reduced almost 18%. Generalizing this issue, the response to water stress coefficient of the product or Ky become 1.14. The regression equation also indicated that in case of being precipitation about 300 mm, wheat yield in the area will be near one tone. For 25% decrease in precipitation, wheat yield reduced about 22%. According to figures, the fitted model was significant and well-acted; and during the growth period of these crops, lack of water supply caused water stress as well as severe yield loss. The occurrence of high tension time percentage of 25% for wheat and 20% for lentil in all years of the study period confirms that the climatic potential of the region does not reach the standard requirements of plants, so it is not possible to achieve optimal performance. Hence, new methods of tillage, conservation agriculture and supplemental irrigation are recommended for achieving high yield, as well as the expected cultivation method for crops.

Soil temperature (Ts) is a key factor controlling the soil physical, chemical and biological properties and processes and consequently affects agricultural crop productions. The objective of this study was to estimate Ts from meteorological data using different machine learning methods. For this purpose, meteorological data and soil temperature at different depths (5, 10, 20, 30, 50 and 100 cm from the soil surface) for 25 years (1994-2018) were collected from 17 synoptic stations in Khuzestan province, Iran. Air temperature, wind speed, relative humidity, evaporation, precipitation, and vapor pressure were used as inputs to train the models. Multiple Linear Regression (MLR), Multilayer Perceptron Neural Network (MLPNN) and Extreme Learning Machine (ELM) models were used to predict soil temperature from metrological data. The results indicated that all models predicted temperature of the top layer (0-30 cm) better than the ones in sublayers. on the other hand, by increasing soil depth the accuracy of the models diminished; so that, the best and worst Ts predictions were belong to 5 cm and 100 cm depth, respectively. The results revealed that MLR, MLPNN and ELM models provided desirable performance in modeling Ts at all depths, with R2 values ranging 0.700-0.864, 0.967-0.997, and 0.967-99, RMSE values ranging 2.557–2.873, 0.034–0.072, and 0.028–0.078 °C, and MAE values ranging 1.398–1.529, 0.023–0.063, and 0.023–0.065 °C, respectively. Overall, the results showed that MLPNN and ELM models had approximately similar performance and better accuracy than MLR model. However, because of the high computational speed of the ELM model, it is recommended to use MLPNN model for estimation of soil profile Ts.

Whileland resource management needs detailed and accurate information about soil properties and distribution, this kind of data is limited in Iran. In this research, we tested performance of three digital soil mapping (DSM) approaches including Multiple Linear Regression (MLR), Cubist (CU) and Random Forest (RF) to map the spatial 3D distribution of soil organic carbon (SOC) in Saadat Shahr plain in Fars Province. Latin hypercube sampling (LHS) was used to determine locations of soil profiles in the field. The soil profiles were sampled and SOC was measured. Different environmental covariates including terrain attributes, remote sensing auxiliary variables, and maps of soil, geoform and distance from rivers were used in this research as auxiliary data. According to the link of the environmental covariates and soil organic carbon contents in the framework of each model in combination with equal-area spline algorithm, soil organic carbon maps were produced at five standard depths of soils in the whole study area. Model performance was evaluated by root-mean-square error (RMSE), mean error (ME) and normalized root-mean-square error (NRMSE). Among the used models, RF model showed the highest performance to predict organic carbon in depths of 0-5 and 60-100 cm. Meanwhile, MLR and CU had the lowest error for prediction in depths of 5-15 and 15-30 cm, respectively. In spite of these results, RF model was considered as the best model for its power to explain the spatial distribution of soil organic carbon in all soil depths in the study area.

To prevent water stress in plants and have sustainable water management in the field, fast and accurate determination of irrigation time is one of the most important issues. Measuring soil moisture and leaf surface temperature are two methods of determining time of irrigation. In this research, by combination of these two methods, a model for planning and management of forage maize irrigation (cultivar SC-701) is presented. The air relative humidity (RH) and temperature (Ta), leaf surface temperature (TL), and soil moisture content (SM) were measured in 2013 and, by using artificial neural network model and multiple stepwise method, a regression model was developed. Experiments were carried out in 2014 with five treatments including 100%, 85%, 75%, 65%, and 35% total available water (TAW), with four replications, Irrigation was carried out when soil moisture content reached the treatments moisture level. Measurements of the previous year were repeated and the model was calibrated. The results of the first year showed a correlation (R2=0.87) between the parameters RH, Ta, TL, Ta-TL as independent variable and SM as the dependent variable. Then, using three input parameters of air temperature, leaf surface temperature, and relative humidity, Determination Coefficient of soil moisture content model was calculated as R2= 0.92. In this model, soil moisture has an inverse relation with (Ta) and (TL-Ta) variables, but a direct relation with RH. Soil moisture content was compared using the model for the second year treatments and compared with the measured values. The difference in soil moisture content measured and estimated by the model at the peak solar radiation time (at noon) was less than ±10%. The model estimated 75% TAW treatment data well, with very small difference compared to the measured value.

The object of this research is the estimation of spatial distribution of three air temperature parameters at daily and monthly scales including minimum, maximum and mean temperatures at the Kurdistan province using MODIS sensor images setted on Aqua and Terra satellites. For this object, 8 synoptic stations at Kurdistan province were selected and for these 8 stations at 2013 and 2014, daily minimum, maximum and mean air temperature data and also land surface temperature at 4 daily times at these 8 stations for 2013 and 2014 years were extracted by processing on the MODIS sensor images. Afterwards, the multiple linear regression method was used to extract regional regression models for Kurdistan province between these 3 air temperature parameters and land surface temperature and to assesing the errors, cross validation based on the Mean Absolute Error, Mean Bias Error and Nash-Sutcliffe Efficiency coefficient was adopted. The results showed that it is a powerful relation between all of these 3 air temperture and land surface temperatures extracted by sattelite images. The results of cross validation showed an approperiate and reseanable agreement between the measured and estimated values of these 3 parameters at both daily and monthly scales so that the mean absolute error for minimum, maximum and mean temperatures were 2.7, 2.1 and 1.6 °C at daily scale and 1.9, 2. 1 and 1.1 °C at monthly scale, respectively. These results showed that it is possible to estimate these air temperature parameters at the places witout any meteorological stations with an approperiate and acceptable accuracy by extracting land surface temperatures of MODIS sensor for these places and applying the extracted regression models.

Statistical analysis and forecast discharge data play an important role in management and development of water systems. The most fundamental issues of statistical analysis and forecast discharge in Iran are lack of data in long term period and lack of stream flow data in gauging stations. Considering the issues mentioned in this study, we tried to estimate the daily data flow (runoff) of Santeh gauging station in Kordestan province using the nearby hydrometric and meteorological stations data. This estimation occurred based on the sixteen different input combinations, including data of daily flow of hydrometric stations Safakhaneh and Polanian and daily runoff in Santeh precipitation gauging station. In this research, the daily flow estimation of the Santeh station in each of the months of the year was evaluated for sixteen different combinations and artificial neural network models and multiple linear regressions. The performance of each model was evaluated with the indicators RMSE, CC, NS and t-student statistic. The results showed good performance of both models but the performance of the artificial neural network model was better than the regression model in estimation of the daily runoff in the most months of the year. Mean error of artificial neural network and multiple linear regression models was respectively estimated as 6.31 and 8.07 m3/s in the months of the year. It should be noted that the artificial neural network, for each sixteen combination used, had better result than the regression model.

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.
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.
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.
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.

Accuracy of the pedo-transfer functions can be improved using more flexible equations. The objective of this study was to compare pedo-transfer functions with different flexibility [e. g. multiple linear regression (MLR)، the physic-empirical model of Arya and Paris (AP)، artificial neural network (ANN)، and group method of data handling (GMDH)] for predicting soil water contents at field capacity and permanent wilting point. Pedo-transfer functions were developed from data of particle size distribution، organic carbon، bulk density، and water contents at 0،. 33 and 1500 kPa. The accuracy and reliability of the GMDH algorithm was superior to the other pedo-transfer functions for predicting the soil volumetric water contents at field capacity and permanent wilting point، because of fewer lower roots mean squared error (RMSE) and AIC criteria and more larger agreement index (D-index). It seems that the GMDH preference superiority is due to its higher GMDH capability to determine nonlinear and complex relationships between soil factors affecting factors the soil water contents at field capacity and permanent wilting points.

In order to assessment of water quality and characterize seasonal variation in 18O and 2H in relation with different chemical and physiographical parameters and modelling of effective parameters, an study was conducted during 2010 to 2011 in 30 different ponds in the north of Iran. Samples were collected at three different seasons and analysed for chemical and isotopic components. Data shows that highest amounts of δ18O and δ2H were recorded in the summer (-1.15‰ and -12.11‰) and the lowest amounts were seen in the winter (-7.50‰ and -47.32‰) respectively. Data also reveals that there is significant increase in d-excess during spring and summer in ponds 20, 21, 22, 24, 25 and 26. We can conclude that residual surface runoff (from upper lands) is an important source of water to transfer soluble salts in to these ponds. In this respect, high retention time may be the main reason for movements of light isotopes in to the ponds. This has led d-excess of pond 12 even greater in summer than winter. This could be an acceptable reason for ponds 25 and 26 (Siyahkal county) with highest amount of d-excess and lowest amounts of δ18O and δ2H. It seems light water pumped from groundwater wells with minor source of salt (originated from sea deep percolation) in to the ponds, could may be another reason for significant decrease in the heavy isotopes of water (18O and 2H) for ponds 2, 12, 14 and 25 from spring to summer. Overall conclusion of multiple linear regression test indicate that firstly from 30 variables (under investigation) only a few cases can be used for identifying of changes in 18O and 2H by applications. Secondly, among the variables (studied), phytoplankton content was a common factor for interpretation of 18O and 2H during spring and summer, and also total period (during a year). Thirdly, the use of water in the spring was recommended for sampling, for 18O and 2H interpretation compared with other seasons. This is because of function can be explained more by variables and there are more variables compare with other two seasons. Fourthly, potassium concentration in spring and total volume of water in summer would be most appropriate variables for interpretation of data during these seasons.

Climate variability is the most effective parameter on crop productivity especially rainfed crops.in this study, the anomaly index was calculated for meteorological elements such as maximum temperature(T max), minimum temperature(T min), precipitation(P) and crop yield(Y) for Razavi Khorasan province in Iran, which is located in arid and semi arid region of the world. Precipitation Anomaly Percentage (PAP) is a traditional drought monitoring index, it shows the precipitation as a percentage of the long-term average or normal Thepurpose of this study is finding significant relation between various indexes such as yearly precipitation anomaly (PA) with yearly maximum temperature anomaly(T max A) and yearly minimum temperature anomaly(T min A) for the period (1989-2005).At first, moving average of 5, years old,were examined for 10 stations. In addition, yearly precipitation anomaly, probability and return period of it were calculated using sigmoidfunction. Pearson correlation was examined to find relationship between change percent of yearly precipitation anomaly(PAP), yearly minimum(TminAP), maximum temperature anomaly(TmaxAP) and yearly rainfed crop yield anomaly of wheat (YAP). Multiple linear regression was calculate between YAP (as dependent parameter) and climate variability(indpendent parameter). Sarakhs, Ghuchan and Mashhad had significant correlation (P <0.05) 0.63,0.55 and 0.54 respectively. In other station, no relation found between PAP and YAP. A significant relation between TmaxAP and YAP was obtained in Mashhad (P < 0.05, r= -0.48) and TminAP and YAP was found only in Ghuchan(r=0.56).

The present study discusses a method used to produce updated information about vegetation cover in arid and semi-arid zones, using RS data and GIS technique. In this method, Landsat ETM+ data in 2002 was collected in an area of about 60000 ha in Nodoushan basin, Yazd, Iran. To collect the necessary ground data, 50 sites of different vegetation types were selected and the percentage of vegetation cover in each one was determined. Also, different vegetation and soil indices were derived and crossed with located sampling points using ILWIS software capabilities. To get the best fitted curve, the relationship between vegetation cover, as a dependent variable, and satellite data bands, vegetation indices and environmental factors, as independent variables were assessed. Therefore, a multiple linear regression model was established for the prediction of vegetation cover percentage in the studied area. Finally, a vegetation cover map with high a precision was produced. As a conclusion, it can be said that mapping of vegetation cover via remote sensing is possible even if its vegetation cover is sparse.