stepwise regression

To evaluate the genetic diversity and relationships among traits of herbaceous milkvetch, 10 genotypes belong to five species including Astragalus vegetus. A. brevidens, A. brachyodontus, A. effusus and A. cyclophyllus in Sanandaj in a randomized complete block design with three replications were studied. Seed sowing was done in November 2018 and morphological traits of plant in spring 2020 were measured. Analysis of variance released significnant (p≤0.01 and 0.05) difference for milkvetch populations (G) for all studied traits. In 2020 (two years old plants), the mean of plant establishment was about 26%, the number of primary branches was 3.2, plant height, the tallest stem length and crown diameter were 16.9, 19.2 and 15.4 cm respectively, the length and width of the root were 17.4 and 6.2 cm, respectively, the average of fresh weights of the whole plant, arial and root parts were 2.3, 1.9 and 0.4 g and the average of dry weights of the whole plant, arial and root parts were were 0.9, 0.7 and 0.2 g respectively, average dry to fresh weight ratio was 0.38. Average of leaflet per leaf was 7.7, so mean of length and width of leaf and leaflet were 7.5, 3.7, 2.3 and 0.5 cm respectively. Up to this stage, the accession of A. effusus Darreh Shohada (West Azerbaijan) and three accession of A. vegetus (Zereshk of Qazvin, Saral of Kurdistan and Qarehbagh of West Azerbaijan) were superior ones in terms of forage production and other studied traits. The weight of the plant was positively correlated with plant vigor, length and width of leaf (p≤0.05) and with plant height, crown diameter, number of primary branches, length and width of root, arial and root weight at 1% probality level. The results of stepwise regression of forage yield on other traits showed that plant weight, establishment percentage, plant height and crown diameter at 1% and the length of leaf and root at 5% probality level (R2adj = 91.1%) were forage yield components. Therefore, in order to achieve higher forage, the genotypes with larger aerial parts and deeper roots should be selected.

Increased concentration of atmospheric carbon dioxide due to human activities has resulted in accelerated global warming process. Forestation is the most effective way to absorb atmospheric carbon dioxide and store it in terrestrial ecosystems in order to reduce and mitigate global warming. This study aimed to investigate the comparison of soil carbon sequestration in various stands in ​​Kuhdasht Aquifer and to present the best regression model for carbon sequestration based on all soil characteristics. In each stand (20-year old) of Ficus carica, Punica granatum, Pistacia vera, Amygdalus lycioides and Cupressus arizonica species as well as control rangeland, a number of 10 sample plots (5×5 m) were randomly selected and soil samples were taken in each plot at 0-10, 10-30 and 30-50 cm depths. All soil samples were transferred to the laboratory in order to measure soil characteristics including soil texture, organic carbon, bulk density, electrical conductivity, lime percentage and soil acidity. The results showed that the highest value of carbon sequestration in soil of Pistacia vera stand (54.94 tha-1) significantly (P <%1) higher compared to other stands, followed by Ficus carica (50.23 tha-1), Amygdalus lycioides (31.53 tha-1), Punica granatum (27.09 tha-1), Cupressus arizonica (24.17 tha-1) and control rangeland (9.01 tha-1) stands. Results also showed significant differences (P <%1) between the studied stands in terms of soil texture, acidity, organic carbon, electrical conductivity and soil bulk density. Also, the result of stepwise regression indicated that soil texture and acidity were the most important components affecting soil carbon sequestration, respectively.

Identification of climatic characteristics affecting the annual yield of Rheum Ribes can be useful in management and development of this species in the rangelands. In this research, the annual yield of this species in Khorasan-Razavi province based on 74 climatic parameters during a ten-year period evaluated and affecting climatic parameters extracted using data mining methods. First, the role of climatic parameters associated with temperature, humidity, rainfall and sunny hours analyzed using correlation and regression methods. Then, 11 classification algorithms in MATLAB software programmed and compared. The results showed that the Rheum Ribes yield has a positive relationship with the average of maximum temperatures in the summer, the range of high temperature in May to September, the maximum of summer temperatures and the relative humidity and rainfall of the spring. Evaluation of the algorithms using the indices of coefficient of determination and mean square error showed that in estimation of the annual yield based on climatic factors, the pattern recognition method at the testing stage with a coefficient of determination equal to 0.46 and regression methods, classification discrimination and K nearest neighbor (KNN) at the training stage (coefficient of determination equal to 1) had the best performance. With regard to the effective factors in stepwise method, the linear regression method at the testing stage (coefficient of determination equal to 0.74) and K nearest neighbor method at the training stage with coefficient of determination equal to 1, estimate the Rheum Ribes yield more accurately. Also, the proposed K nearest to mean (KNM) method for k values equal to 6 and 7 with regard to all factors and the effective factors resulted from stepwise method, respectively, had higher accuracy in yield estimation. So, application of data mining methods and the proposed model, in recognition of climate parameters affecting different rangeland species could be a practical approach.

Droughts are intrinsically regional. The critical conditions in the dry periods occur when there is a severe shortage of water resources for a long period on a wide area. Therefore, an investigation of the regional and not of a local drought offers better understanding of this phenomenon. This study aims at a regional analysis of the hydrological drought in East Azarbaijan Province for a period of 45 years (1972– 2017) using the data collected at 25 hydrometric stations with proper spatial distribution across the province. Homogeneous hydrological areas was identified based on hierarchical analysis method in the SPSS software using the following independent variables: Area, Perimeter, Length of the main basin, Main channel gradient, Drainage density, Basin shape factor, Total hydrographic network, Weighted average slope of the basin, Maximum, Minimum and Median height of the basin, and its length. Frequency analysis and probability distribution of the dependent variables (duration and intensity of hydrological drought spells) was performed on the basis of the two year return period for each hydrometric station. The factor analysis was performed using the PCA method in the SPSS software. The suitability of the data for the factor analysis was evaluated using the KMO coefficient. The value of this coefficient was estimated as 0.546, which indicates that the data were suitable for factor analysis. The result of factor analysis showed that the contribution of perimeter, height and shape of the basin were 40.13, 26.42 and 13.12% (Sum 79.68 % ) respectively in explaining the variance of the variables (i.e. the most important factors in zoning). Results of regional analysis using multivariable regression (stepwise method) indicated that perimeter, avrage height and the basin shape coefficient have a decisive role in estimating the drought severity and its duration with two-year return period in East Azarbijan. In the scarcity of hydrometric stations in most of the watersheds of the East Azarbaijan province, the regional equation presented in this study can be used to estimate hydrological drought characteristics of watersheds without observed surface runoff data.

Aggregate stability is a fundamental property influencing upon soil erosibility and main hydraulic and physical characteristics. Knowledge of soil components controlling aggregate stability is very important for soil structure conservation. Also, identification of the critical soil properties affecting aggregate stability is central to understanding where erosion might be located within a catchment as a result of natural patterns in aggregate stability. The objective of this study, carried out in surface soils of Taleghan, was to investigate the relationship between aggregate stability and some selected soil properties. In this study, based on homogeneity in slope, aspect, and lithology work units were prepared, and then by relying on results of geostatistical studies in homogeneous work units in soils of Taleghan watershed, with an area of 3260 hectares, 84 points were selected as the soil sampling points. The applied indices of stability were the mean weight diameter (MWD) and wet aggregate stability (WAS). Correlation between mean weight diameter and measured properties were extracted using Pearson correlation coefficient and evaluated in terms of their statistical significance. Then, based on the results of correlation and using stepwise regression, models of aggregate stability prediction were developed. The unexpected results have been obtained from correlation between soil properties and aggregate stability so that the correlation between aggregate stability and the percentage of Caco3, clay and saturation were negative while that of fine sand and very fine sand were positive. Thus, the improvement of aggregate stability and soil erosion control in Taleghan watershed depends on management of these soil properties and soil conservation projects in order to adjust and optimize them.

The aim of this study was to determine some factors affecting dust storms phenomenon using different methods. In order to determine the best-input combination, variable reduction techniques such as factor analysis (maximum likelihood, principal component analysis), Gama test, and multivariate forward regression analysis were used. Each of these methods presented different combinations used by feedforward neural network model, with Levenberg–Marquardt algorithm and multivariate forward regression with R²=0.87 and RMSE=0.04 was selected as the best suitable combination of neural network model. In addition, monthly and seasonal data were applied by neural network using the best-input combination, and the simulation of dust storm phenomenon was done in summer and spring during the months of April, May, June, July, August and September with a higher correlation coefficient and lower mean square error, due to the good distribution of the dust storm data. The results showed that based on these methods used in this study, dominant wind speed, horizontal visibility, continuity and average of wind speed were the most important factors affecting dust storm phenomenon in Yazd province.

In order to reduce sediment and soil conservation plan generation، the design and calculation of the exact size of the dam at dams، evaluation and estimation of sediment yield in the watershed of the dam، is essential. In general erosion and sediment transport of the most complex issues that determine the exact hydrodynamic equations for the various variables، it is not easy. The aim of this study was to simulate suspended sediment load using a neural-fuzzy model is combined This is necessary for the various input combinations using stepwise regression، genetic algorithm model and gamma test and then determine the most suitable model for the input. Karaj Dam Sierra station data in the period 1974-2011 (37 years) was used so that the 15 variables as input and output variables as was considered. Results indicated neuro-fuzzy models in a suitable combination of the number of training data to select the most appropriate modeling of the genetic algorithm to be achieved. It should be noted that in the case of all input variables، modeling neuro-fuzzy model was not appropriate Therefore، methods are necessary to determine the appropriate combination of inputs. Finally، the Genetic Algorithm model as the most appropriate method and stepwise regression، second، and third order gamma test as appropriate methods to determine combination of neuro-fuzzy model inputs selected. The neuro-fuzzy model when using a selected combination of genetic algorithms to simulate suspended sediment load 99/0 correlation coefficient and the mean square error as the most appropriate modeling 0000009/0 has been selected.

Soil carbon sequestration and storage is one of the important methods to reduce the atmospheric carbon dioxide and mitigating climate change effects. Therefore, with identification of the effective factors on carbon storage can be helpful to improve soil fertility and environmental protection. Therefore, this study was conducted to investigate the effects of some soil physical factors on soil organic carbon (SOC) storage content in drylands of Sarfiroozabad watershed, Kermanshah province, Iran. So, the watershed was separated into homogeneous zones and soil sampling was performed using a randomized systematic method in the units. The samples were transported to the laboratory and texture, TNV%, gravel% and organic carbon were analyzed to determine the SOC storage. Then, using stepwise regression, cluster and factor analysis, the effective factors on carbon storage were studied. The results of cluster analysis showed three variables such as TNV%, Silt% and SP%, are effective on SOC storage and the regression model with this three variables can predict 33 percent of the soil carbon storage variability with a correlation coefficient of 0.574. Also, cluster analysis to predict the effects of physical factors on carbon storage is important with EF= 0.32. Therefore, it was proven that the soil factors are more effective than climate and topography factors to predict SOC storage. It shows that improving the soil conditions and correct management, to keep soil carbon stocks should be given more attention.