Wheat yield prediction modeling by soil properties: a case study in orth-west of Iran

Crop yields are dependent on a number of factors such as soil type, weather conditions and farmingpractices. crop yield estimates in different soil types are required to meet the needs of farmers, landappraisers, and governmental agencies in iran as around the world. this study was conducted to modelthe wheat-grain yields [triticum aestivum l.] by soil properties in khoy area, the north-west of iran.the wheat yields (mean of 5 years) were applied to predict and model the wheat yields under anaverage level of management used through the area. the prerequisite data on main soil physicochemicalcharacteristics was collected and measured to clarify the correlation and multiple regression analysiswhich are used to establish the relationships between the soil properties and the wheat-grain yields.based on the calculated soil index, the general equation (ge) taking the soil index ranging from 0 to100 % into account was proposed to predict the wheat-grain yields applicably. the results hereinmarkedly proposed other two regression equations for the areas having soil index higher and lower than70 %, respectively. the results indicated that within three obtained regression models, the equationsuggested for the area having soil index higher than 70 % is appreciably more accurate than the modeloutlined by the fao and potentially could be recommended for predicting the wheat yield in study area.moreover, the ge regression model and the proposed model for the area having the soil index lowerthan 70 % showed the same accuracy compared with the fao model but calibrated based on the studyarea condition. therefore, our proposed regression models for the wheat-grain yields prediction couldbe used instead of performing the fao models across the country with approximately same soil andclimate status.