مقایسه روش رگرسیون خطی چند متغیره و روش RSM در تعیین داده های آب و هوایی روی نوسانات فصلی سن گندم Eurygaster integriceps Puton (Hemiptera., Scutelleridae)

The Sunn pest, Eurygaster integriceps Put. (Heteroptera: Scutelleridae) is the most important harmful insect in Iran, that feeds on the leaf, stem and seed of wheat and reduces the quantity and quality of the crop. If it is not managed, the amount of damage will reach 100%. This study aimed to predict population fluctuation of Sunn pest in the field using Response Surface Method (RSM) and Multiple Linear Regression (MLR) in 2015 and 2016, in two farms each with an area of one hectare in Chadegan city, Isfahan province. In this models sampling date, average temperature, average relative humidity, wind speed, wind direction, rainfall and height from sea level were the input variables and total growth stages were used as the output variables. After sensitivity analysis due to the easy of the model and extraction of effective factors, four factors including sampling date, average temperature, humidity and wind speed were selected. Data analysis in RSM by Design expert software was performed. Data analysis with MLR was performed by using SPSS statistics. In the response surface method, the best model was obtained from the Quartic type. The amount of coefficient of determination after sensivity analysis for predict the density of total growth stages in RSM and MLR was obtained 0.9 and 0.47, respectively. The amount of root mean square error for predict total growth stages in RSM and MLR was obtained 0.093 and 0.26, respectively. The results showed that RSM to predict population density total growth stages sunn pest was more fast and reliable methods than multi linear regression.