Assessment of Non-Linear Regression Models to Evaluate Response of Seedling Emergence Rate to Temperature in Three Oil Seed Rape Seed Cultivars

In order to evaluate the response of emergence rate of oil seed rape seedlings to temperature, a pot experiment was conducted to assess the relationship between emergence rate of three rape seed cultivars (apetalous flower, RGS003 and Syn3) by using a non-linear regression against temperature variation. Six models (tooth shape, segmented, beta, curve, quadratic and cubic) were used and the superior model was selected according to square root of MSE, coefficient of determination, coefficient of correlation and simple linear regression coefficients (a, b) between observed and predicted emergence rates. Results showed that the seedling emergence response to temperature explained by beta function in RGS003, by beta and cure functions in Syn3, and by segmented function in apetalous flower cultivar. Prediction of cardinal temperatures by using the best models showed that the base temperature varied between 7.89 and 10.94, the optimum temperature was between 20.1 and 24 and the ceiling temperature was between 36 and 37.19°C for rape seed cultivars. The biological day estimation for seedling emergence by using segmented and beta models showed that there were significant differences between rape seed cultivars. According to the results of this experiment, beta and segmented models could be used for quantification of seedling emergence response of rape seed cultivars to temperature. It seems that these models and parameters (base, optimum and ceiling temperatures) could be used for prediction of seedling emergence in the field. Cardinal temperatures and its related model could be used for modeling of seedling emergence of rape seed cultivars.