Evaluation of Nonlinear Regression Models to Describe Seed Germination Response of Mountain Rye (Secale mountanum) to Temperature

Abstract

The present study sought to evaluate the effect of different temperatures on germination and to determine cardinal temperatures (i.e., base, optimum and maximum) of Secale mountanum at temperatures of 3, 5, 10, 15, 20, 25, 30 and 35 o C. Three nonlinear regression models (i.e., segmented, dent-like and beta) were used for quantifying the response of germination rate to temperature. The results showed that in addition to germination percentage, temperature has a significant impact on germination rate. Given the root mean square of errors (RMSE) of germination time, the coefficient of determination (R 2 ), the simple linear regression coefficients a and b, and the relationship between the observed and the predicted germination rates, the best models for determination of cardinal temperatures of Secale mountanum were dent-like and beta models. Base, optimum and maximum temperatures were estimated to be about 2.70 to 3.17, 21.27 to 30.00 and 35.00 to 35.05°C, respectively for the dent-like model. However, given the high value of SE for temperature base and negative estimate of the base temperature of the beta model, one can report the dent-like model as the right model. Therefore, by using the dent-like model and the estimated parameters, it is possible to use this model for predicting germination.