Quantification of the Priming Effect of Canola (Brassica napus cv. Zafar) Response to Temperature Using Nonlinear Regression Models

Abstract

Introduction: Temperature is one of the primary environmental regulators of seed germination. Seed priming technique has been known as a challenge to improving germination and seedling emergence under different environmental stresses. Quantification of germination response to temperature and priming is possible, using non-liner regression models. Therefore, the objective of this study was to evaluate the effect of temperature and priming on germination and determination of cardinal temperatures (base, optimum and maximum) of Brassica napus L. Material and Methods: Treatments included priming levels (non-priming, priming with water, gibberellin 50 and 100 mg/l) and temperature (5, 10, 15, 20, 30, 35 and 40 °C). Germination percentage and time to 50% maximum seed germination of Brassica napus L. were calculated for different temperatures and priming by fitting 3-parameter logistic functions to cumulative germination data. For the purpose of quantifying the response of germination rate to temperature, use was made of 3 nonlinear regression models (segmented, dent-like and beta). The root mean square of errors (RMSE), coefficient of determination (R 2 ), CV and SE for the relationship between the observed and the predicted germination percentage were used to compare the models and select the superior model from among the methods employed. Results: The results indicated that temperature and priming were effective in both germination percentage and germination rate. In addition, the results showed that germination percentage and rate increase with increasing temperature to the optimum level and using priming. As for the comparison of the 3 models, according to the root mean square of errors (RMSE) of germination time, the coefficient of determination (R 2 ), CV and SE, the best model for the determination of cardinal temperatures of Brassica napus L. for non-primed seeds was the segmented model. For hydro-priming and hormone-priming with 50 mg/l GA, the best models were segmented and dent-like models and for hormone-priming with 100 mg/l GA, the dent-like model was the best. The results showed that for non-priming, hydropriming with water, gibberellin 50 and 100 mg/l treatments, the segmented model estimated base temperature as 3.54, 2.57, 2.34 and 2.34 °C and dent-model estimated base temperature as 3.34, 2.45, 2.21 and 2.83 °C, respectively. The segmented model estimated optimum temperature as 24.62, 23.23, 23.69 and 24.38 °C. The dent-model estimated lower limit of optimum temperature and upper limit of optimum temperature as 20.01, 19.62, 16.25, 19.87 and 28.81, 27.38, 29.58 and 27.31 °C. Conclusion: Utilizing non-liner models (segmented, dent-like and beta) for quantification of germination of Brassica napus L. response to different temperatures and priming produced desirable results. Therefore, utilizing the output of these models at different temperatures can be useful in the prediction of germination rate in different treatments.