Comparison of probability distribution functions in thermal-time models for modeling of spring oilseed rape germination to temperature

The models based on thermal-time concept are useful tools for predicting germination in relation to time and temperature. In this study, conducted in 2016 at Ramin Agriculture and Natural Resources University, thermal-germination model was developed based on seven probability distribution function (Logistic, Weibull, Gumbel, Loglogistic, Inverse-Normal, Log-Normal and Gamma) and predicted germination time courses by these models for two spring oilseed rape cultivars (RGS003, Sarigol) were compared with the Normal thermal-germination outputs. Germination test were conducted at eleven constant temperature regimes of 8, 12, 16, 20, 24, 28, 32, 33, 34, 35 and 36 ºC. Results indicated that the Log-Normal thermal-germination model gave best fit to germination time courses of both cvs. RGS003 (AICc=-1173) and Sarigol (AICc=-1180). Based on the outputs of this model, base temperature for germination of cvs. RGS003 and Sarigol were estimated to be 5.85 and 5.60 ºC, respectively. The suboptimal thermal-time to initiate germination were predicted as 118.40 ºC h in cv. RGS003 and 120.00 ºC h in cv. Sarigol, While thermal-time required to complete germination at supra-optimal temperatures were estimated to be 29.07 ºC h in cv. RGS003 and 31.47 ºC h in cv. Sarigol. Also, both oilseed rape cultivars showed thermoinhibition beyond averaged temperature of 33.17 ºC. Estimated parameters in this study can be used in crop simulation models.