Nonlinear Regression Analyses of Sugar Beet Germination Parameters under High Temperatures

Global warming is a serious problem in many areas of the world, including Iran. Sugar beet (Beta vulgaris) is a strategic crop for Iran with an important role in sugar production. The purpose of this study was to investigate the germination behavior of sugar beet cultivars at high temperatures. In addition, the accuracy of different regression models in predicting maximum germination, temperature to reach 50% of maximum germination, and optimum germination temperature, was compared. Germination tests were conducted using four cultivars, Aria, Paya, FD-415 and Rosaflor, at constant temperatures ranging from 20–44 °C, with 3 °C intervals. Percentage germination and seed vigor index were calculated, and radicle, plumule and seedling length were measured. Different Hill, sigmoid, logistic, Gompertz, symmetric and Gaussian models were used to predict germination characteristics. All germination characteristics decreased with temperature increase. Paya was the most resistant to high temperature. The highest optimum temperature for germination, predicted by the 4-parameter symmetric model, was 24.80 °C for Paya, and the lowest was for Aria (21.64 °C). Optimum temperature ranges predicted by 4-parameter symmetric and two types of Gaussian models for radicle length, plumule length and seedling length were higher than predicted optimum temperatures for germination. Nonlinear models with higher parameters were more accurate. Use of regression models was reliable for screening sugar beet germplasm resistant to heat at the germination stage.