Temperature and water potential effects on Carthamus tinctorius L. seed germination: measurements and modeling using hydrothermal and multiplicative approaches

Abstract

Seed germination is a biological process that is strongly affected by temperature and water potential. Our objective was to measure experimentally and model this combined effect and estimate robust parameter values that will assist researchers to estimate safflower germination rate under variable experimental conditions. A laboratory experiment was conducted to investigate the combined effect of seven temperatures regimes (10, 15, 20, 25, 30, 35 and 40 °C) and five water stress levels (0, −0.4, −0.8, −1.2 and −1.6 MPa) on safflower seed germination. The derived dataset was analyzed using two modeling approaches that combine temperature and water potential effects: the multiplicative and the hydrothermal time models. The associated parameter estimates for each model were determined through statistical optimization and model performance evaluated against an independent dataset. The hydrothermal time parameters were 493.3 MPa h, 8.2 °C, and −1.34 MPa for θ HT (hydrothermal time constant) T b (base temperature), and ψ b(50) (median base water potential) in sub-optimal temperatures, respectively. The parameter estimates for the multiplicative model were determined as 7.9 °C for T b, 21.4 °C for T o1 (lower optimal temperature), 29 °C for T o2 (upper optimal temperature), and 40 °C for T c (ceiling temperature); 0 MPa for WPc (critical water potential) and 1.18 h−1MPa−1 for water potential sensitivity coefficient (WPS); and 17.9 h for g o (physiological hours for seed germination). Model evaluation showed that the multiplicative model predicted time to 50 % of seed germination more accurately (RMSE = 4.3 h and R 2 = 0.98) than the hydrothermal time model (RMSE = 9.5 h and R 2 = 0.93).