Nonlinear Modeling of Temperature-Driven Mycelial Growth Reveals Divergent Thermal Niches in Multinucleate and Binucleate Rhizoctonia Isolates

Temperature fundamentally governs fungal growth and pathogenic potential, yet conventional polynomial approaches often produce biologically unrealistic cardinal temperature estimates. Robust thermal performance characterization is crucial for disease risk prediction and elucidating the ecological adaptations of Rhizoctonia spp., a soilborne pathogen of substantial economic and ecological significance. We conducted a systematic comparison of 11 nonlinear regression frameworks to describe temperature-dependent mycelial growth dynamics across 17 isolates, encompassing 11 binucleate (BN) Rhizoctonia and six multinucleate (MN) R. solani anastomosis groups (AGs). We evaluated model performance using a multicriteria approach that combined goodness-of-fit statistics (adjusted R2, RMSE, SE) with information-theoretic measures (AICc, Akaike weights ωi). No single model proved universally superior. However, asymmetric models consistently outperformed symmetric ones in capturing nonlinear thermal responses. Thermal characterization using the best-fit models revealed divergent ecological strategies: BN Rhizoctonia isolates showed broad thermal tolerance ranges (base temperature, Tb: 5.43–13.86 °C; optimal temperature, Topt: 19.42–31.03 °C), indicative of generalist adaptation. Conversely, MN R. solani isolates exhibited restricted, elevated-temperature preferences (Tb: 7.18–15.47 °C; Topt: 24.70–28.39 °C), reflecting a specialized, highly aggressive pathogenic phenotype. Bootstrap resampling (n = 1,000) confirmed overwhelming statistical significance for all cardinal parameters (p < 10–9), with optimal temperatures exhibiting the highest precision (median SE = 0.28 °C). Our findings highlight the value of nonlinear, biologically grounded models─notably Segmented and Weibull formulations─for resolving thermal growth kinetics in Rhizoctonia spp. The multicriteria model selection strategy we present has wide-ranging applicability to ecophysiological investigations and facilitates climate-adaptive approaches to disease forecasting and integrated management.