Mathematical modeling of the germination and growth of Leucaena leucocephala under different substrates and nursery conditions

Abstract

Livestock production in tropical regions is predominantly extensive and relies heavily on native or monoculture pastures, which often prove insufficient for ruminant nutrition. The incorporation of Leucaena leucocephala into silvopastoral systems represents a promising strategy due to its high forage quality; however, information on its early establishment under nursery conditions remains limited. This study aimed to model the germination dynamics and early seedling growth of L. leucocephala under different substrate compositions during the nursery phase. Germination percentage and daily plant height were recorded over a 30-day period. Treatment effects were evaluated using analysis of variance (ANOVA) and growth dynamics were described using non-linear sigmoidal models (Gompertz, Logistic, von Bertalanffy, and Brody). Significant differences in germination rate among substrates were detected (p<0.05), whereas no significant effect of substrate on plant height was observed during the evaluation period (p>0.05). Among the evaluated models, von Bertalanffy, Gompertz, and Logistic functions provided the best fit for plant height based on R² and AIC criteria. Although some models showed high R² values for germination, elevated AIC values suggest limited biological adequacy. These findings highlight the usefulness of predictive modeling to support nursery management decisions, optimize substrate selection, and facilitate the establishment of L. leucocephala in sustainable silvopastoral systems.