Seed production, infestation, and viability in Acacia tortilis (synonym: Vachellia tortilis) and Acacia robusta (synonym: Vachellia robusta) across the Serengeti rainfall gradient

Abstract

Tree recruitment in savannas proceeds in multiple stages characterized by successive filters occurring at the seed and seedling stages. The “demographic bottleneck” hypothesis suggests that such filters ultimately restrict tree density and prevent trees from dominating grasses in savannas, but many of the demographic transitions underlying this assumption have not been quantified. We investigated how short- (1–2 years) and long-term (40 + years) rainfall patterns influenced seed production, infestation, and viability for two dominant species, Acacia robusta and Acacia tortilis across the Serengeti ecosystem mean annual precipitation gradient over a two-year period. We found that neither production, nor infestation, nor viability was influenced by rainfall. Pod production differed between species and increased with tree height in A. robusta. Mean infestation proportion in 2013 was higher (mean ± SE; 0.28 ± 0.08) in A. tortilis than in A. robusta (0.11 ± 0.05) but the trend reversed in 2014, when A. tortilis (0.33 ± 0.10) had lower infestation than A. robusta (0.61 ± 0.09). Under laboratory conditions, A. tortilis and A. robusta seeds had maximum germination (= viability) proportions of 70 and 20%, respectively. Mean seed viability was more than five-fold higher (0.46 ± 0.19) in A. tortilis than in A. robusta (0.08 ± 0.10). Our study has produced important estimates for seed stage demographic dynamics that can be used for modeling tree dynamics in Serengeti system, and savannas in general.