Author(s): KLIMAS, C. A.; KAINER, K. A.; WADT, L. H. de O.; STAUDHAMMER, C. L.; RIGAMONTE-AZEVEDO, V.; CORREIA, M. F.; LIMA, L. M. da S.

Summary: During 5 y, we monitored reproductive activity and seed production of Carapa guianensis in two forest types to test the hypothesis that seed production is influenced by multiple factors across scales (regional climatic cues, local habitat heterogeneity and individual tree attributes). Variability in seed production was moderate at the population (CVp =1.25) and individual level (xCVi =1.24). Amixed modelwith a Poisson regression revealed that seed production was explained by variables at all scales. Total seed production was significantly higher in occasionally inundated forests. Diameter at breast height, dbh2, crown cross-sectional area, liana load, density, dry-season rainfall and mean maximum temperature were also significant in explaining seed production variation. Seed production increased with dbh until 40?50 cm, then decreased. Liana load demonstrated a negative relationship with seed production, but only in terra firme forests. Climatic cues (rainfall and temperature parameters) were central to setting overall patterns in reproductive activity and seemed to best explain why years with high seed production were consistent across the two forest types (habitats) examined. Dry-season rainfall was positively correlated with seed production.

Publication year: 2012

Types of publication: Journal article

Unit: Embrapa Acre

Keywords: Andiroba, Análise estatística, Análisis estadístico, Carapa guianensis, Clima, Desarrollo estacional, Desempeño del crecimiento, Distribución de Poisson, Erva daninha, Fator de crescimento, Fenologia, Fitomejoramiento, Floresta, Growth performance, Liana, Phenology, Plant breeding, Poisson distribution, População de planta, Producción de plántulas, Produção de sementes, Regressão de Poison, Reprodução vegetal, Seasonal development, Seedling production, Statistical analysis, Temperatura