Main Article Content
Studies concerning the diameter structure are essential for understanding the development of native forests, provide growth and yield parameters sufficient to generate estimates that support sustainable management. However, the mathematical modeling of probability functions, such as density, becomes difficult to implement in multi-modal distributions, found in forest species and communities. The species Araucaria angustifolia (Bertol.) Kuntze, of social, environmental and economic features, presents multimodal distribution pattern forming units called demographic cohorts, which arise due to natural and anthropogenic agents that operate in niches. Therefore, taking inventory database of trees with a diameter at breast height equal to or greater than 9.5 cm (DBH≥9.5 cm), from the National Forest of São Francisco de Paula, RS, it was possible to test different probability density functions. The best fit to the data series consists of a truncated polynomial function of seventh degree, which besides maintaining fitted values very close to those observed for total and diameter classes, maintained the multimodal distribution characteristic.