Effectiveness of competition indices for understanding growth in an overstocked stand

•Competition indices were evaluated to explain tree growth.•Distance-independent indices were more sensitive.•Including crown variables did not improve prediction ability. The objective of this study was to select and discuss indices to explain competition among trees in an unmanaged stand in order to support decision-making in forest management. The work was carried out in a plantation of Araucaria angustifolia (Bertol.) Kuntze located in a national forest in southern Brazil. The plantation was established in 1946 in 2 × 2 m spacing, with only one thinning that happened sometime between 1970 and 1980. Forty-six sample plots were established using an adaptation of the Prodan method. We tested distance-independent and distance-dependent competition indices that included or excluded crown area variables. An equation was generated to describe the relation between diameter at breast height and crown diameter, since these were highly correlated (r = 0.87); this permitted calculation of the growth space required for the trees in the stand. It is important to carefully choose the variables and competition indices to be tested in each study case. Including crown variables did not improve prediction ability to overcome the performance of the Stage distance-independent index. In general, the distance-independent indices performed better than the distance-dependent ones, even the stand arrangement in our study area was modified through thinning, mortality, and natural regeneration. When the goal is to better understand the ecological significance of a competition index and not only its statistical ability to predict short-term growth, simpler indices are recommended.