Variation in competition in eucalypt forests: the importance of standardization in pattern analysis

Abstract

Abstract. Contrasting views have been put forward that competition intensity should either increase with community standing crop or remain constant. Testing these hypotheses with long‐lived plants in natural communities is possible using pattern analysis, involving statistical summaries of the relationships between the total size and distance of nearest neighbors. This approach, however, may produce a spurious negative correlation between competition intensity and standing crop in cases where individual size increases with standing crop. A method for overcoming this problem using data standardization is illustrated using data from 15 stands in a variety of forest types (subalpine woodland, dry forest, warm wet forest, cool wet forest, and near‐rainforest) in southeastern Australia. Standing crop, estimated from stand volume, and water availability, estimated from evapotranspiration, varied significantly among forest types. Total size and nearest neighbor distance was measured for 50 pairs of trees in three stands of each forest type. The sum of nearest neighbor sizes increased significantly with interneighbor distance in all forest types, indicating that competition contributed to pattern at all levels of standing crop. The slopes of the regression equations describing these relationships varied significantly among forest types and with evapotranspiration, suggesting that competition intensity varied. This result, however, was shown to be the product of a significant positive correlation between individual size and standing crop. Once this correlation was removed by using standardized measures of tree size, there was no evidence for variation in competition intensity among forest types or with evapotranspiration.