Thinning strategies for Eucalyptus dunnii population: balance between breeding and conservation using spatial variation and competition model

Abstract

For continuous genetic gains over time, a balance between genetic gain and maintaining the genetic base must be a constant concern of forest breeders. This study aims at determining the best thinning strategies for a population of Eucalyptus dunnii, by incorporating the effects of environmental heterogeneity and competition in the analysis, as well as the best growth trait regarding precision and accuracy. The population studied consisted of 160 open-pollinated families. The survival and growth (height, HT; diameter at breast height, DBH; and volume, VOL) were evaluated 4 years after planting. The growth rate data were analyzed and compared by four mixed models. Selection and thinning strategies were simulated by varying the number of families, individuals within families, and selected individuals, considering the estimated genetic gains and the effective size. The species showed good survival (89.7%) and productive performance (mean annual increment = 42 m3 ha−1 y−1). The Spatial+Competition Model provided the best fit for DBH and VOL. The strategies that allow a balance between improvement (genetic gains) and genetic conservation (effective size) consist of keeping 36 to 50% of the individuals in the test (370 to 510 trees ha−1), by reducing more intensively the number of individuals from the worst-performing families. The selection of 100 individuals with a restriction of at most one individual per family generates the largest number of effective size (Ne), with more than double the Ne obtained without restricting the individuals per family, with a small drop in genetic gain.