How important are provenance and pedigree in later cycles of tree breeding? an example using Acacia auriculiformis

Abstract Low input breeding programs, based primarily on open-pollinated mating, have been operating for Acacia and Eucalyptus species over the last thirty years. These programs have now progressed to later-generation breeding cycles after one or more rounds of selection, breeding and infusion of additional wild or exchanged later-generation selections. They typically lack the capacity for advanced genetic analysis that incorporates extensive marker-based genotyping. Using an Acacia auriculiformis breeding program in Vietnam as a case study, we compare simple models of advanced-generation breeding populations with an individual-tree model that incorporates all available pedigree information. We also examine the importance of including ancestral provenance in these models. Although laborious, assembling the pedigree was worthwhile in this example: we advocate commencing this work as soon as possible after domestication commences while records are more readily accessible. A key benefit is more realistic estimation of breeding values and genetic parameters. Additionally, access to the pedigree allows calculation of group co-ancestry and status number, enabling better management of inbreeding and the accumulation of relatedness in the breeding population. These benefits will likely increase in subsequent generations as kinship accumulates in the breeding population. In the absence of pedigree records, inclusion of ancestral provenance in family models can give acceptable estimates of heritability.