Optimisation
Tree breeders make efforts to get their operation efficient by optimising tree breeding. Scientists develop tools aimed at improvement of the efficiency of tree breeding programmes. Optimising can mean adapting strategies and methods to certain species, group of populations, structure of genetic variation and mode of inheritance of the important traits to obtain the highest benefit per unit of time. Optimising is usually carried out at the following levels: breeding strategy (appropriate intensity of breeding, breeding population structure and size, plan for maintenance of genetic diversity), breeding methods (mating type, testing and selection methods, testing population size and time) and deployment methods of the genetically improved material (seed orchards and clonal forestry: genetic contribution, size). Computer simulators are frequently used: stochastic – based on defined and random algorithms; and deterministic – based on defined algorithms.
Contents
Selection strategies have been compared for annual progress in long-term breeding at a given annual cost considering genetic gain, gene diversity, cost components, and time components. For Norway spruce it seems favourable to clone full sib families and then select based on clonal performance while for Scots pine a two-stage strategy seems best, first phenotypic pre-selection and then progeny-testing the selections.
Tree improvement
A genetically variable population and a method of selecting genetically superior individuals provide the basis for tree improvement by breeding. In essence, a tree improvement program sets out to isolate and evaluate the genetic component of variation in one or more characters of interest. In the simplest procedure, cycles of selection reduce the available population in a particular direction to enhance desirable traits, then breeding from selections to expand the population with improved characteristics. Breeding strategies vary with species and objectives, but all use mating designs to generate information and new material. Choice of a suitable breeding strategy and mating design is a key decision in any breeding program. Kiss (1986) used a 2-level design in British Columbia to study variation within and between separate populations of white spruce, both within British Columbia and from eastern North America.
The breeding program for white spruce initiated in 1986 by the Canadian Forestry Service in the Maritimes employed 2 kinds of mating: polycross, to test clones for general combining ability; and pair-mating, to generate material for second generation selections (Fowler et al. 1988).
Newton’s (2003) systematic review of yield responses of white spruce and 3 other North American conifers to forest tree improvement practices indicated that correct provenance-progeny selection could yield juvenile height growth gains of about 12% at 20 years for white spruce, and a corresponding merchantable productivity (mean annual merchantable volume increment) gain of 26% at 50 years for plantations established at nominal initial densities on medium-to-good quality sites. Also, preliminary estimates derived from individual case studies indicated that first generational selection strategies for white spruce could increase merchantable productivity by approximately 20% at 45 years.