Investing in planting genetically improved silver birch (Betula pendula Roth) in Swedish plantations requires understanding how birch stands will develop over their entire rotation. Previous studies have indicated relatively low production of birch compared to Norway spruce (Picea abies (L.) Karst.) and Scots pine (Pinus sylvestris L.). This could result from using unrepresentative basic data, collected from unimproved, naturally-regenerated birch (Betula spp.) growing on inventory plots often located in coniferous stands. The objective of this study was to develop a basal area development function of improved silver birch and evaluate production over a full rotation period. We used data from 52 experiments including planted silver birch of different genetic breeding levels in southern and central Sweden. The experimental plots were established on fertile forest sites and on former agricultural lands, and were managed with different numbers of thinnings and basal area removal regimes. The model best describing total stand basal area development was a dynamic equation derived from the Korf base model. The analysis of the realized gain trial for birch showed a good stability of the early calculated relative differences in basal area between tested genotypes over time. Thus, the relative difference in basal area might be with cautious used as representation of the realized genetic gain. On average forest sites in southern Sweden, improved and planted silver birch could produce between 6–10.5 m3 ha–1 year–1, while on fertile agriculture land the average productivity might be higher, especially with material coming from the improvement program. The performed analysis provided a first step toward predicting the effects of genetic improvement on total volume production and profitability of silver birch. However, more experiments are needed to set up the relative differences between different improved material.
Timber production and profitability were evaluated for spontaneously-regenerated mixtures on two formerly clearcut areas. The abandoned areas developed into birch-dominated (Betula pendula Roth and Betula pubescens Ehrh.) stands with successional ingrowth of Norway spruce (Picea abies (L.) H. Karst.). An experiment with randomized treatments within blocks was established, using three management strategies and one unthinned control, resulting in variation in optimal rotation age, merchantable volume and species composition. The management strategies were evaluated based on total production (volume) by using measured growth data 42 years after clearcutting and the modelled future stand development. The long-term effects of spontaneous regeneration and management strategies were evaluated based on land expectation value (LEV) and compared with a fifth management strategy using artificial regeneration and intense thinnings. 12 years after treatment, at a stand age of 42 years, the unthinned control had produced the highest total stem volume. At interest rates of 2% or higher, the unmanaged forest was an economically viable strategy, even compared to an intensive management strategy with a preferred merchantable timber species. Interest rates clearly impacted the profitability of the different management strategies. This study shows that when spontaneous regeneration is successful and dense, the first competition release can have a high impact on the development of future crop trees and on the species mixture.
Genetically improved Norway spruce (Picea abies (L.) Karst.) and Scots pine (Pinus sylvestris L.) are used extensively in operational Swedish forestry plantations to increase production. Depending on the genetic status of the plant material, the current estimated genetic gain in growth is in the range 10–20% for these species and this is expected to increase further in the near future. However, growth models derived solely from data relating to genetically improved material in Sweden are still lacking. In this study we investigated whether an individual tree growth model based on data from unimproved material could be used to predict the height increment in young trials of genetically improved Norway spruce and Scots pine. Data from 11 genetic experiments with large genetic variation, ranging from offspring of plus-trees selected in the late 1940s to highly improved clonal materials selected from well performing provenances were used. The data set included initial heights at the age of 7–15 years and 5-year increments for almost 2000 genetic entries and more than 20 000 trees. The evaluation indicated that the model based on unimproved trees predicted height development relatively well for genetically improved Norway spruce and there was no need to incorporate a genetic component. However, for Scots pine, the model needed to be modified. A genetic component was developed based on the genetic difference recorded within each trial, using mixed linear models and methods from quantitative genetics. By incorporating the genetic component, the prediction errors were significantly reduced for Scots pine. This study provides the first step to incorporate genetic gains into Swedish growth models and forest management planning systems.