Fig. 1: Reliable data is available for the Douglas fir as introduced tree species, which can therefore be seen as suitable object for scientific conclusions. (photo: FVA/Haas)
Fig. 2: Position of the experiments with introduced tree species in Baden-Württemberg.
The state forest of Baden-Württemberg has been managed for decades according to the principles of close-to-nature forestry, which includes the prioritisation of native tree species. Inversely, this implies a purposeful bias against introduced tree species. However, when considering climate change, the long-term suitability of native tree species may be questionable.
The experimental cultivation of introduced tree species has a long tradition in Baden-Württemberg. Beginning in 1840 such plantations were started in the former forest district of Güglingen. More extensive trials were installed in Nagold, Weinheim, Breisach and Reutlingen. Generally, the impetus can be traced back to the commitment of individual officers, and their continuance was strongly dependent on the interest of their successor, so that the data quite often did not fulfil scientific standards.
Compared to these practice trials, the experiments with introduced tree species that are available at the department of growth and yield of the FVA have the advantage of continuous documentation and exact data recordings. To scrutinise the suitability of these data from a statistical point of view on one hand and to get an idea about their growth performance on the other, these experiments were recently analysed, and their growth performance compared to that of the main tree species located in the direct neighbourhood of these experiments.
Data
The data base comprises more than 350 experimental trials with 40 introduced tree species. For scientifically sound conclusions, the data had to satisfy certain defined requirements, which was only the case for 15 tree species (fig. 2). Reliable data are available for Pseudotsuga menziesii, Quercus rubra and Larix kaempferi with 67, 49 and 40 experimental plots respectively. For Abies grandis, Picea sitchensis, Thuja plicata, Chamaecyparis lawsoniana and Pinus nigra between 10-17 experimental plots could be used in the analysis, whereas there were only for four to eight plots of Carya ovata, Pinus strobus, Abies nordmanniana, Picea omorika, Sequoiadendron giganteum, Juglans regia, and Liriodendron tulipifera. The principle relationships between age and top height as well as between top height and total volume production that were found for these tree species are therefore only of limited validity.
Results
Using experimental data for each tree species, regression analyses were performed on top height versus age and total volume production versus top height. The resulting equations allowed us to construct top height and age related site index curves on one hand and the yield comparison of some of the main tree species (comparative tree species) on the other.
For example, the site index curves of Abies grandis, plotted together with the top height development of the experimental plots, is presented in fig. 3. Fig. 4 shows the relationship between total volume production and top height. For Abies grandis, Larix kaempfferi and Quercus rubra the existing top height- and total volume increment relationships could be updated and improved; for other tree species, this relationship was made available for the first time.
Fig. 3: Top height development of Abies grandis and top height site index curves (reference age 50 years).
Fig. 4: Total volume production versus top height of Abies grandis.
Fig. 5: Top height site class of the comparative tree species in relation to the introduced tree species.
Fig. 6: Total volume production of the comparative tree species in relation to the introduced tree species.
The yield comparison shows that Abies grandis and Pseudotsuga menziesii are clearly superior to the comparative tree species regarding the top heights reached at the reference age (fig. 5) as well as the total volume production (fig. 6) (with Abies grandis being slightly superior to Pseudotsuga menziesii). Whereas the top height site index of Larix kaempfferi under similar growing conditions is comparable to Picea abies and Abies alba. However, total volume production is significantly lower. Picea sitchensis and Abies nordmanniana lag behind Picea abies and Abies alba in terms of site index and total volume production. Regarding both parameters, Quercus rubra equals the level of Fagus silvatica and is superior to Quercus robur by 20% (top height at the reference age) and 40% (total volume production). Unfortunately, sufficient data was not available for a yield comparison of the other introduced tree species.
Conclusions
The study confirms the superior yield performance of Abies grandis, Pseudotsuge menziesii and Quercus rubra, which has already been stated by other authors. Due to their impressive growth potential, these tree species are generally acknowledged as possible growing alternatives. It is difficult to predict how this assessment may develop under the ongoing trend in climate change, especially when one considers that the estimation of the viability of a tree species can also change under current environmental conditions. Without doubt, the growth potential of Abies grandis is impressive, but there are hints that this species is considerably exposed to risks due to droughts and fungi.
Moreover, this study demonstrates that experiments that are scientifically worthwhile require much time, space and, finally, money. It might be disappointing that only 15 of the 40 introduced tree species could be evaluated in a statistically sound manner since the data base contained useful data from only a few tree species. This underlines the big challenge for implementing and documenting long term experiments with its attendant high risk of failure. In the long-term nature of forestry production, there is the danger of assessing the characteristics and suitability of a specific tree species without knowing how to assess failures on the basis of surviving stocks so as to derive a general validity from individual case studies. In order to avoid such mistakes, growth and yield trials have to be designed thoroughly, which is not an easy task with regard to the diversity of forest growth conditions. This problem might be even greater for experiments that are focused on the suitability of tree species in the face of changing climate conditions. Using tree species, of which solid knowledge is available from their country of origin, is therefore advisable in this context.
