Annals of Warsaw University of Life Sciences - SGGW Forestry and Wood Technology 83, 2013: 32-36 (Ann. WULS - SGGW, For. and Wood Technol. 83, 2013) Linear relations between defect frequency and volume of beech logs ZBIGNIEW KARASZEWSKI 1), MARIUSZ BEMBENEK 2), PIOTR S. MEDERSKI 2), ANNA SZCZEPAŃSKA-ÁLVAREZ 3), DIETER. F. GIEFING 1), ANDRZEJ WĘGIEL 4) 1) Wood Science and Application Department, Wood Technology Institute 2) Department of Forest Utilisation, Poznań University of Life Sciences 3) Department of Mathematical and Statistical Methods, Poznań University of Life Sciences 4) Department of Forest Management, Poznań University of Life Sciences Abstract: Linear relations between defect frequency and the volume of beech logs. The classification of round wood is based on both the length and diameter of the wood as well as the presence and size of its inherent defects. Certain defects occur in particular places along the tree trunk e.g. red heart appears at butt level whereas sound knots will typically be seen within the top section of the tree. These defects can have a significant effect on the quality/grade of the assortment. Due to the possibility of these defects appearing throughout the structure of the trunk, it can be asserted that their presence will have an influence on tree volume within the assortment. The aim of this research paper was to verify which of the analysed defects, which lead to the down grading of beech, will be associated with the largest volume. An analysis of this dependency was carried out in 15 beech stands on three different quality sites. 1389 logs with a combined volume of 1029.32 m 3 were classified. The linear relationship between the frequency of a given defect and the volume of the tree inflicted by this defect was investigated. No significant differences were observed between the number of assortments with a given defect and volume of trees burdened with that defect within the confines of the three sites. The customer of round beech wood in Poland can expect defects of a similar size and nature, irrespective of the quality of the tree s habitat. Keywords: wood defects, beech wood quality, Fagus sylvatica, hardwood INTRODUCTION The presence of wood defects and the measurement of assortment size form the basis for the classification of round wood. During the rating of round wood quality, the type of defect, as well as its size, are taken into consideration. The causes of defects are complex and diverse. Defects like sweep or red heart may arise during the natural growth of the tree, scar or shakes could appear due to external factors. Anthropogenic damages are directly related to the type of forest operations [1, 2] being administered at the time. The distribution of defects on the tree trunk can, to a certain extent, be seen as being methodical. Open knots dominate the upper parts of the tree, dead knots the mid section, whereas red heart can be expected to be seen at the butt. Norms predict the simultaneous presence of several defects in hardwood. A hypothesis was made that since different defects that lower the quality of the wood are present in different parts of the trunk, volume will be related to these defects. Moreover, the volume of an assortment inflicted by e.g. sweep will be significantly different to the volume of an assortment affected by for example red heart. The aim of this paper was to verify which defects, which lead to a reduction in quality of logs, are associated with the largest tree volume. This relationship was analysed in beech stands in three different sites qualities. MATERIALS AND METHODS Beech logs were classified by foresters in 4 different Forest Districts: Gryfino, Karwin, Trzebież (Regional Directorate of the State Forests (RDSF) Szczecin) [3, 4, 5] and Jugów (RDSF Wrocław) [6]. Data were collected from logs harvested on 15 sample plots, in mature beech stands (91-160-years old). The plots were of four different site qualities (I-IV, in the scale I-V). The sample plots were divided into 3 groups/classes according to site quality: 1 site quality I (the most productive), 2 site quality II, and 3 site qualities III and 32
IV (the least productive). In site quality class 1 there were 565 logs and 403.13 m 3 of wood, in class 2 and 3: 370 and 454 logs with 246.48 m 3 and 379.71 m 3 of wood, respectively. 1389 logs of 1029.32 m 3 were classified in sections. Eight wood defects (the most frequent) were taken into consideration during the research: sound knots, unsound and dead knots, covered knots, shakes, sweep, scars, double pith and red heart. The remaining defects, also mentioned in norms for timber classification [14] were omitted due to their negligible presence. The linear relationship between the frequency of a given variety of defect and the volume of the tree on which that defect was observed was analysed. A linear regression was set and the significance of the regression coefficient was investigated. Programme R was used to carry out the calculations. The statistical significance level was set at 0.05. To check if stand quality classes are significant in the regression model, ANOVA was considered. Moreover, to detect outliers and influence points, Bonferonni Test and the influence measure were used. The regression model was determined and significant regression coefficients were given. RESULTS No significant differences were observed between the number of assortments with a given type of defect and the volume of the tree inflicted with that given defect within the confines of the three site qualities. It was found that, irrespective of the production abilities of a given habitat, tree assortments with the same given defect will be of a similar volume. The customer of round beech wood in Poland can expect defects of a similar size and nature, irrespective of the quality of the tree s habitat. This observation is supported by research conducted on red heart in beech, the presence of which is not dependent on habitat conditions but rather on the age of the tree and how the forest is managed [7, 8, 9, 10]. For each of the given defects, a relationship was observed between the number of logs and their volume [Fig.1]. In every case the regression coefficient was significant. The assigned regression equations reflected positively on the analysed relationship between the variables which is reflected by the size of the R 2 value (Adjusted R-squared) (tab. 1). Tab. 1 Regression coefficients for wood defects Defect Regression co-efficient p R 2 Sound knots 0.5245 3.03 10-6 * 0.8864 Dead and unsound knots 0.7332 4.14 10-4 * 0.7405 Covered knots 0.6259 3.85 10-7 * 0.9245 Shakes 0.5784 9.91 10-10 * 0.9673 Sweep 0.4246 6.80 10-6 * 0.8389 Scars 0.5786 1.70 10-8 * 0.9299 Double pith 0.3794 4.01 10-7 * 0.9239 Red heart 0.7861 1.84 10-8 * 0.9445 As was expected, defects generally affecting the upper parts of the log affect the volume of the graded wood to a lesser extent. The lowest value attained for double pith was 0.3794 which means that the next beech log with double pith would have a volume of 0.38 m 3. Dead and unsound knots, as well as covered knots appear in the mid-sections of the trunk, in logs generally classified as WC0. The significant length of these average quality sawlogs (WC0) as well their relatively big diameter results in a volume between 0.63-0.73 m 3. 33
Fig. 1 Linear relationships for analysed defects. Shakes as well as scars are typical defects located in the lowest parts of the tree. Their predicted volume was 0.58 m 3.The largest volume (0.79 m 3 ) was recorded in logs which were downgraded as a result of red heart. Red heart doesn t usually appear to such an extent as to completely downgrade round wood other than sawlogs. Moreover, cross sections taken from 34
the butt of the beech tree, revealed red heart with a smaller surface area in comparison to cross sections taken from the same trunk, but 6 to 9 m higher [11]. CONCLUSIONS 1. No significant differences between the analysed dependencies (namely defects and the volume of the downgraded assortment) were confirmed in the analysed site qualities. A lack of differences was prevalent in all the pre-analysed types of defects. 2. Some defects, which contribute to the downgrading of logs, were associated with a greater tree volume. 3. Red heart, dead and unsound knots as well covered knots were the defects most detrimental to volume size. 4. The quality of round wood can be significantly improved with the undertaking of appropriate actions such as the pruning of trees, altering competition as well as the shortening of the final felling age during the cultivation of the forest. These actions can improve the quality of the wood by eliminating and/or reducing the size as well as the frequency of defects. REFERENCES 1. KARASZEWSKI Z., GIEFING D. F., MEDERSKI P. S., BEMBENEK M., DOBEK A., STERGIADOU A. 2013: Stand damage when harvesting timber using a tractor for extraction. (Forest Research Papers), 74 [1]: 27-34. 2. KARASZEWSKI Z., BEMBENEK M., MEDERSKI P.S., SZCZEPAŃSKA- ALVAREZ A., BYCZKOWSKI R., KOZŁOWSKA A., MICHNOWICZ K., PRZYTUŁA W., GIEFING D.F. 2013: Identifying beech round wood quality distributions and the influence of defects on grading. Drewno. Pr. Nauk. Donies. Komunik., (56) 189: 39-54. 3. KOZŁOWSKA A. 2010: Wpływ wad u buka na jakość surowca drzewnego w nadleśnictwie karwin. Maszynopis. Katedra Użytkowania Lasu Up. Poznań. 4. MICHNOWICZ K. 2011: Wpływ poszczególnych wad drewna na możliwość i klasyfikację surowca wielkowymiarowego buk. Maszynopis. Katedra Użytkowania Lasu Up. Poznań. 5. PRZYTUŁA W. 2012: Wpływ poszczególnych wad drewna na możliwość zakwalifikowania surowca drzewnego u buka zwyczajnego. Maszynopis. Katedra Użytkowania Lasu Up. Poznań. 6. BYCZKOWSKI R. 2011: Wpływ poszczególnych wad drewna na możliwość zakwalifikowania surowca drzewnego u buka. Maszynopis. Katedra Użytkowania Lasu Up. Poznań. 7. GOMEZ N., BOCK J. 2010: Comprendre et maîtriser le coeur rouge du hêtre. Rendez-Vous Techniques, 27-28: 3-7. 8. KNOKE T. 2002: Value of complete information on red heartwood formation in beech (Fagus sylvatica L.). Silva Fennica, 36 [4]: 841-85. 9. KNOKE T. 2003: Predicting red heartwood formation in beech trees (Fagus sylvatica l.). Ecological Modelling, 169 [2]: 295-312. 10. TRENČIANSKY M., KOLENKA I. 2006: Analýza upyvu rubného veku na vznik a vývoj nepravého jadra bukového dreva. Acta Facultatis Forestalis Zvolen, [XLVIII]: 453-466. 11. DZIEWANOWSKI S., JORASZ A., DEREK D. 1978: Określenie jakościowej struktury drewna liściastego. Zadanie 1. Określenie jakościowej struktury drewna buczyn pomorskich. Maszynopis. 35
12. GIEFING D. F. 1999: Podkrzesywanie drzew w lesie. Wydawnictwo Akademii Rolniczej Im. Augusta Cieszkowskiego W Poznaniu. Poznań. 13. HEIN S. 2008: Knot attributes and occlusion of naturally pruned branches of Fagus sylvatica. forest ecology and management, 256 [12]: 2046-2057. 14. PN-92/D-95008. Surowiec drzewny. Drewno wielkowymiarowe liściaste. Wspólne wymagania i badania. Streszczenie: Związki liniowe pomiędzy częstością występowania wady a miąższością sortymentów bukowych. Klasyfikacja drewna okrągłego oparta jest na rozpoznaniu wad, ich rozmiaru oraz na pomiarze długości i średnicy drewna. Występowanie poszczególnych wad związane jest z określonym miejscem na pniu. Celem pracy było sprawdzenie, które wady prowadzą do deklasacji drewna o jak największej miąższości. Dokonano analizy tej zależności dla drzewostanów bukowych trzech różnych bonitacji. Badania przeprowadzono w 15 dojrzałych drzewostanach bukowych różnej bonitacji. Sklasyfikowano 1389 kłód o miąższości 1029,32 m 3. Zbadany został liniowy związek między liczbą występowania defektu danego typu a miąższością drewna, na której ten defekt się znajdował. Nie stwierdzono istotnych różnic między liczbą występowania sortymentów z wadą danego typu a miąższością drewna obarczoną tą wadą w obrębie badanych trzech klas bonitacji. Odbiorca bukowego drewna okrągłego w Polsce może spodziewać się wad na zbliżonym poziomie, niezależnie od jakości siedliska. Acknowledgements: The authors would like to thank Anna Kozłowska, Rafał Byczkowski, Klaudiusz Michnowicz and Wojciech Przytuła for their help with the field research. Corresponding author: Zbigniew Karaszewski Wood Technology Institute, Winiarska Street, no. 1, 60-654 Poznań, Poland e-mail address: z_karaszewski@itd.poznan.pl 36