1. Assessing the effects of water availability on selected wood properties of E. grandis in South Africa Sasha Naidoo, Anton Zbonak, Norman Pammenter and Fethi Ahmed Forestry and Forest Products Research Centre CSIR, Durban IUFRO Taipei October 29 – November 2 (2007)

2. Background Previously: Plantation yield quantified in terms of stem volume Currently: Focus on wood quality as determinants of plantation productivity

3. Background Environmental factors Significant effects on wood properties of eucalypts Subtle interactions among environmental factors Soil type, depth and nutrient status influence water retention and water availability Eucalyptus grandis – grown across a range of sites dependant on water availability

4. General aim To assess the response of wood properties of Eucalyptus grandis to varying levels of water availability To improve understanding of factors that affect wood properties

5. Outline of presentation Review experimental approach used Illustrate some results and outcomes Summarize the next phase in this study

6. MACRO ZONESMAT  Cool Temperate< 16 o C  Warm Temperate16.1- 19 o C  Sub-Tropical> 19.1 o C Cool Temperate MEAN ANNUAL PRECIPITATION (MAP) 15.1-16 o C DRYMOISTWET 14.1-15 o C DRYMOISTWET < 14 o C DRYMOISTWET Sub-tropical Warm Temperate LOW, MEDIUM and HIGH soil water storage (SWS) values form 3 rd level of classification ICFR site classification system Sub-tropical Warm temperate Cool temperate HOWEVER: E. grandis is not grown in all the macro zones: INCOMPLETE DESIGN

7. 3 compartments per cell 5 trees per compartment Ages: ~ 6-12 years No coppiced material Experimental design MAP SWS DRYMOISTWET LOW 3 2 3 MED. 3 3 HIGH 2 3 MAP SWS DRYMOISTWET LOW 31 2 MED. HIGH 3 3 3 WARM TEMPERATE (16.1-19 o C) SUB-TROPICAL (> 19.1 o C)

8. Sample preparation for measuring wood properties pith N breast height (1.3 m) Twin-blade saw Pith-to-bark strip (x 2) Image analysis – vessel and fibre characteristics Density and NIRS (predicted cellulose and lignin) WOOD PROPERTIES MEASURED x 2 cores per tree

9. Vessel and fibre measurements 20-25 µm thick section Fibres measured every alternate 0.5 mm Vessel diameter Vessel frequency Vessel percentage Fibre diameter Fibre lumen diameter Cell wall thickness Vessels measured every 0.5 mm

10. Gamma-ray densitometry and near- infrared spectroscopy Density measured every 0.5 mm Gamma ray densitometer XDS NIR Spectrometer NIR spectra obtained every 5 mm NIRS – prediction of cellulose and lignin contents

11. DENSITY (g.cm -3 ) BARK PITH BARKPITH Constructing radial maps

12. DRYMOISTWET LOW HIGH Mean density (MD) (g.cm -3 )0.40-0.49 0.50-0.590.60-0.69 Key Density in the sub-tropical region SWS MAP MD = 0.57 a MD = 0.47 b MD = 0.46 b MD = 0.55 a MD = 0.54 a n=15 n=14 n=5 n=15 MD = 0.58 a Proportion of lower density wood increases with increasing MAP

13. DRYMOISTWET LOW MED. HIGH SWS MAP MD = 0.52 a MD = 0.49 a MD = 0.51 a MD = 0.53 a MD = 0.50 a MD = 0.53 a MD = 0.51 a n=14 n=15 n=8 n=10n=14 Mean density (MD) (g.cm -3 )0.40-0.49 0.50-0.590.60-0.69 Key Density in the warm temperate region

14. SUB-TROPICAL WARM TEMPERATE Increase in fibre diameter with increase in MAP Similar pattern of response to MAP in both regions LOW SWSHIGH SWS DRYMOISTWETDRYMOIST WET Mean fibre diameter (µm) LOW SWSHIGH SWSMED. SWS DRYMOISTWETMOISTWETMOISTWET Fibre diameter Mean fibre diameter (µm) bc a ab c bc abc bc a ab c abc c 32 H 33 M 333 L WMD SWS MAP 32 H 33 M 323 L WMD SWS MAP 333 H M 323 L WMD SWS MAP 333 H M 313 L WMD SWS MAP

15. SUB-TROPICAL WARM TEMPERATE LOW SWSHIGH SWS DRYMOISTWETDRYMOIST WET NIR-predicted cellulose (%) LOW SWSHIGH SWSMED. SWS DRYMOISTWETMOISTWETMOISTWET NIR-predicted cellulose Significantly higher NIR-predicted cellulose values with higher MAP b a a b a a a b bc bcd cd d d 32 H 33 M 333 L WMD SWS MAP 32 H 33 M 323 L WMD SWS MAP 333 H M 323 L WMD SWS MAP 333 H M 313 L WMD SWS MAP NIR-predicted cellulose (%)

16. SUB-TROPICAL WARM TEMPERATE LOW SWSHIGH SWS DRYMOISTWETDRYMOIST WET NIR-predicted lignin (%) LOW SWSHIGH SWSMED. SWS DRYMOISTWETMOISTWETMOISTWET NIR-predicted lignin a b b a b ab a b a a a a a 32 H 33 M 333 L WMD SWS MAP 32 H 33 M 323 L WMD SWS MAP Significantly lower NIR-predicted lignin values with higher MAP 333 H M 323 L WMD SWS MAP 333 H M 313 L WMD SWS MAP NIR-predicted lignin (%)

17. Summary of results At low MAP - higher density, smaller vessel and fibre diameters, lower predicted cellulose and higher predicted lignin As MAP increases - Lower density, larger vessel and fibre diameters, higher predicted cellulose and lower predicted lignin Sub-tropical region Response of wood properties followed similar trends with an increase in MAP Differences in wood properties measured usually only significant at low SWS Warm temperate region

18. Future work planned for study Use bark to pith profiles and climate to identify growth rings climate Separate portions of bark to pith profiles into age classes compare wood characteristics among similar age classes Model wood properties with climatic variables

19. Significance of study Assess existing sites Improved use of existing material Evaluate future sites for planting

20. Acknowledgements Eucalyptus Co-operative – Center for Scientific and Industrial Research (CSIR), Mondi and Sappi Institute for Commercial Forest Research (ICFR) – Site Classification System

21. THANK YOU http://ffp.csir.co.za 983179111@ukzn.ac.za