Does initial spacing influence crown and hydraulic architecture of Eucalyptus marginata?
A. H. Grigg (1), C. Macfarlane (2, 3), C. Evangelista (1), D. Eamus (4) and M. A. Adams (5)
1. Alcoa World Alumina Australia, P.O. Box 172, Pinjarra, WA 6208, Australia / 2. CSIRO Forest Biosciences, CSIRO Centre for Environment and Life Sciences, P.O. Box 5, Wembley, WA 6913, Australia / 3. Corresponding author () / 4. Institute for Water and Environmental Resource Management, University of Technology, Sydney, NSW 2000, Australia / 5. School of Biological, Earth & Environmental Sciences, University of New South Wales, Sydney, NSW 2000, Australia / Received July 8, 2007; accepted October 22, 2007; published online March 3, 2008
Summary
Long-term declines in rainfall in south-western Australia have resulted in increased interest in the hydraulic characteristics
of jarrah (Eucalyptus marginata Donn ex Smith) forest established in the region’s drinking water catchments on rehabilitated bauxite mining sites. We hypothesized
that in jarrah forest established on rehabilitated mine sites: (1) leaf area index (L) is independent of initial tree spacing; and (2) more densely planted trees have less leaf area for the same leaf mass, or
the same sapwood area, and have denser sapwood. Initial stand densities ranged from about 600 to 9000 stems ha–1, and trees were 18 years old at the time of sampling. Leaf area index was unaffected by initial stand density, except in
the most sparsely stocked stands where L was 1.2 compared with 2.0–2.5 in stands at other spacings. The ratio of leaf area to sapwood area (Al:As) was unaffected by tree spacing or tree size and was 0.2 at 1.3 m height and 0.25 at the crown base. There were small increases
in sapwood density and decreases in leaf specific area with increased spacing. Tree diameter or basal area was a better predictor
of leaf area than sapwood area. At the stand scale, basal area was a good predictor of L (r2 = 0.98, n = 15) except in the densest stands. We conclude that the hydraulic attributes of this forest type are largely independent
of initial tree spacing, thus simplifying parameterization of stand and catchment water balance models.
