Relative performance of coppice versus seedlings of 16 eucalypt taxa over two rotations in northern coastal Zululand, South Africa

From the 1980s, Eucalyptus grandis × E. camaldulensis (GC) and E. grandis × E. urophylla (GU) were planted as alternatives to E. grandis for improved productivity in the Zululand coastal regions of South Africa. Although these hybrid combinations provided a degree of drought and/or disease tolerance together with the ability to regenerate the stand via the management of coppice growth, there was the need to test better-suited species for the drier, northern parts of coastal Zululand. Species, provenances and inter-specific hybrid clones of the genera Eucalyptus, Corymbia, Acacia and Casuarina were tested as alternatives to GC and GU in four site × species trials over one rotation (1996 to 2006). Although E. longirostrata, Corymbia henryi, C. citriodora subsp. variegata and E. argophloia showed promise as alternatives to GC and GU, their coppicing ability remained unknown. To determine this, one of the trials (Palm Ridge) was re-implemented and managed for regeneration via coppice over a second rotation (2R). Coppicing ability, growth and susceptibility to pests and disease data were compared for the 16 species at Palm Ridge over the two successive rotations (1R — planted = 8.9 years; 2R — coppiced = 8.2 years). Except for E. argophloia and one E. camaldulensis provenance, all treatments had equivalent, or improved stem stocking in 2R relative to 1R (trial mean: 1R = 1 174 stems ha-1; 2R = 1 288 stems ha−1). The two commercial hybrid clones had the highest volumes in both the 1R (GU = 174 m3 ha−1; GC 163 m3 ha−1) and 2R (GU = 149 m3 ha−1; GC 124 m3 ha−1) rotations. The mean annual volume increments (MAI in m3 ha−1 y−1) were not significantly different between rotations (1R = 10.74 m3 ha−1 y−1; 2R = 9.95 m3 ha−1 y−1), nor were there any significant rotation × treatment interactions indicating that the treatments performed in a similar manner over both rotations. Although there were differences among provenances within taxa, these were not as great as the differences amongst the taxa tested: GC/GC > C. citriodora subsp. variegata > E. logirostrata > C. henryi, with E. tereticornis, E. camaldulensis, E. argophloia and E. watsoniana not performing well over both rotations. Despite the higher rainfall recorded over 2R (837 mm y−1) relative to 1R (765 mm y−1), the incidence of pests and diseases that was not recorded in 1R most likely contributed to the reduced growth of the coppiced trees. Most notable of these were Leptocybe invasa on the GC, E. tereticornis and E. camaldulensis treatments, Botryosphaeria spp. on all but the Corymbia treatments, kino on the Corymbia treatments, and Coniothyrium spp on the GU treatment. Relative to the two commercial hybrid clones (GC and GU), E. longirostrata, and C. henryi were the most promising of the alternative species tested over both rotations (planted and coppiced), with C. citriodora subsp. variegata also showing promise. Although all treatments were affected by pest(s) or disease(s), some of the species showed resistance and provenances within species tolerance, indicating the potential for selection for a reduction in biotic risk.