Productivity gains by fertilisation in Eucalyptus urophylla clonal plantations across gradients in site and stand conditions

摘要

Nutrition management in Eucalyptus plantations is fundamental for sustaining high production. Fertilisation is routinely used to improve tree nutrition, providing profitable returns on large investments. Growth responses to fertilisation differ dramatically among sites, however, so efficient investment decisions in fertilisation is important. The twin-plots design characterises the fertilisation response in a short period of time, providing the information needed to landscape-scale silvicultural prescriptions. This method entails the establishment of pairs of plots, with one control and one treated plot at each location. The control plot may typically be a permanent plot of an inventory network, providing representative information for a company's decision-making. The paired twin-plot receives intensive management (high fertilisation and weed control) to minimise (or remove) these constraints to forest productivity. We used this approach with 131 blocks of twin-plots to represent an area of 34 540 ha in the state of Sao Paulo, Brazil. Clonal plantations of Eucalyptus urophylla were remeasured one and two years after treatment. Fertilisation increased wood growth by 15% for two years (4.0 t ha super(-1) y super(-1), or 8.1 n super(3) ha super(-1) y super(-1)), with the current annual biomass increment reaching 31.6 t ha super(-1) y super(-1) (64.2 m super(3) ha super(-1) y super(-1)) versus 27.6 t ha super(-1) y super(-1) (56.1 m super(3) ha super(-1) y super(-1)) of the control plots. Twin-plots located on sandier and less fertile soils showed twice the fertilisation response of other plots, increasing wood growth by 8.5 t ha super(-1) y super(-1) (16.9 m super(3) ha super(-1) y super(-1)). The two predominant clones responded similarly to fertilisation. Older stands showed higher responses than younger stands (1.7 t ha super(-1) y super(-1) for each additional year), and the greater response in older stands probably represented increase in fertilisation rates over time, rather than a feature of he age of the stands per se. Fertilisation response correlated negatively with site index (base age 7), soil clay content, and soil base nutrient levels (Ca, Mg and K). Models for the prediction of fertilisation response can be used to develop regional- and site-specific fertiliser prescriptions to maximise financial gain from fertilisation.