region. For example, natural forest production in Uganda was 0.7 m3 ha" 1year_1, while nearby Eucalyptus plantations produced more than 40 m3 ha-1year_1 (Laurie 1962; cited by Wadsworth 1997). In many parts of the tropics plantations yield several times the quantity of wood of most natural tropical forests and most forest in temperate regions (Evans 1992). This is true for plantations where species match the sites and when plantations are given the proper management, including genetic selection, as discussed below. Fertilization and pesticide application also contribute to the high production of plantations in comparison with natural forests.
On average, Eucalyptus and Pinus species, which dominate industrial plantations in developing countries, have similar mean annual increments (MAI) in volume of 10-20 m3 ha"1 year"1 (Table 6.2; FAO 2000 b). However, many species can achieve much faster growth rates. For example, Eucalyptus grandis can yield 40-50 m3 ha-1 year-1 and under good conditions with advanced tree improvement it can yield much more. Other widely planted tropical hardwoods including Casuarina equisetifolia, Tectona grandis, and Dalber-gia sissoo have MAIs of less than 15 m3 ha-1 year-1 and frequently under 10 m3 ha-1 year-1 (FAO 2001a).
Plantation yield, whether it is measured in biomass or volume, is a function of the species' genetics, the silvicultural practices used, and site factors. Climate and site have a very large impact on growth rates. The humid tropics and more fertile sites are more conducive to higher growth rates than locations with long dry seasons or infertile or degraded soil. On many sites in India, for example, teak frequently has an MAI of only 4-8 m3 ha-1 year-1, partly because of drought combined with poor soils (FAO 2000b). In Costa Rica, in contrast, on good sites teak yields an average of 18 m3 ha-1 year-1 (Barquero 2004). Some species, such as Gmelina arborea and some of the Eucalyptus species, are very site sensitive. The average yield of Gmelina arborea worldwide is 25 m3 ha-1 year-1, while in Costa Rica it yields an average of 40 m3 ha-1 year-1 (Barquero 2004). Pinus species, in contrast, generally tolerate adverse conditions better.
Both tree breeding and silviculture have improved growth rates of several industrial species of eucalypts and pines. Good examples are Eucalyptus grandis and E. urophylla in Brazil (FAO 2000b). Much genetic improvement has been advanced by private companies, especially for the most frequently used species of pines and eucalypts. At Aracruz in the state of Espiritu Santo in southern Brazil, hybrids of E. grandis and E. urophylla (called "urogran-dis") yield an average of 44 m3 ha-1 year-1 and can be harvested at 7 years for pulp, or at 15 years for sawn wood (C.A. Roxo, pers. comm., www.ara-cruz.com.br).
Research on other species, including indigenous plantation trees, is ongoing at universities and other research institutions. For some native species in the humid tropics, genetic improvement has advanced to the stages of
trials of seed origin and progenies, the first step in the domestication of indigenous species (Mesén et al. 2000). For example, for Cordia alliodora, Vochysia guatemalensis, and a few other native species in Central America, CATIE in Costa Rica has determined the best provenances (specific origin of the seed in a region or locality in a given country) that suit most planting conditions. In addition, progeny studies have helped to assess specific sources of seed (parent trees growing in a seed orchard or collection) for selected species such as Vochysia guatemalensis, Acacia mangium, Eucalyptus grandis, and other species of interest (Fig. 6.1).
Genetic diversity of a particular tree species is an important consideration in tropical plantations. Genetic diversity within a species helps to ensure that the species will be able to adapt to stresses and changing environmental conditions. The tendency among plant breeders is to select for individuals with a few desirable characteristics. Usually these characteristics are rapid growth and good form. Characteristics that are often neglected include: large roots; efficient association between roots and mycorrhizal fungi; production of root exudates that improve nutrient recycling; strategic reproductive patterns that take advantage of natural seed dispersal processes and avoid seed predation; high resistance to insects and disease; and allelopathic repression of other plants. When selecting individuals for reforestation using native or non-native species, it is important that the characteristics necessary for survival in the wild are included in the genome so that if it is not feasible to fertilize the plantations and spray them against insect attack, they will still be able to survive. When selecting trees to grow in specific sites such as in degraded lands, other characteristics that may be selected for should include the ability to grow on acid soils, withstand drought, and grow at low nutrient concentrations.
The length of the rotation period strongly affects both end-use and economics. Many fast-growing Eucalyptus, Acacia, Casuarina species, and Gmelina arborea are grown on short rotations of under 15 years, as they are used primarily for pulp or fuelwood. Usual rotations in Kenya for E. grandis are 6 years for domestic fuelwood, 7-8 years for telephone poles, and 10-12 years for industrial fuelwood (FAO 2000b). In Brazil this species is largely grown for pulp or charcoal on 5- to 10-year rotations. Species being grown for high-value saw logs usually have longer rotations; teak (Tectona grandis) is grown in Asia on 50-to 70-year rotations (although in Costa Rica and other regions in Latin America much shorter rotations, about 25 years, are expected); and high-value conifers such as Araucaria angustifolia are grown on 40-year rotations. Generally, pines are grown on medium-length rotations of 20-30 years, unless grown solely for pulpwood, when shorter rotations may be adopted.
An important rule for deciding on the length of a rotation for plantations is the following: as long as diameter growth continues steadily, or does not decrease significantly, the yield will increase exponentially. The landowner's rate of return is increasing. A landowner with a stand of trees that average 50-cm diameter may be tempted to cut them and replant with new seedlings. The landowner should realize that by waiting, the yearly biomass increment (and hence profit) will increase much faster than the increment from newly planted seedlings. Once a forest stand has been cut and secondary succession begins, primary productivity drops way down. The reason is that biomass increases exponentially as diameter increases linearly. An example is as follows, assuming that tree trunks have a parabloic shape: A tree of 20 cm diameter and 10 m height that is growing at a rate of 2 cm diameter and 50 cm height year-1 (assuming a wood density of 1.0) will increase its weight by 42 kg year-1. However, a tree of 40-cm diameter and 20-m height growing at the same rate will gain 164 kg year-1.
Biomass increment of trees can be estimated by various means. If we assume that the trunk of the tree has a parabolic shape, then volume of the stem (PV) equals one-half the basal area times tree height (Whittaker and Marks 1975). Production for a given time period is the biomass at the final time minus the biomass at the time of the first measurement. Biomass of trees can also be estimated using allometric equations, where biomass is calculated as a function of relatively easily measured parameters such as diameter at breast height (DBH) and height of trees. Allometric equations are developed from field measurements of tree DBH, height, and biomass. Several allometric equations for estimating biomass of tropical trees have been developed, both for forests and plantation species (Brown and Iverson 1992; Montero and Kanninen 2002; Pérez and Kanninen 2002). The forms of the equations vary widely, including various polynomials and models incorporating DBH and height of trees, but the most common equation is the power function B=aDBH , where B is biomass. The power function is most often fitted by linear regression on log-transformed data (Dudley and Fownes 1992).
Biomass increment may not be the only consideration in deciding the length of a plantation rotation. There are also economic considerations. The increase in yield due to delaying a cut must be weighed against having the money now and having it in the future, at a value discounted to the present.
Was this article helpful?