Do tree plantations support higher species richness and abundance than pasture lands? (systematic review)
Increased worldwide demand for wood products, coupled with public concern over the loss or degradation of natural forests, has lead to a steady increase in plantation establishment throughout most regions of the world. Most of the world’s new plantations are generally established on former agricultural lands that are often of declining economic value for grazing or cropping. There is an expectation that when established within intensively used landscapes, plantations can contribute positively to biodiversity conservation.
We conducted a systematic global review of differences between timber plantations and pasture lands in terms of animal and plant species richness and abundance, and assessed the results using meta-analysis techniques. Our principle aim was to test the hypothesis that plantations contain higher species richness or abundance than pasture.
We searched multiple electronic databases and the internet using different combinations of Boolean search-terms. Search terms were run in separate or limited combinations depending on the requirements or limitations of the database used. We also obtained papers from colleagues and through reference lists from published studies including major review articles and books on plantations. Furthermore, we obtained information from some government studies and reports. Data were available for meta-analyses comparing the species richness and abundance of plantations and pasture lands for five taxonomic groups: plants, invertebrates, reptiles/amphibians, mammals, and birds. Studies that provided estimates of mean species richness and/or abundance, and the corresponding estimates of standard deviations and sample sizes, were included in the meta-analysis.
Our systematic literature search identified 1,967 articles of potential relevance to our study. Of these articles, 66 provided biological monitoring information for plantations and pasture lands. Of these, 30 articles were excluded from the meta-analysis due to their lack of provision of information necessary for the analysis (eg. sample size, mean, or standard deviation. No articles were excluded due to problems with experimental design, which were not already excluded on other grounds. In total, 36 primary studies met our criteria for inclusion within the meta-analysis (Table 1). Studies varied widely in the information provided about factors affecting the species richness or abundance of different taxa within pastures and plantations. We were limited to assessing those factors that were consistently reported in the literature. The majority of studies provided multiple contrasts of species richness and/or abundance between pasture lands (control) and plantations (treatment). Some studies contrasted multiple treatments to a common control, and others contrasted multiple controls to a common treatment, hence creating divisions within studies.
Within each taxon there was considerable variation in the difference between species richness and abundance between plantations and pasture lands. Birds and reptile/amphibians exhibited significantly higher species richness, and mammals exhibited significantly higher abundance, in plantations than in pasture lands which lacked remnant vegetation. Reptile/amphibian species richness was significantly higher in plantations in general. No significant differences in species richness were found for mammals, plants, or invertebrates, and no significant differences in abundance were found for birds, reptiles/amphibians, invertebrates, or plants.
We found that for most taxa, plantations and pasture lands were not sufficiently consistent in their impact on species richness or species abundance to allow for general conclusions regarding their relative biodiversity value. Some taxa did have higher species richness or abundance in plantations than in pasture lands. However, it is only within the presence of taxonomic caveats (ie. reptiles/amphibians), or specific landscape features (ie. absence of remnant vegetation within pasture), that it can be concluded that plantations support higher species richness or abundance than pasture land. We emphasize that caution is warranted when making general statements about the inherent biodiversity value of diverse and broadly-defined land-uses.
Increased worldwide demand for wood products, coupled with public concern over the loss or degradation of natural forests (Lamb et al. 2001; Lindenmayer and Hobbs 2004), has lead to a steady increase in plantation establishment throughout most regions of the world (FAO 2007). Plantations are being established globally at a rate of 3 million ha per year (2000-2005, FAO 2006) and currently provide almost 50% of the world’s wood production (FAO 2007). In some nations, plantations comprise a substantial proportion of national forest area (FAO 2006). The principle benefit of plantations is that they enable large volumes of wood products to be produced per unit of land area (Sedjo 1999), although their capacity to sequester carbon has made this land-use a potential contributor to climate change mitigation efforts (Laclau 2003; Miehle et al. 2006; Paul et al. 2008; Redondo-Brenes 2007).
There is a large literature assessing the relative biodiversity value of plantations versus natural forests (see Barlow et al. 2007; Hartley 2002; Lindenmayer and Hobbs 2004). In almost all cases, plantations contain fewer native fauna and flora relative to that found within natural forests, with a corresponding increased abundance and species richness of exotic species (Barlow et al. 2007; Hartley 2002; Lindenmayer et al. 2002). However, most of the world’s new plantations are generally established on former agricultural lands (Sedjo 1999), that are often of declining economic value for grazing or cropping (Lamb et al. 2001). Under these circumstances, plantation establishment may provide both economic and environmental benefits. For instance, plantations can be used to sequester carbon and thereby reduce net greenhouse gas emissions (Jackson and Schlesinger 2004); lower water tables to help reduce dry land salinisation (Walker et al. 2002); and under some circumstances, relieve some of the pressure of timber demands from natural forests (Hartley 2002).
There is an emerging expectation that when established within intensively used landscapes (eg. agriculture), plantations can contribute positively to biodiversity conservation (Hartley 2002; Lugo 1997; Moore and Allen 1999). For instance, the flora and fauna of industrial scale plantations can compare favorably to that found within intensive land uses such as annual crop and pasture lands (Carnus et al. 2006; Hartley 2002; Moore and Allen 1999). For this reason, there has been promotion of the view that plantations provide higher environmental benefits, associated with increased biodiversity value, than agricultural landscapes (Moore and Allen 1999). We suggest that part of this expectation arises from plantations providing increased structural complexity relative to agricultural landscapes, which increases the variety of available resources upon which greater species diversity can rely (August 1983; Brokaw and Lent 1999; McElhinny et al. 2005). There is empirical and theoretical support for the positive relationship between increasing structural complexity and increases in biodiversity (but see Erdelen 1984; MacArthur et al. 1966; MacArthur and MacArthur 1961; McElhinny et al. 2005). However, if increased structural complexity is to enable plantations to support higher species richness than agricultural areas, then this one factor must dominate other contributing factors to species richness, such as habitat heterogeneity and the presence of native versus exotic vegetation. If generalizations are warranted, and these are to be incorporated into environmental policy and planning, it is important that the form and direction of changes in species richness, abundance and composition associated with these land- uses are identified, as plantations are increasingly replacing a significant percentage of many nations’ agricultural lands (Kanowski et al. 2005).