Rainforests in eastern Australia have been extensively cleared over the past two centuries. In recent decades, there have been increasing efforts to reforest some of these cleared lands, using a variety of methods, to meet a range of economic and environmental objectives. However, the extent to which the various styles of reforestation restore structure, composition and ecological function to cleared land is not presently understood. In this study, we develop and apply a method for quantifying the structural attributes of reforestation sites in tropical and subtropical Australia. The types of reforestation studied were plantation monocultures, mixed species cabinet timber 2 plots, diverse restoration plantings and unmanaged regrowth. Two age classes of reforestation were examined: 'young' (5-22 years), incorporating sites from all categories, and 'old' (30-70 years), in which only monoculture plantations and regrowth were represented. A total of 104 sites were surveyed including reference sites in intact rainforest and pasture. Intact rainforest was characterised by a suite of complex structural features including abundant special life forms (vines, epiphytes, hemi-epiphytes and strangler figs), a dense stand of trees in a range of size classes, a closed canopy, a shrubby understorey and a well-developed ground layer of leaf litter and woody debris. These features were lost on conversion to pasture. While all types of reforestation returned some elements of structural complexity to cleared land, young plantation monocultures, cabinet timber plots and young regrowth had a relatively simple structure. These sites typically had a low density of woody stems, a relatively open canopy and grassy ground cover, and lacked large trees, coarse woody debris and most special life forms. Restoration plantings and old regrowth were more complex, with a high density of woody stems, a relatively closed canopy and shrubby understorey. Old monoculture plantations in the tropics had acquired many of the structural attributes of intact forest, however this was not the case in the subtropics, where plantations were subject to more intensive management. The marked differences in structural complexity between sites suggest that the different types of reforestation practiced in eastern Australia are likely to vary considerably in their value as habitat for rainforest biota.

In Australia, as in many countries, there has been a shift in timber production from native forests to plantations. While plantations are primarily considered an efficient means of producing timber, there is increasing interest in their potential contribution to biodiversity conservation. Plantations may have both positive and negative consequences for biodiversity, at a range of scales. We compiled a list of these consequences from the literature, and used them to assess plantation scenarios proposed for cleared rainforest landscapes in tropical and subtropical Australia. The scenarios were monocultures of: (i) hoop pine, (ii) exotic pine and (iii) eucalypts; (iv) mixed species plantations; (v) a mosaic of monoculture plantations; and (vi) a mosaic of plantations and ecological restoration plantings. Of these scenarios, plantations of eucalypts and exotic pines have the least positive consequences for biodiversity: they have little or no intrinsic value in rainforest landscapes, provide poor quality habitat for rainforest biota, and (particularly eucalypts) are characterised by a relatively open canopy which in cleared landscapes favours the recruitment of grasses and other weeds. The three scenarios based on plantations of rainforest trees have similar, moderately positive consequences for biodiversity, while a mosaic of plantations and restoration plantings has the most positive consequences for biodiversity in cleared rainforest landscapes. All scenarios may have negative impacts on biodiversity conservation if plantations replace remnant forest, provide habitat for weeds, or the tree species used in plantations or their genes escape into native forests. In practice, the relative importance of positive and negative impacts, and hence the ranking of scenarios may vary with landscape forest cover. Scenarios with strongly positive consequences for biodiversity would be favoured for the reforestation of heavily cleared landscapes, whereas scenarios with few negative consequences for biodiversity would be favoured in well-forested landscapes. Consequently, any plantation of rainforest trees may have acceptable consequences for biodiversity in well-forested landscapes, provided the trees are not invasive or carrying exotic genotypes, and plantations are managed to control weeds and feral animals. With the same caveats, plantations of exotic pines may also be acceptable from a biodiversity conservation perspective in well-forested landscapes. At present, our capacity to design and manage rainforest plantations for both timber and biodiversity objectives is limited by a lack of information on factors affecting www.elsevier.com/locate/foreco Forest Ecology and Management 208 (2005) 359-372 timber production, biodiversity values and trade-offs or synergies between these objectives. Obtaining this information will require the integration of large-scale, long-term biodiversity research in broadscale plantation projects. #

Following two centuries of land clearing, the past two decades have seen growing efforts to reestablish forest on formerly-cleared sites. While the immediate goals of reforestation vary, there is also a widespread expectation that one of its effects will be an improvement in "biodiversity value". However, agreed standards concerning how this can be measured, and against what benchmarks it should be judged, are lacking. This paper describes a study of biodiversity development in different types of rainforest reforestation in tropical and subtropical eastern Australia. It provides information on the responses of rainforest fauna and also discusses key issues of survey design and methodology that, if ignored, may limit the effectiveness of monitoring programs. The nature of rainforest, its history within Australia, and its role as fauna habitat are briefly reviewed. Modern deforestation and human land use, and various reforestation pathways (including regrowth, timber plantation, and ecological restoration projects) are described. Then, design principles, biodiversity measurements, and issues relating to their choice are discussed. These include: spatial scale; site selection and replication in relation to environmental variation; reference sites; the array of potentially measurable biotic and process variables, and spatio-temporal measurement scales. Finally we explore analytical options and present selected findings, using univariate and multivariate approaches, and comparing simple species richness, functional groupings, and analyses of taxonomic composition. Rainforest biodiversity value is defined as the development of a rainforest-like set of biota and ecological processes. Reforested sites were generally intermediate between pasture and rainforest reference sites in the measured components of rainforest biodiversity value. Many components had been rapidly (by around 10 years) acquired by ecological restoration sites, although it is clear that some components will take decades or longer to develop. The results also show: (1) the existence of production/ biodiversity trade-offs, in that sites managed for timber production acquired less biodiversity value than those planted for ecological restoration; (2) moderate correspondence across different indicator taxa when they are analysed for assemblage composition; (3) very little agreement among indicator taxa when overall richness is used; (4) a likelihood of important landscape and context effects. Long-term conservation of rainforest fauna will require rainforest restoration over substantial areas of currently denuded land. However, although rainforest restoration may often show reasonable success, it should not be viewed as an alternative to conserving existing remnants and advanced regrowth.