Convention on biological diversity
|
|  Yüklə 1.31 Mb.
|
III.Ecosystem and landscape diversityAn ecosystem concerns scale relative to the organism or group of organisms or even the ‘question’ being considered. For forests, an ecosystem generally refers to a clearly identifiable and distinct area of certain species and functional and abiotic components. Ecosystem function is the total of all processes that occur within the ecosystem including production, nutrient cycling and water movement. Many forest ecosystem classifications exist, usually based on an ordination of soils and moisture regimes, which combine to support a particular kind of forests, level of functions and associated organisms. The role that species may play in ecosystems was briefly discussed above; considerably more information is needed with respect to this important concept (see Chapter II). Several hypotheses have been advanced to explain the role of species in ecosystems. These include (Vitousek and Hooper, 1993; Lawton, 1994): a) a null hypothesis, where ecosystem function does not change with gain or loss of species; b) the idiosyncratic hypothesis, which predicts that there will be a change in function related to species loss or gain, but the magnitude and direction may not be predictable; c) the rivet hypothesis (Ehrlich and Ehrlich, 1981), which predicts that all species contribute to function and that loss/gain in function is equitable for each, or that thresholds in numbers of species exist that once crossed, a disproportionate amount of function is lost; d) the redundant species hypothesis that predicts only some species contribute to function and the roles of all others are redundant (Walker, 1992; Lawton and Brown, 1993). There is little evidence to support any of these hypotheses, particularly for vertebrate species, and certainly we generally do not know the form or shape of the relationship. As a general principle however, theory and experimentation suggest that complex ecosystems with large leaf areas are more productive and maintain more niche space, and hence more species, than less productive ecosystems with minimal richness in primary producers (Tilman et al., 1996, 1997). However, rich systems are also likely to maintain a high species redundancy (Lawton, 1994). Although Sugihara (1980) suggested communities form around a hierarchical niche structure (‘sequential breakage model’) that obviates any redundancy, work by Tilman and Downing (1994) revealed that reducing the number of primary producers from 25 to 10 did not alter net primary productivity in a field experiment. Opposite results were achieved by Symstad et al. (1998) who showed a decline in plant biomass with species reduction in grassland systems. They also found that nitrogen retention (a measure of productivity) changed unpredictably depending on the species removed, supporting the idiosyncratic hypothesis. A further unknown is the role that rare species may play, under extreme events, in buffering an ecosystem against change (Schulze and Mooney, 1993). The relationship between stability in systems and species diversity and richness is still obscure and is a source of concern with respect to the use of forests. Concerns over resource use and effects on biodiversity relate to questions about ecosystem resiliency. In other words, can these systems be used in a manner that permits them to return to a state that is similar to the state existing prior to harvesting and, hence, maintain the same species diversity (sustainable use)? If not, then they may return to an altered state, with little or no stability, which cannot support the biological diversity that existed prior to use. Because forests take many years to grow, this question of sustainability remains open. It is likely that using ecosystems results in less than optimal production in the system and that constant use impairs maximum production by altering the ecosystem state beyond certain thresholds. However, predicting thresholds, and hence levels of sustainable use, is problematic. A second set of questions with regards to stability in ecosystems pertains to the functional roles that species may play within the system that are not directly related to productivity. For example, numerous species are responsible for pollination of plants, the primary producers. Without these species, which live off excess energy in the system, the system could not exist. This role may be a ‘keystone’ role as referred to above. Often, exotic (or non-native) species may affect stability in ecosystems by altering processes such as these. Landscape is one of several scales of forest biological diversity (Noss, 1990), although landscape by itself is a concept of scale. As with ecosystems, a particular landscape is defined relative to the organisms, populations or processes under discussion. In the case of a forest landscape, the size reflects the broad-scale processes that cause variation in forest ecosystems. Processes that affect forest community structure are largely historical and operate at large-scales (Cornell and Lawton, 1992). An understanding of how past forest landscapes were formed and how processes have changed through history is needed to provide a perspective on new landscape structure that is inevitably altered by humans. Contemporary forestry and land-use practices can substantially alter forest landscape structure, often irrevocably, through land-clearing, changes to patch size and distribution, through losses of fine-scale structures and microhabitats (Haila et al., 1994), away from heterogeneity towards homogeneity (e.g., Pickett and White, 1985), and through the creation of dynamics that differ from those under more natural disturbance regimes (Hunter, 1990; Virkkala and Toivonen, 1999). Many animal species, particularly those with large body sizes, respond to forest landscape structure. In addition, the effects of forest fragmentation on species, discussed above, operate at the landscape scale. Therefore, the distribution of forests across the landscape has implications for maintaining biological diversity. Suffling (1991, 1995) has advanced the theory that landscapes are in “constant disequilibrium”, in other words, stability at the scale of forest landscape never occurs. This view is shared by Baker (1995) on the basis of considerable modelling of northern temperate mixed forests of the United States. According to this theory, landscapes are constantly “catching up” with disturbance regimes that change at two temporal scales, the longer one in the order of several hundreds of years. Current climate data can not provide the basis for predicting that present northern temperate and boreal forest ecosystems, formed following disturbances 100-500 years ago, will recur at any particular point in the future, because forest response to disturbance is a complex process involving numerous interacting factors and time lags (Drake, 1990; Bonan, 1992). The ultimate factors that control landscape structure act in a “top-down” manner, affecting ecological processes over large areas and long time scales, while more proximate factors influence individual patch sizes and their rates of change over shorter time periods. The local distribution of forest ecosystems is influenced by site conditions, as modified by disturbance events. The development of forest vegetation across a landscape is influenced by three ultimate factors acting at large spatial and temporal scales: physical factors (including soils, lithology, elevation and relief), climate (rainfall and temperature) and anthropogenic factors (including history and livelihood). Physical geography and climate set fundamental limits on forest development, while history represents the net (or ‘combined’ or ‘cumulative’) effect of specific events of forest growth, change and destruction over time. Coupled with the ultimate factors are proximate factors, those that act at more limited spatial and temporal scales to influence stand development. Palaeoecological research implicates climate as a key factor in temperate and boreal forest change during the post-glacial period (Webb, 1986; Ritchie, 1987) and suggests that plant migration can generally track climate change, regardless of the speed of that change (Pitelka et al., 1997). Payette (1992) notes that although climate and ecosystem processes can explain the migration and retraction of forest types in boreal and temperate biomes, competition is also an important factor in structuring plant communities. Forest development at large scales not only responds to climate, disturbance and forest management, but also to the ability of individual species to compete, to given specific historical sequences and stochastic events and to specific soil conditions. At large scales, tree species richness correlates with many climatic variables, such as number of growing-degree-days, mean annual temperature, mean annual insolation and total annual precipitation. Net primary productivity is a variable that integrates many aspects of climate, such as length of the growing season, soil temperature and moisture regimes. A study testing this ‘climate affects species richness’ hypothesis for Ontario, Canada (an area of 1 x 106 km2, over 13o of latitude) revealed a highly significant positive relationship between tree species richness and net primary productivity (Thompson, 2000). The association between topography, soils and forest types (or individual ecosystems) has been reasonably well studied (e.g. Sucachev, 1928; Cajander ,1948; Daubenmire, 1968) and it has been shown that topography and soil types exert strong influences on forest development. Relationships exist among slope, soil type, microclimate and soil moisture content, all of which can be used to infer local productivity (Brady, 1984). Graumlich and Davis (1993) reported that substrate governs the distribution of pines (Pinus spp.) and birches (Betula spp.) in the North American Great Lakes Basin over an area of hundreds of square kilometres. Knowledge of the associations between soil types and plants can be used to predict and model forest landscape structure. Proximate factors, as stated earlier, are those that influence the development of forest vegetation at a smaller scale. The principal types of proximate factors are fires, wind, insect infestation, diseases, flooding and human interventions, as well as periodic changes in weather, competition among species and grazing by herbivores. At intermediate scales, human intervention is important to the structure of forest landscapes. However, it is doubtful that the complex inter-relationships of natural systems can be replicated by management (Hansen et al., 1991; Hunter, 1993). Although one of the goals of current forestry practices in many jurisdictions is to emulate natural processes, the concept that fire or hurricane effects can be emulated through logging practices - a cornerstone of sustainable forest management - is unlikely to be valid. Forest harvesting, coupled with fire suppression, has altered forest stand composition. Reforestation strategies and policies have changed over time, as have silvicultural practices. These changes have led to the redirection of forest successional trends, particularly compared to natural pathways, and ultimately to changes in landscape pattern. An important factor affecting the distribution of individual species, and ultimately the landscape pattern, is the reduction of seed sources as a result of past logging over extensive areas. A further example of human intervention in natural processes can be found in the role, which introduced fauna species, including ungulates now play in altering or inhibiting successional pathways in many forest types in North America, Australia, New Zealand and Europe. Although there is a hierarchy of scales of processes and factors that affect forest landscape patterns, in which factors at the upper level impose constraints on the next, there are also interactions among these factors from each level. Humans appear to be altering landscape-level processes at intermediate scales by forestry practices and, to a larger extent, by modern silvicultural practices such as controlling fires, logging forests and planting forests of types not expected from natural succession They are also causing changes at very large scales by altering climate. The result is altered landscape patterns at all scales, which may have eventual effects on forest plant species and their survival. This remains an important topic for research. Forest landscape patterns are quantified by remote sensing from aerial photograph or satellite image information that has been entered into a geographic information system (GIS), and analysed using any one of several software tools, such as FRAGSTATS (McGarigal and Marks, 1995) and Patch Analyst (Elkie et al., 1999), to quantify landscape variables. Common variables used include amounts of forest types by patch size and age-class, edge to interior ratio, fractal dimension, distance between patches, amount of non-forest, total edge, shape index, degree of interspersion and road density. The value of these statistics is that they enable comparisons between disturbed and undisturbed landscapes (or any two landscapes) under null models of intrinsic spatial patterns. Results of such comparisons can lead to comparisons and hypotheses with respect to biological diversity at large scales.
Several factors limit the global knowledge of forest biological diversity including insufficient taxonomy, a poor understanding of traditional knowledge, insufficient autecological and synecological knowledge of species, communities and ecosystems and the lack of infrastructure and capacity in many countries for inventory and monitoring programmes. Monitoring of biological diversity is a hierarchical procedure requiring a range of technologies and information. The scale can be depicted in terms of knowledge from forest types down to genetic resources with the associated required technologies and research to accomplish and develop the knowledge base. Where an individual country or agency is along that scale depends on its history of conservation policies, the dedication of the management agency to sustainable use, the availability of funding and the technical capacity and research available. This chapter suggests that there is the need for a broad science research programme and data collection programme to understand the mechanisms affecting forest biological diversity and the relationship between biodiversity and forest goods and services. The research and management agenda, located in Annex lII, provides a path towards the sustainable management of forest resources. There is a large number of sources of information pertaining to forest biological diversity, a partial list of such sources is located in Annex IV.
Forest ecosystems provide a wide array of goods and services at a range of scales from local to global. As well as sustaining commodities such as timber, traditional goods for local populations and services with economic return such as eco-tourism, forests perform a key role in providing vital services, which usually have no clear market value, notably global climate regulation and watershed protection. (Box 4). The first part of this chapter aims to identify the key functional mechanisms for each of the three main forest biomes (boreal, temperate and tropical) and how they relate to biodiversity. It then assesses how human activities affect ecosystem functions and biodiversity and the ability of forests to deliver goods and services. The second part of the chapter assesses the values to people of forest goods and services that are linked to forest biological diversity and the impacts of human activities on those values. There are many different perspectives on forest values, so it is important to distinguish these and analyse the implications of changes in forest biological diversity for the range of stakeholder interests.
|