Wetland connectivity: understanding the dispersal of organisms that occur in Victoria’s wetlands draft
|
|  Yüklə 2.04 Mb.
|
The importance of biological connectivityUnderstanding biological connectivity is fundamentally important because it underlies landscape-scale ecological processes (Nathan and Muller-Landau 2000, Hubbell 2001, Wright et al. 2003, Leibold et al. 2004). For example, dispersal provides opportunities for both native and introduced species to expand their range and migrate in response to local and regional changes in habitat conditions such as climate change. High rates of dispersal can allow species with narrow habitat requirements to reach suitable habitats in the landscape. Dispersal can also rescue populations in suboptimal habitats from extinction and, where disturbance events have caused local extinctions, dispersal can help species to recolonise these sites. Dispersal also promotes gene flow among populations and prevents populations becoming reproductively isolated. Maintaining gene flow among populations increases genetic diversity, which facilitates adaptations to the environment on ecological and evolutionary time scales (Whitlock and McCauley 1999). Where dispersal links populations of multiple species across the landscape, they interact to form a ‘metacommunity’. The concept of metacommunities forms the basis for all of the recognised models explaining community assembly — neutral, species sorting, source–sink, and patch dynamics (Leibold et al. 2004). These models consider how community assemblages are influenced by the frequency of dispersal, species habitat requirements, interaction among species, and trade-offs between traits that increase dispersal and those that confer a competitive advantage.
The neutral model is based on the premise that species are ecologically equivalent; that is, they do not differ in fitness, niche specialisation or dispersal ability, and this renders the environment neutral. Under these conditions, species distribution patterns are not influenced by environmental heterogeneity in the landscape but by distances among habitats, which determine the probability of colonisation. The neutral model has sparked considerable debate in the ecological literature on the relative importance of environment and dispersal in determining species distribution patterns. A number of studies have shown that differences in environmental conditions among habitats explains as much, if not more, of the variation in community structure explained by distances among habitats (McGill 2003, Thompson and Townsend 2006). It is evident from these landscape-scale studies that in some systems both the environment and dispersal contribute to community structure.
The species sorting model incorporates the influence of both environmental heterogeneity and dispersal limitation. In this model, dispersal determines the probability of a species arriving at a site, but variation in environmental conditions among sites, coupled with species niche specialisation, acts as a filter on species distributions (Leibold et al. 2004). Under conditions where dispersal is not limiting, the structure of communities at a landscape-scale is shaped predominantly by environmental conditions, and species distributions are strongly correlated with environmental variables and not spatial elements. In contrast, where dispersal is limited, species reach only a small subset of suitable habitats and both spatial elements and environment variables explain patterns of distribution in the landscape.
The source–sink or mass effects model posits that high rates of dispersal from ‘source’ populations maintain species in ‘sink’ habitats, when they would otherwise be displaced by stronger competitors, suboptimal conditions or high levels of disturbance (Pulliam 1988, Leibold et al. 2004)
The patch dynamic model applies when habitats are considered to be identical. As habitats are identical in this model, species cannot coexist at a regional scale via niche separation, and instead coexist through trade-offs between competition and colonisation (Hanski 1998, Leibold et al. 2004). In this scenario a species that is a weak competitor for resources can persist among stronger competitors if it is a better colonist, and this is achieved by maintaining higher rates of dispersal or fecundity (Gardner and Engelhardt 2008). These models illustrate the complexity of predicting the ecological outcomes of altered rates of dispersal. Changes in connectivity may favour some species, and at the same time exert negative effects on others. As pointed out by Taylor et al. (2006), connectivity is ‘inherently neither good nor bad’. Therefore the aim of management should not necessarily be to maximise connectivity, but to assess how changes in the landscape will alter patterns of connectivity, and how this will affect the persistence of species.
Wetland habitats are among the most modified ecosystems in the world (Pringle 2006). Globally more than half of the world’s wetlands may have been destroyed (Ramsar Convention Bureau 1996). Similar losses have been incurred in Australia: Victoria has lost 30% of its natural wetlands and New South Wales up to 70% of some wetland types, and in Western Australia 70% of wetlands on the Swan Coastal Plain have been lost (Wasson et al. 1996). Compounding the impact of habitat loss has been the over-abstraction of water for human use, which has degraded rivers and isolated wetlands. Humans have already acquired more than half of the Earth’s accessible freshwater resources, a figure predicted to climb to 70% by 2025 (Postel 1996, Pringle 2006). In the USA almost all streams have been modified by dams or other water diversion schemes (Poff et al. 1997). In Australia river regulation has produced highly modified flow regimes and alienated floodplain wetlands (Kingsford 2000, Arthington and Pusey 2003). These landscape changes have reduced populations of aquatic plants and animals directly through habitat loss, but also indirectly because they impinge on the ability of organisms to move among habitat patches. Reduced connectivity, coupled with multiple environmental threats such as eutrophication, salinisation and invasive species, have driven many aquatic systems into a perilous state. The World Wildlife Fund reported that, on a global scale, populations of 200 freshwater fauna have declined by 50% in just over 30 years (1977–1999). Master et al. (1998) reported that freshwater species in the USA are in greater peril than terrestrial species. In Australia, 14% of frog species and 9% of freshwater fish species are now endangered or threatened. A third of Australia’s rivers have lost 20–100% of the aquatic invertebrate species that should be present, and 80% of River Red Gum trees in South Australia are stressed because of reduced flooding regimes and saline groundwater (Saunders et al. 1996, Wasson et al 1996). Current threats to aquatic biodiversity are likely to be compounded as regions of Australia shift towards a more arid climate in response to climate change (Murphy and Timbal 2007). With increasing aridity it is likely that the number and size of wetlands will decline, and the duration, frequency and magnitude of flood events that periodically link aquatic habits in the landscape will be compromised. These changes are likely to profoundly affect the movement of organisms among habitat patches, limiting their ability to relocate to more favourable climate zones. For organisms that are intolerant of desiccation, recovery from periods of drought will depend on their ability to access refuges and to recolonise following these events. Addressing threats to aquatic systems requires a landscape-scale management approach. Although the principles of landscape conservation have been applied for some time in terrestrial systems (e.g. forests and grasslands), such an approach has not been adopted in aquatic systems (Doerr et al. 2010). In Australian systems the focus has been largely on assessing habitat connectivity for native mammals and birds living in woodland and forest ecosystems; the processes that connect aquatic systems are less understood (Weins 2006, Doerr et al. 2010). Developing a landscape-scale approach for the conservation of aquatic systems is a challenging task. Aquatic systems are diverse, and this necessitates a multi-species approach (Haig et al. 1998) that considers the processes that allow the exchange of plants, invertebrates, fish, amphibians and waterbirds among habitats. The dispersal potential of these organisms differs greatly, from waterbirds capable of moving vast distances to sessile organisms such as plants and many invertebrates that rely on vectors to disperse their propagules (e.g. seeds and vegetative fragments of plants, and resting eggs of invertebrates). In the following chapters a conceptual framework for understanding how dispersal operates in wetland systems is outlined, followed by a synthesis of the dispersal biology of the main groups of wetland-dependent taxa, including invertebrates, plants, frogs and fish. For each group the objective has been to identify where possible:
|