Australian flora evolution

Source:

• The evolutionary history of the Australian flora and its relevance to biodiversity conservation

• Chapter · January 2015 DOI: 10.1017/CBO9781139519960.01

Summary

• Australia is a repository of a unique flora once widespread throughout Gondwana, the southern supercontinent

• As the connectivity to other landmasses was lost, an indigenous Australian flora evolved in response to changing environmental conditions from prevalently wet to prevalently dry

• Although the Tertiary fossil record suggests the continent- wide dominance of rainforest communities, it also documents an adaptive trend towards cooler, drier and more seasonal climates

• More recently, the Quaternary saw a succession of glacial cycles that produced transitional plant communities that were considerably different from current ones

• On the whole, major natural disturbances and long-term landscape modifications have steadily contributed to Australia’s floristic diversity

• However, the arrival of people has introduced increasing and more rapid pressure on natural habitats through intensive land usage and degradation, as well as the introduction of pests and weeds resulting in a new set of threats to biodiversity

• Protecting and conserving species as well as the condition of natural ecosystems is becoming and increasingly pressing concern that requires an integrated approach

• Describing the extent and distribution of diversity and endemism is important, but should represent only a preliminary step towards the development of conservation criteria

• Biodiversity management is increasingly reliant on our understanding of the relative vulnerability of species and communities to current, past and future threats

• As we better understand how species and assemblages respond to temporal changes in environmental disturbance, it will become increasingly possible to explore proactive conservation and restoration strategies

1.1 Origins and distribution of Australia’s floristic diversity

• Australia is a unique repository of plant diversity representative of a flora once widespread throughout Gondwana, the southern supercontinent

• An indigenous Australian flora evolved as the connectivity to other landmasses was lost and the continental plate moved from polar to equatorial latitudes

• A continent-wide change from prevalently wet to prevalently dry conditions stimulated floristic diversification by driving the rapid radiation of dry- adapted vegetation types

• However, conditions have not been consistently uniform across the continent

• Environmental and biogeographic barriers have impacted on connectivity and diversification, particularly during the recent climatic cycles of the Quaternary

1.1.1 A very brief account of the major events that shaped the early Australian flora

• The earliest recognised vascular flora appeared around the Early Silurian (over 400 Mya) and by the end of the Devonian lycophytes (an ancient plant group including club-mosses and quillworts) started to develop latitudinal and regional differentiation across emerged landmasses, attaining astonishing size and emulating the structure of trees within the Tropics 

• These early vascular plants dominated the flora through to the later part of the Carboniferous (to 300 Mya) when a shift of the southern landmasses toward polar latitudes cooled the climate and triggered the ascendancy of seed ferns

• During the early Permian (251 - 359 mya), the retreat of glacial ice-sheets brought about further global transformations that contributed to the southern radiation of the distinctive Glossopterids

Through the Permian and into the Mesozoic (between 300 and 200 Mya), emerged landforms across the globe were connected into a large, continuous landmass, Pangaea

The considerable latitudinal extent of this supercontinent became one of the primary drivers of floristic differentiation, resulting in distinct southern and northern plant assemblages. At that time, landscape-level transformations instigated major floristic restructures in the south, and the radiation of ferns, seed ferns, cycads and conifers resulted in a net increase in biodiversity. This floristic transition represented the culmination of a long period of geological and environmental transformation.

The physiological innovations that provided embryos with greater protection from the elements were critical to the evolution of our modern flora. The Gymnosperms (seed- producing plants) were dominant and most diverse throughout the Mesozoic. Yet, although the fossil record contains an abundance of conifer lineages affiliated to extant ones (such as Aracauriaceae and Podocarpaceae), many of the Gymnosperms that domi- nated the Jurassic went extinct towards the end of the Cretaceous. A set of unique environmental conditions combining warmer climates with extreme photoperiods (Australia and Antarctica were connected at very high latitudes) were conducive to further ecological and evolutionary innovations and the increasing dominance of the flowering plants (Angiosperms), as exemplified by the diversification of the ancient genus Nothofagus (Hill et al., 1999; Figure 13.1, Plate 32).

13.1.2 The rise of flowering plants

During the Tertiary the Southern-Gondwanan flora was dominated by broad-leaved vegetation. A phase of global warming during the Eocene (between 55 and 35 Mya)

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 261 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 261

      Figure 13.1 (Plate 32) Nothofagus moorei in the Border Ranges (NSW). This is the most northerly Australian member of a quintessential Gondwanan genus extensively represented within the fossil record. The genus is still present from South America, through to New Zealand and Papua New Guinea. (Photo by M. Rossetto.)

resulted in the continent-wide dominance of rainforest communities. Some of the lineages present in these extensive southern forests can still be found within rainforest remnants scattered along the eastern coast of Australia (Weston & Kooyman, 2002). This surprising taxonomic conservatism has sometimes inspired incorrect narratives of Australian rainforests being static replicates of those ancient communities.

The morphological characteristics of macrofossils from the Australian flora of the Tertiary also document a trend towards the cooler, drier and more seasonal climates that still shape our flora (Greenwood & Christophel, 2005). By the end of the Eocene, the separation of Australia from Antarctica resulted in the establishment of circumpolar currents and in the formation of polar ice caps. As a result, the post-Eocene flora was increasingly variable across the landscape with dry-adapted (scleromorphic) and at a later stage arid-adapted (xeromorphic) vegetation types becoming more common (Hill, 1998).

Although typical southern families such as the Proteaceae (currently including banksias, grevilleas and waratahs) appear in the fossil record from around 90 Mya, it was only by the Early Oligocene (around 30 Mya) that floristic assemblages resembling extant communities started to appear, such as heaths and peat-swamps. As the

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 262 [259–277]12.7.20141:09PM

262 Maurizio Rossetto

    Box 13.1 Recent insights on the origins of the Australian flora

Gondwanan origins Recent discoveries have expanded and sometimes modified our understanding of evolutionary patterns within the Gondwanan flora. The fossil record that represents the Mesozoic and Cenozoic vegetation of what is now Patagonia (South America) is uncovering taxa that are unexpectedly similar to those now almost exclusively restricted to Australian mesic forests (Wilf et al., 2005). These findings support the increasingly popular notion that Australia’s unique flora is representative of ancient vegetation types once widely distributed across the southern supercontinent. Interestingly, the fossil deposits from southern Argentina have also uncovered the oldest known Eucalyptus macrofossils (Gandolfo et al., 2011). Such findings stress the complex evolutionary history of Australia’s characteristic dry-adapted flora, and its links to ancient marginal Gondwanan habitats.

Ancient families, recent radiations The available fossil record shows that ancient plant groups, such as the cycads, have preserved consistent morphological patterns through time. This has lead to the general acknowledgment that maximum cycad diversity in Australia evolved during the Jurassic and has since decreased under competitive pressure from the Angiosperms. Surprisingly, recent molecular studies revealed that most of the existing cycad diversity in Australia originated as recently as 12 Mya (Nagalingum et al., 2011). Similarly, population-level studies in Macrozamia found that current distributional patterns in Central Australia are the result of even more recent dispersal events (Ingham et al., 2013). These studies demonstrate that ancient clades can still actively respond to new evolutionary opportunities.

Major diversification episodes As the Antarctic Circumpolar Current was established after the separation of Australia and Antarctica in the Oligocene, closed forests were gradually replaced by sclerophyllous open vegetation continent-wide. Molecular dating studies have shown that this period of increased continental drying corresponded to the timing of major evolutionary radiations among some of the most iconic dry-adapted lineages (Allocasuarina, Banksia, Eucalyptus, Tetratheca for exam- ple) and to the timing of extinction of many rainforest lineages (Crisp et al., 2004; Crayn et al., 2006). Molecular dating also identified coincident patterns between East–West speciation and diversification, and the aridification and elevation of the Nullarbor Plain (13–14 Mya), an important barrier to East–West migration (Crisp & Cook, 2007).

      Australian continent (Sahul) came into contact with the Philippines sea-plate around 25 Mya, and progressively closed-in on the Sundaland continent to the north, opportunities for floristic exchange emerged (Sniderman & Jordan, 2011). In the Late Oligocene–Early Miocene, Australia crossed into subtropical latitudes and the marked shift in dominance from Nothofagus to Myrtaceae and Casuarinaceae symbolizes the extent of relentless environmental transformation. The aridification of Central Australia started around 15Mya and intensified during the past 4My. By then, forests had changed from closed to open canopy across the continent, and warm/wet plant communities were mostly relegated to coastal areas. See also Box 13.1.

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 263 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 263

 13.1.3 The cyclical events of the Quaternary

During the past 2 My, a succession of glacial cycles has brought about rapid worldwide shifts in habitat suitability and availability. During the past 700 kya climatic fluctuations were particularly intense (Hays et al., 1976). Globally cooler, dryer climates became more prevalent and tropical deserts expanded. In the northern hemisphere the ice-ages were characterised by the expansion of extensive ice sheets that reduced the habitat available to whole communities, a process that left distinct signatures on the fossil record and in the genetics of populations (Petit et al., 2005).

The ecological and physiological responses of single species to these extreme cycles have played an important role in defining current floristic patterns. Some species adap- ted to changing conditions or dispersed to more suitable locations, while others only persisted in marginal portions of their former range or became extinct (Hewitt, 2000). Consequently, much can be learned about species- and community-level resilience to ongoing change from the exploration of historical dynamics.

In Australia glacial maxima did not bring about extensive ice sheets; however, an overall decline in moisture shaped the distribution of vegetation by increasing aridity, fire frequency and fire intensity. Well-dated fossil pollen records from northern Queensland suggest that increased climatic instability in the past 200 ky brought about the replacement of araucarian forests by true sclerophyllous vegetation dominated by eucalypts (Kershaw et al., 2007). The last glacial maximum was particularly intense, as evidenced by the appearance of large dune fields throughout the continent, and by the severity of habitat fragmentation, exemplified by Eucalyptus intrusions in tropical rainforest as recently as 8 kya (Hopkins et al., 1993).

The brief interglacial periods (such as the current one) brought about warmer, moister and more predictable climatic conditions. Populations constrained to localised refugia by the previous glacial peak could take advantage of these new circumstances to expand into newly available habitat. These cyclical expansion/contraction events produced transitional plant communities that were considerably different from current ones. For instance, the combination of genetic and environmental data suggests that some species were able to survive in habitats otherwise considered as unsuitable (Worth et al., 2009). Interestingly, an increasing number of studies investigating temporal changes in distri- butional patterns suggest that the loss of a species’ ‘core’ habitat does not necessarily imply its localised extinction (although it certainly indicates a greater level of vulner- ability; Stewart & Lister, 2001).

13.1.4 Patterns of diversity: why are species distributed and assembled the way they are?

The absence of direct land-bridges between Australia and other continents for over 20 My has led to the endurance of a distinctive local flora (over 90% of the 30 000+ flowering plants are endemic; Chapman, 2005). The prevalence of low-nutrient soils, the scarcity of altitudinal relief, and the continuing impact of large-scale disturbance events (such as fires, floods, droughts and cyclones) contributed to diversification, but also resulted in an uneven spread of diversity (Box 13.2). At present most of the arid centre remains diversity-poor, while coastal regions such as the Wet Tropics and the Southwest are particularly species-rich (Hopper & Gioia, 2004).

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 264 [259–277]12.7.20141:09PM

264 Maurizio Rossetto

    Box 13.2 Learning from the past to predict the future: what regulates rainforest diversity?

The distribution and diversity of the Australian rainforests have been transformed since the breakup of the Gondwanan supercontinent and through the extreme climatic cycles of the Quaternary. Broad-leaved vegetation has contracted continent-wide and is now restricted to a ‘continental archipelago’ of remnants along the east coast where the altitudinal relief provided by the Great Dividing Range preserves suitable habitat. Yet, despite occupying only 1% of the continental landmass, rainforests are still characterised by high floristic diversity and endemism.

While assemblage-level richness decreases significantly to the South and to the West, many rainforest species have distributions that transcend those of areas tradi- tionally identified as rainforest refugia (Plate 10). Such patterns show how the selec- tive filtering brought about by large-scale environmental change has impacted differently on species and assemblages through time.

Palaeoecological evidence suggests that rainforest vegetation contracted during the cooler, drier conditions of the glacial peaks (Kershaw et al., 2007). The impact of these recurrent events varied according to landscape-level stability and to the availability of micro-refugia (Graham et al., 2006; Mellick et al., 2012). It is possible to differentiate between stable refugia and re-colonisation areas by interpreting regional measures of phylogenetic diversity in the light of recurrent disturbance events (Kooyman et al., 2011). Fine-scale population genetic studies can then ascribe unambiguously con- flicting distributional patterns in closely related species to differences in dispersal potential (Rossetto et al., 2009).

Functional groups are likely to respond to large-scale environmental disturbances in different manners (Rossetto & Kooyman, 2005). At a species level, the survival of rainforest tree populations is dependent on their capacity to persist locally (by resprouting after disturbance for example; Rossetto et al., 2004a) or to move to newly available habitat (Rossetto et al., 2004b). Interestingly, functional fitness varies with local conditions. For instance, the absence of large frugivorous vertebrate (such as the cassowary) in northern NSW limits the distribution of taxa that are otherwise more widely dispersed in the Wet Tropics (Rossetto et al., 2008).

Continental-scale functional biogeographic patterns can also reveal local disturb- ance history. A comparative study between regional floristic pools discovered a prevalence of easily dispersed rainforest lineages within areas that have endured significant levels of long-term disturbance (Kooyman et al., 2011). These associative patterns highlight how re-colonisation processes can vary according to local environ- mental and ecological histories.

Overall, the Australian rainforest flora demonstrates a considerable degree of resilience despite having been significantly impacted by climatic and geological change. As direct and indirect anthropogenic pressures increase, understanding how temporal transformations in selective pressures impact on species and func- tional groups can help us assess their longer-term vulnerability.

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 265 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 265

 By investigating the distribution and assemblage of species across the landscape, we can identify the major biotic and abiotic factors filtering the distribution of lineages. The study of biodiversity turn-over across environmental gradients is becoming an increas- ingly popular research topic for evolutionary and conservation ecologists. Three main measures of biodiversity are commonly referred to: gene diversity, lineage diversity and assemblage diversity. Each of these measures has unique evolutionary significance and informs conservation in a distinct manner.

Quantifying biodiversity at the gene level can identify the distribution of important adaptive variation (Hoffmann & Willi, 2008). Lineage diversity can be used to cir- cumscribe the geographic boundaries of evolutionary differentiated provenances (Evolutionary Significant Units, ESUs; Moritz, 1994) and guide translocation and restoration processes. At a broader geographic level, quantifying species richness or measuring phylogenetic diversity and endemism can help identify regions of high conservation priority (Forest et al., 2007). Finally, the definition of floristic and functional assemblages can be used to explore repeated landscape-level patterns and identify vulnerable communities.

It must be remembered, though, that describing the distribution of diversity is only an initial step. In order to interpret the distribution of biodiversity and develop adequate management strategies, the relative impact of current (adaptive) vs historical (biogeographic) processes needs to be investigated.

13.2 Conservationchallenges

The Gondwanan heritage is still a conspicuous element of Australia’s flora, with major natural disturbances and long-term landscape modifications steadily contributing to current floristic diversity. The arrival of humans between 65 kya and 40 kya brought further and arguably more rapid and extreme change. The landing of early hunter- gatherer communities has impacted directly (via locally altered fire regimes) and indi- rectly (via the extinction of the megafauna; Rule et al., 2012) on native vegetation. The latest wave of colonization over the past 200 years has introduced increasing pressure on natural habitats through intensive land usage and degradation, as well as the introduc- tion of pests and weeds (see also Chapter 6) resulting in a new set of threats to biodiversity.

13.2.1 Major threats to the Australian flora

The development of effective biodiversity conservation strategies relies on the identifica- tion of the biotic and abiotic processes that are a threat to the abundance, distribution and survival of taxa (Table 13.1). Key threatening processes are defined as those directly causing the decline of natural plant populations (Falk, 1990). These include the destruction of habitat, competition by invasive species, the loss of associated fauna (pollinators, dispersal vectors and other symbionts), the loss of genetic diversity and adaptive potential, and direct extirpation (via clearing, foraging or disease).

In order to develop adequate flora management strategies, it is important to appreciate the broader temporal context of threats, as the impact of external stresses can change through time, particularly as a result of ever increasing anthropogenic pressures.

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 266 [259–277]12.7.20141:09PM

266

Table 13.1

Maurizio Rossetto

Listed threatening processes: an example from New South Wales (Australia). The list of 26 key threatening processes currently (August 2013) listed as impacting on native flora in NSW under the Threatened Species Conservation Act 1995 (shown in alphabetical order) www.environment.nsw.gov.au/threatenedspecies

    Threatening process impacting on native flora

 Alteration of habitat following subsidence due to longwall Introduction and Establishment of Exotic Rust Fungi mining of the order Pucciniales pathogenic on plants of the

family Myrtaceae

  Alteration to the natural flow regimes of rivers and streams Introduction of the large earth bumblebee (Bombus and their floodplains and wetlands terrestris)

  Anthropogenic climate change Invasion and establishment of exotic vines and scramblers

  Bushrock removal Invasion and establishment of Scotch broom (Cytisus scoparius)

  Clearing of native vegetation Invasion and establishment of the cane toad (Bufo marinus)

  Competition and grazing by the feral European rabbit Invasion of native plant communities by African Olive (Oryctolagus cuniculus) Olea europaea L. subsp. cuspidata

  Competition and habitat degradation by feral goats (Capra Invasion, establishment and spread of Lantana hircus) camara

  Competition from feral honey bees (Apis mellifera) Invasion of native plant communities by Chrysanthemoides monilifera (bitou bush and

boneseed)

  Forest Eucalypt dieback associated with over-abundant Invasion of native plant communities by exotic psyllids and bell miners perennial grasses

  High-frequency fire resulting in the disruption of life cycle processes in plants and animals and loss of vegetation structure and composition

Loss and degradation of native plant and animal habitat by invasion of escaped garden plants, including aquatic plants

  Herbivory and environmental degradation caused by feral Predation by the ship rat (Rattus rattus) on Lord Howe deer Island

  Importation of red imported fire ants (Solenopsis invicta) Predation, habitat degradation, competition and disease transmission by feral pigs (Sus scrofa)

  Infection of native plants by Phytophthora cinnamomi Removal of dead wood and dead trees

       The combined effect of threatening processes needs also to be considered at the com- munity as well as at the species-level. For instance a number of Australian ecosystems, such as tropical savannas and temperate eucalypt forests, have been identified as partic- ularly vulnerable to multiple combined threats including changes in hydrology, extreme weather events, salinisation, sea level rise, pollution and overexploitation (Laurance et al., 2011).

Protecting and conserving species as well as the condition of natural ecosystems requires an integrated approach. This involves support from pertinent research, the development of long-term monitoring strategies, the upgrade of existing protected areas (including their location and status), the use of predictive conservation management approaches (that incorporate climate change scenarios), and the continuing development of adequate regulatory and educational tools. Increasing con- sideration has been given to human-induced climate change, and Lindenmayer et al. (2010) suggested a range of landscape-level biodiversity conservation strategies specifically targeted to this issue. These include the significant reduction of greenhouse

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 267 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 267

 gas emissions, improved tackling of pre-existing stressors, preparing for the effects of major natural disturbances, significant improvements to off-reserve conservation, and making the existing reserve system more comprehensive, representative and forward-looking.

13.2.2 Conservation of threatened plants

Within a setting of limited time and resources, an important issue arising from conservation planning is the prioritization of what needs to be conserved. Such priority has traditionally been given to species and communities that are rare and potentially already in trouble from an evolutionary perspective.

A central management tool employed by environment protection agencies is the development of recovery plans listing and prioritizing conservation actions for individual threatened species. In principle, such plans provide a comprehensive and strategic inventory of actions such as threat abatement, surveys, monitoring and research that should ultimately lead to the removal of the species from the threatened species list. The listing of recovery actions relies on a basic understanding of the ecological, genetic and physiological factors likely to influence long-term survival. Unfortunately, a compromise needs often to be struck between the urgency for action, the availability of resources and the need for critical research-based information.

Recovery activities for threatened species usually involve one or more of the following steps: habitat protection (including grazing, fire and weed control), disease contain- ment, ex-situ storage and propagation (including seed banking; Offord & Meagher, 2009), habitat restoration, and finally various forms of translocation (Box 13.3; See also Figure 13.2, Plate 49). These activities are normally prioritised according to the degree of impact they will have on the recovery of this and other species, and according to resource requirements and feasibility. Generally though, the recovery of a species is never a simple short-term project, and successful down-listing can rarely be achieved without the early establishment of monitoring strategies and performance measures (in NSW, for example, no rare plants have been down-listed due to recovery actions alone).

Because of the limited amount of resources available for the recovery of such a considerable number of threatened species (612 plant species and 103 ecological communities are currently listed in NSW alone), single-species recovery plans and actions are not always the best option. In fact, recent studies have shown that the development of a recovery plan does not necessarily influence the status of a threatened species, mostly because of the paucity of resources dedicated to its implementation, and the failure of planning for long-term maintenance (Bottrill et al., 2011). Consequently, investing in actions that benefit more than one species is increasingly being recommended as an important means to improve biodiversity conservation outcomes.

However, multi-species recovery plans should not bring together a random group of threatened taxa. The approach through which species are combined within a single recovery plan needs to be considered carefully, because even closely related taxa can be ecologically different. Consequently the development of truly shared recovery actions can be difficult. The recognition that not all species contribute equally to ecosystem function has shifted the focus of ecological and conservation research from taxonomic to functional grouping (Westoby & Wright, 2006).

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 268 [259–277]12.7.20141:09PM

268 Maurizio Rossetto

Box 13.3 Major considerations for translocation as a conservation tool for rare plants (from Vallee et al., 2004)

Translocations can be defined as the transfer of plants from an ex-situ collection or a natural population to a location in the wild. There are three main types of translocation actions for threatened species.

* Enhancement: where an existing population is increased in size and in genetic diversity via the planting of additional individuals.

* Re-introduction: where a new population is established in a site where it formerly occurred but it is now extinct.

* Conservation introduction: where a new population is established in an environ- mentally and ecologically suitable site outside its known range.

Many translocation attempts have been unsuccessful in the past, consequently the need to translocate and the likelihood of success need to be carefully considered before starting (Godefroid et al., 2011). The main objective of a translocation program for a threatened plant should be to directly support the conservation of the target species through the establishment and maintenance of self-sustaining populations. Translocation can be deemed as successful by the following criteria.

* Management and control of the threats responsible for the rarity of the species (including site management and protection).

* Long-term establishment of translocated individuals.

* Replication across multiple sites in order to avoid risks associated with stochastic

events.

* Successful recruitment within translocated populations.

Once it has been decided that translocation is justified and likely to succeed, a number of preliminary steps need to be followed. These include organising a relevant recovery team with the necessary skills, and gathering relevant information includ- ing collating research findings where necessary. The type of biological information needed includes the following.

* Understanding the species’ taxonomic status and evolutionary potential. This includes the species’ population genetics measuring diversity, inbreeding, as well as current and historical dynamics.

* Understanding the species’ reproductive biology. This includes the species’ primary mode of regeneration, breeding system, pollination and dispersal mechanisms, seed biology, and demography.

* Understanding the species’ ecology. This includes the species’ distribution range, its edaphic and climatic requirements, associated flora and fauna, relevant func- tional and life history characteristics, response to fire, and its susceptibility to local threats.

* Understanding optimal propagation and collection strategies. This includes being aware of ex-situ propagation and seed storage techniques, and in situ planting methodologies.

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 269 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 269

      Figure 13.2 (Plate 49) Successful reintroduction of the Corrigin grevillea (Grevillea scapigera) in the Western Australian wheatbelt. This site was weeded and partially restored. Reintroduced plants are successfully growing and reproducing. (Photo by Robert Dixon. Reproduced with permission.)

Once these issues have been considered, recipient sites have been selected and a suitable experimental design has been developed, a translocation proposal can be prepared and the necessary resources can be gathered. Source material can then be collected based on the available understanding of the species’ biology, sites can be prepared by removing threats and restoring native vegetation (if necessary), and plants can be planted being mindful of timing and experimental design. Finally, a translocation project can only be successful if appropriate monitoring, evaluation of success and ongoing management are planned from the start (Monks, 2008).

As an additional note, it is important to remember that climate change brings to the fore further considerations when planning rare species translocations. Prioritization criteria should consider the ability of the species to track predicted change based on its functional capacities and the characteristics of its ideal habitat. ‘Stranded’ species will raise a range of issue relating to planting outside the species’ natural range. The risks involved with the potential of hybridisation and localized invasions needs to be balanced with extinction risk. These issues further highlight the importance of including evolutionary research and long-term monitoring in threatened species conservation plans.

Research based on ecological and functional traits indicates that if the recovery objective is to reinstate functioning ecosystems, community diversity (species richness) is not as relevant as functional diversity. Within this context, functional grouping of threatened species can be used to identify shared threatening processes and to coordi- nate recovery efforts across entire regions, while simultaneously responding to the

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 270 [259–277]12.7.20141:09PM

270 Maurizio Rossetto

requirement for cost-effectiveness (Kooyman & Rossetto, 2008). Such progressive multi- species approaches are likely to provide more resource-efficient information gathering and lead to more focused implementation strategies.

13.2.3 Biodiversity and climate change

Human-induced climate change is likely to further impact on a biosphere already heavily perturbed by human-derived threatening processes such as decreased availability of water, clearing and fragmentation, invasive species and pollution (Steffen et al., 2009). Rapid climate change can also act directly on the physiology and life cycle of plants. One of the critical issues of increasing concern for conservation biologists is to understand how plant species and floristic assemblages will respond to rising CO2 levels. For instance, simulated future scenarios project strong floristic shifts towards more woody communities across Africa (Higgins & Scheiter, 2012). These predictive models support the expectation that atmospheric CO2 has been and will be a major factor shaping future vegetation patterns (Manea & Leishman, 2011).

Changing temperature gradients can directly impact on the reproductive phenology of species and establish new barriers to gene flow among geographically separated populations (Elzinga et al., 2007). This, in turn, can cause genetic isolation and localized reproductive failures. As a consequence, the longer-term evolutionary context needs to be carefully considered when outlining new protected areas in preparation for climate change, so that selected sites can improve connectivity and anticipate possible modifi- cations in key environmental and climatic criteria (Bellard et al., 2012).

Plants have a range of options to counteract climate change. They can take advantage of existing genetic plasticity, or they can rely on gene flow from other localities to complement their adaptive response. Either of these two options relies on the presence of sufficient genetic variation accessible within short timeframes. Alternatively, individ- uals will need to be able to disperse to more suitable habitats, provided that these are available and accessible (Corlett & Wescott, 2013). Such range shifts have already been described (Parmesan, 2006) but these are only possible for functional groups that are suited to dispersal (such as species with wind- or vertebrate-dispersed fruits for example). Furthermore, landscape-level connectivity can be particularly difficult in areas where habitat fragmentation and land development are increasingly impacting on natural processes.

Overall, it can be expected that complex case-specific interactions between the func- tional capacities of species and local environmental conditions will result in significant changes in the composition and distribution of plant assemblages. As local environ- mental conditions change, species with a sufficiently broad evolutionary spectrum will be more likely to persist locally either through plasticity or through adaptive changes. If and when species run out of options, direct management actions (including assisted colonization, which is the focus of growing conservation debate) will need to be con- sidered before anthropogenic needs exclusively dictate landscape conservation criteria.

13.2.4 Restoring communities: achieving a balance between integrity and evolutionary potential

Strictly speaking, ecological restoration projects endeavor to return a community to a perceived historical state. Unfortunately, good planning and long-term monitoring are

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 271 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 271

 not always considered in restoration projects, as these often have numerical or aesthetic targets rather than ecological ones. A more realistic (and arguably more evolutionarily fitting) goal may be to move a degraded community towards a less disturbed state that will enable the restored ecosystem to recover compositional structure and function, as well as dynamics (Falk et al., 2006). Consequently, restoration projects should aim to re-establish the dominant species of key functional groups first (such as nitrogen fixers or early successional woody species), rather than focus on number of species alone. Targeting key functional elements becomes the first step towards the recovery of the ecosystem services that are critical to broader-level biodiversity conservation (seed dispersal, pollination, pest control, for example).

Restoration practice and research have made significant advances in fields such as ecophysiology and functional ecology, food webs and the distribution of resources, the impact of invasive species and biodiversity dynamics (see relevant chapters within Falk et al., 2006). There has also been increasing consciousness about the importance of including an evolutionary context within restoration ecology. Although the objective of restoration projects is to maximize the self-sustainability and adaptive potential of re-established populations, this does not necessarily translate to trying to maximize local genetic diversity. Inbreeding is not always a negative factor and heterozygosity is not always a positive one. The introduction of novel diversity can potentially reduce the fitness of locally adapted populations (a process known as outbreeding depression; Templeton, 1994) and, accordingly, the sourcing of re-vegetation material should be considered in light of explicit evolutionary criteria. However, because of the limited resources devoted to obtaining relevant data (as well as the inevitable practical limita- tions), the selection of provenance material is often based on basic distributional patterns.

Much has been written about provenance sourcing and it is now evident that the definition of ‘suitable’ provenance varies from species to species (Broadhurst et al., 2008). The issue is not only with identifying provenance boundaries but also with defining the relative evolutionary impact of those boundaries. The perceived benefits of retaining the genetic ‘integrity’ of a site are based on the assumption that locally sourced plants are better adapted to local conditions and will therefore survive longer, grow faster and have reproductive advantages over non-local plants. However, plants that are locally adapted now may not be so in the future.

Furthermore, the risks associated with the mixing of material from geographically distant populations are generally overestimated, particularly within recently fragmented landscapes (Frankham et al., 2011). Excessively conservative sourcing criteria could in fact increase the risk of inbreeding and further erode an already low genetic pool. As a consequence, the availability of information on genetically diverse and climatically suitable non-local provenance material will become ever more important in the future (Hancock & Hughes, 2012).

Rapid climate change can accentuate existing disequilibria between a species’ geographic spread and the availability of suitable habitat, with or without direct human intervention. Under the current circumstances of high anthropogenic disturb- ance, biodiversity conservation should increasingly focus on managing change as restor- ing past assemblages will become gradually more difficult. Thomas (2011) argued that a philosophy based on restoring the composition of biological communities as they were is out of sync with the reality of current environmental and biological change. An

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 272 [259–277]12.7.20141:09PM

272 Maurizio Rossetto

alternative option for restoring evolutionary fitting ecosystems and conserve bio- diversity is the translocation of species outside their normal distributional range. Although assisted colonisation is generally considered only if traditional conservation strategies are insufficient, the risks involved might be contained if planting takes place within the same broad geographic region and the destination sites lack related endemics.

Populations stranded in increasingly unsuitable refugia, particularly those surrounded by areas mainly dedicated to human use, might be suitable targets for active manage- ment. Moving individuals from warm-adapted populations to cooler locations may increase the probability of more widespread resilience under a warming environment (Webber et al., 2011). Such approaches can potentially be applied to species naturally distributed across wide environmental gradients or to species that have been shown to display genetically based clines under different climatic conditions.

These are still relatively untested concepts and the current level of scientific debate on the benefits and pitfalls of artificial modifications in a species’ distribution is under- standable and justified (Ricciardi & Simberloff, 2009; Hannah, 2010). Interestingly though, reshuffling events have been relatively common in natural populations and communities, particularly during the current glacial cycles (Willis & Niklas, 2004). An increasing body of scientific work is showing that current distributional ranges some- times describe past dynamic process rather than represent a state of equilibrium with current habitat availability (Box 13.4).

Box 13.4 Reshuffling of genes, individuals and assemblages: evidence from the Proteaceae

A novel set of analytical tools can tease apart associations between the distribution of genetic diversity and a variety of relevant environmental variables. Combining molecular data with environmental niche models can track how temporal variations in habitat suitability are associated with changes in connectivity and gene flow.

A study combining molecular, morphological and bio-climatic analyses across the entire distribution of Telopea speciosissima (the NSW waratah, Proteaceae) discovered strong associations between population-level differentiation and localised climatic conditions (Rossetto et al., 2011). Fine-scale population genetics showed that as post- glacial warming broke down the phenological barriers that separated upland and lowland populations, gene flow was established between these morphologically and genetically differentiated populations.

A follow-up investigation extending to the other three continental waratahs demonstrated that within- and between-species divergences were caused by the instability of their respective climatic envelopes during glacial cycles (Rossetto et al., 2012; Plates 8 and 9). These studies recognised the allopatric basis of speciation in Telopea, as well as the transitional nature of species-level differentiation. Within such systems, the adaptive morphological and genetic segregation between lineages is strengthened during periods of environmentally enforced isolation, while lineage- level plasticity is enhanced during times of admixture.

C:/ITOOLS/WMS/CUP-NEW/5159390/WORKINGFOLDER/STOW/9781107033542C13.3D 273 [259–277]12.7.20141:09PM

The evolutionary history of the Australian flora 273

   During the Quaternary, recurring adjustments in the availability of suitable habitat impacted on the distribution and assemblage of many other species. These reshuf- flings brought together closely related species and lead to landscape-level hybrid- isation events, such as those described for a number of eucalypts from the subgenus Symphomyrtus in Tasmania (McKinnon et al., 2001).

More recent phylogeographic research has uncovered evidence of hybridisation among species that are highly morphologically and ecologically distinct. By mapping the distribution of chloroplast haplotypes shared among continental Lomatia species (Proteaceae), it was possible to identify refugial areas that protected diversity during unfavourable climatic conditions, distributional overlap zones where hybridisation was common, and biogeographic breaks that consistently prevented genetic connec- tivity (Milner et al., 2012). Such information greatly advances our understanding of the evolutionary history of this genus, as well as being critical to the detection of geographic areas of high conservation value.

Further fine-scale morphological and genetic studies on selected species of Lomatia are showing that genetic exchange at hybrid fronts continues for multiple genera- tions. However, despite continuing gene exchange the strength of the selective drivers that influence differentiation among species is such that the adaptive archi- tecture of the original species is eventually restored (McIntosh et al., 2014).

These and other studies highlight the transitional nature of species-level differ- entiation in plants. Understanding the causational processes responsible for the establishment of distinct evolutionary units is a critical step towards the development of predictive biodiversity conservation strategies.

13.3 Conclusion

Historically geological and climatic instability have been important factors influencing the local reshuffle of genes, species and communities. Yet, despite the fact that the Australian flora has evolved considerable resilience to change, the speed and intensity of human-induced disturbance presents significant threats to the preservation of biodiversity.

As anthropogenic pressure is accelerating and intensifying, there is increasing aware- ness about the need to move towards more ‘evolutionary aware’ biodiversity manage- ment strategies. Describing the extent and distribution of diversity and endemism is important, but should only represent a preliminary step towards the development of conservation criteria. Biodiversity management is increasingly reliant on our under- standing of the relative vulnerability of species and communities to current, past and future threats.

As changes in the distribution and assemblage of species take place in response to local selective pressures, it is important to differentiate between long-term dynamic processes and a state of equilibrium with currently available niche space. As manage- ment actions and landscape conservation criteria are increasingly impacted by urgent anthropogenic needs, more proactive restoration strategies will need to be explored more regularly