Biological Conservation 191 (2015) 830–838

Contents lists available at ScienceDirect

Biological Conservation

j ourna l homepage: www.e lsev ie r .com/ locate /b ioc

Discussion

Wildlife restoration: Mainstreaming translocations to keep commonspecies common

David M. Watson ⁎, Maggie J. WatsonInstitute for Land, Water and Society, Charles Sturt University, Australia

⁎ Corresponding author.E-mail address: dwatson@csu.edu.au (D.M. Watson).

http://dx.doi.org/10.1016/j.biocon.2015.08.0350006-3207/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 5 June 2015Received in revised form 13 August 2015Accepted 24 August 2015Available online 14 September 2015

Keywords:ReintroductionTranslocationNovel ecosystemsLandscape ecologyEnvironmental planningClimate change adaptation

Inmost urban and agricultural landscapes, remnants of native vegetation are surrounded by an inhospitable ma-trix. Although vagile species come and go, many reptiles, amphibians and small mammals are effectively strand-ed and declining towards local extinction. In the same landscapes, other areaswhere these species are absent areimproving in habitat quality, both through natural regeneration and active restoration efforts. So, for many spe-cies in many domesticated landscapes, there are too many individuals in some patches of decreasing quality andno individuals in patches of increasing quality. One solution to this situation is to move animals from those areaswhere there are plenty to suitable areas where there are none. These targeted translocations apply lessonslearned from revegetation to dispersal-limited animals to in-fill distributional ranges, increase population sizeand improve both demographic and genetic connectivity, pushingnonequilibrialmetapopulations away fromex-tinction via an imposed mass effect. In contrast to conventional reintroduction schemes—expensive, reactive in-terventions involving highly-trained specialists and captive-raised endangered species—these inexpensive,proactive, community-driven initiatives aim to avert future declines by keeping common species common. Havingintroduced the wildlife restoration vision, we use two scenarios to illustrate the benefits of the approach—tospecies, ecosystem function, ecological understanding, restoration practise and public engagement. As well asadhering to best-practise reintroduction techniques to ensure animal welfare is not compromised and avoiddetrimental effects to source populations or release sites, we emphasize community participation, data qualityand long-term accessibility as paramount to maximize learning opportunities.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Habitat loss, fragmentation and land-use intensification have taken aheavy toll on most ecosystems—species lost, processes disrupted andfunctions diminished. As global climate systems are further disrupted,the human population continues to grow and production agricultureaccounts for an ever increasing proportion of primary productivity(Watson et al., 2013), these pressureswill be exacerbated and ecosystemspushed progressively further into no-analogue conditions (Williams andJackson, 2007). Despite the swelling literature on range-shifts (seePatiño and Vanderpoorten, 2015), microrefugia (Sublette Mosblechet al., 2011) and assisted colonization (Gallagher et al., 2015 and refer-ences therein), our basic understanding of the microclimatic tolerances,microhabitat requirements and other fine-scale distributional determi-nants remain incomplete for most species in most systems (Lunt et al.,2013), precluding predictions (Clark et al., 2003) and severelyconstraining on-ground interventions to avert further losses.

Despite these knowledge gaps, extensive areas of many ecosystemshave been revegetated, these actions ameliorating many threateningprocesses while improving our understanding of how to minimize

further disruption from accelerating climate change (Maschinski andHaskins, 2012; Fischman et al., 2014). To date, however, restoration ef-forts and practitioners focus almost exclusively on plants (Clewel andAronson, 2013) and, when animals are considered, only rare or endan-gered species are targeted. Consequently, restored ecosystems typicallycomprise revegetated rural and urban landscapes that are either depau-perate, or overrun with invasive or weedy animal species(Munyenyembe et al., 1989; McKinney, 2008). Although supportingan increasingly complex set of plants and recreated habitats, manyurban and agricultural areas contain increasingly simplified animalcommunities (Werner, 2011). In addition to suboptimal ecological out-comes, these practises largely ignore people living andworking in theseareas (Miller andHobbs, 2002)—communities considered to be increas-ingly disconnected from nature (Colding, 2011).

Here, we advocate incorporating translocations of dispersal-limitedanimals into mainstream environmental management, focusing ontwo classes of domesticated landscapes: urban and agricultural areas.Rather than being the sole prevail of rare species or protected areas,we see animal translocations to be philosophically and practically com-parable to revegetation—useful interventions when natural recoloniza-tion is too slow and/or too patchy. By seeding highly modifiedlandscapes with recruits of locally-extirpated animals, wildlife restora-tion is a proactive scheme to maintain ecosystem function and avert

831D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

future declines while providing opportunities for local people to engagewith nature, thereby building environmental literacy in the wider com-munity. Having articulated our vision and considered how it comple-ments existing theoretical and practical approaches, we considercandidate species and landscapes best suited to wildlife restoration,and population genetic, implementation and policy factors. We usetwo illustrative scenarios to showcase how wildlife restoration couldbe applied to landscape-scale conservation and close with five princi-ples to guide further discussion. Although relevant to policy formula-tion, environmental planning and restoration practise, this perspectiveis intended primarily for conservation biologists, challenging them tore-imagine the domesticated landscapes where we live and work asvanguards for ecological restoration.

2. Wildlife restoration

Originally involving the planting of non-native species for erosioncontrol and aesthetic improvement (Burton and Burton, 2002), revege-tation has become the main tool for restoration and remediation. Thisshift in motivation is reflected in advanced techniques to harvest seed,grow seedlings and maximize establishment success (of rare and com-mon species; ruderal and climax communities alike), all guided byclearly-articulated objectives to effect lasting improvements to ecologi-cal processes and function (Burton and Burton, 2002; Maschinski andHaskins, 2012). By striving to emulate site-specific successional dynam-ics, current restoration efforts and practitioners rarely consider animals(Clewel and Aronson, 2013), operating under the tacit hypothesis thatas habitat structure and resource availability increase, animals will pro-gressively return (Palmer et al., 1997). Empirical tests of this hypothesishave foundmoremobile species aremore likely to recolonise, but many

Fig. 1. Simplified landscape illustrating how targeted translocations of dispersal-limited animallandscape, presumedhabitat suitability for a lizard reflecting overall productivity, ranging fromgPanel 2 depicts the same landscape after initial clearing—habitat in the most productive land(coloured lizard denotes occupied patch; white lizard = unoccupied). Panel 3 depicts the fragmnant but has increased in unoccupied remnants (through both deterministic and stochastic proeration. Panel 4 (bottom) depicts the outcome of translocating lizards into unoccupied habitat pto increasing patch occupancy, habitat suitability in the source area increases as density-depenregenerated site refines habitat suitability criteria for future releases, the colour of this habitat

animals are either unable or unwilling to navigate less hospitable inter-vening areas (Watson, 2010; Craig et al., 2012; Vergnes et al., 2013).

To illustrate how metapopulation dynamics can affect animals inhighly-modified landscapes, imagine a checkerboard, shaded squaresdenoting patches where a species is present, white where it is absent(Fig. 1). Rather than being static through time, occupied andunoccupiedhabitats may change, especially for vagile species for which patchesblink on and off across a landscape. For sedentary species or taxa of lim-ited vagility, patterns of occupancy may be more stable, but the trajec-tory of individual populations will vary—some populations growing,others declining. For those patches surrounded by an inhospitable ma-trix, recruits will be unable to leave the patch, often leading to crowdingand lower resource availability, diminishing habitat quality over time.So, rather than uniform, occupied cells may change shade as populationtrajectories rise and fall (Fig. 1). But many white cells—those unoccu-pied areas—also vary. Through time, an increasing proportion of unoc-cupied patches may contain all the resources required by a species—allthat is missing are recruits. Where habitats are less fragmented andfunctional connectivity is high, animals eventually locate these unoccu-pied areas (Huxel and Hastings, 1999). But, for highly-fragmented sys-tems and dispersal-limited animals, the likelihood of recruitscolonizing unoccupied patches is low and, for many woodland- andforest-dependent species, that probability is very close to zero (Craiget al., 2012 and references therein). So, in many landscapes, land-usechange, improved agricultural practises and the passage of time (e.g., in-dustrial areas rezoned as residential estates, broad-acre ploughing re-placed with no-till cropping; street trees maturing and forminghollows) have resulted in large areas switching from unsuitable to suit-able, yet a suite of limited-mobility animals are unable to colonize andinhabit them (Watson, 2010). Faunal relaxation through both

s can be used to maximize occupancy of suitable habitat. Panel 1 (top) depicts the originalreen (productive, high habitat suitability) to red (lowproductivity, lowhabitat suitability).forms was disproportionately cleared and the lizard persists only in the largest remnantented landscape ~100 years later—habitat suitability has decreased in the occupied rem-cesses) and a new patch of potential habitat has become available through natural regen-atches (two high suitability remnants and one low suitability regenerated site). In additiondent effects are alleviated by reducing crowding. Successful establishment of lizards in thepatch changing from red to green.

832 D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

deterministic and stochastic processes represents a one-way ratchet to-wards progressively emptier habitats performing fewer ecosystem ser-vices becoming more vulnerable to invasion by exotic species.

A solution to this situation is to move individual animals from thoseareas where there are plenty into areas where there are none (Fig. 1).Rather than occasional interventions undertaken only when regionalpopulations become scarce, we envision community-driven environ-mentalmanagement involving large numbers of translocations of abun-dant andwidespread speciesmotivated by the axiom “keeping commonanimals common”. By taking excess individuals from remnants where aspecies persists and moving them to unoccupied but otherwise suitablehabitat patches, a range of both pure and applied benefits can be real-ized. Considered in population terms, these actions equate to givingnonequilibrial metapopulations a push away from extinction via an im-posed mass effect (after Leibold et al., 2004), while temporarily reliev-ing density-dependent effects in source populations. From anecological perspective, these actions collectively comprise a coordinateddistributed experiment, testing current ideas about habitat preferencesand resource requirements for candidate species. In ecosystemmanage-ment terms, wildlife restoration increases the functional diversity ofremnant habitats, making themmore resistant to invasion. From a con-servation perspective, this intervention spreads risk, increasing both de-mographic and genetic connectivity, thereby safeguarding againstregional loss. From an education and engagement perspective, thiscommunity-driven initiative involves the wider public in active conser-vation management in their local area, empowering landholders andproviding a new range of incentives to reward best-practisemanagement.

If this visionwere to be realized under steady-state conditions, best-case scenario outcomes would be the founding of a few subpopulationsof already common andwidespread species. Many introductions wouldfail, while most reintroduced subpopulations would likely blink off in ageneration or two. But steady-state conditions are unlikely for manylandscapes—species distributions are shifting under novel climatesand no analogue communities are already developing (Williams andJackson, 2007). Land-use patterns are changing, with entire districtspreviously used for production agriculture becoming better suited tocarbon farming and other non-extractive uses (Watson et al., 2013).By introducing locally-extirpated species to pockets of suitable vegeta-tion in the areas where we live and farm, we are augmenting the num-ber of candidate native species able to persist in these alteredlandscapes over the long term. In addition to in-filling distributionalranges, seeding novel ecosystems and safeguarding regional popula-tions, wildlife restoration embraces the essential human element ofconservation biology. Identifying and conserving species of local signif-icance is an effective way to promote environmental advocacy and in-crease linkages between people and nature (MacDonald and Johnson,2003). Returning salamanders to an isolated wetland, skinks to a golfcourse, or shrews to a cemetery gives communities their own localchampions while increasing their environmental literacy (Cairns,2002) and contextualizing national and continental-scale visionswithinwhich these actions are embedded (Foreman, 2004).

Wildlife restoration complements the existing portfolio of conceptsand associated tools available to conservation researchers and restora-tion practitioners. It relies on many of the lessons learned from rare,captive-raised animals in reintroduction biology, but applies them tofree living individuals of common species. It shares the same goal asassisted colonization/managed relocation (moving individuals to avoidan unacceptably high number of extinctions), but addresses this objec-tive by infilling gaps in current distributions rather thanmoving speciesbeyond current ranges. Our vision responds directly to the growing rec-ognition that natural recolonization is too slow, too patchy, or both(Gallagher et al., 2015). Our vision embracesManning et al. (2009) con-cept of landscape fluidity, with anticipatory restoration actions neededin many systems to maximize retention of species and maintenance ofecosystem function. Our approach complements the emerging concept

of rewilding (Foreman, 2004), downscaling from continents, tigersand wolves to parks, lizards and frogs. Wildlife restoration aligns withthe growing understanding of the importance of climatically-determined microrefugia and their key role as sources of recruitsunder rapidly changing climates (Sublette Mosblech et al., 2011;Weeks et al., 2011). Finally, our approach borrows from aquatic ecology,specifically the emphasis of recruitment limitation as a key factor driv-ing community assembly (Roughgarden et al., 1987). By treating trans-locations of common animals as a coordinated distributed experiment(after Fischman et al., 2014), we can expand our understanding of hab-itat requirements and distributional constraints whichwe can then pro-gressively apply to restored, rehabilitated, and possibly recreatedsystems. By taking a proactive approach and managing species whilethey are still common, this approach emulates preventative medicine,achieving better outcomes more efficiently and cost effectively thanwaiting until problems arise (Huxel and Hastings, 1999). Finally andmost importantly, rather than being just another tool for restorationecologists to mend broken ecosystems, wildlife restoration is drivenby thewider community, involving the general public in initial selectionof candidate species and sites, eventual monitoring of establishmentsuccess and all intermediate stages.

3. Lessons from reintroduction biology: what we know andwhatwedon't know

Humans have been moving wild animals around for millennia,motivated by a variety of nutritional, recreational and aesthetic objec-tives (Griffith et al., 1989; Seddon, 2010). Although the majority ofrecent interventions involve restocking and enlarging the distributionof game species for hunting andfishing, they are increasingly being con-ducted for conservation management (Griffith et al., 1989). Transloca-tion for conservation purposes relates almost exclusively to rare orendangered species, with many programmes using captive-raised ani-mals (Armstrong and Seddon, 2008). Originally conducted on an adhoc basis (often as last ditch attempts to avert extinction) the field of re-introduction biology hasmatured and developed into its own discipline(Serena and Williams, 1995), complete with synthetic reviews andcomparisons of successful and unsuccessful initiatives (Fischer andLindenmayer, 2000; Sheean et al., 2012).

Several factors consistently emerge as predictors of successful rein-troduction programmes, directly informing how to design wildliferestoration programmes for maximum benefit. Griffith et al. (1989)found habitat quality to be critical to successful establishment, andthat reintroductions to core areas within prior distributional rangesare more successful than to peripheral areas. Veitch (1994) noted thatthe most important prerequisite to successful reintroduction was that“the original cause of their extirpation from that locality must bereversed or other repair activity undertaken to a point where thereintroduced species can survive” (see also IUCN, 1996). But identifyingthe driver of extirpation is rarely straightforward, and often relies on in-tuition and hearsay (Ewen and Armstrong, 2007). In addition to partic-ular threats or resource requirements, this difficulty also relates tohabitat suitability—where a species presently occurs need not be a reli-able basis for determining the full range of habitats where that speciesonce occurred. Moreover, given climatic changes and dynamic shifts inthe gradients defining distributional ranges, prior occurrence maypaint a misleading picture of future suitability. The difficulties ofassessing habitat suitability underscore the need for experimentalapproaches, treating introductions as trials to test existing ideas abouthabitat preferences and refine our understanding of both bottom-upand top-down factors constraining realized distributions (Morton,1987). Although researchers have long advocated using this approachwith common species (Serena and Williams, 1995), reintroductionprogrammes remain focused on rare and endangered taxa, with neitherthe numbers of animals nor willingness to risk losing animals to im-prove our understanding of their needs.

833D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

Plants, however, are regarded very differently by researchers andrestoration practitioners alike. Large-scale reintroductions of commonspecies are conducted on a routine basis, often using a variety ofmethods to determine which factors determine establishment success(Clewel and Aronson, 2013). As well as revegetation, transplant exper-iments are a standard tool to determine distributional constraints anddiscern genotypic from environmental effects on plant growth (Lesicaand Allendorf, 1999). This traditional approach has undergone a renais-sance as interest in plant distribution has switched from understandingdistributional patterns to predicting the effects of climate change(Savolainen et al., 2007), with protocols refined and standardized tomaximize comparability (see Kremer et al., 2012).

The only group of animalswhere reintroductions of common speciesare routinely conducted is fish. Although the primary goal of mostprogrammes is to maintain stocks for commercial and/or recreationalfisheries, conservation management and maintaining ecosystem func-tion are increasingly driving restoration efforts (Downs et al., 2002).Aside from habitat restoration for spawning substrate, fish introduc-tions typically entail releasing large-numbers of hatchery-bred juve-niles. Translocation of adults is occasionally conducted (e.g., movingadults of anadromous adults upstream of dams and other barriers) butconnectivity with headwaters is typically achieved using fish laddersand other engineering solutions. Interestingly, the way fish are treatedand the language used to describe these actions have more in commonwith plants than terrestrial animals—adults are “transplanted” (Olssonet al., 2006); hatchlings are “grown” in “nurseries”. Many of the ethicaland philosophical aspects of translocation have been discussed and re-solved in the fisheries management literature, and moving animalswithin and between watersheds is now widely practised. Althoughfisheries management may have been the original objective for mosttranslocations, conservation management typically follows, with riversunifying diverse stakeholder groups and catalysing whole-of-catchment restoration efforts (Perrow et al., 2002).

Whether through fear of upsetting the ‘natural balance’ or simplybecause this approach is foreign tomost terrestrial ecologists, systemat-ic and replicated translocation of common animals is rarelycountenanced, let alone conducted (Serena andWilliams, 1995). Ratherthan being relevant only for threatened species management or neces-sarily involving costly programmes centred on captive individuals andextensive reserves, we see translocation as a valuable tool in routinenatural resourcemanagement. The greatest factor constraining assistedcolonization and integrating animals explicitly into restoration is a fun-damental lack of understanding of the factors defining habitat suitabili-ty (Lunt et al., 2013; Gallagher et al., 2015). In addition to testing ourfragmentary understanding of habitat requirements and distributionalconstraints, mainstreaming translocations with common species willarrest future population declines while addressing all ten ofArmstrong and Seddon's (2008) key questions in reintroduction biolo-gy. By building capacity and incorporating organizational learning op-portunities for management agencies, our vision will generate theexpertise and social licence required to realize the potential of assistedcolonization and other bold interventions presently deemed prohibi-tively risky.

4. Practicalities

4.1. Candidates

Instead of rare, charismatic or highly-interactive taxa, wildliferestoration focuses on common species that are dispersal limited, patch-ily distributed and locally abundant. Although traditionally disregardedin conservation biology, common species are disproportionately impor-tant in ecosystem structure and function (Gaston, 2010). Thus, subtledeclines in their local abundance and distribution can drive pervasivecascades, leading to loss of less common species and diminished provi-sion of ecosystem services. Dispersal ability can be difficult to quantify

in absolute terms, but identifying those species within a local biotathat are unable to move between remnant habitats is more straightfor-ward (e.g., using expert opinion;Watson, 2003).Many amphibians, rep-tiles and small mammals are ideal candidates, especially those speciesdependent on vernal pools/wetlands (i.e., home ranges constrained byreproductive strategies) or those species reliant on burrows/hollows(i.e., central-place foragers). Rather than necessarily vertebrates, variousarthropods and molluscs would also be candidates for translocation(e.g., Sherley, 1995). For long-lived animals, or those groups with lowreproductive rates and high site fidelity, comprehensive surveys of po-tential source areas are required to ensure that removal of individualswill not have lasting detrimental impacts.

In terms of candidate areas for release sites, selection criteria will bedefined by the species beingmoved and by the motivations of the com-munities involved. In addition to habitat extent and quality, careful con-sideration of threats is needed, even though the reasons underlyinglocal absences may be unclear. Baseline surveys to confirm absencesand establish site suitability will become progressively more refined asoutcomes of earlier translocations are integrated and determinants ofhabitat suitability aremore finely resolved. For sites deemed unsuitable,these surveysmay identifymanagement interventions required tomax-imize suitability for future releases (e.g., installation of nest boxes or re-moval of incompatible exotic species; see Groenewald, 1992). Althoughsource populations closest to release sitesmight seemmost appropriate,peripheral populations (relative to range boundaries and microclimatictolerances) should be considered, both tomaximize retention of uniquehaplotypes but also recognizing that the habitats presently occupied bythe species need not reflect the full range of suitable habitats (Ewen andArmstrong, 2007). Finally, althoughwe envisagewildlife restoration be-comingpart of revegetation and remediation initiatives, initial interven-tions should relate to remnant habitat, both to maximize the likelihoodof establishment and to build up a picture of the factors defining habitatsuitability.

4.2. Population genetic considerations

Most of the issues surrounding provenance, out-breeding depres-sion and how best to select founders to avoid inbreeding depressionhave been thoroughly explored in the revegetation and reintroductionliterature, but several matters warrant exploration here. Although re-vegetation practitioners have long championed the need to uselocally-sourced seed, there is little evidence for superior growth oflocal ecotypes or “home site advantage” (Galloway and Fenster, 2000,see also Lesica and Allendorf, 1999). Referring explicitly to out-breeding depression in reintroductions, Weeks et al. (2011) wrote:“We suggest that perceived, but generally unsubstantiated, risks placetoo much constraint on current management options, commonly lead-ing to inaction”, while Burton and Burton (2002) noted: “Indeed advo-cacy for the use of ‘local only’ plant materials can take on nearreligious proportions even when scientific imperatives are lacking”. Incontrast, inbreeding avoidance requires explicit consideration, relatingto the source populations used, as well as the individuals selected fortranslocation. Current approaches to growing native plants for revege-tation use a “multi-lineal approach”—using mixtures of seed stockfrommultiple sources to ensure genetic breadth and population robust-ness, under the working assumption that if local genetic adaptation re-ally matters, local environmental filters will select from this broadermix, ensuring a close fit (Burton and Burton, 2002). Hence, by drawinganimal recruits from multiple source populations, both the amount ofgenetic variation represented and the adaptive potential increases, re-ducing the likelihood of both genetic and demographic losses acrossthe wider region. Just as with habitat suitability, successivereintroductions represent valuable opportunities to improve our under-standing of how genetic factors influence population persistence(Fenster and Dudash, 1994, see also Pääbo, 2000).

Fig. 2.Boston is a large citywith both high-density urbande-velopment and a series of interconnected greenways knownas the Emerald Necklace. These remnants and restored sitesrepresent valuable habitat for many birds and other vagiletaxa, but a suite of dispersal-limited taxa is missing. Manyspecies of bumble bee are common but declining (Collaet al., 2011): translocations of red-belted bumble bee(Bombus rufocinctus) would bolster populations, resultantinformation then applied to less common species. Urbanareas could also provide refuges for bumble bees from agri-cultural pesticides. Red-backed salamanders (Plethodoncinereus) thrive in urban environments, lower predationpressures in urban wetlands leading to increased survi-vorship of aquatic larvae. Finally, northern short-tailedshrews (Blarina brevicauda) are ideal candidates for ur-ban translocation, more hospitable climatic conditionsincreasing survival over winter enabling rapid expansionfrom initial release sites into established gardens. Photo-graphs by Larry Connolly (background), Sean McCann(red-belted bumble bee), Gilles Gonthier (short-tailedshrew) and Dave Huth (red-backed salamander).

834 D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

4.3. Implementation

Instead of being conducted solely by researchers,wildlife restorationrelies on the participation of contributors with a much broader com-bined skill-set. Baseline surveys of potential release sites and sourceareas would be designed and conducted by natural resource manage-ment agency staff with at least a Masters-level education in wildlifeecology (or equivalent). While researchers may need to be involved inassisting with initial development of protocols and advice on speciesand site selection, these officers would be responsible for makingdecisions, conducting the translocations and monitoring their success.Rather than imposed, this work would necessarily be conductedin close consultation and direct participation of local stakeholders.Whether involving university students to help capture recruits, land-holders contributing to baseline surveys, or citizen scientists assistingwith subsequent monitoring, opportunities for community involve-ment will enable greater efficiency but also greater longevity ofprogrammes by building and maintaining the necessary social licence.Careful consideration of the perceptions regarding translocation out-comes would be needed; e.g., how best to communicate the scientificvalue of failed efforts? In many jurisdictions, all of these proposedactions can be conducted within current legislative and managementframeworks, but policies and regulations may need adjustment. Thesechanges to regulatory frameworks need not be problematic, but couldcatalyse discussions with policy makers and enforcement officers atlocal and regional levels, adding their expertise and integrating theselocal actions with catchment and regional scale conservation planning.

5. Scenarios

Initially, we envisage wildlife restoration to be most applicable intwo classes of domesticated landscapes—those dominated by urbanand agricultural development. In both settings, remnants of native veg-etation are surrounded byhigh-contrastmatrices (sensuWatson, 2002),making inter-patch movements possible only by vagile taxa (e.g., birds,bats, butterflies). With time since initial habitat loss, the matrix (i.e.,suburbia and farmland) will become progressively more hospitable as

vegetation (parks and gardens; roadsides and on-farm plantings) ma-tures and becomes more complex (e.g., Munyenyembe et al., 1989;Manning et al., 2009).

These similarities notwithstanding, the emphasis and medium- tolong-term objectives of wildlife restoration in these two classes of land-scape differ. For urban areas, the high density of people coupled withthe small size of most potential release sites means that establishinglarge populations is a secondary goal. Rather, engaging the wider com-munity with biodiversity in their local area is the aim, and this willneed to be considered in selecting potential species and candidateareas for release. By integrating wildlife restoration with existingprogrammes centred around particular parks or reserves and building-in opportunities for volunteers, the overall benefits will extendwell be-yond the particular species and places involved. Given that cities charac-teristically develop in places with high fertility soils, reliable rainfall andmore hospitable climates than surrounding areas (Werner, 2011), weanticipate establishment of new populations to be most successful forspecies associated with high productivity landforms. By contrast, thelower density of people living in agricultural landscapes and larger rem-nants of vegetation associated with low productivity landforms meansthat establishing new populations and filling in distributional gaps willbe themain priority. Rather than being conducted in isolation,we envis-age wildlife restoration to become part of existing natural resourcemanagement approaches, assisting in efforts to restore functionalconnectivity to minimize local extinctions as regional climates change.Engagement will be a secondary aim and may best be achieved viatargeted programmes through local schools, allowing students to recon-cile their understanding of apex predators and rewilding initiativeswithtangible examples of local, familiar animals (MacDonald and Johnson,2003).

5.1. Scenario 1—American city

Most cities, especially those in the Northern hemisphere that havebeen urbanized for centuries, contain networks of small habitat patches(i.e., parks and reserves, Fig. 2) initially set aside or specifically plantedto provide aesthetic amenity and recreation (see McKinney, 2008;Colding, 2011) linked by smaller habitat patches (i.e. street trees and

Fig. 3. The south-west slopes region of New South Wales (south-eastern Australia) is representative of many agriculturallandscapes—habitats on the more productive soils disproportion-ately cleared andmost remaining habitat restricted to low fertilitylandforms and linear strips beside creek-lines and roads. As land-use intensifies, the area used for cultivated crops and livestock pro-duction decreases, with both extent and quality of native habitatsincreasing in many areas. Of the dispersal-limited, patchily-distributed and locally abundant candidates, two reptiles andone mammal are ideal candidates for translocation within this land-scape. Blue-tongue lizards (Tiliqua scincoides) are awidespread skinkand, although common in gardens and large remnants, they are fre-quently killedbyvehicles on roads and thusunable to recolonize oth-erwise suitable remnants (Koenig et al., 2001). Squirrel glidersdepend on large hollow-bearing trees as denning sites but, althoughthey readily use nest boxes, they do not cross open ground so are un-able to access isolated woodlands. Lace monitors are large-bodiedpredatory lizards that persist in disturbed landscapes but were inad-vertently extirpated from many farming areas through secondarypoisoning (from historic rabbit control efforts). Photographs byDavid M. Watson (background), Roger Smith (lace monitor),Alex Clarke (blue-tongue lizard) and Michael Todd (squirrelglider).

835D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

backyards). As suburbs age, gardens mature and structural connectionsdevelop (e.g., transport corridors beside roads and railways), habitat ex-tent and quality increase for many species, with birds, butterflies andbats typically increasing in diversity (McKinney, 2008; Werner, 2011).Only the more vagile subset of the original biota can recolonise theseimproving habitats, however with under-exploited resources potential-ly facilitating invasive species and subsidizing native mesopredatorsand other opportunistic generalists (e.g., raccoons Kluza et al., 2000).Those dispersal-limited taxa are represented by relictual populationsalong city margins and riparian buffers, threatened both by density-dependent factors and habitat loss as cities expand and new develop-ments encroach.

Of the dispersal-limited species distributed along the fringes ofurban areas, bumble bees are one of the more speciose groupsthat tend to thrive in urban habitat, are not aggressive, are easilytranslocated using artificial nest boxes (Hobbs et al., 1960) and can ben-efit both native vegetation and gardens through pollination services. Asan initial trial, ten schools add andmonitor ten bee nest boxes in nearbygreen spaces or school grounds (e.g., Golick et al., 2003). Each nest boxlocation is assessed (plant diversity, temperature, aspect) prior totranslocation, and continually monitored by students for two years.This trial provides a large scale, high sample-size translocation experi-ment where all aspects—from nest box building to release and subse-quent monitoring—are conducted by the community. South-facingnest boxes within 50 m of native flowering shrubs and within 100 mof a reliable water source may prove more successful, refining criteriafor homeowners who wish to join into the city-wide bumble beenetwork.

After this proof of concept trial, school and community groups can berecruited to participate in a larger-scale release of salamander eggs acrossthe whole city. Relocating the eggs of common salamander species intourban environments spreads the risk of regional extinction and teachesstudents about nature while potentially founding additional breedingpopulations. Releases can be timed to coincide with water quality assess-ments or rainfall events, thus providing experimental data to determinethe exact requirements for egg translocations,while providing recreation-al opportunities for local groups and reinforcing ties to their local naturalspaces. Nomatter the exact species chosen, the improvements to local di-versity will increase regional population stability and seed new

microrefugia and source populations from which these species canspread. Most of all, projects such as these create incentives for increasedpublic engagement in ecosystem management, forging partnerships be-tween disparate community groups and increasing awareness and litera-cy in local environmental issues (Cairns, 2002; Colding, 2011).

As a final candidate for translocations,moving shrews into parks andcemeteries will facilitate expansion from relocation sites to adjoiningbackyards (Vergnes et al., 2013), and pave the way for future transloca-tions of other predatory species. As with many small mammals, shrewsare morphologically conservative (Fig. 2), and are frequently confusedwith rodents. These identification issues will encourage educationalopportunities at local schools and community groups to learn aboutthe diversity of small mammals found in the area. Indeed, such a pro-gramme might lead to the discovery of previously undocumentedrelictual populations of small mammals persisting in the area. Byselecting a small, non-threatening predator, discussions of other carniv-orous species could be initiated, potentially catalysing the social changenecessary to develop future translocation projects that contribute tolarger-scale rewilding programmes.

5.2. Scenario 2: Australian farmland

Most southern Australian agricultural regions are managed forwheat and sheep production, but large-scale revegetation has becomepart of many properties' ongoing land management. Initially undertak-en to combat gulley erosion around drainage lines and salinization infloodplains, these plantings are augmented by shelterbelts and farm for-estry plantations. These activities, coupled with natural regenerationand regrowth on abandoned farmland and adjacent roadsides has re-sulted in the proportion of native vegetation increasing inmany farmingdistricts (Geddes et al., 2011; Watson et al., 2013). Although mostuncleared vegetation is restricted to ridges and other low productivitylandforms (Watson, 2011; Fig. 3), revegetation and regeneration has in-creased tree cover in riparian areas, valley floors and other high produc-tivity sites, improving habitat quality and extent for many woodlanddependent animals (Geddes et al., 2011). Reintroductions are increas-ingly being used as management tools for rare and threatened species(e.g., Manning et al., 2011), often in concert with feral animal controland other threat abatement measures.

836 D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

Of the dispersal-limited, patchily-distributed and locally abundantcandidates, skinks are one of the more diverse groups, most species re-liant on coarse woody debris, litter and outcrops (Koenig et al., 2001;Michael et al., 2011; see also Craig et al., 2012). Preliminary communica-tions with potential landholders explain the need for trials with multi-ple species to improve understanding of habitat requirements, therebyprogressively increasing the likelihood of success. As an initial trial,five species of skink are translocated to five properties, each containingfive patches of remnant woodland. Having conducted surveys to deter-mine suitability and locate nearby source populations, all translocationsare carried out in aweekend. Although a relativelyminor exercise for anenvironmental manager, to a quantitative animal ecologist, this repre-sents 125 separate experimental trials to determine habitat preferences,25 replicates for each species (providing the basis for two PhDprojects).Two years after release, repeat surveys find skinks were more likely toestablish in those woodlands with standing dead trees, crash grazing(rather than set stocking or no livestock grazing) and if recruits weredrawn from more than one source population. In addition to refinedprotocols for future releases, set stocking and removal of dead treesfor firewood are prioritized by the regional land management agencyas issues to emphasize when considering applications for small grantsto maximize biodiversity values and carbon sequestration on farms.

After this proof-of-concept trial, another set of translocations is un-dertaken, moving two groups of eight squirrel gliders (Petaurusnorfolcensis; sourced from four adjacent populations) to unoccupiedroadside remnants where ongoing fox control and retention of largehollow-bearing trees meet the habitat suitability criteria. The animalsonly establish in one of the two release sites, which differs from the un-successful site in having twice as many available hollows and greaterdensity of scattered trees in adjacent fields. In addition to refining thehabitat suitability criteria, this gives landholders an additional incentiveto install nest boxes on their properties andmaintain fox control, there-by increasing habitat quality for a wide range of other woodland-dependent species. By scaling up strategic translocations to encompassentire catchments, these local actions rapidly build momentum, contin-ually becoming more successful as lessons learned from previous trialsare applied.

The final candidate for translocation in this system is the Lace Mon-itor (Varanus varius), a large predatory lizard found throughout easternAustralia in woodland and forests. Eight gravid females are moved intotwo farming landscapes where monitors hadn't been seen for decades:an established grazing property raising sheep with extensive woodlandalong the ridgelines and creeks; and a group of adjoining lifestyle prop-erties and gardens with remnant roadside woodland. Pre-release sur-veys confirm the presence of termite mounds (used as egg-layingsites) and high densities of European rabbits (an invasive species andfrequent prey item). Follow up surveys conducted by undergraduatestudents confirm the presence of at least five individual monitors inboth landscapes three years after release,with dietary analysis revealingrabbits, carrion, cats and locusts are important dietary items. Althoughlandholders in both landscapes report monitors occasionally takingpoultry, they also note a decrease in rabbit activity, both in woodlandand adjoining pastures used for livestock grazing (Fig. 3). By selectinga charismatic predator, the pros and cons of larger future translocationprojects and rewilding schemes become part of the lexicon of theseland owners, making wildlife restoration less of an imposition andmore of an element of best-practise property management.

6. Principles

Building on established principles underpinning successfulreintroductions of rare and threatened species (e.g., the five principlesof Fyfe, 1978; the 10 principles of the IUCN, 1996), we propose the fol-lowing five guiding principles to maximize the lasting benefits of wild-life restoration, ranked in order of importance.

1. Ensure source populations are not adversely impactedGiven that one of primary aims of wildlife restoration is to ensure thelong-term viability of common species by founding new subpopula-tions, adverse impacts to existing subpopulationsmust be avoided. Inaddition to determining theminimum size of populations to be usedas source areas and maximum proportions of populations to be re-moved, additive impacts need to be given careful consideration.Thus, minimizing trapping for monitoring purposes, reducing han-dling of individuals to aminimum, and instituting hygienic protocolsto ensure no pathogens or potential invasive species are introducedto source areas during fieldwork. Inadvertent impacts could beassessed by comparing source areas used with potential source pop-ulations not used, both to identify potential adverse impacts beforethey become threats, but also to determine whether populationsbenefit from reduced density-dependent factors by removing indi-viduals. Finally, potential on-ground actions to improve habitats insource areas should be actively investigated, applying the same ap-proach used in revegetation to ‘seed orchards’ to maximize futurerecruits.

2. Ensure release sites are not adversely impactedThe same precautionary approach needs to be applied to all areas re-ceiving individuals, ensuring best practise animal welfare principlesare upheld and no pathogens or invasive species are accidentally in-troduced. As well as the initial release of animals, this relates to sub-sequent surveys to evaluate success of the intervention. As part ofbaseline surveys to determine habitat suitability for the targetspecies, potential interactions should be given careful consideration,especially if potential release sites contain relictual populations ofplants or animals poorly represented elsewhere. By comparing re-lease sites visited frequently for post-release monitoringwith poten-tial release sites only visited for initial surveys of habitat suitability,any unforeseen consequences (positive and negative) of transloca-tions will be detected early, informing management of affectedsites and refining protocols for future interventions.

3. Monitor, monitor, monitorThe importance ofmonitoring every stage of every translocation can-not be overstated. Rather than necessitating annual visits to all sites,monitoring must balance flexibility with consistency, working withlandholders and other community members to ensure adherenceto the same set of practises.Minimally, thismonitoring should followbasic BACI principles, comparing before and after interventions insites where translocations were conducted and sites where no ani-mals were released. In addition to measuring the successes centralto wider uptake, this monitoring will maximize learning opportuni-ties from inevitable failures. Rather than simply relating to sites andspecies, parallel monitoring of ecological functions and aspects ofthe wider community should be undertaken concurrently, enablingevaluation of howwildlife restoration alters community perceptionsand values.

4. Wildlife restoration is by the people, for the peopleIn stark contrast to interventions involving rare or endangered spe-cies, these translocations cannot be driven by biologists. To realizemaximum benefits to species, sites and communities, wildlife resto-ration actions must be driven by the same kinds of people who haverallied behind revegetation, restoration and rewilding. Rather thangiving carte blanche for anyone to move anything anywhere, thisprinciple draws on sociological research on the determinants ofbehavioural change,maximizing the likelihood ofmeaningful changeby ensuring inclusion at all stages (Cairns, 2002). Rather thansupplanting other approaches, we envisage wildlife restoration tocomplement traditional revegetation efforts, combining transloca-tions with indigenous plantings to restore connectivity and enhancethe flora and fauna of existing remnants. By adapting existing regula-tory frameworks, adjusting relevant policy instruments where need-ed and incorporating training opportunities for key stakeholders,consistency, efficiency and broader uptake will be entrained.

837D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

5. All outcomes must be sharedThe learning opportunities from these coordinated distributed ex-periments will only be realized if outcomes of all translocations arereported and widely accessible (Fischman et al., 2014). Rather thanconstraining individual groups or enforcing conformation to rigidprotocols developed elsewhere, these reporting requirements willenable early detection of determinants of successful initiatives,thereby ensuring iterative improvement of techniques and on-ground outcomes.Web-accessible distributed databases can be read-ily constructed to facilitate both information storage and retrieval—ofboth data related to outcomes and associated metadata relating toactions (who, what, where and when). By establishing these over-arching frameworks from the outset, large-scale conservation man-agement can become progressively more evidence-based.

7. Prospect

Conservation Biology has moved well past discussions of what is‘natural’, preservationist ideals (see Smith andWinslow, 2001) replacedwith themore pragmatic ‘novel ecosystems’ paradigm, with functional-ity and purpose paramount. Although we share the views of Hoegh-Guldberg et al. (2008) that assisted colonization and other bold initia-tivesmust be considered ifwe are tominimize the loss of entire lineagesfrom extinction, we also recognize the caution with which these ideashave been received (Lunt et al., 2013; Gallagher et al., 2015).We simplydo not know enough about the resource needs and habitat require-ments of species to complete risk management procedures for assistedcolonization with due diligence—we cannot be sure that the treatmentwill end up being more destructive than the underlying disease. If weadopt interventionist management with common species in the land-scapes where we live, work and grow our food, valuable lessons canbe learned to foster this much-needed confidence. By building capacityand gaining the trust of the wider community, we can generate the req-uisite momentum to start pushing the population extinction ratchet inthe opposite direction, one local good news story at a time.

8. Conclusions

Restoration ecology has undergone a rapid transformation fromreactive, conservation-driven practise to a theoretically-groundedscientific discipline, achieving improved on-ground outcomeswhile ad-dressing fundamental questions about ecological process. Here, weidentify the next key challenge for restoration ecology to embrace: toexplicitly include animals in restoration efforts. Many animals aredispersal-limited, many sites cannot be reconnected and many suitablehabitats are missing key biotic elements. Applying lessons learned fromrevegetation and principles of reintroduction biology developed withrare and endangered species, we outline a novel approachwe callWild-life Restoration. Bymoving dispersal-limited animals stranded in isolat-ed remnants to unoccupied but otherwise suitable habitat patches,distributional ranges can be in-filled, local extinctions avoided and re-gional populations of common species safeguarded. As well as beingfar more cost-effective than reactive management to reverse declines,this proactive approach has several other benefits. By designing eachset of translocations using consistent protocols, these individual conser-vation actions collectively represent a coordinated distributed experi-ment, revealing the mechanistic basis of observed habitat preferencesand identifying the climate and resource-based determinants of realizeddistributions. By returning native species, functional diversity and eco-system service provision are augmented, thereby reducing the likeli-hood of colonization by invasive species. Finally, all aspects of WildlifeRestoration involvemembers of the local community, building inmean-ingful opportunities for engagement and increasing connectedness be-tween people and the places where they live and work. The ongoingsuccess of revegetation initiatives has demonstrated the great willing-ness of landholders to repopulate their gardens, neighbourhoods and

production-dominated properties with the plants that used to growthere—why not give them the opportunity to put some of the animalsback as well?

Acknowledgments

Discussions with Amos Bouskila, Larry Connolly, Michael Craig,Damian Michael, Dale Nimmo, Euan Ritchie and Rodney van der Reeclarified the ideas expressed herein. MJW recognizes the support ofthe Institute for Land Water and Society, Conservation Evidence andthe Technion University.

References

Armstrong, D.P., Seddon, P., 2008. Directions in reintroduction biology. Trends Ecol. Evol.23, 20–25.

Burton, P.J., Burton, C.M., 2002. Promoting genetic diversity in the production of largequantities of native plant seed. Ecol. Restor. 20, 117–123.

Cairns Jr., J., 2002. Rationale for restoration. In: Perrow, M.R., Davy, A.J. (Eds.), Handbookof Ecological RestorationPrinciples of Restoration 1. Cambridge University Press,Cambridge, pp. 10–23.

Clark, J.S., Lewis, M., McLachlan, J.S., HilleRisLambers, J., 2003. Estimating populationspread: what can we forecast and how well? Ecology 84, 1979–1988.

Clewel, A.F., Aronson, J., 2013. Ecological Restoration: Principles, Values, and Structure ofan Emerging Profession. 2nd ed. Island Press, Washington DC.

Colding, J., 2011. Creating incentives for increased public engagement in ecosystem man-agement through urban commons. In: Boyd, E., Folke, C. (Eds.), Adapting Institutions:Governance, Complexity and Social-Ecological Resilience. Cambridge UniversityPress, Cambridge, pp. 101–124.

Colla, S., Richardson, L., Williams, P., 2011. Bumble Bees of the Eastern United States.USDA Forest Service.

Craig, M.D., Hardy, G.E.S.J., Fontaine, J.B., Garkakalis, M.J., Grigg, A.H., Grant, C.D., Fleming,P.A., Hobbs, R.J., 2012. Identifying unidirectional and dynamic habitat filters to faunalrecolonisation in restored mine-pits. J. Appl. Ecol. 49, 919–928.

Downs, P.W., Skinner, K.S., Kondolf, G.M., 2002. Rivers and streams. In: Perrow, M.R.,Davy, A.J. (Eds.), Handbook of Ecological RestorationRestoration in Practice 2.Cambridge University Press, Cambridge, pp. 267–296.

Ewen, J.G., Armstrong, D.P., 2007. Strategic monitoring of reintroductions in ecologicalrestoration programmes. Ecoscience 14, 401–409.

Fenster, C.B., Dudash, M.R., 1994. Genetic considerations for plant population restorationand conservation. In: Bowles, M.L., Whelan, C.J. (Eds.), Restoration of EndangeredSpecies: Conceptual Issues, Planning and Implementation. Cambridge UniversityPress, Cambridge, pp. 34–62.

Fischer, J., Lindenmayer, D.B., 2000. An assessment of the published results of animal re-locations. Biol. Conserv. 96, 1–11.

Fischman, R.L., Meretsky, V.J., Babko, A., Kennedy, M., Liu, L., Robinson, M., Wambugu, S.,2014. Planning for adaptation to climate change: lessons from the US national wild-life refuge system. Bioscience 64, 993–1005.

Foreman, D., 2004. Rewilding North America: a Vision for Conservation in the 21stcentury. Island Press, Washington.

Fyfe, R.W., 1978. Reintroducing endangered birds to the wild: a review. In: Temple, S.A.(Ed.), Endangered Birds: Management Techniques for Preserving Threatened Species.Croom Helm, London, pp. 323–329.

Gallagher, R.V., Makinson, R.O., Hogbin, P.M., Hancock, N., 2015. Assisted colonization as aclimate change adaptation tool. Austral Ecol. 40, 12–20.

Galloway, L.F., Fenster, C.B., 2000. Population differentiation in an annual legume: localadaptation. Evolution 54, 1173–1181.

Gaston, K.J., 2010. Valuing common species. Science 327, 154–155.Geddes, L.S., Lunt, I.D., Smallbone, L.T., Morgan, J.W., 2011. Old field colonization by native

trees and shrubs following land use change: could this be Victoria's largest exampleof landscape recovery? Ecol. Manag. Restor. 12, 31–36.

Golick, D.A., Schlesselman, D.M., Ellis, M.D., Brooks, D.W., 2003. Bumble boosters: studentsdoing real science. J. Sci. Educ. Technol. 12, 149–152.

Griffith, B., Johnston, C.A., Scott, J.M., Carpenter, J.W., Reed, C., 1989. Translocation as aspecies conservation tool: status and strategy. Science 245, 477–480.

Groenewald, G.H., 1992. The relocation of Cordylus giganteus in the Orange Free State,South Africa: pitfalls and their possible prevention. J. Herp. Ass. Afr. 40, 72–77.

Hobbs, G.A., Virostek, J.F., Nummi, W.O., 1960. Establishment of Bombus spp.(hymenopetera: apidae) in artificial domiciles in Southern Alberta. Can. Entomol.92, 868–872.

Hoegh-Guldberg, O., Hughes, L., McIntyre, S., Lindenmayer, D.B., Parmesan, C.,Possingham, H.P., Thomas, C.D., 2008. Assisted colonization and rapid climate change.Science 321, 345–346.

Huxel, G.R., Hastings, A., 1999. Habitat loss, fragmentation and restoration. Restor. Ecol. 7,309–315.

IUCN, 1996. IUCN/SSC guidelines for re-introductions. 41st Meeting of the IUCN Council,Gland Switzerland, May 1995. http://iucn.org/themes/ssc/pubs/policy/reinte.htm.

Kluza, D.A., Griffin, C.R., Degraaf, R.M., 2000. Housing developments in rural New England:effects on forest birds. Anim. Conserv. 3, 15–26.

Koenig, J., Shine, R., Shea, G., 2001. The ecology of an Australian reptile icon: how do blue-tongued lizards (Tiliqua scincoides) survive in suburbia? Wildl. Res. 28, 214–227.

838 D.M. Watson, M.J. Watson / Biological Conservation 191 (2015) 830–838

Kremer, A., Ronce, O., Robledo-Arnuncio, J.J., Guillaume, F., Bohrer, G., Nathan, R., Bridle,J.R., Gomulkiewicz, R., Klein, E.K., Ritland, K., Kuparinen, A., Gerber, S., Schueler, S.,2012. Long-distance gene flow and adaptation of forest trees to rapid climate change.Ecol. Lett. 15, 365–377.

Leibold, M.A., Holyoak, M., Mouquet, N., Amarasekare, P., Chase, J.M., Hoopes, M.F., Holt,R.D., Shurin, J.B., Law, R., Tilman, D., Loreau, M., Gonzalez, A., 2004. Themetacommunity concept: a framework for multi-scale community ecology. Ecol.Lett. 7, 601–613.

Lesica, P., Allendorf, F.W., 1999. Ecological genetics and the restoration of plant communi-ties: mix or match? Restor. Ecol. 7, 42–50.

Lunt, I.D., Byrne,M., Hellmann, J.J., Mitchell, N.J., Garnett, S.T., Hayward, M.W.,Martin, T.G.,Mcdonald-Madden, E., Williams, S.E., Zander, K.K., 2013. Using assisted colonisationto conserve biodiversity and restore ecosystem function under climate change. Biol.Conserv. 157, 172–177.

MacDonald, D.W., Johnson, P.J., 2003. Farmers as custodians of the countryside: links be-tween perception and practice. In: Tattersall, F.H., Manley, W. (Eds.), Farming andMammals. Occasional Publication of the Linnean Society, pp. 2–16.

Manning, A.D., Fischer, J., Felton, A., Newell, B., Steffen, W., Lindenmayer, D.B., 2009. Land-scape fluidity—a unifying perspective for understanding and adapting to globalchange. J. Biogeogr. 35, 193–199.

Manning, A.D., Wood, J.T., Cunningham, R.B., McIntyre, S., Shorthouse, D.J., Gordon, I.J.,Lindenmayer, D.B., 2011. Integrating research and restoration: the establishment ofa long term woodland experiment in south-eastern Australia. Aust. Zool. 35,633–648.

Maschinski, J., Haskins, K.E. (Eds.), 2012. Plant Reintroduction in a Changing Climate.Island Press, Washington D.C., U.S.A.

McKinney, M.L., 2008. Effects of urbanization on species richness: a review of plants andanimals. Urban Ecosyst. 11, 161–176.

Michael, D.R., Cunningham, R.B., Lindenmayer, D.B., 2011. Regrowth and revegetation intemperate Australia presents a conservation challenge for reptile fauna in agriculturallandscapes. Biol. Conserv. 144, 407–415.

Miller, J.R., Hobbs, R.J., 2002. Conservation where people live and work. Conserv. Biol. 16,330–337.

Morton, E.S., 1987. Reintroduction as a method of studying bird behavior and ecology. In:Jordan III, W.R., Gilpin, M.E., Aber, J.D. (Eds.), Restoration Ecology: A Synthetic Ap-proach to Ecological Research. Cambridge University Press, pp. 166–172.

Munyenyembe, F., Harris, J., Hone, J., Nix, H., 1989. Determinants of bird populations in anurban area. Austral Ecol. 14, 549–557.

Olsson, I.C., Greenberg, L.A., Bergman, E., Wysujack, K., 2006. Environmentally inducedmigration: the importance of food. Ecol. Lett. 9, 645–651.

Pääbo, S., 2000. Of bears, conservation genetics and the value of time travel. Proc. Natl.Acad. Sci. U. S. A. 97, 1320–1321.

Palmer, M.A., Ambrose, R.F., Poff, N.L., 1997. Ecological theory and community restorationecology. Restor. Ecol. 5, 291–300.

Patiño, J., Vanderpoorten, A., 2015. How to define nativeness in organisms with high dis-persal capacities. A response to Essl et al. J. Biogeogr. 7, 1360–1362.

Perrow, M.R., Tomlinson, M.L., Zambrano, L., 2002. Fish. In: Perrow, M.R., Davy, A.J. (Eds.),Handbook of Ecological RestorationPrinciples of Restoration 1. Cambridge UniversityPress, Cambridge, pp. 324–354.

Roughgarden, J., Gaines, S.D., Pacala, S.W., 1987. Supply-side ecology. The role of physicaltransport processes. In: Gee, J.H.R., Giller, P.S. (Eds.), Organization of Communities:Past and Present. Blackwell Scientific Press, London, pp. 491–518.

Savolainen, O., Pyhäjärvi, T., Knürr, T., 2007. Gene flow and local adaptation in trees.Annu. Rev. Ecol. Evol. Syst. 38, 595–619.

Seddon, P.J., 2010. Moving along the conservation translocation spectrum. Restor. Ecol.18, 796–802.

Serena, M., Williams, G.A., 1995. Wildlife conservation and reintroduction: an Australianperspective. In: Serena, M. (Ed.), Reintroduction Biology of Australian and NewZealand Fauna. Surrey Beatty & Sons, Chipping Norton, pp. 247–252.

Sheean, V.A., Manning, A.D., Lindenmayer, D.B., 2012. An assessment of scientificapproaches towards species relocations in Australia. Austral Ecol. 37, 204–215.

Sherley, G., 1995. Translocations of the mahoenui giant weta Deinacrida n. sp. andPlacostylus land snails in New Zealand: what have we learned? In: Serena, M. (Ed.),Reintroduction Biology of Australian and New Zealand Fauna. Surrey Beatty & Sons,Chipping Norton, pp. 57–63

Smith, S.E., Winslow, S.R., 2001. Comparing perceptions of native status. Nativ. Plants J. 2,5–11.

Sublette Mosblech, N.A., Bush, M.B., van Woesik, R., 2011. On metapopulations andmicrorefugia: palaeoecological insights. J. Biogeogr. 38, 419–429.

Veitch, C.R., 1994. Habitat repair: a necessary prerequisite to translocation of threatenedbirds. In: Serena, M. (Ed.), Reintroduction Biology of Australian and New ZealandFauna. Surrey Beatty & Sons, Chipping Norton, pp. 97–104.

Vergnes, A., Kerbirious, C., Clergeau, P., 2013. Ecological corridors also operate in an urbanmatrix: a test case with garden shrews. Urban Ecosyst. 16, 511–525.

Watson, D.M., 2002. A conceptual framework for the study of species composition inislands, fragments and other patchy habitats. J. Biogeogr. 29, 823–834.

Watson, D.M., 2003. Long-term consequences of habitat fragmentation: highland birds inOaxaca, Mexico. Biol. Conserv. 111, 283–303.

Watson, D.M., 2010. From pattern to process: towards understanding drivers of diversityin temperate woodlands. In: Lindenmayer, D., Bennett, A., Hobbs, R. (Eds.), Temper-ate Woodland Conservation and Management. CSIRO Publishing, Collingwood,pp. 159–166.

Watson, D.M., 2011. A productivity-based explanation for woodland bird declines: poorersoils yield less food. Emu 111, 10–18.

Watson, S., Luck, G., Spooner, P., Watson, D.M., 2013. Human-induced land-cover change:incorporating the interacting effects of frequency, sequence, time-span and magni-tude of changes on biota. Front. Ecol. Environ. 12, 241–249.

Weeks, A.R., Sgro, C.M., Young, A.G., Frankham, R., Mitchell, N.J., Miller, K.A., Byrne, M.,Coates, D.J., Eldridge, M.D.B., Sunnucks, P., Breed, M.F., James, E.A., Hoffmann, A.A.,2011. Assessing the benefits and risks of translocations in changing environments:a genetic perspective. Evol. Appl. 4, 709–725.

Werner, P., 2011. The ecology of urban areas and their functions for species diversity.Landsc. Ecol. Eng. 7, 231–240.

Williams, J.W., Jackson, S.T., 2007. Novel climates, no-analog communities, and ecologicalsurprises. Front. Ecol. Environ. 5, 475–482.