Longterm Shifts in Abundance and Distribution of a Temperate Fish Fauna

Global Ecology and Biogeography, (Global Ecol. Biogeogr.) (2011) 20, 58–72

RESEARCH Long-term shifts in abundance and PAPER distribution of a temperate ﬁsh fauna: a response to climate change and

ﬁshing practicesgeb_575 58..72

Peter R. Last1,2*, William T. White1,2, Daniel C. Gledhill1,2, Alistair J. Hobday1,2, Rebecca Brown3, Graham J. Edgar3 and Gretta Pecl3

1Climate Adaptation Flagship, CSIRO Marine ABSTRACT and Atmospheric Research, GPO Box 1538, Aim South-eastern Australia is a climate change hotspot with well-documented Hobart TAS 7001, Australia, 2Wealth from Oceans Flagship, CSIRO Marine and recent changes in its physical marine environment. The impact on and temporal Atmospheric Research, GPO Box 1538, Hobart responses of the biota to change are less well understood, but appear to be due to TAS 7001, Australia, 3Tasmanian Aquaculture influences of climate, as well as the non-climate related past and continuing human and Fisheries Institute, Private Bag 49, Hobart impacts. We attempt to resolve the agents of change by examining major temporal TAS 7001, Australia and distributional shifts in the fish fauna and making a tentative attribution of causal factors. Location Temperate seas of south-eastern Australia. Methods Mixed data sources synthesized from published accounts, scientific surveys, spearfishing and angling competitions, commercial catches and underwater photographic records, from the ‘late 1800s’ to the ‘present’, were examined to determine shifts in coastal fish distributions. Results Forty-five species, representing 27 families (about 30% of the inshore fish families occurring in the region), exhibited major distributional shifts thought to be climate related. These are distributed across the following categories: species previously rare or unlisted (12), with expanded ranges (23) and/or abundance increases (30), expanded populations in south-eastern Tasmania (16) and extra- limital vagrants (4). Another 9 species, representing 7 families, experienced longer- term changes (since the 1800s) probably due to anthropogenic factors, such as habitat alteration and fishing pressure: species now extinct locally (3), recovering (3), threatened (2) or with remnant populations (1). One species is a temporary resident periodically recruited from New Zealand. Of fishes exhibiting an obvious poleward movement, most are reef dwellers from three Australian biogeographic categories: widespread southern, western warm temperate (Flindersian) or eastern warm temperate (Peronian) species. Main conclusions Some of the region’s largest predatory reef fishes have become extinct in Tasmanian seas since the ‘late 1800s’,most likely as a result of poor fishing practices. In more recent times, there have been major changes in the distribution patterns of Tasmanian fishes that correspond to dramatic warming observed in the local marine environment. *Correspondence: Peter Last, CSIRO Marine Keywords and Atmospheric Research, GPO Box 1538, Hobart TAS 7001, Australia. Climate change, fishing, south-eastern Australia, spatial shift, Tasmania, E-mail: [email protected] temperate fishes, temporal shift.

fauna has been described in a series of checklists and guides INTRODUCTION dating back to the late 1800s and early 1900s (e.g. Lord & Scott, Changes in faunal assemblages over time have been recorded in 1924). In the ‘1980s’, these fishes were surveyed extensively and many marine communities and are often linked to anthropo- good intra-regional information was obtained on the distribu- genic impacts, such as fishing (e.g. Jackson et al., 2001), pollu- tions of most species (e.g. Last et al., 1983). However, since the tion (e.g. Schiel et al., 2004), invasive species (e.g. Currie & mid-1990s many unusual records and sightings have been Parry, 1999) and climate change (e.g. Holbrook et al., 1997; reported, including coastal species not previously reported from Sagarin et al., 1999; Hiddink & Hofstede, 2008). Significant the region, poleward extensions of ‘1980s’ ranges, and increased changes in aquatic communities have been strongly associated local abundances in southern parts of Tasmania. with global warming, suggesting that these ecosystems are Stuart-Smith et al. (2009) concluded that Tasmanian subtidal extremely vulnerable to climate change (Richardson & reef communities have remained stable over the past decade Poloczanska, 2008). The effect of climate change on temperate despite evidence of localized ocean warming (Ridgway, 2007). inshore marine communities, specifically in rocky intertidal Nevertheless, they detected some species-level responses consid- zones, has been demonstrated in a number of studies (e.g. ered to be symptomatic of ocean warming, including southward Sagarin et al., 1999; Helmuth et al., 2006), with studies of the range extensions of some fishes. They suggested that major effects of climate on temperate fish communities in the North changes in faunal structure probably occurred in the decade Atlantic (e.g. Perry et al., 2005; Hiddink & Hofstede, 2008) and prior to commencement of their study; for example, forests of eastern (Holbrook et al., 1997) and south-west Pacific (Stuart- giant kelp (Macrocystis pyrifera) initially declined in the ‘1980s’ Smith et al., 2009). The Antipodean region is recognized as a (Edyvane, 2003). Hence, by examining shifts over a longer time ‘hotspot’ for ocean warming (Ridgway, 2007). Since 1944, the frame we should be able to gain better insights into the extent of East Australian Current (EAC) has extended poleward down the change. Stuart-Smith et al. (2009) also observed that less abun- coast of Tasmania by approximately 350 km and the sea surface dant species often exhibited greater shifts than the dominant temperature has warmed at an average rate of 2.28°C per elements of the fauna, presenting challenges when monitoring century (Ridgway, 2007; see also Fig. 1). Thus, biological climate impacts. responses to climate change in the region are expected to have Since 2006, when the senior author made an oral presentation already begun (Hobday et al., 2007; Ling et al., 2009). describing range changes in south-eastern Australian fishes, pat- The relatively species rich, cool temperate coastal ichthyo- terns of distributional shift in the fauna have become widely fauna of Tasmania is ideal for investigating temporal and spatial accepted (e.g. Hobday et al., 2007) but have not been rigorously shifts in species composition in this hotspot. The region is well- described. In addition, global warming may not be the only separated geographically from mainland Australia, but the driver of change in the region – changes in the fauna may have respective faunas are loosely interconnected through the Bass occurred due to exploitation. Resolving these agents of change is Strait by a series of island groups that potentially form ‘stepping important because not all distributional shifts are necessarily stones’ for the southward dispersal of adult fishes from northern climate related. Recolonization may occur in response to bioregions. Although complex, the biogeography of south- changes in fishing pressure, allowing a return of fishes from eastern Australia, which has been the subject of detailed inves- populations in less heavily exploited regions. Attribution of tigation for the purpose of regional marine planning (IMCRA, change is important for improving coastal management strate- 1996), is now reasonably well understood. The Tasmanian fish gies, as well as recording and predicting the impacts of climate change. The following hypotheses are evaluated using a combination of historical and present-day data sources: (1) the coastal ichthyofauna has experienced significant temporal losses of some elements since the ‘late 1800s’ evident from local extirpation or major reduction in population size and distribution of some species, and (2) the ichthyofauna has exhibited a marked change in composition since the ‘1980s’, with these differences represented by range extensions of warm-temperate provincial elements from the north. Herein, we define the coastal ichthyofauna as an assemblage of fishes (about 300 species) that have been recorded from the coastal zone of Tasmania.

Figure 1 The average annual sea surface temperature anomaly METHODS from the long-term average calculated from the HadlSST dataset (monthly, 1 °C; Rayner et al., (2003); http://hadobs.metofﬁce.com/ Data sources hadisst/) for the area illustrated in the inset map [40–44° S, 147–151° E] for each year (dotted line) and for each decade (open The Tasmanian coastline is situated south of latitude 39°12′ Sin bars) over the period 1880–2009. the Australian Exclusive Economic Zone (EEZ) and includes the

Figure 2 The Maugean region of south-eastern Australia, including Tasmania. Arrows indicate the influence of adjacent bioregions, i.e. Flindersian (western warm temperate) and Peronian (eastern warm temperate). islands of the central (Kent and Hogan groups), eastern Allport), Johnston compiled the first annotated checklists of (Furneaux Group) and western (King Group) Bass Strait, and Tasmanian fishes. His work was thorough and comments on mainland Tasmania, including its adjacent islands (see Fig. 2). A occurrence and commonness were provided for most of the 188 temporal reconstruction of the coastal ichthyofauna was synthe- species known to exist in the region. This list was soon super- sized from a rich variety of sources because a complete time seded by a revised version which included 214 species series of distribution and abundance data is lacking. Data were (Johnston, 1890). Johnston’s studies are particularly relevant to accessed from historical documents (published manuscripts, this investigation as data were obtained primarily from coastal checklists and books), spearfishing competition results, field assemblages, caught by beach seine, gillnet and hook and line, surveys, commercial fishing records, underwater observations, with some additional information on offshore species taken by photographic records and other unpublished anecdotal infor- pelagic fishing and bottom lining methods using hook, line and mation. Data sources varied in type, content, completeness and lure. In 1923, Lord (1923) produced a revised checklist, includ- validity, but when combined, provided a robust overview of the ing 259 species, which was supplemented by additional species composition and general distribution of species during each taken by exploratory trawl surveys of the region by the FRV period. This information was summarized to produce distribu- Endeavour. Soon afterwards, he published the first guide to ver- tional dossiers on candidate species for each of three periods: tebrates of the region (Lord & Scott, 1924), which included ‘late 1800s’ (1880–1925), ‘1980s’ (1970–85) and ‘present’ notes on 262 fish species. These data have been summarized here (1995–now)–asummaryofthese data is provided in based on a nomenclatural history and our knowledge of the Appendices S1–S3 in the Supporting Information and will be contemporary fauna. made available from the Tasmanian Coastal Climate Change Range Expansion Database and Mapping project (REDMAP) at ‘1980s’ http://www.redmap.org.au/. Common names of species referred to in this study follow the Australian Fish Names standard The 1970s and first half of the 1980s were important periods for (Yearsley et al., 2006; http://www.seafood.net.au). Main datasets the exploration of marine biodiversity in this region with several relevant to each period are discussed briefly below. surveys of inshore areas (e.g. Last, 1979; Edgar, 1984) and deeper parts of the continental shelf and slope (e.g. Last & Harris, 1981). Compositions of Tasmanian coastal fish assemblages and ‘Late 1800s’ the distributions of their species were defined during the prepa- Coastal fish communities were not exploited by indigenous ration of two regional guides (Edgar et al., 1982; Last et al., people (Luckman & Davies, 1978) and prior to the arrival of 1983). The most valuable supporting data for these publications Europeans in the early 1800s were likely to be near pristine. By came from field surveys, spearfishing competitions, fisheries sta- the mid 1800s, most distributional information on Tasmanian tistics and underwater photographic records, the most impor- fishes had been obtained incidentally by naturalists. Soon after, tant of which are documented below. fishing activities inshore increased dramatically and some popu- 1. Field surveys. In 1982, Dr Barry Hutchins, an ichthyologist lations soon displayed evidence of localized depletion from the Western Australian Museum, completed an under- (Johnston, 1883). Based on his own research and using an water survey of Tasmanian fishes as part of a broader investiga- unpublished manuscript list (compiled by a local naturalist, M. tion of the ichthyofauna of southern Australia. This 2-month

60 Global Ecology and Biogeography, 20, 58–72, © 2010 Blackwell Publishing Ltd Long-term shifts in a temperate fish fauna survey was conducted in summer when seasonal vagrants are ‘Present’ most abundant in the region. He used a combination of ichthy- opoison samples and underwater observations and photographs Our ‘present’ knowledge of the coastal ichthyofauna is based on to compile a list of about 140 species from 9 generalized sites fishery catches (both commercial and recreational), diver obser- located around the coastline of Tasmania. This survey led to the vations and information provided by scientists, scuba divers and discovery of new cryptic species and provided important distri- recreational and commercial fishers to REDMAP. Species distributional information on the coastal ichthyofauna. Dr Hutchins butions have been mapped in bioregionalization studies for had an exceptional knowledge of temperate Australian inshore regional marine planning of the Australian EEZ (IMCRA, 1996), fishes, so species missing from this survey are as important as and their biogeographic affinities have been determined from those recorded, because their absence suggests that they are most these datasets, which are held within the Codes for Australian likely either uncommon or absent on reefs in the region. Addi- Aquatic Biota (CAAB) database (http://www.cmar.csiro.au/ tional lists of species, based on underwater observations, were caab/). Broader-scale information on the distribution of warm published by two of the authors (Last, 1979; Edgar, 1984), who temperate species was obtained from important regional guides also compiled additional distributional information on fishes in (Hutchins & Swainston, 1986; Gomon et al., 2008). Newly Tasmanian waters. recorded species and those with changes in the distributional 2. Spearfishing competitions. Underwater fishing is an efficient profiles were identified; these included species with recently method of estimating the abundances of rare reef fishes established breeding populations in Tasmanian waters, latitudi- (Lincoln Smith et al., 1989); hence, catches from competitions nal range shifts and/or increases in abundance, or direct range are a valuable but under-utilized historical data source. Catch expansions in south-east Tasmania where the former range was data, collated from 12 competition sites held in northern, restricted or patchy. Eight classification categories (Table 1) were eastern and southern Tasmania in the 1970s, were pooled at each used to characterize these changes. site to provide a total catch and to determine species composition for each site. Pooled data represented effort from 315 diver Analysis days (each about 4 h day-1) culminating in c. 1300 diver hours and the collection of 2502 specimens of 50 coastal species; this Data sources, as listed above, were used to compile a list of represents some 17% of the known Tasmanian coastal fish Tasmanian fishes exhibiting temporal shifts in their distribution fauna. Numerous factors may have influenced the likelihood of and/or abundance. Information was initially assembled for the capture for targeted species, including their presence at the site, entire fish fauna, with detailed data collation focused on the diver experience, suitability of local habitat and weather condi- coastal assemblages which are best represented across the three tions. While failure to capture a species at any site cannot be time periods (i.e. ‘late 1800s’,‘1980s’ and ‘present’). By combin- definitively interpreted as a non-occurrence, this sampling ing available data we reconstructed change profiles for candidate method (competition scoring) is based on capturing as many species at each of these time intervals and assigned them to one large species as possible, so moderately abundant and common or more of the eight main qualitative categories of distributional fishes are typically thoroughly represented and a selection of change as defined in Table 1. These categories consider the rare species are usually also caught (Coll et al., 2004). change status of species using the ‘1980s’ as a ‘control’ period 3. Underwater photographic records. The first photographic (preceding recent observed changes in the‘present’ fauna), based guide to coastal fishes of Tasmania and the Bass Strait (Edgar on whether species are new to the region, were once considered et al., 1982) was based on images taken by a local group of rare, have recovering populations or now have expanded local underwater photographers and scientists. Fourteen members ranges and/or abundances (Categories 1, 3, 4, 6). A separate of the Tasmanian Underwater Photographic Society dived category was provided for species exhibiting distributional throughout the region over a 2-year period to source images of changes at the southern limit of the Tasmanian Province, the as many species as possible; they captured images of 124 of the Bruny bioregion (Category 5). Other species, which were 290 coastal fishes then thought to occur in the region, as well as recorded from the ‘late 1800s’,disappeared from the fauna by the obtaining important information on their distributions and ‘1980s’ (Category 2) or are now threatened (Category 7); still abundances. Effort data were not recorded but would have others only occur in the region as extra-limital vagrants (Cat- amounted to several hundred diver hours. Hence, species not egory 8). Species missing since the ‘late 1800s’ were further photographed were considered to be cryptic, difficult to photo- subdivided into those considered to be extinct locally, now con- graph or very rarely encountered during this period. fined to remnant populations, occurring periodically in the 4. Complete regional guide. A guide to all 461 fishes then region, or doubtful records (based on erroneous data or misi- known to occur in Tasmanian waters was prepared soon after dentifications such as confusion with sibling species). (Last et al., 1983). Much of the data were obtained through the Some species conform to more than a single category of sources described above, as well as government funded fishery change. For example, a species may have experienced a serious surveys and personal records of commercial and recreational population decline since the ‘late 1800s’ to be considered as fishermen. Hence, notes on the local distributions and abun- locally extinct by the‘1980s’ (Category 2a), but is now recovering dances of species, where specified, were used to characterize and (Category 6) with significant changes in its regional abundance summarize the fauna in the ‘1980s’. since the ‘1980s’ (Category 4). Evidence of change was inferred

Table 1 Classiﬁcation of distributional change in the Tasmanian coastal ichthyofauna and numbers of families and species of ﬁshes in each of these different categories based on the 37 families and 61 species with observed distributional and/or abundance changes.

Category Deﬁnition Families Species

1. Previously rare or unlisted species Not listed or very rare in the ‘1980s’ but now resident or becoming established; 812 primarily includes colonizers from biogeographic provinces to the north 2. Missing species Recorded in the ‘late 1800s’ but very rare or absent from the fauna by the ‘1980s’: a Localextinctions Nolongeroccursintheregion 3 3 b Remnants Remains in the region but with an extremely contracted or remnant distribution 2 3 c Periodical residents Occurs periodically as mid-term residents (c. 2–10 years) 1 1 d Doubtful records Likely misidentiﬁcation; probably confused with sibling species 9 9 3. Expanded ranges Resident exhibiting poleward range extension (usually southward) in part(s) of 15 23 Tasmania since the ‘1980s’; extra-limital vagrants were excluded from this group 4. Abundance increases Resident exhibiting an obvious abundance increase (usually southward) in part(s) of 21 30 Tasmania since the ‘1980s’; where they were once not well represented or rare 5. Expanding in the south Rare or known from small populations in south-east Tasmania in the ‘1980s’ but 14 16 now with much broader ranges; this scenario is usually accompanied by a general abundance increase 6. Recovering species Recorded from the region in the ‘late 1800s’, absent or at low abundance in the 33 ‘1980s’, and now increasing; climate change may not be the sole attribution of change for these species 7. Now threatened species No discernible change in distribution before the ‘1980s’ but now extremely restricted 12 or declining in abundance and range 8. Extra-limital vagrants Previously unlisted transient species from tropical Australian biogeographic 34 provinces; probably migrated to Tasmania with the assistance of the East Australian Current (EAC)

from a detectable decline or increase in either abundance, fre- Flindersian, western warm temperate species; Peronian, eastern quency of occurrence or geographic range. A spreadsheet pro- warm temperate species; Southern Australia (S. Aust), wide- viding supporting information and a dossier for each candidate spread throughout temperate southern Australia; Tropical, species is provided in Appendices S1–S3. mainly widespread off northern Australia; Widespread, ubiqui- The main faunal lists for the ‘late 1800s’ (Johnston, 1883; Lord tous in Australian seas. & Scott, 1924) and ‘1980s’ (Edgar et al., 1982; Last et al., 1983) were used to assess the local distributional status for each species RESULTS during these time periods. Other historical data sources, discussed above, were used to iterate these assessments, and the Some 37 families (over a third) and about 61 species (about a statuses of species is summarized in Appendix S4, according to fifth) of the coastal ichthyofauna of Tasmania, which now con- criteria outlined in Appendices S2 and S3. Secondly, distribu- sists of about 300 species, have, or appear to have, undergone tional details were mapped for each species (Appendix S3) to important compositional shifts over the three periods consid- eight Tasmanian bioregions (IMCRA, 1996) for the ‘1980s’ and ered (i.e. ‘late 1800s’, ‘1980s’ and ‘present’) leading to overall ‘present’. Abbreviations for variables, bioregions, levels of increased species richness and faunal complexity (Appen- change and the putative causes of change and their estimated dix S4). Changes were recorded in all eight of our classification confidence levels are provided in Appendix S1. Historical ranges categories (Table 1). These changes include the loss or drastic of species in the ‘1980s’, based primarily on Hutchins & Swain- reduction in relative abundance of at least five large predatory ston (1986), including specific distributional information fishes, and possibly another 14 unresolved or restricted species, largely from Tasmania (Last et al., 1983) and lists from the ‘late treated in ‘late 1800s’ checklists (Johnston, 1883, 1890) 1800s’, are summarized in Appendix S2. Present ranges, where (Table 2). More recently, there is evidence of present-day distri- they differ from the situation in the ‘1980s’, are compiled from butional shifts in 52 species (c. 17% of the coastal fish fauna), the broad variety of sources specified above (Appendix S2). including the recent range extension of warm temperate species Other summary data compiled included primary habitat and from the north that were either previously rare or unrecorded indices of commonness, occurrence and breeding status from the region (12 species), new extra-limital vagrants (4 (Appendix S3). Each species was also assigned to one of the species), those that now have expanded southern populations following classes based on their biogeographic affinities in the (23 species), those that have extended their ranges in south- Australian region (IMCRA, 1996): Bassian, primarily in the Bass eastern Tasmania (16 species) and/or increased their abun- Strait; Tasmanian, restricted cold temperate Australian species; dances in Tasmania (30 species) (Table 1). Of these, 45 species,

Table 2 Coastal species recorded by Lord & Scott (1924) exhibiting major changes in their occurrence status over the three temporal periods referred to in this study.

Name (Lord & Scott, 1924) Current name (likely) Status late 1800s Status 1980s Present status

Gyropleurodus galeatus Heterodontus galeatus Present Locally extinct Locally extinct Carcharias arenarius Carcharias taurus Not uncommon Locally extinct Locally extinct Orectolobus maculatus (Orectolobus halei) Common Extremely restricted Extremely restricted Pseudobatrachus dubius Batrachomoeus dubius Present Dubious record Dubious record Histrio histrio Histrio histrio Present Dubious record Dubious record Brachionichthys hirsutus Brachionichthys hirsutus Not uncommon Common and restricted Rare and restricted Brachionichthys politus Sympterichthys politus Common and restricted Restricted Extremely restricted Hemirhamphus intermedius (Hyporhamphus melanochir) Seasonal Dubious record Dubious record Trachichthys australis Trachichthys australis Probably common Absent Recovering Centropogon australis (Gymnapistes marmoratus) Infrequent Dubious record Dubious record Neosebastes panda (Neosebastes scorpaenoides) Not common Dubious record Dubious record Sillago maculata (Sillago bassensis) Doubtful Dubious record Dubious record Trachurus novaezelandiae (Trachurus declivis) Common Dubious record Dubious record Sparus australis (Acanthopagrus butcheri) Common Dubious record Dubious record Sciaena antarctica Argyrosomus hololepidotus Rare Locally extinct Locally extinct Dactylopagrus morwong Nemadactylus valenciennesi Not common Possibly extinct Recovering Latridopsis ciliaris Latridopsis ciliaris Rare Locally extinct Periodical resident Achoerodus gouldii (Achoerodus viridis) Common Locally extinct Recovering Notolabrus celidotus (Notolabrus tetricus) Present Dubious record Dubious record

Names in brackets are likely valid names of misidentifications. representing 27 families (about 30% of the inshore fish families ern Bass Strait coast, has established populations in the occurring in the region), exhibited major distributional shifts Furneaux Group. This genus was not previously represented in thought to be climate related. Changes in each of these catego- Tasmanian waters. All other species are reef dwellers, over a third ries are described further below. are wrasses (Labridae), and seven are Peronian species that are dominant members of a fish assemblage normally associated with barrens formed by the longspine sea urchin (Centrostepha- Previously rare or unlisted species nus rodgersii). Compositional knowledge of the Tasmanian fish fauna has expanded since the ‘1980s’ as additional species are discovered. Missing species Most recent additions to the checklist are fishes from the less comprehensively explored deep sea or the open ocean. Non- Distributional information from the ‘late 1800s’ (Johnston, cryptic, coastal species are less likely to escape detection, so in 1883; Lord & Scott, 1924) varies in its detail, but at least 16 this region, which has been extensively surveyed, their recent, species ‘went missing’ from the Tasmanian fauna before the simultaneous appearances are likely to reflect ‘real’ change. ‘1980s’ (Last et al., 1983), and another three recovering species Members of a group of 12 ‘previously rare or unlisted’ Tasma- suffered major declines in their ranges (Table 2). The group of nian inshore fishes have expanded their distributions since the ‘missing species’ includes three species that are now considered ‘1980s’ (Table 3). They include: three indigenous colonizing to be extinct locally, three with remnant distributions, a peri- species from the north (previously unlisted); six species whose odical resident and another nine species whose listings appear occurrence in the region in the ‘1980s’ was based on a few to be based on misidentifications and doubtfully occur in the specimens (rare or extremely restricted); and three species that region. had become rare by the ‘1980s’ but are now recovering (see Large predatory fishes, such as the greynurse shark (Carchar- ‘Recovering species’ section below). There is no definitive evi- ias taurus) (Fig. 3b), crested hornshark (Heterodontus galeatus) dence that any of these species have established self-sustaining and mulloway (Argyrosomus hololepidotus), have not been populations, but as most are present as either adults or in size- observed in Tasmanian waters for nearly a century so we con- able aggregations it is likely that some are capable of breeding in sidered them to be extinct locally; whereas other large predators, the Tasmanian region. the queen snapper (Nemadactylus valenciennesi) and eastern Most new regional records of these species come from the blue groper (Achoerodus viridis) (Fig. 3f), are now showing weak Furneaux Group and/or the eastern Tasmanian coast. A single signs of recovery. These are all distinctive fishes, mostly well soft-bottom species, the eastern shovelnose stingaree (Try- regarded as recreational or commercial species, and are unlikely gonoptera imitata) (Fig. 3a), which is abundant along the north- to have been overlooked if present.

Table 3 Coastal species previously unlisted (U), rare (R) or extremely restricted (R&R) in Tasmanian seas during the ‘1980s’ that are now more widespread or recovering (RE); includes their biogeographic afﬁnity, location observed, occurrence, primary habitat type, association with longspine sea urchin barrens and local breeding status.

Scientiﬁc name Status Biogeographic afﬁnity Location Occurrence Habitat Urchin barrens Breeding

Trygonoptera imitata U Bassian Flinders Localized Soft bottom Yes? Aplodactylus lophodon U Peronian C Bass Patchy Reef Yes No Notolabrus gymnogenis U Peronian S Bass Patchy Reef Yes No Achoerodus viridis RE Peronian S Bass/NE Patchy Reef Yes No Trachichthys australis RE S. Aust Flinders Common Reef Yes? Nemadactylus valenciennesi RE Flindersian W Patchy Reef Yes? Gymnothorax prasinus R S. Aust NE Patchy Reef Yes Yes? Latropiscus purpurissatus R S. Aust S Bass Patchy Reef Yes? Nemadactylus douglasii R Peronian NE Common Reef Yes? Chromis hypsilepis R&R Peronian E Patchy Reef Yes Yes? Ophthalmolepis lineolatus R&R Peronian E/SE Patchy Reef Yes Yes? Eupetrichthys angustipes R&R Peronian NE/SE Patchy Reef Yes Yes?

? = unconﬁrmed. Location: C Bass, central Bass Strait islands; E, east; Flinders, Flinders Island; NE, north-east; S Bass, southern Bass Strait; SE, south-east; W, west.

Thebluemoki(Latridopsis ciliaris)isprimarilyaNew Species with an expanded range Zealand species that appears to recruit periodically across the Tasman Sea into southern Australian waters (Yearsley et al., Twenty-three species display recent southward shifts in their 1999). It was recorded from Tasmanian waters in the late 1880s range limits, and about half of these (12 species) also show (Johnston, 1890) but its occurrence locally was questioned by subtle increases in their abundances within their extended Lord & Scott (1924). The species was omitted from Last et al. ranges (Table 4). Almost all of these fishes are reef dwellers (1983) because there had been no confirmed records from the (91%) and about half are dominant coastal elements of the 20th century despite a comprehensive search for information or Peronian Province (11 species); other biogeographic groups specimens. However, soon after (c. 1985), juveniles were pre- include widespread southern Australian (seven species), Flinder- sented to the senior author from eastern Tasmania, presumably sian (four species) and Bassian (single species) fishes. This following a large spawning event in the Tasman Sea. Sightings of pattern exists in 14 families with labrids (four species) and juveniles were frequent in the late 1980s but this cohort disap- kyphosids (three species) being the most important groups. peared shortly after an adult fish was captured in southern Warm temperate surf-zone fishes, the silver drummer (Kyphosus Tasmania in the early 1990s. This species has not been recorded sydneyanus) (Fig. 3c) and rock blackfish (Girella elevata), which from the region since, suggesting that its occurrence in the once had very restricted distributions in the region (Last et al., region is periodical. 1983), are now more abundant in the north and east, particu- Three species occur in remnant populations that have not larly in the Furneaux Group where schools of G. elevata some- recovered in recent times. A wobbegong (assumed to be the gulf times include gravid individuals. Girella elevata is one of the five wobbegong, Orectolobus halei) was considered common by most commonly speared fishes in competitions off Sydney Johnston (1883), but by the ‘1980s’ was confined to a handful of (Lincoln Smith et al., 1989). Other reef species, such as the resident adults, observed regularly at a very small, isolated reef onespot puller (Chromis hypsilepis), the snakeskin wrasse off Flinders Island (Furneaux Group). According to local divers (Eupetrichthys angustipes) and the southern Maori wrasse (Oph- this population still exists (c. 2009) but its range has barely thalmolepis lineolatus), which were restricted locally to the Kent expanded beyond this patch. Similarly, two handfishes have suf- Group (central Bass Strait), have now been recorded from fered serious range reductions and are now threatened (see‘Now several localities to the south. While seven species have broad- threatened species’ section below). ened their ranges in Bass Strait (four in the southern Bass Strait, The ‘late 1800s’ lists contain records (nine taxa) that are two off Flinders Island and one off the central Bass Strait doubtful and probably resulted from the erroneous application islands), the most striking expansions occur along the south- of names to well-known Tasmanian species; most of these are east (12 species) and/or east (five species) coasts. names of close relatives or sister species. However, three distinc- A large cheilodactylid, the blue morwong (Nemadactylus tive fishes, the eastern frogfish (Batrachomoeus dubius), the valenciennesi), is now caught infrequently off the northern west sargassum fish (Histrio histrio) and the eastern fortescue coast and remains the only species known to display a poleward (Centropogon australis), are not easily confused with other local shift off the west coast. Weaker trends in this largely remote species and may have once existed in the region; three of these region probably reflect poorer sampling effort rather than sta- genera are not otherwise recorded locally. bility in community structure.

Figure 3 Underwater photographs representing species from each of the eight change categories: (a) eastern shovelnose stingaree (Trygonoptera imitata), a previously unlisted species; (b) greynurse shark (Carcharias taurus), a missing species; (c) silver drummer (Kyphosus sydneyanus), an expanded range species; (d) Castelnau’s wrasse (Dotalabrus aurantiacus), a species with an increased abundance; (e) snapper (Pagrus auratus), a species with expanded range in south-eastern Tasmania; (f) eastern blue groper (Achoerodus viridis), a recovering species; (g) spotted handﬁsh (Brachionichthys hirsutus), a now threatened species; (h) Queensland groper (Epinephelus lanceolatus), an extra-limital vagrant. Photographs (b), (e), (f) and (h) were taken outside Tasmanian seas.

eastern coast (23 species) as along the northern coast (12 More abundant species species). These species occur in a mixture of habitats but two- Thirty species exhibit an increase in abundance in some part of thirds are reef fishes (20 species) with a smaller mix of soft their Tasmanian range (Table 4). This group comprises families, bottom (four species), seagrass (three species), pelagic (two with more than half being perch-like fishes (c. 63%) and a sixth species) and inshore demersal fishes (one species). elasmobranchs (c. 17%). Generalized increases were observed Half of the fishes in this category are warm temperate across the region: Tasmania-wide (four species), southern Bass Australian endemics and half of them are widespread in south- Strait and north-east (12 species), east (nine species), south-east ern Australia (15 species). The warm temperate Peronian (eight (13 species) and west (one species). Twice as many species species) and Flindersian (seven species) provinces are about appear to have undergone abundance increases along the evenly represented. At least 14 species (e.g. Atypichthys strigatus,

Table 4 Coastal species that have undergone recent poleward shifts and/or increases in abundance in Tasmanian waters, including the location observed, biogeographic afﬁnity, primary habitat type and local breeding status.

Scientiﬁc name Shift Abundance increase Biogeographic afﬁnity Habitat Breeding

Heterodontus portusjacksoni SE S. Aust Reef No Trygonorrhina dumerilii S Bass S Bass Flindersian Soft bottom No Dasyatis brevicaudata SE S. Aust Soft bottom No? Trygonoptera imitata Flinders Bassian Soft bottom Yes? Urolophus paucimaculatus SE S. Aust Soft bottom Yes Myliobatis australis SE S. Aust Soft bottom No? Gymnothorax prasinus NE S. Aust Reef Yes? Latropiscis purpurissatus S Bass S Bass S. Aust Reef Yes? Trachichthys australis Flinders Flinders S. Aust Reef Yes? Platycephalus laevigatus SE Flindersian Seagrass Yes Hypoplectrodes maccullochi SE S Bass Peronian Reef Yes? Hypoplectrodes nigroruber S Bass S. Aust Reef Yes? Pomatomus saltatrix Tas S. Aust Pelagic No Seriola lalandi Tas S. Aust Pelagic No Pagrus auratus Tas S. Aust Inshore No Girella elevata SE N/E Peronian Reef Yes Girella tricuspidata SE Peronian Reef Yes Girella zebra SE E/SE Flindersian Reef Yes Kyphosus sydneyanus SE NE S. Aust Reef No? Atypichthys strigatus SE N/E Peronian Reef Yes Scorpis aequipinnis E/SE Flindersian Reef Yes Scorpis lineolata SE N/E Peronian Reef Yes Enoplosus armatus ES.AustReefYes Chromis hypsilepis E Peronian Reef Yes? Parma microlepis SE Peronian Reef No Chironemus marmoratus Tas Peronian Reef Yes Aplodactylus lophodon C Bass Peronian Reef No Cheilodactylus nigripes E/SE Flindersian Reef Yes Dactylophora nigricans E/SE Flindersian Reef Yes Nemadactylus douglasii NE Peronian Reef Yes? Nemadactylus valenciennesi W W Flindersian Reef Yes? Achoerodus viridis NE Peronian Reef No Dotalabrus aurantiacus E/SE Flindersian Seagrass Yes Eupetrichthys angustipes NE/SE S. Aust Reef Yes? Notolabrus gymnogenis S Bass Peronian Reef No Ophthalmolepis lineolatus E/SE Peronian Reef Yes? Haletta semifasciata E/SE S. Aust Seagrass Yes Heteroscarus acroptilus SE S. Aust Reef Yes Odax cyanomelas E/SE S. Aust Reef Yes Eubalichthys mosaicus S Bass S Bass S. Aust Reef No Omegophora armilla E/SE SE Peronian Reef Yes

? = unconﬁrmed. Location: C Bass, central Bass Strait islands; E, east; Flinders, Flinders Island; N, north; NE, north-east; S Bass, southern Bass Strait; SE, south-east; Tas, Tasmania; W, west.

Scorpis lineolata, Dotalabrus aurantiacus (Fig. 3d) and Haletta Species with expanded populations in the south-east semifasciata) now occur as both adults and juveniles where their abundances have recently increased, and are likely to be breeding Sixteen species appear to have expanded their ranges in south- at these locations (Table 4). However, another six species (i.e. east Tasmania by exhibiting either a broader distribution or Pagrus auratus (Fig. 3e), Pomatomus saltatrix, Seriola lalandi, general increase in abundance in the Bruny bioregion (Table 5). Eubalichthys mosaicus, Heterodontus portusjacksoni and Trygon- These ﬁshes are generally large to medium-sized, with elasmo- orrhina dumerilii), are either known to breed elsewhere or are branchs well represented (four species). Most were known from unlikely to breed in Tasmanian waters. the bioregion in the ‘late 1800s’ and ‘1980s’, but were typically

Table 5 Coastal species previously known from south-east Recovering species Tasmania that have recently expanded their range and/or abundance in this region, classified by biogeographic affinity, A small group of species, which were once well represented in primary habitat type and local breeding status. the region based on early historical information but were missing or represented by remnant populations during the Biogeographic ‘1980s’, are now showing signs or recovery or have re-recruited Scientific name affinity Habitat Breeding to the region. These species include: Trachichthys australis, Heterodontus portusjacksoni S. Aust Reef No Nemadactylus valenciennesi and the eastern blue groper, Acho- Dasyatis brevicaudata S. Aust Soft bottom No erodus viridis (Fig. 3f). The recent rediscovery of juveniles of A. Urolophus paucimaculatus S. Aust Soft bottom Yes viridis, a major predator of urchins, in the north and north-east, Myliobatis australis S. Aust Soft bottom No may have important ecological implications in the future. This Pomatomus saltatrix S. Aust Pelagic No large wrasse was once considered to be a ‘common species Seriola lalandi S. Aust Pelagic No around the rocky section of the coast’ (Lord & Scott, 1924), but Pagrus auratus S. Aust Inshore No was later excluded from the Tasmanian fish fauna (Last et al., Haletta semifasciata S. Aust Seagrass Yes 1983) as it had not been seen for more than 50 years and was Odax cyanomelas S. Aust Reef Yes considered to be extinct locally. The long-term survival of N. Platycephalus laevigatus Flindersian Seagrass Yes valenciennesi and A. viridis in the region is likely to be dependant Girella zebra Flindersian Reef Yes on the implementation of prudent management plans associ- Scorpis aequipinnis Flindersian Reef Yes Cheilodactylus nigripes Flindersian Reef Yes ated with fishing. Dotalabrus aurantiacus Flindersian Seagrass Yes Similarly, the unmistakable cave-dwelling southern roughy Girella tricuspidata Peronian Reef Yes (Trachichthys australis), which was said to ‘inhabit rocky Omegophora armilla Peronian Reef Yes reefs’ (Lord & Scott, 1924) but was not observed during extensive underwater surveys around Tasmania in the ‘1980s’, is now common off Flinders Island (P. Nichols, personal communication). more widespread and abundant in the Bass Strait than in areas to the south. The level of range expansion in this bioregion varies between Now threatened species species. For example, the smooth stingray (Dasyatis brevicau- Handfishes (family Brachionichthyidae), which are endemic to data) has gone from being an occasionally occurring, seasonal Australia, are represented by at least 14 small species, and all but transient to a more widely distributed species that is abundant three of these occur in Tasmanian seas (Last & Gledhill, 2009). and possibly resident at some locations. The rock flathead Two species, Brachionichthys hirsutus (Fig. 3g) and Sympterich- (Platycephalus laevigatus), abundant in parts of the Bass Strait thys politus, both occurred in healthy populations in the ‘1980s’, but missing from the west and east, was represented in the primarily in the south-east, but have now undergone serious south-east in the ‘1980s’ by a small extra-limital population in population declines and are now listed as threatened (e.g. Last one small embayment; it has now expanded its range to become et al., 2007). Data are insufficient to ascertain the cause(s) of widespread throughout large sheltered bays of the south-east. these declines, but appear, at least in some instances, to be Reef species, such as the sea sweep (Scorpis aequipinnis) and related to critical habitat loss or damage. herring cale (Odax cyanomelas), which were typically abundant only in the Bass Strait, are now widespread on reefs along the eastern and south-eastern coastlines. Extra-limital vagrants The snapper (Pagrus auratus) was thought to be confined to the north and east coasts in the ‘late 1800s’ (Lord & Scott, 1924). Four widely distributed transient species represent new records However, while they have been caught infrequently by local for the Tasmanian region. All of these fishes occur in tropical fishermen in the south in recent decades, their abundance seas, and while two of them are typically pelagic, all occur in appears to have increased widely across their entire range and coastal habitats elsewhere in Australia, e.g. the Queensland they are now targeted by fishers in parts of southern Bass Strait groper (Epinephelus lanceolatus) (Fig. 3h). There are few histori- and the Furneaux Group. cal records of the tiger shark (Galeocerdo cuvier) from south of Abundances appear to have increased for all 16 species near Sydney but there have been several captures in the last decade off the southern limits of their ranges. This assemblage consists of north-east Tasmania. Tropical vagrants are rare in the listed nine widespread southern Australian, five Flindersian and two fauna (11%), so the recent arrival of these species is not incon- Peronian species. These fishes occupy a variety of niches exhib- sistent with a southward distributional shift in composition. iting preferences for either soft bottoms, seagrasses, reefs or While all of these species were of reproductive size, they are inshore demersal habitats. Coexisting adults and juveniles indi- unlikely to breed in southern seas. Another tropical offshore cate that at least 10 of these species are now likely to breed in the demersal species, the thorny tinselfish (Grammicolepis brachius- bioregion. culus), was also captured from the region in 2003.

populations and ranges in the south-east. Also, a small suite of DISCUSSION previously unrecorded vagrants from tropical habitats has been The effects of climate change on marine communities can be recorded, providing further evidence of climate-induced manifested as shifts in the latitudinal distributions of species. change. However, not all latitudinal boundaries are set by climate, and A large proportion of the species (c. 40%) showing distribu- localized hydrographic features need to be considered (Helmuth tional and/or abundance changes in Tasmanian waters have et al., 2006). Oceanographic models have demonstrated that widespread distributions encompassing all of the southern Aus- climate change is highly likely to cause the EAC flow to tralian biogeographical provinces (see IMCRA, 1996). These strengthen and extend southward, in turn resulting in further species are successful inhabitants of warm temperate Australian warming of the Tasman Sea. Increasing sea surface temperatures seas and are likely to become better represented off Tasmania as over the last 60 years off eastern Tasmania are primarily the sea temperatures increase. Eight species (17%) exhibiting distri- result of changes in the EAC rather than global surface fluxes butional and/or abundance changes in Tasmanian waters are of (Ridgway, 2007). The surface temperatures in the time-scale a Flindersian (western warm temperate) origin. All of these examined by Ridgway (2007) equate to an increase of about species displayed either abundance increases or a southward 2.3°C/100 years and a corresponding poleward advance by the range expansion, but only one had been recorded previously as EAC of about 350 km. Over the time-scale of this investigation rare in Tasmanian waters (Tables 3 & 4). Twelve species with (‘late 1800s’ to ‘present’) there have been strong increases in Peronian (eastern warm temperate) affinities appear to have water temperature, above the long-term average for the region, shifted their ranges in association with the recent southward in each decade since 1970 (Fig. 1). Increased water temperatures extension of the longspine sea urchin (Centrostephanus rodger- in the Tasman Sea are likely to have a cascading effect through sii). However, an additional five species with Peronian distribu- local marine ecosystems, probably most evident in, but not tions, but not associated with barrens created by these urchins, exclusive to, inshore marine ecosystems. For example, biological have recently expanded their ranges in Tasmanian waters. Inter- responses to these changes have been recorded off the east coast estingly, only two of the species with expanded ranges and/or of Tasmania for sea urchins (Ling et al., 2009) and shore crabs abundances in south-eastern Tasmania (the Bruny Bioregion) (Thresher et al., 2003). are Peronian species. Only one Bassian (Bass Strait) species, Long-term empirical data on the distributions of marine Trygonoptera imitata, has extended its range southward into biota collected using comparative methods are rarely available. Tasmanian waters. In their absence, we are forced to make best use of often dispar- Colonizing species might be expected to first appear closest ate and largely non-quantitative data sets – in our case mixtures to the source of recruitment from adjacent regions. Hence, the of literature, field, photographic and other observational Bass Strait Islands act as ‘stepping stones’ or distributional path- records, which together provide a useful basis for comparing ways south. In the ‘1980s’, three typical Peronian species distributional patterns. The strength of our argument in iden- (Chromis hypsilepis, Ophthalmolepis lineolatus and Eupetrichthys tifying change rests on the quality and reliability of accumulated angustipes) were confined to the Kent Group (central Bass information for each of the three periods, particularly the Strait), but all of these now have expanded ranges east and ‘1980s’, which we consider to be a tipping point for potential south. Other southern Bass Strait species have extended their changes. These data were compiled from a period of high scien- ranges southward along the east coast, and often into bays of the tific activity where consistent absences of non-cryptic species south-east. across the region are reliable indicators of extreme rarity or Breeding range data, definitive evidence that species are estab- non-occurrence. As discussed above, the argument for ‘real’ lished or resident in a region rather than extra-limital or tran- change is based on the regularity of this pattern, shared across sient, is not widely available for Australian fishes. However, supraspecific taxa. populations of some species (e.g. Girella elevata), until only Our synthesized data strongly support changes in ichthyofau- recently considered to be rare in the region, have been observed nal distribution over the span of three decades (‘1980s’ to with gravid individuals (P. Last, unpublished data). Other non- ‘present’). Multiple cases of latitudinal shifts in geographic migratory species (e.g. Urolophus paucimaculatus, Haletta semi- range, and corresponding increases in abundances of 52 Tasma- fasciata) presumably breed within their new extended ranges as nian coastal fish species, provide strong inferential evidence of a they now occur there as both adults and juveniles. These species changing environment (Root et al., 2003). Stuart-Smith et al. are not known to undertake long migrations so their adult (2009) concluded that there has been no major change in the populations, which are often widely separated, are unlikely to be structure of Tasmanian reef communities in the last decade, but connected reproductively; this is particularly evident for Bass did note that some species changed in abundance; one of these, Strait fishes that have disjunct populations in the south-east. Scorpis lineolatus, exhibited a five-fold increase in abundance. Not all the changes observed can be attributed to climate We report changes over longer time-scales than those of Stuart- change. Most changes in faunal structure before the ‘1980s’ are Smith et al. (2009), including species now present that were once likely to be due to anthropogenic rather than environmental considered rare or unrecorded from the region, others primarily factors. Some apical predators, including sharks (e.g. Carcharias confined to Bass Strait that have become resident or occur fre- taurus, Orectolobus halei) and large teleosts (e.g. Argyrosomus quently further south, as well as those that now have expanded hololepidotus, Nemadactylus valenciennesi, Achoerodus viridis),

68 Global Ecology and Biogeography, 20, 58–72, © 2010 Blackwell Publishing Ltd Long-term shifts in a temperate fish fauna appear to have experienced serious range reductions or regional Fishery impacts are also evident in offshore assemblages. extirpation since the 1880s. Johnston (1883) observed that Commercial fishes, the gemfish (Rexea solandri) and blue-eye estuaries of the south-east once contained an abundance of trevalla (Hyperoglyphe antarctica), which typically occur in deep commercial fishes and juveniles of other species. However, water along the continental slope, were once caught trolling at unprotected areas near population centres were soon ‘rendered the surface off south-east Tasmania (Lord & Scott, 1924). In the almost barren’ by intensive beach seining. At the same time, the case of gemfish, populations have been subjected to overfishing graball (a type of gillnet), was used to catch reef fishes in nearby off south-eastern Australia (Punt & Smith, 1999) and this bays and over shallow reefs, and this method is still used by both species is no longer abundant, even in its primary habitat. Thus, commercial and recreational fishers in Tasmania (Metcalf et al., a number of the changes to components of the fauna of this 2008). Edgar & Barrett (1999) have shown that large fishes are region are best attributed to non-climate stressors. particularly vulnerable to gillnetting practices; hence, this We have highlighted examples of both non-climate and method is likely to be destructive to fish communities over long climate-related impacts on a wide diversity of fishes in this time intervals. While inconclusive, it is likely that apparent peri- study, with 27 families (including six elasmobranch families) odic or regional extinctions of large sharks and teleosts during displaying recently altered distributions off Tasmania. The the 1900s were due primarily to the effects of fishing. majority of species showing distributional and/or abundance Some of the species putatively impacted are particularly vul- changes in Tasmanian waters are reef-dwellers (66%). Most of nerable to certain fishing practices (Pogonoski et al., 2002). By these families are represented by a single species, but there 1969, populations of eastern blue groper (Achoerodus viridis) are multiple species from the families Cheilodactylidae, had been seriously reduced off New South Wales and a total ban Kyphosidae, Labridae and Serranidae, of which all but one are on their capture was implemented in 1980 because large catches reef dwellers. Previous studies have shown that reef fishes are were being taken by commercial gillnets (Pogonoski et al., useful indicators of environmental change (e.g. Holbrook et al., 2002). This species, which was marketed in Tasmania in the 1997); however, changes are also evident within the pelagic 1800s (as adults exceeding 1 m) and described as common in the fauna. The frigate mackerel (Auxis thazard) was recorded in the 1920s, subsequently went missing until a few juveniles were 1800s but has been rarely seen in Tasmanian waters, until discovered off the northern coast in 2004. However, the likeli- schools were observed recently off north-eastern Tasmania. hood of A. viridis re-establishing breeding populations in Tas- Anecdotal information provided by recreational fishers indi- mania is low; males take about 10 years to mature (Gillanders, cates that abundances of other warm water tunas and billfishes 1995) and are unlikely to reach this stage due to fishing pressure have increased in recent years. unless protective measures are implemented. While there have been major range shifts in some dominant Elasmobranch fishes are particularly vulnerable to some coastal fishes, particularly those that occur primarily in the Bass fishing methods, particularly gillnetting (Walker et al., 2005). Of Strait, many other species have not undergone any obvious dis- the 10 species exhibiting some form of change in Tasmanian tributional shift. For example, no distributional changes have waters, one is a new record, one is a seasonal vagrant, five have been observed for the common Bass Strait reef species, the expanded their ranges or increased their abundance and three horseshoe leatherjacket (Meuschenia hippocrepis), the yellow- appear to have suffered significant population declines. Elasmo- striped leatherjacket (Meuschenia flavolineata), the pencil weed branchs have K-selected life-history strategies (i.e. they have low whiting (Siphonognathus beddomei) and the scalyfin (Parma productivity, low fecundity, mature at a late age and are slow victoriae), and a soft-bottom species, the southern bluespotted growing), making them especially vulnerable to threatening flathead (Platycephalus speculator). The absence of poleward processes such as intense fishing pressure (Stevens et al., 2000). movement in these taxa, despite obvious shifts in other The IUCN Species Survival Commission recognized this issue co-occurring species, suggests that it is difficult to predict which and established the Shark Specialist Group (SSG) to provide species will exhibit induced range shifts. In situ changes in abun- assessments for all sharks and rays as part of their Red List of dance, physiology or phenology (Hobday et al., 2007) of these Threatened Species® program (IUCN, 2008). Two sharks appear species may still mean that climate change is having an impact, to have undergone significant population declines in Tasmanian but without more detailed study impacts remain unknown. The waters. Carcharias taurus is listed as vulnerable by the IUCN Red distributions of these important, non-cryptic species need to be List due to its extremely low reproductive rates and population monitored in the coming years to assess their future responses to declines recorded off eastern Australia and South Africa (Pollard environmental change. & Smith, 2000). Catches of wobbegong sharks (including While we have focused on changes in fish distributions, Orectolobus halei) off south-eastern Australia, particularly habitat shifts in response to environmental change can also have New South Wales, have declined markedly in recent decades a considerable cascading effect in a biological community; an (Huveneers et al., 2007). These declines resulted in populations example of this off the Tasmanian coastline is the longspine sea of two wobbegong species in New South Wales initially being urchin. This shallow-water urchin, which is found also off listed as Vulnerable by the IUCN Red List (Cavanagh et al., northern New Zealand and the Kermadec Islands (Andrew & 2003). Thus, it is likely that threatening processes acting on these Byrne, 2007), has a primarily Peronian-derived distribution in species have significantly reduced, or possibly eliminated, the Australia. Before 1978, it was only known in Tasmanian waters populations of these sharks in Tasmanian waters. from the central northern Bass Strait but is now established

Global Ecology and Biogeography, 20, 58–72, © 2010 Blackwell Publishing Ltd 69 P. R. Last et al. along most of the east coast of Tasmania, a change attributed to least five, possibly up to 19, species have undergone serious a strengthening of the EAC and associated ocean warming (Ling declines and are possibly extinct locally. Localized extinctions, et al., 2009). Newly formed urchin barrens are likely to result particularly of apical predators such as the greynurse shark and from both climate change and reduced predation pressure asso- mulloway, are likely to be attributed to fishing pressure. Other ciated with commercial harvesting of the urchin’s key predator, species, such as the gulf wobbegong, now have extremely the southern rock lobster (Jasus edwardsii). Our study indicated restricted distributions and are likely to face a similar fate that 7 of the 12 species that were rare, previously unlisted or without intervention. Thus, the fauna of this region was not recovering in Tasmanian waters are strongly associated with ‘pristine’ before this new invasion associated with warmer tem- Centrostephanus rodgersii barrens (Table 3). Of these seven peratures began. species, six have a Peronian origin, providing strong circumstan- This paper highlights, however, that many warm temperate tial evidence that these species have shifted their distributions species have colonized the cool temperate Tasmanian region or southward in relation to latitudinal shifts in the distribution of substantially expanded their ranges, consistent with a strength- C. rodgersii or its associated barrens. Latitudinal shifts in the ening of the EAC and associated rises in sea temperature. distribution of biological facies and their associated biota have Although climate change is considered beneficial to a number of not been well researched, but some coralline algal communities species, it will also be detrimental to others (Poloczanska et al., also appear to have expanded their ranges southward in recent 2008). The effect of environmental change on species endemic years (P. Last, unpublished data). to cool temperate Australia has not been fully investigated, but Other non-climate-related change within the fauna is exem- the lack of refugia south of Tasmania (i.e. the southernmost plified by a small group of introduced species. Two New Zealand limits of the Australian continental shelf), especially for coastal coastal reef fishes, the triplefins Forsterygion gymnota and For- species with population centres in southern Tasmania, is of sterygion varium, were introduced into the Derwent Estuary and concern and should be a focal point of future research. Reliable D’Entrecasteaux Channel, south-eastern Tasmania (Clements baselines are needed to evaluate further change, and to that end, et al., 2000). These species were initially thought to be recent bioregionalization studies using national fish datasets Tasmanian endemics (Last et al., 1983). At least four New (e.g. IMCRA, 1996) can be used for this purpose. Similarly, Zealand marine invertebrate species have been introduced into long-term scientific and community-based monitoring will be Tasmanian waters, probably with shipments of live oysters important in documenting changes in the composition of around the 1920s (Edgar, 1997), and the triplefins may have marine ecosystems. The paucity of documented impacts of been transported in the same way. climate change in the marine realm of the Southern Hemisphere The importance of habitat availability should not be under- is due to the resource-intensive and expensive nature of marine estimated when attempting to predict the responses of species to monitoring (Richardson & Poloczanska, 2008). Volunteer environmental change. The flathead Platycephalus laevigatus,a recording schemes, or ‘citizen science’ projects such as Flindersian species, has expanded its range in the bays of south- REDMAP, allow for detailed monitoring to continue and east Tasmania and is now more widespread and abundant in expand with comparatively little financial investment, but offer seagrass habitats of this region. However, despite these changes, scope for the generation of much-needed baseline data to it has never been recorded off the central east coast, possibly as enhance monitoring efforts of science organizations. a consequence of the unavailability of appropriate habitat. Thus, future changes in faunal distribution may occur, not only in ACKNOWLEDGEMENTS response to direct changes in the physical environment (water temperature) but to indirect changes in preferred habitat Numerous colleagues, divers, fishers and the general public have (Hobday et al., 2007). provided unpublished data and anecdotal records over a number of decades. We acknowledge the wealth of information that has been forwarded so enthusiastically from such sources. CONCLUSIONS In particular, we acknowledge assistance from professional fish- This paper set out to investigate two hypothetical changes in the erman Bill Smedley, and divers, including: Karen Gowlett- ichthyofauna off Tasmania in the south-eastern Australia Holmes, Mike Nichols, Andrew Pender, Mike Sugden, Malcolm climate change hotspot. We hypothesized that the coastal fish Wells and numerous divers from the Australian Underwater assemblages have exhibited a marked change in composition Federation. Neville Barrett (Tasmanian Aquaculture and since the ‘1980s’. The results of this study showed that about a Fisheries Institute, TAFI), Barry Hutchins (formerly of the fifth (61 species) of the Tasmanian coastal fish fauna have, or Western Australian Museum) and Rick Stuart-Smith (TAFI) appear to have, undergone important compositional shifts since have kindly provided unpublished data and advice. John the 1800s. Examination of longer-term fish composition data Pogonoski (CMAR) assisted with editing of the manuscript. revealed substantial changes to the coastal ichthyofauna of Tas- Alastair Graham and Louise Conboy (CMAR) have provided mania, strongly supporting the first hypothesis. In addition, to assistance in the collation of data and collection support. illustrate that not all change might be climate-related, we Thanks also to Nick Mooney and Ian Banks for use of under- hypothesized that certain elements of the Tasmanian ichthyo- water photographs. Finally, the authors wish to collectively fauna have been lost or severely depleted since the ‘late 1800s’.At thank the three anonymous referees, as well as GEB editors

David Currie and Julian Olden, for providing constructive sug- Australian marine life. CSIRO Marine and Atmospheric gestions and comments on earlier drafts of this manuscript. Research. Report to the Australian Greenhouse Office, Canberra, Australia. September 2006. Holbrook, S.J., Schmitt, R.J. & Stephens, J.A. Jr (1997) Changes in an assemblage of temperate reef fishes associated with a REFERENCES climate shift. Ecological Applications, 7, 1299–1310. Andrew, N.L. & Byrne, M. (2007) Ecology of Centrostephanus. Hutchins, B. & Swainston, R. (1986) Sea fishes of southern Edible Sea urchins: biology and ecology (ed. by J.M. Lawrence), Australia. Swainston Publishing, Perth. pp. 191–204. Developments in aquaculture and fisheries Huveneers, C., Walker, T.I., Otway, N.M. & Harcourt, R.G. science, Vol. 37. Elsevier, Amsterdam. (2007) Reproductive synchrony of three sympatric species of Cavanagh, R., Kyne, P., Fowler, S.L., Musick, J.A. & Bennett, M.B. wobbegong shark (genus Orectolobus) in New South Wales, (2003) The conservation status of Australasian chondrichthy- Australia: reproductive parameter estimates necessary for ans. Report of the IUCN Shark Specialist Group Australia and population modelling. Marine and Freshwater Research, 58, Oceania Regional Red List Workshop. Queensland, Australia, 765–777. 7–9 March 2003. The University of Queensland, Brisbane. IMCRA (1996) Interim marine bioregionalisation for Australia: Clements, K.D., Jawad, L.A. & Stewart, A.L. (2000) The New towards a national system of marine protected areas. CSIRO, Zealand triplefin Grahamina signata (Teleostei; Tripterygi- Hobart. idae): a junior synonym of G. gymnota from Tasmania. IUCN (2008) 2008 IUCN Red List of threatened species. Available Journal of the Royal Society of New Zealand, 30, 373–384. at: http://www.redlist.org (accessed June 2009). Coll, J., Linde, M., García-Rubies, A., Riera, F. & Grau, A.M. Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., (2004) Spear fishing in the Balearic Islands (west central Botsford, L.W., Bourque, B.J., Bradbury, R.H., Cooke, R., Mediterranean): species affected and catch evolution during Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., Lange, C.B., the period 1975–2001. Fisheries Research, 70, 97–111. Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S., Currie, D.R. & Parry, G.D. (1999) Changes to benthic commu- Tegner, M.J. & Warner, R.R. (2001) Historical overfishing and nities over 20 years in Port Phillip Bay, Victoria, Australia. the recent collapse of coastal ecosystems. Science, 293, 629– Marine Pollution Bulletin, 38, 36–43. 638. Edgar, G.J. (1984) General features of the ecology and biogeog- Johnston, R.M. (1883) General and critical observations on the raphy of Tasmanian subtidal rocky shore communities. Papers fishes of Tasmania. Papers and Proceedings of the Royal Society and Proceedings of the Royal Society of Tasmania, 118, 173–186. of Tasmania, 1882, 51–143. Edgar, G.J. (1997) Australian marine life: the plants and animals Johnston, R.M. (1890) Further observations upon the fishes and of temperate waters. Reed Books, Victoria. fishing industries of Tasmania, together with a revised list of Edgar, G.J. & Barrett, N.S. (1999) Effects of the declaration of indigenous species. Papers and Proceedings of the Royal Society marine reserves on Tasmanian reef fishes, invertebrates and of Tasmania, 1889, 1–25. plants. Journal of Experimental Biology and Ecology, 242, 107– Last, P.R. (1979) First records of the one spot puller (Chromis 144. hypsilepis) and the spotted stingaree (Urolophus gigas)from Edgar, G.J., Last, P.R. & Wells, M.W. (1982) Coastal fishes of Tasmanian waters with an annotated list of fishes recorded Tasmania and Bass Strait. Cat and Fiddle Press, Hobart. from Kent Islands, Bass Strait. Tasmanian Naturalist, 59, Edyvane, K.S. (2003) Conservation, monitoring and recovery of 5–12. threatened giant kelp (Macrocystis pyrifera) beds in Tasmania. Last, P.R. & Gledhill, D.C. (2009) A revision of the Australian Final report for Environment Australia. Department of handfishes (Lophiiformes: Brachionichthyidae), with descrip- Primary Industries, Water and Environment, Hobart. tions of three new genera and nine new species. Zootaxa, Gillanders, B.M. (1995) Reproductive biology of the protogy- 2252, 1–77. nous hermaphrodite Achoerodus viridis (Labridae) from Last, P.R. & Harris, J.G.K. (1981) New locality records and pre- south-eastern Australia. Marine and Freshwater Research, 46, liminary information on demersal fish faunal assemblages in 999–1008. Tasmanian waters. Papers and Proceedings of the Royal Society Gomon, M., Bray, D. & Kuiter, R. (eds) (2008) Fishes of of Tasmania, 115, 189–209. Australia’s southern coast. Museum Victoria, Melbourne. Last, P.R.,Scott, E.O.G. & Talbot, F.H. (1983) Fishes of Tasmania. Helmuth, B., Mieszkowska, N., Moore, P. & Hawkins, S.J. (2006) Tasmanian Fisheries Development Authority, Hobart. Living on the edge of two worlds: forecasting the responses of Last, P.R.,Gledhill, D.C. & Holmes, B.H. (2007) A new handfish, rocky intertidal ecosystems to climate change. Annual Review Brachionichthys australis sp. nov. (Lophiiformes: Brachionich- of Ecology, Evolution and Systematics, 37, 373–404. thyidae), with a redescription of the critically endangered Hiddink, J.G. & Hofstede, R. (2008) Climate induced increases spotted handfish, B. hirsutus (Lacepède). Zootaxa, 1666, in species richness of marine fishes. Global Change Biology, 14, 53–68. 453–460. Lincoln Smith, M.P., Bell, J.D., Pollard, D.A. & Russell, B.C. Hobday, A.J., Okey, T.A., Poloczanska, E.S., Kunz, T.J. & (1989) Catch and effort of competition spearfisherman in Richardson, A.J. (2007) Impacts of climate change on southeastern Australia. Fisheries Research, 8, 45–61.

Ling, S.D., Johnson, C.R., Ridgway, K., Hobday, A.J. & Haddon, thyans), and the implications for marine ecosystems. ICES M. (2009) Climate driven range extension of a sea urchin: Journal of Marine Science, 57, 476–494. inferring future trends by analysis of recent population Stuart-Smith, R.D., Barrett, N.S., Stevenson, D.G. & Edgar, G.J. dynamics. Global Change Biology, 15, 719–731. (2009) Stability in temperate reef communities over a decadal Lord, C. (1923) A list of the fishes of Tasmania. Papers and time scale despite concurrent ocean warming. Global Change Proceedings of the Royal Society of Tasmania, 1922, 60–73. Biology, 16, 122–134. Lord, C. & Scott, H.H. (1924) A synopsis of the vertebrate animals Thresher, R.E., Proctor, C., Ruiz, G., Gurney, R., MacKinnon, C., of Tasmania. Oldham, Beddome and Meredith, Hobart. Walton, W., Rodriguez, L. & Bax, N. (2003) Invasion dynamics Luckman, J.S. & Davies, K. (1978) They called it Transylvania. of the European shore crab, Carcinus maenas, in Australia. The south west book: a Tasmanian wilderness (ed. by H. Gee, Marine Biology, 142, 867–876. J. Fenton and G. Hodge), pp. 7–16. Australian Conservation Walker, T.I., Hudson, R.J. & Gason, A.S. (2005) Catch evaluation Foundation, Hawthorn, VIC. of target, by-product and by-catch species taken by gillnets Metcalf, S.J., Dambacher, J.M., Hobday, A.J. & Lyle, J.M. (2008) and longlines in the shark fishery of south-eastern Australia. Importance of trophic information, simplification and aggre- Journal of Northwest Atlantic Fisheries Science, 35, 505–530. gation error in ecosystem models. Marine Ecology Progress Yearsley, G.K., Last, P.R. & Ward, R.D. (1999) Australian seafood Series, 360, 25–36. handbook: an identification guide to domestic species. CSIRO Perry, A.L., Low, P.J., Ellis, J.R. & Reynolds, J.D. (2005) Climate Division of Marine Research, Hobart. change and distribution shifts in marine fishes. Science, 308, Yearsley, G.K., Last, P.R. & Hoese, D.F. (2006) Standard names of 1912–1915. Australian fishes. CSIRO Marine and Atmospheric Research Pogonoski, J.J., Pollard, D.A. & Paxton, J.R. (2002) Conservation Paper 009. CSIRO Marine and Atmospheric Research, Hobart. overview and action plan for Australian threatened and potentially threatened marine and estuarine fishes. Environment Australia, Canberra. SUPPORTING INFORMATION Pollard, D. & Smith, A. (2000) Carcharias taurus. 2009 IUCN Red List of threatened species. Version 2009.1. Available at: Additional Supporting Information may be found in the online http://www.iucnredlist.org (accessed 2 July 2009). version of this article: Poloczanska, E.S., Hawkins, S.J., Southward, A.J. & Burrows, Appendix S1 Abbreviations and keys to supporting M.T. (2008) Modeling the response of populations of com- information. peting species to climate change. Ecology, 89, 3138–3149. Appendix S2 Summary data for candidate species detailing Punt, A.E. & Smith, A.D.M. (1999) Harvest strategy evaluation changes in abundance and distribution. for the eastern stock of gemfish (Rexea solandri). ICES Journal Appendix S3 Summary of historical records and biogeographic of Marine Science, 56, 860–875. affinities of candidate species. Rayner, N.A., Parker, D.E., Horton, E.B., Folland, C.K., Alex- Appendix S4 Coastal species exhibiting distributional change in ander, L.V., Rowell, D.P., Kent, E.C. & Kaplan, A. (2003) Tasmanian waters since the ‘late 1800s’ and their current occur- Global analyses of sea surface temperature, sea ice, and night rence status and change category. marine air temperature since the late nineteenth century. Journal of Geophysical Research, 108, D14, doi: 4407 10.1029/ As a service to our authors and readers, this journal provides 2002JD002670. supporting information supplied by the authors. Such materials Richardson, A.J. & Poloczanska, E.S. (2008) Ocean science: are peer-reviewed and may be reorganized for online delivery, under-resourced, under threat. Science, 320, 1294–1295. but are not copy-edited or typeset. Technical support issues Ridgway, K.R. (2007) Long-term trend and decadal variability of arising from supporting information (other than missing files) the southward penetration of the East Australian Current. should be addressed to the authors. Geophysical Research Letters, 34, L13613. Root, T.L., Price, J.T., Hall, K.R., Schneider, S.H., Rosenzweigk, BIOSKETCH C. & Pounds, J. (2003) Fingerprints of global warming on wild animals and plants. Nature, 421, 57–60. Peter Last is leader of the Biogeography and Sagarin, R.D., Barry, J.P., Gilman, S.E. & Baxter, C.H. (1999) Taxonomy team of CSIRO Marine and Atmospheric Climate related change in an intertidal community over short Research in Hobart, Australia and is curator of the and long time scales. Ecological Monographs, 69, 465–490. Australian National Fish Collection (ANFC), with Schiel, D.R., Steinbeck, J.R. & Foster, M.S. (2004) Ten years of interests in biogeography, systematics and phylogeny of induced ocean warming causes comprehensive changes in Indo-West Pacific marine fishes as well as bioregional marine benthic communities. Ecology, 85, 1833–1839. marine planning for the Australian region. Stevens, J.D., Bonfil, R., Dulvy, N.K. & Walker, P.A. (2000) The effects of fishing on sharks, rays, and chimaeras (chondrich- Editor: Julian Olden