journal of the Royal Society of Western Australia, 82:131-136,1999 Escape of cultured barramundi (Lates calcarifer Bloch) into impoundments of the Ord River system. Western Australia R G Doupe1'3 & A J Lymbery2'3 1 Centre for Ecosystem Management, Edith Cowan University, Joondalup, WA 6027 2 Agriculture Western Australia and 3Murdoch University, Murdoch, WA 6150 email: [email protected] Manuscript received February 1999; Accepted July 1999 Abstract Mitochondrial DNA sequences were used to compare barramundi found in impounded parts of the Ord River system to their cultured counterparts from a fish farm in one dam (Lake Argyle). Two haplotypes were common to all fish sampled, indicating the high probability that farmed fish are escaping, and could potentially interact with the wild population of the lower Ord River. Although the Lake Argyle barramundi farm currently produces only about 50 tonnes of fish per year, there are plans to expand barramundi production at Lake Argyle by up to 200 times this amount. If that was to occur and no efforts were made to reduce the escape of cultured fish, then sufficient numbers of escaped barramundi may survive to reproduce in the wild fishery. Maintaining the genetic integrity of wild barramundi populations will be best achieved by setting and enforcing standards to minimise escapes from fish farms. Introduction and these mostly concern salmonids in the northern hemisphere (Fleming & Gross 1993; Fleming et al. 1996; Fish are among the world's most intensively cultured Einum & Fleming 1997). organisms. Many species are being propagated for both restocking programs and as founders of novel cultured Barramundi (Lates calcarifer Bloch) is a highly fecund populations elsewhere, and large numbers of farmed fish and euryhaline fish with a tropical Indo-West Pacific are also escaping unintentionally from aquaculture distribution including northern Australia. There are facilities (Bergen et ai 1991; Lund et al. 1991; Gausen & around 30 commercial barramundi farms in Australia, Moen 1991; Webb et al. 1993; Windsor & Hutchinson producing approximately 500 tonnes in 1996 (Brown et al. 1995). A significant frequency and scale of escape of 1997). Both the progeny for grow-out operations and the farmed fish has been described by Webb et ai (1991) as broodstock themselves are typically descendants of those "inevitable", with numbers of escaped salmonids broodstock used to begin the industry in government approaching or exceeding the numbers of naturally facilities in Queensland and the Northern Territory some produced fish in certain places (Saunders 1991). Thus, the years ago. Therefore, comparatively few broodstock may global expansion of fish farming has implications for both have founded much of the present barramundi culture cultured and natural fish populations (Fleming & Gross industry in Australia. For example, in its few years of 1993; Hayes et al. 1996). operation, the aquaculture facility at Lake Argyle in Western Australia (Fig 1) has reared the progeny of Captive rearing conditions combined with artificial broodstock supplied by the government hatchery in the selection, intentional or not, cause fish to diverge from Northern Territory. No more than four broodstock their wild phenotype through environmental and genetic captured from the coast near Darwin have been used in processes (Cross & Challanain 1991; Swain et al. 1991). group larval production for the Lake Argyle farm, with Further, the small numbers of broodstock typically used one highly fecund female producing about 75% of all in fish culture encourage genetic divergence through progeny so far supplied (G Schipp, Hatchery Manager, inbreeding (Tave 1993). There is evidence that fish Aquaculture Branch, Department of Primary Industry derived from cultured populations may have both and Fisheries, Northern Territory, pers comm). ecological (e.g. competition, disease introductions) and genetic (e.g. loss of genetic adaptation, genetic In recent years, there has been an increasing incidence homogenization) impacts on wild populations (e.g. Rinne of barramundi caught as a bycatch of the Lake Argyle & Minckley 1985; Loudenslager et al. 1986; Ferguson 1990; catfish (Arius spp) fishery, and by recreational fishers in Bartley & Gall 1991; Hindar et al. 1991; Waples 1991; Spillway Creek, to which waters from Lake Argyle flow Wilde & Echelle 1992; Heggberget et al. 1993; Fleming et al. when the lake exceeds the bank-full stage. Spillway Creek 1996; Einum & Fleming 1997). There remain, however, joins a natural watercourse (Stonewall Creek) some few data on exactly how cultured and wild fish interact. distance downstream, and then the Lake Kununurra irrigation diversion dam (i.e. the Ord River; Fig 1). At © Royal Society of Western Australia 1999 least two explanations can account for the presence of 131 Journal of the Royal Society of Western Australia, 82(4), December 1999 132 R G Doupe & A J Lymbery: Escape of cultured barramundi barramundi in these impoundments. Either these fish Australia (Shaklee et al. 1993; Keenan 1994; Chenoweth were captured from the natural barramundi population et al. 1998), and Doupe et ah (1999) showed substantial below the diversion dam wall and have been released genetic differentiation between barramundi from the into Lake Kununurra, where they have moved upstream Kimberley (Ord and Fitzroy Rivers), Darwin and Cairns as part of their ecological migration, or they are cultured (Queensland). Since the depiction of Kimberley fish that have escaped from farm netpens into Lake barramundi as a genetically differentiated stock, the Argyle and then into Spillway Creek. From these proprietors at Lake Argyle have been encouraged to perspectives, this study aimed to compare the genetic collect broodstock from the Ord River region, but the identity of barramundi found in the impounded sections program is at a pilot stage and the Darwin hatchery of the Ord River to cultured barramundi from the Lake remains the source of broodstock for the Lake Argyle fish Argyle fish farm. farm. Are barramundi escaping? Materials and Methods The identical mtDNA haplotypes found among barramundi of the fish farm. Lake Argyle and Spillway Barramundi were sampled from the Lake Argyle Creek provide strong evidence that barramundi have barramundi farm (7 fish), from the open waters of Lake escaped from the farm at Lake Argyle. An alternative Argyle (8 fish) and from Spillway Creek (4 fish). For each explanation is that they have migrated to Spillway Creek fish, tissue samples comprising a caudal fin clip and/or and Lake Argyle from the Ord River. Haplotypes A and muscle tissue (Doupe & Chandler 1998) were placed in B found in this study were not detected in the Ord River specimen vials containing 80% ethanol for 2 hours. Once population sampled by Doupe et al. (1999). It is possible the alcohol had diffused through the tissue, it was that haplotypes A and B are present in the natural Ord replaced with 70% ethanol for sample storage. Samples River population, but were missed by the sample of were then transported to Perth and processed at the Doupe et ah (1999). Nevertheless, it seems a reasonable laboratories of the Centre for Human Genetics at Edith assumption that the haplotypes would occur no more Cowan University, using the methods of Doupe & frequently than ]/u or 0.07. If this is so, then the Chandler (1998) and Doupe et ah (1999). probability that the 8 fish sampled from Lake Argyle Briefly, DNA was extracted from the tissue samples originated from the Ord River is < 5.76 1010 (0.078), and and primers described by Chenoweth et ah (1998) were the probability that the 4 fish sampled from Spillway used to amplify a 290 bp fragment within Region 1 of the Creek originated from the Ord River is < 2.40 10'5 (0.074) barramundi mitochondrial control region by the Clearly, these are highly unlikely events. The most polymerase chain reaction. The amplified product was parsimonious explanation, then, is that the fish in Lake sequenced using thermal cycle sequencing, interpreted Argyle and Spillway Creek escaped from the fish farm. with the aid of the Sequence Navigator 1.0.1 software package. Sequences from different fish were aligned by Hybridization between escaped and wild barramundi eye using MacClade 3.03 (Maddison & Maddison 1992). The presence of escaped farmed fish in Spillway Creek Barramundi mtDNA control region sequences were suggests that they can potentially migrate from Lake compared among those localities surveyed in this study Argyle to Lake Kununurra. A trial stocking of 124 tagged and with the wild population of the lower Ord River barramundi in Lake Kununurra recaptured fish from sampled by Doupe et al. (1999). below the diversion dam (n = 9), and as far downstream as the Ord River estuary (C Bird, Fisheries Western Australia, pers comm). This indicates that barramundi Results can move from Lake Kununurra into the lower Ord River, and suggests that escaped cultured fish have the This study identified a total of 19 barramundi mtDNA potential to genetically contribute to the natural Ord control region sequences, with each being 231 nucleotide River population. The risk of hybridization then depends bases in length. Two distinct mtDNA haplotypes were upon the rate of escape of cultured fish, how many found (haplotypes A & B) and are shown in Table 1. Both survive to reproductive maturity, and then how many haplotypes were detected at each locality; 5 A and 2 B achieve reproductive success. The long-term effects of haplotypes at the fish farm, 2 A and 6 B haplotypes in hybridization depend upon the effective population size Lake Argyle, and 3 A and 1 B haplotypes in Spillway (N) of the existing Ord River population, compared to Creek. Neither the A nor the B haplotype found in this the size of the cultured population that survives and study was detected in the natural Ord River population contributes genetic material to it. N is extremely difficult sampled by Doupe et al (1999). to measure, but is usually very much smaller than the actual population size (Waples 1990). About 300-500 barramundi have been caught in Lake Argyle as a Discussion bycatch of the professional catfishery (C Ostle, Fisheries Western Australia, pers comm), and about 12-15 The Western Australian Department of Environmental individual fish have been seen in Spillway Creek (Doupe, Protection and Fisheries Western Australia are signatories unpublished observations). How many have escaped to a Memorandum of Understanding concerning the beyond there into Lake Kununurra is not known. translocation of aquatic species into Western Australia and between watersheds within Western Australia. The effect of hybridization on local adaptation Barramundi display significant population genetic There is concern about genetic interaction between structure eastwards from the Ord River in Western wild stocks and fish bred for aquaculture (Ryman 1991), 133 Barramundi mtDNA control region sequences showing haplotypes A and B. A matching nucleotide base is indicated by AL - Lake Argyle; SC - Spillway Creek; FF - Fish Farm. J o u r n a l o f th e R o y a l S o c ie ty o f W 1 e 3 s 4 ter n A u s tr a lia , 8 2 ( 4 ) , D e c e m b e r 1 9 9 9 CT... R G Doupe & A ] Lymbery: Escape of cultured barramundi including barramundi (Keenan &c Salini 1990; Shaklee et al. Acknowledgements: This work was funded through the Fish and Fish 1993; Doupe 1997). There is a perceived risk that Habitat Protection Program of Fisheries Western Australia. C Ostle and C Chalmers were largely responsible for the initiation of the project. S translocation of hatchery-reared fish could result in the Grandison, D Macleod, A McEwen, S McIntosh, C Ostle, J Roe, T Sinclair, establishment of less fit haplotypes, thus reducing stock N Stewart, G Wellington and J Williams assisted in the collection of fitness through the disruption of coadapted genomes samples. G Schipp provided information for Darwin broodstock origin (Keenan 1994). There is no clear relationship between the and usage, and D Chandler was of great assistance in the laboratory. M Johnson, R Lenanton, G Maguire and H Recher kindly read a draft of this extent of genetic differences among populations and the work. degree of local adaptation; the issue is largely one of whether genetic differentiation has occurred through genetic drift or through selection (Johnson 2000). References Keenan (1994) thought that population genetic differentiation in barramundi is due to genetic drift as a Bartley D M & Gall G A E 1991 Genetic identification of native cutthroat trout (Oncorhynchus clarki) and introgressive result of small effective population sizes. There is, hybridization with introduced rainbow trout (O. mykiss) in however, wide variability in life history traits among streams associated with the Alvord Basin, Oregon and barramundi populations. For example, wild barramundi Nevada. Copeia 1991:854-859. display geographic variation among localities for sex Bergen P I, Gausen D & Hansen L P 1991 Attempts to reduce the change (Moore 1979; Davis 1982), the presence of primary impact of reared Atlantic salmon on wild stocks in Norway. females (Moore 1979; Maneewong 1987), variable length/ Aquaculture 98:319-324. sex ratio relationships (Patnaik & Jena 1976; Moore 1979; Brown D, van Landeghem K & Schuele M 1997 Australian Davis 1982) and the possibility of fully marine life aquaculture: Industry profiles for selected species. ABARE histories (Pender & Griffin 1996). Such flexible life report to the Fisheries Resource Research Fund. ABARE histories may be due to non-adaptive genetic differences Research Report 97.3, Australian Bureau of Agricultural and Resource Economics, Canberra. or phenotypic plasticity, or they may be real adaptive differences. Much of the ecological diversity found Chenoweth S F, Hughes J M, Keenan C P & Lavery S 1998 Concordance between dispersal and mitochondrial gene among geographically disparate barramundi populations flow: isolation by distance in a tropical teleost, bates calcarifer. coincides with the population genetic differentiation Heredity 80:187-197. found in the summary studies of Shaklee et al. (1993) and Cross T F & Challanain D N 1991 Genetic characterisation of Keenan (1994). As yet, there is little ecological data for Atlantic salmon (Salmo salar) lines farmed in Ireland. Kimberley barramundi (Doupe et al. 1999). Aquaculture 98:209-216. Davis T L O 1982 Maturity and sexuality in barramundi, Lates Stock management calcarifer (Bloch), in the Northern Territory and South-eastern Gulf of Carpentaria. Australian Journal of Marine and Although demographic and ecological information is Freshwater Research 33:529-545. required if the influence of local selection pressures and Doupe R 1997 Conserving barramundi genetic resources: A role population genetic differentiation are to be clarified, it for aquaculture. Austasia Aquaculture 11:46-47. has been argued that sustainable barramundi aquaculture Doupe R G & Chandler D 1998 Fin clipping the barramundi: in the Kimberley region of Western Australia should look Showing the old dog some new tricks in gathering population toward the propogation of local stocks (Doupe 1997). genetic data. In: Proceedings of the Tropical Aquaculture However, the use of local stocks for culture will not Seminar held at Lake Argyle, Western Australia. The Tropical necessarily safeguard the genetic integrity of the wild Aquaculture Association and Fisheries Western Australia, population. Aquaculturalists restrict the captive gene Perth, 10-20. pool, either unintentionally through limited broodstock Doupe R G, Horwitz, P & Lymbery, A J 1999 Mitochondrial usage or by selecting for favoured production traits, like genealogy of Western Australian barramundi: Applications growth rate and disease resistance. The result of this of inbreeding coefficients and coalescent analysis for separating temporal population processes. Journal of Fish artificial selection will be a captive population that is Biology 54:1197-1201. genetically divergent to its wild counterparts, irrespective Einum S & Fleming I A 1997 Genetic divergence and interactions of its origin. in the wild among native, farmed and hybrid Atlantic The simplest and most pragmatic management salmon. Journal of Fish Biology 50:634-651. approach would be to ensure that cultured fish do not Ferguson M M 1990 The genetic impact of introduced fishes on escape. This could be achieved by requiring farms to native species. Canadian Journal of Zoology 68:1053-1057. provide better security, apart from predator nets, and by Fleming I A & Gross M R 1993 Breeding success of hatchery and installing a mesh or netting mechanism at the Spillway wild coho salmon (Oncorhynchus kisutch) in competition. Creek outlet that inhibited barramundi movement. A Ecological Applications 3:230-245. barramundi barrier at Spillway Creek would ideally be Fleming I A, Jonsson B, Gross M R & Lamberg A 1996 An sized so as not restrict the movement of other aquatic experimental study of the reproductive behaviour of farmed and wild Atlantic salmon (Salmo salar). Journal of Applied species, such as black bream (Hephaestus jenkinsi), which Ecology 33:893-905. are a significant component of the recreational fishery in Gausen D & Moen V 1991 Large-scale escapes of farmed Atlantic Lake Kununurra, and the catfishes that support a small salmon (Salmo salar) into Norwegian rivers threaten natural commercial fishery in its upper reaches. Further populations. Canadian Journal of Fisheries and Aquatic consideration should also be given to the expansion of Sciences 48:426-428. commercial catfishing licences in Lake Argyle to allow Hayes J P, Guffey S Z, Kriegler F J, McCracken G F & Parker C R the sale of barramundi caught as a bycatch; presently 1996 The genetic diversity of native, stocked, and hybrid catfishermen are required to return all non-target species populations of brook trout in the Southern Appalachians. to the water. Conservation Biology 10:1403-1412. 135 Journal of the Royal Society of Western Australia, 82(4), December 1999 Heggberget T G, Johnsen B O, Hindar K, Jonsson B, Hansen L P, Rinne J N & Minckley W L 1985 Patterns of variation and Hvidsten N A & Jensen A J 1993 Interactions between wild distribution of Apache trout (Salmo apache) relative to co¬ and cultured Atlantic salmon: A review of the Norwegian occurrence with introduced salmonids. Copeia 1985:285-292. experience. Fisheries Research 18:123-146. Ryman N 1991 Conservation genetic considerations in fishery Hindar K, Ryman N & Utter F 1991 Genetic effects of cultured management. Journal of Fish Biology 39 (Supp A):211-224. fish on natural fish populations. Canadian Journal of Fisheries and Aquatic Sciences 48:945-957. Saunders R L 1991 Potential interaction between cultured and wild Atlantic salmon. Aquaculture 98:51-60. Johnson, M S 2000 Measuring and interpreting genetic structure to minimize the genetic risks of translocations. Aquaculture Shaklee J B, Salini J & Garrett R N 1993 Electrophoretic Research 31:133-143. characterization of multiple genetic stocks of barramundi Keenan C P 1994 Recent evolution of population structure in (hates calcarifer) in Queensland, Australia. Transactions of the Australian barramundi, hates calcarifer (Bloch): An example of American Fisheries Society 122:685-701. isolation by distance in one dimension. Australian Journal of Swain D P, Riddell B E & Murray C B 1991 Morphological Marine and Freshwater Research 45:1123-1148. differences between hatchery and wild populations of coho Keenan C & Salini J 1990 The genetic implications of mixing salmon (Oncorhynchus kisutch): Environmental versus genetic barramundi stocks in Australia. In: Introduced and origin. Canadian Journal of Fisheries and Aquatic Sciences Translocated Fishes and their Ecological Effects (ed D A 48:1783-1791. Pollard). Proceedings of the Australian Society for Fish Tave D 1993 Genetics for Fish Hatchery Managers. AVI, New Biology, 8. Department of Primary Industries and Energy and York. Bureau of Rural Resources, Canberra, 145-150. Loudenslager E J, Rinne J N, Gall G A E & David R E 1986 Waples, R S 1990 Conservation genetics of Pacific salmon. III. Biochemical genetic studies of native Arizona and New Estimating effective population size. Journal of Heredity 81: 277-289. Mexico trout. Southwestern Naturalist 31:221-234. Lund R A, 0kland F & Hansen L P 1991 Farmed Atlantic salmon Waples R S 1991 Genetic interactions between hatchery and wild (Salmo salar) in fisheries and rivers in Norway. Aquaculture salmonids: Lessons from the Pacific Northwest. Canadian 98:143-150. Journal of Fisheries and Aquatic Sciences 48 (Supp 1):124-133. Maddison W P & Maddison D R 1992 MacClade: Analysis of Webb J H, Hay D W, Cunningham P D & Youngson A F 1991 Phylogeny and Character Evolution. Sinauer Associates Inc, The spawning behaviour of escaped farmed and wild adult USA. Atlantic salmon (Salmo salar L) in a northern Scottish river. Maneewong S 1987 Induction of spawning of sea bass (hates Aquaculture 98:97-110. calcarifer) in Thailand. In: Management of Wild and Cultured Webb J H, Youngson A F, Thompson C E, Hay D W, Donaghey Sea Bass/Barramundi (hates calcarifer) (eds J W Copland & D M J & McLaren, I S 1993 The spawning of escaped farmed L Grey). ACLAR Proceedings No 20. Australian Centre for Atlantic salmon, Salmo salar L, in western and northern International Agricultural Research, Canberra, 119-129. Scotland rivers: egg deposition by females. Aquaculture and Moore R 1979 Natural sex inversion in the giant perch (hates Fisheries Management 24:663-670. calcarifer). Australian Journal of Marine and Freshwater Wilde G R & Echelle A A 1992 Genetic status of pecos pupfish Research 33:803-813. populations after establishment of a hybrid swarm involving Patnaik S & Jena S 1976 Some aspects of biology of hates calcarifer an introduced congenor. Transactions of the American (Bloch) from Chilka Lake. Indian Journal of Fisheries 23:65- Fisheries Society 121:277-286. 72. Windsor M L & Hutchinson P 1995 Minimising the impacts of Pender P J & Griffin R K 1996 Habitat history of barramundi salmon aquaculture on the wild salmon stocks. In: hates calcarifer in a north Australian river system based on Sustainable Fish Farming (eds H Reinertsen & H Haaland). A barium and strontium levels in scales. Transactions of the A Balkema, Rotterdam, 149-166. American Fisheries Society 125:679-689. 136