|Scientific Name:||Lamna nasus (Bonnaterre, 1788)|
|Infra-specific Taxa Assessed:|
|Red List Category & Criteria:||Vulnerable A2bd+3d+4bd ver 3.1|
|Assessor(s):||Stevens, J., Fowler, S.L., Soldo, A., McCord, M., Baum, J., Acuña, E., Domingo, A. & Francis, M.|
|Reviewer(s):||Cavanagh, R.D., Heupel, M. & Simpfendorfer, C. (Shark Red List Authority)|
The porbeagle is a wide-ranging, coastal and oceanic shark, but with apparently little exchange between adjacent populations. Low reproductive capacity and high commercial value (in target and incidental fisheries) of mature and immature age classes makes this species highly vulnerable to over-exploitation and population depletion. This depletion, despite variations in availability of data and degree of depletion between the northern and southern hemispheres, is considered to meet Vulnerable criteria globally. The eastern and western North Atlantic populations have both been seriously over-exploited by directed longline fisheries. Collapse of the Northeast Atlantic population led to intensive target fishing in the well-documented Northwest Atlantic fishery in the 1960s, with most of the virgin biomass removed in just six years. Renewed target fishing in the 1990s led to a further population decline to ~11 to 17% of virgin biomass within the three generation period for this species. Recently improved management in the Northwest Atlantic should now help stocks to recover, however the Northeast Atlantic population has been subject to unrestricted fishing pressure ever since its earlier crash. Data are lacking, but stock depletion is considered to be much greater than in the Northwest Atlantic. Longline tuna and swordfish fleets in the southern hemisphere take a significant partially-utilised bycatch. Only limited trend data are available, including over 90% declines in landings by the Uruguayan longline fleet in the southwest Atlantic.
|Previously published Red List assessments:|
|Range Description:||The porbeagle shark is a wide-ranging coastal and oceanic species found in temperate and cold-temperate waters worldwide (1 to 18°C, 0 to 370 m), more common on continental shelves. Coastal and oceanic, amphitemperate, with centres of distribution in the North Atlantic and in a circumglobal band of temperate water of the southern Atlantic, southern Indian Ocean, southern Pacific and Antarctic Ocean (Compagno 2001). In the Southwest Atlantic Ocean it is found below 26°S (A. Domingo, pers.comm.), in the Southeastern Pacific, between 23 and 37°S (E. Acuña, pers. comm). Tagging studies in the Northwest Atlantic have shown mainly short to moderate distances of up to 1,500 km along continental shelves (Francis et al. in press). Distances travelled by 143 porbeagle tagged in a US study ranged 4 to 1,005 nautical miles (nm), with a mean distance of 234 nm, with over 90% moving less than 500 nm from their original tagging location (Kohler et al. 2002). Porbeagles tagged off southern England have been recaptured off Spain, Denmark and Norway (2,370 km away), and a porbeagle tagged in Ireland travelled 4,260 km Kohler and Turner 2001), suggesting mixing throughout their range in the Northeast Atlantic (Stevens 1976, 1990). According to the Canadian Department of Fisheries and Oceans (DFO 2001a), mature porbeagle sharks are rarely seen in winter and spring, with monthly catches exhibiting a seasonal and sex-specific migration of mature sharks towards the southern Newfoundland mating grounds in spring. The birthing grounds for the porbeagle shark in the Northwest Atlantic remain unknown (DFO 2001a). There is little known about the porbeagle shark in subequatorial Africa. No information on mating or nursery grounds is available. Porbeagles in the Northwest Atlantic make annual migrations along the coast between the Gulf of Maine and Newfoundland (Campana et al. 1999, Campana and Joyce 2004). |
See the Supplementary Material for Figure 1: Distribution map for Lamna nasus.
Native:Argentina; Australia; Bermuda; Brazil; Canada (Newfoundland I, Nova Scotia); Chile; Denmark; France; Germany; Gibraltar; Greenland; Guinea; Ireland; Morocco; Namibia; Portugal; South Africa (Eastern Cape Province, Western Cape); Spain; Sweden; United Kingdom; United States (Maine, Massachusetts, New Jersey, New York, Rhode Island, South Carolina); Uruguay
|FAO Marine Fishing Areas:|
Atlantic – southwest; Atlantic – southeast; Atlantic – western central; Atlantic – eastern central; Atlantic – Antarctic; Atlantic – northeast; Atlantic – northwest; Indian Ocean – eastern; Indian Ocean – western; Indian Ocean – Antarctic; Mediterranean and Black Sea; Pacific – southwest; Pacific – southeast
|Range Map:||Click here to open the map viewer and explore range.|
|Population:||North Atlantic tagging studies (DFO 1999, Kohler et al. 2002, Stevens 1990) indicate only one trans-Atlantic movement (Kohler and Turner 2001), implying that the two north Atlantic populations are distinct. There is no evidence of genetic exchange between the North Atlantic and the Southern Hemisphere population(s), which are separated by warm water. The number of subpopulations in the southern oceans is unknown. |
Biomass of the Northwest Atlantic porbeagle population was estimated at 4,409 t (11% of virgin biomass) and female spawners estimated at 6,075 (10% of the virgin abundance) (Campana et al. 2001).
Populations studied in the North Atlantic and Mediterranean segregate by size and sex. Although adults are very rare in the Mediterranean, it appears to be a nursery ground.
See the Supplementary Material for Figures 1 to 7 (catch rates and landings data).
|Current Population Trend:||Decreasing|
|Habitat and Ecology:||This section is taken mainly from Stevens (in press) species assessment for Lamna nasus (in Fowler et al. in press). |
The porbeagle reaches a maximum reported size of 355 cm TL (Francis et al. in press). Males mature at about 165 cm TL in the South Pacific and 195 cm TL in the North Atlantic. Females mature at about 195 cm TL in the South Pacific and 245 cm TL in the North Atlantic (Jensen et al. 2002, Francis and Duffy 2005, Francis et al. in press).
Reproduction is oophagous with litters of 1 to 5 pups (average four) produced, which are 68 to 78 cm TL at birth (Compagno 1984a, Gauld 1989, DFO 2001a, Francis and Stevens 2000, Francis et al. in press). Aasen (1963) estimated that the gestation period was about eight months in the North Atlantic and that individual females breed each year. However, Shann (1923) found two distinct size groups of embryos present in the December to February period and suggested that gestation may last 18 to 24 months. Gauld (1989) noted that a resting period may be present between parturition and fertilisation. Francis and Stevens (2000), Jensen et al. (2002) and Francis et al. (in press) estimate an 8 to 9 month gestation period. Birth occurs in spring off Europe, spring-summer off North America and winter in Australasia (Aasen 1963, Francis and Stevens 2000, Jensen et al. 2002) and the Eastern Pacific off Chile (Acuña et al. unpublished data).
Natanson et al. (2002) and Campana et al. (2002) examined age and growth in the North West Atlantic population and reported a maximum age of 26 years. However they estimated longevity might be as high as 46 years in an unfished population. Ages at 50% maturity for North Atlantic males and females are 8 and 13 years respectively (Jensen et al. 2002).
Porbeagles feed mostly on teleost fish, both pelagic and demersal species, and on cephalopods (Compagno 1984a). In the North West Atlantic, pelagic fish and squid are the main diet in deep water, and pelagic and demersal fish are important in their diet in shallow water (Joyce et al. 2002).
Like other mackerel sharks, the porbeagle is endothermic, maintaining its muscle and visceral temperatures above that of the surrounding seawater. It prefers temperatures below 18°C and has been caught in water temperatures as low as 2°C on the bottom (3°C on the surface) at high latitudes (Svetlov 1978). The preferred temperature range in the Northwest Atlantic is 5 to 10°C (Campana and Joyce 2004).
See the Supplementary Material for Table 1: summary of life history parameters for Lamna nasus.
This section is taken primarily from the review in Anonymous 2004, which refers in detail to other key sources.
The main threat to the porbeagle shark is unsustainable fisheries (target and bycatch) utilising its very high value meat. It is also a valued target game fish species for recreational fishing in Ireland and UK. The low reproductive capacity and high commercial value of both mature and immature age classes makes this species highly vulnerable to over-exploitation and population depletion. Well documented declines in this extremely valuable species during the past fifty years in the North Atlantic, which is the major reported source of world catches, has taken place during a period of rising fishing effort and market demand, and improved fisheries technology.
Porbeagles are a valuable secondary target of many fisheries, particularly longline fisheries, also gill nets, driftnets, pelagic and bottom trawls, and handlines. Examples include the demersal longlines for Patagonian toothfish Dissostichus eleginoides in the southern Indian Ocean and by the Argentinean fleet (Victoria Lichtstein, CITES Management Authority of Argentina, in litt. to TRAFFIC Europe, 27 October 2003), and longline swordfish and tuna fisheries in international waters off the coasts of Argentina and Uruguay (Domingo undated). Despite the large amount of fishing activity capturing porbeagles in the Southern Hemisphere, New Zealand is the only country that reports landings to FAO, indicating that the southern catch is largely unreported.
The high value of porbeagle shark meat means that most "bycatch" is exploited. The exception is in those high seas tuna and billfish fisheries where vessels' holding space is too limited to enable even valuable shark carcasses to be retained; in these cases the fins alone may be retained (e.g., the Japanese longline fishery for southern bluefin tuna off Tasmania and New Zealand, the pelagic fishing fleets of other countries in the southern Indian Ocean and probably elsewhere in the Southern Hemisphere (Compagno 2001)). Northeast Atlantic.
Porbeagle has been fished in this region by many European countries, principally Denmark, France, Norway and Spain. There has never been any restriction on fishing effort on this stock. The Northeast Atlantic fishery began when Norway started targeting porbeagle in the 1930s using longlines. Norwegian landings first reached a peak of 3,884 t in 1933. About 6,000t were taken by the Norwegian fleet in 1947, when the fishery reopened after the Second World War, followed by a progressive drop in landings to between 1,200 and 1,900 t from 1953 to 1960. The collapse of this fishery led to the redirection of fishing effort by Norwegian and Danish longline shark fishing vessels into the Northwest Atlantic. Norwegian landings from the Northeast Atlantic subsequently decreased from 160 to 300 t per annum in the early 1970s to only 10 to 40 t per year in the late 1980s/early 1990s, while average Danish landings fell from over 1,500 t in the early 1950s to less than 100 t throughout the 1990s (DFO 2001a, Gauld 1989).
French and Spanish longliners have operated directed fisheries for porbeagle since the 1970s. Reported landings from the main French fishing grounds in the Celtic Sea and Bay of Biscay decreased from over 1,092 t in 1979 to 3 to 400 t in the late 1990s. Spanish vessels appear to have taken porbeagle opportunistically both in the early and late 1970s and since 1998. Landings off Spain tend to be greater during the spring and autumn, with a drop in the summer (Mejuto 1985, Lallemand-Lemoine 1991). It is unclear, however, whether the very variable early landings data from the Spanish fleet (from nil to nearly 4,000 t per year) represents huge variations in catches, possibly the result of "boom and bust" fisheries removing different segments of the stock, or differences in catch reporting. Bonfil (1994) estimated that 50 t of porbeagle were taken as a supplementary catch in the Spanish longline swordfish fishery in the Mediterranean and Atlantic during 1989. The long line fishery in the Bay of Biscay (ICES Area VIII), directed at the more abundant blue shark, also landed about 30t of mainly porbeagle and some shortfin mako Isurus oxyrinchus during 1998 to 2000. ICES data (Heessen 2003) indicate that annual landings from Area IXa into mainland Portugal peaked at almost 3,000 t in 1987 to 1988 and have since declined (these records do not appear in the FAO statistics).
Reported landings from the historically most important fisheries, around the UK and in the North Sea and adjacent inshore waters have decreased to very low levels during the past 30 to 40 years, while catches from the offshore ICES sub-regions west of Portugal, west of the Bay of Biscay and around the Azores have increased since 1989. This is attributed to a decline in heavily fished and depleted inshore populations and redirection of effort to previously lightly exploited offshore stocks.
This species has virtually disappeared from Mediterranean records. In the North Tyrrhenian and Ligurian Sea Serena and Vacchi (1997) reported only 15 specimens of porbeagle during a few decades of observation. Soldo and Jardas reported only nine records of this species in the Eastern Adriatic since the end of 19th century until 2000. Recently two new records were reported (A. Soldo unpublished data). Orsi Relini and Garibaldi (2002) reported two newborn porbeagles were caught as bycatch of the swordfish longline fishery in the Western Ligurian Sea. A young porbeagle, considered to be very recently born was reported in the central Adriatic Sea (Orsi Relini and Garibaldi 2002). A young specimen was also caught in the central Adriatic during big-game fishing. On the basis of its length, it is suggested to be of an age 1 to 17 months (Marconi and De Maddalena 2001). These records indicate a possible nursery area in Central Mediterranean. During research of bycatch in the western Mediterranean swordfish longline fishery no porbeagles were caught (De La Serna et al. 2002). Only 15 specimens were caught during research conducted in 1998 to 1999 on bycatch of sharks in large pelagic fisheries: catches were reported only in the southern Adriatic and Ionian Sea, mainly by driftnets (Megalofonou et al. 2000). Official statistics for Mediterranean area show that the only landings in the Mediterranean were reported in 1996 by Malta: 1 t (FAO 2002). Porbeagles are also popular as recreational species (big game fishing) in some areas of Mediterranean.
Porbeagle fishing in the Northwest Atlantic started in 1961, when the fleet of Norwegian shark longliners began operating off the coast of New England and Newfoundland after the Northeast Atlantic stocks had been depleted. Catches increased rapidly from about 1,900 t in 1961 to more than 9,000 t in 1964. By 1965 many of the vessels had switched to other species or moved to other grounds because of the population decline. The fishery collapsed after only six years, landing less than 1,000 t in 1970. Smaller landings were also reported by Faeroese fishing vessels from around the same time and throughout the 1970s and 1980s. Norwegian and Faeroese fleets have been excluded from Canadian waters since the establishment of Canada's EEZ in 1995. Canadian and US authorities reported all landings after 1995. Porbeagle sharks are now taken almost exclusively by a Canadian directed longline fishery. Bycatch in swordfish longline fisheries and various inshore fisheries are minimal, rarely exceeding 40 t in recent years (DFO 2001a).
Following the collapse of the fishery in the 1960s, it took ~25 years for recovery to ~30% of virgin biomass to take place. Three Canadian vessels entered the targeted Northwest Atlantic fishery in 1994. Catches of 1,000 to 2,000 t per year throughout much of the 1990s reduced population levels to a new low in under ten years: the average size of sharks and catch rates (number/hook) were the smallest on record in 1999 and 2000. By 2000, catch rates of mature sharks were reduced to 10% of the 1992 peak, and immature catch rates to 30% of the 1991 peak. The biomass in 2000 was estimated as 11 to 17% of virgin biomass and fully recruited F estimated as 0.26 (DFO 2001a). The current porbeagle population is seriously depleted and will require a greatly reduced fishing mortality if recovery is to occur (DFO 2001). The 2001 stock assessment by the Canadian Department of Fisheries states: "An annual catch of 200 to 250 t would correspond to fishing at about MSY and would allow population growth." Following this advice, the Canadian quota was reduced to 250 t for the period 2002-2007 to allow population growth and recovery. There is a small quota (92 t) for porbeagle in the US EEZ.
Little information is available for southern ocean porbeagles. Landings from the Southern Hemisphere are only reported to FAO by New Zealand in the Pacific southwest. New Zealand reported landings were 150 to 300 t per year between 1998 and 2003 (Sullivan et al. 2005).
Records of captures of porbeagle by the Uruguayan pelagic tuna longline fleet declined significantly during 1981 to 1998 (A. Domingo undated). During the 1980s, only the two most valuable shark species were retained for their meat: porbeagle and mako Isurus oxyrinchus, representing about 10% of the total longline catch and peaking at 150 t and 100 t landed, respectively, in 1984. By the early 1990s, landings of these two species had fallen to around 10% of that peak, although shark fin prices were rising and blue sharks Prionace glauca and eight other species of large sharks were now also being retained in large quantities. This was accompanied by a decrease in CPUE from 110 kg/1,000 hooks (1988) to 1 kg/1,000 (1999) in the Uruguayan tuna and swordfish fleet. This is not, however, necessarily due to population decrease but to changes in fleet operating maneuvers (area and depth) and an increase in mean temperatures of water masses in the area (A. Domingo pers. comm.). There is bycatch of porbeagle shark in the Chilean artisanal and industrial longline swordfish Xiphias gladius fishery within and outside the Chilean EEZ, between 26 and 36°S (E. Acuña unpublished data; Acuña et al. 2002). There are no good estimates for captures in the Southeastern Pacific off Chile; several times porbeagle has been included as Isurus oxyrinchus. Acuña et al. (2002) determined during a one year sampling period that L. nasus catches were around 3.5% of the total catch of fishes (in numbers) in the Chilean swordfish fishery, with a size range between 75 and 245 cm TL.
Porbeagles recently showed up as bycatch of demersal longlines for the Patagonian toothfish Dissostichus eleginoides in the southern Indian Ocean (Compagno 2001). Porbeagles are rarely seen in South African waters; being occasionally caught in research vessel longline cruises (L.J.V. Compagno pers. comm). An experimental tuna longline fishery operated in South African waters from 1997 until March 2005. Porbeagles were seldom caught in the experimental fishery. Catches were mainly along the southwest coast of South Africa and observed catch rates were approx. 0.014 sharks per 1,000 hooks. Since March 2005 the fishery has been commercialized with the allocation of 17 swordfish-directed vessels and 26 tuna-directed vessels. The current foreign flagged vessels in the commercial fishery target yellowfin tuna in the warm Agulhas current and are unlikely to catch significant quantities of porbeagle. In contrast, foreign flagged vessels, especially Japanese, which target Southern Bluefin Tuna to the south of South Africa, land substantially more quantities of porbeagle as bycatch (C. Smith, pers. comm).
See the Supplementary Material for Table 1: bycatch of porbeagle in South African pelagic longliners; and Table 2: seasonal trend in bycatch.
The porbeagle shark is included on Annex 1 (Highly Migratory Species) of the UN Convention on the Law of the Sea (UNCLOS), which lists "Family Isurida" (former name for Family Lamnidae) among other oceanic sharks. The UN Agreement on Straddling Fish Stocks and Highly Migratory Fish Stocks establishes rules and conservation measures for high seas fisheries resources and has been in force since 2001. It directs States to pursue co-operation in relation to listed species through appropriate sub-regional fisheries management organisations or arrangements. No progress with implementation of shark fisheries management appears to have been achieved.
In 2004, the EU rejected a proposal from Germany for the EU to submit an Appendix II listing proposal to the 13th Conference of Parties to CITES, with the objective of ensuring the sustainability of fisheries supplying international trade. There is no regulation of the international trade demand for meat that is driving many fisheries. However, the porbeagle is one of the subjects of CITES Decision 13.42, directed to Parties, which states that Parties should take note of the species-specific recommendations in document CoP13 Doc. 35 Annex 2 with a view to ensuring that international trade is not detrimental to the status of these species (CITES). The relevant section reads: "The Animals Committee concluded that North Atlantic populations have been severely depleted and noted that quotas in EU waters apply only to non-EU fleets through access agreements. As these quotas are far higher than can be supported by the stock and do not restrict fishing effort they are not considered to be an effective management measure in this case". The Animals Committee recommended the following:
a) ICCAT members are encouraged to collect and report data on catches and discards of porbeagle sharks, as per ICCAT Resolution 95-2, which has yet to be complied with, and undertake stock assessments in order to develop management recommendations. Other relevant Regional Fishery Management Organizations are encouraged to establish and implement similar programmes.
b) The US and Canada are encouraged to enhance existing management for their shared porbeagle stock by establishing a cooperative, bilateral research and fisheries management programme.
c) The World Customs Organization (WCO) is encouraged with urgency to establish a harmonized international code for porbeagle sharks.
Two Regional Fisheries Organisations, ICCAT and IATTC, have adopted finning bans, as have several range states (e.g., USA, EU, Australia, Brazil etc.). More are likely to follow suit.
The 1995 Fisheries Management Plan for pelagic sharks in Atlantic Canada established precautionary catch levels of 1,500 t for porbeagle in the target shark fishery. License limitation, a ban on finning, restrictions on gear, area and seasons, bycatch limits and restrictions to recreational fishers permitting hook and release only were also implemented (Hurley 1998). The porbeagle TAC was reduced to 1,000 t in 1997 (O'Boyle et al. 1998), then to 1,700 t during the two years 2000 to 2001 while additional scientific information was collected (DFO 2001b). As a result of these studies, it was concluded that the population was seriously depleted (to 10 to 20% of virgin biomass) and would require a greatly decreased fishing mortality if recovery is to occur. An annual catch of 200 to 250 t would correspond to fishing at about MSY and would allow population growth. However, annual catch levels of about 1,000 t would be sustainable over the long term once the population has recovered. These DFO analyses, which are the basis for the regional assessment, are contested by COSEWIC (2004), who notes that the quota for 2002 to 2007 of 200 to 250 t represents a substantial reduction from catches in the mid-1990s, but even this amount now corresponds to a high exploitation rate because of the low population abundance. It is highly uncertain if this quota reduction will be sufficient to halt the porbeagle decline, and if so, to what extent the population will recover, given that there is uncertainty in estimating FMSY and the quota, the low number of mature animals remaining in the population, that at its current low abundance the population may experience depensation (Allee effects), and that reduction in fishing pressure is not always sufficient for population recovery (Hutchings 2001). COSEWIC (Committee on the status of endangered wildlife in Canada) designated porbeagle as Endangered in 2004 (COSEWIC 2004).
Since 1993, shark fisheries in Atlantic and Gulf of Mexico waters in the US have been managed under the Fishery Management Plan for Sharks of the Atlantic Ocean. The plan set commercial quotas in the US EEZ for 10 species of pelagic sharks at 580 t dressed weight annually, with recreational bag limits also applied. There is a small quota (92 t) for porbeagle. Commercial fishers require an annual shark permit, and finning is prohibited.
Norway is allocated a quota of 200 t of porbeagle in European Community (EC) waters, reduced in 1985 from the 500 t established in 1982 (Gauld 1989). Since 1985, the Faeroe Islands can also take 125 t from EC waters (originally 300 t in 1982, 150 t in 1984a). These quotas currently exceed total landings from shelf areas in the region and yield no management benefit.
Annex V of the Convention on the Protection and Conservation of the Ecosystems and Biological Diversity of the Maritime Area [also called OSPAR (Oslo-Paris) Convention] requires OSPAR to develop a list of threatened and/or declining species and habitats in need of protection or conservation in the OSPAR maritime area (Northeast Atlantic). OSPAR member states were invited in 2001 to submit proposals for inclusion on this list. In response, Portugal, on behalf of the Azores, proposed to list porbeagle Lamna nasus in the wider Atlantic because of its biological sensitivity, keystone importance and the severe decline in its population. This proposal has not yet been adopted.
The status of the largely unmanaged, unmonitored Northeast Atlantic stock is likely worse than the seriously depleted Northwest stock, with stringent conservation and management action (fisheries closure and stock assessment) needed urgently to enable stocks to rebuild to levels where sustainable commercial and recreational fisheries are possible.
The species is listed on Annex III, "Species whose exploitation is regulated" of the Barcelona Convention Protocol concerning specially protected areas and biological diversity in the Mediterranean, signed in 1995 but not yet ratified. The Mediterranean population of this species was also added in 1997 to Appendix III of the Bern Convention (the Convention on the Conservation of European Wildlife and Natural Habitats) as a species whose exploitation must be regulated in order to keep its population out of danger. No management action has yet followed these listings. The draft action plan for the conservation of cartilaginous fishes (Chondrichthyans) in the Mediterranean Sea (Anonymous 2002) has identified porbeagle as a species that urgently needs development of a management programme for sustainable fisheries.
In 1991, Australia brought in legislation that prevented Japanese longliners fishing in the EEZ from landing shark fins unless they were accompanied by the carcass. Since 1996, these vessels have not fished in the Australian EEZ.
The Southern African population of porbeagles is largely unmonitored and little data is available on bycatch levels in the longline fisheries of this region. However, all foreign flagged vessels in the South Africa fishery are required to have 100% observer coverage, whereas domestic vessels are required to have 20% observer coverage (C. Smith, pers.comm). The bycatch must be monitored and this porbeagle assessment reviewed within five years of the opening of this fishery.
A small, regulated fishery is permitted by New Zealand, which included porbeagles in its quota management system in October 2004, with a TAC set at 249 t (Sullivan et al. 2005). There are no other management measures applicable to the Antarctic and Southern Ocean, since CCAMLR appears not to be monitoring or managing this species.
|Citation:||Stevens, J., Fowler, S.L., Soldo, A., McCord, M., Baum, J., Acuña, E., Domingo, A. & Francis, M. 2006. Lamna nasus. The IUCN Red List of Threatened Species 2006: e.T11200A3261697.Downloaded on 19 February 2018.|