Skip booklet index and go to page content

COSEWIC Assessment and Update Status Report on the Lake Sturgeon in Canada

Population Sizes and Trends

The assessment of sustainability of populations should consider effective population size (Reiman and Allendorf 2001), not just total numbers. A population whose effective population size is too small (< 50) becomes susceptible to inbreeding depression. A population of at least 500 mature individuals is considered necessary to maintain adaptive genetic variation, assuming a population with equal sex ratios and equal contribution of all adults to the next generation. This has important implications for lake sturgeon as in many populations, sex ratios may be unequal, not all individuals breed every year, and there may be variance in the age of maturity (Earle 2002). Therefore effective population sizes for the DUs may be substantially smaller than total population estimates or censuses.

As fisheries population estimates have substantial uncertainty and risk of extinction is serious and potentially irreversible, the preconditions for the application of precaution are met. In such circumstances it is customary to adopt the lower 95% confidence limit for a population estimate as the basis for status assessment. This practice corresponds to recommended practices in fisheries population dynamics as well (Richards and Maguire 1998).


Historic and Geographic Trends

There are several reports on the decline of lake sturgeon populations in Canada (Harkness and Dymond 1961; Houston 1987; Dick and Choudhury 1992). It is uncertain when commercial exploitation of the species commenced, but there are records of Isinglass sales to the Hudson Bay Company at Norway House, Manitoba from 1832 to 1892 (MacDonell 1997, 1998). Commercial harvest for caviar production was initiated in Sandusky, Ohio in 1855 (Harkness and Dymond 1961), and Canadian waters of the Great Lakes were opened to commercial sturgeon fishing in 1879 (Prince 1905). Within a short time, populations of lake sturgeon throughout the Great Lakes and in the surrounding waterbodies on both sides of the International Border were reduced to less than 1% of their former numbers (Hay-Chmielewski and Whelan 1997). As European settlement expanded and infrastructure was developed, prices rose and the sturgeon-rich waters of northwestern Ontario and Manitoba were opened to commercial operators. The collapse of these fisheries followed the pattern of their southern counterparts. As populations in northwestern Ontario and Manitoba began to collapse, commercial harvesting relocated to populations of previously unexploited rivers and lakes (Brousseau 1987; Houston 1987; Sopuck 1987; Patalas 1988). Wherever lake sturgeon have been fished, the result has been the same: a relatively high initial yield followed by a sudden decline to very low levels, which persist thereafter (Harkness and Dymond 1961). This pattern is consistent with life history traits of slow growth and late maturation, and high commercial value products. Commercial catch records demonstrate that lake sturgeon populations over most of their historical range were severely depleted or extirpated by the early 20th century (Tables 5 and 6), and most have not recovered, exceptin the central and eastern parts of the range where some commercial and recreational fisheries in Ontario and Quebec remain (Tables 3-6).

 

Table 3: Total Harvest (kg) by Year for Lake Sturgeon in Quebec
YearTotal Harvest (kg)
1987245 200
1990222 232
1991216 813
1992222 102
1993224 934
1994261 831
1995251 606
1996198 388
1997197 245
1998215 834
1999192 512
2000126 394
2001126 188
2002110 150
200383 042
200481 298
200574 501

Totals for the 1987 to 1998 period are estimates based on a formula using export values. For example, in 1999 lake sturgeon were worth $3.49/kg1. Totals for the subsequent years are based on landings declarations2.

1 Danielle Hébert (Ministère de l‘Agriculture, des Pêcheries et de l’Alimentation du Québec) and Laurette Gagnon (Statistical Services, Fisheries and Oceans Canada.
2 Marcel Bernard (Ministère des Ressources naturelles et de la Faune du Québec).

In Ontario, lake sturgeon still occur in 38 of the old 47 Ontario Ministry of Natural Resources (OMNR) administrative districts. They are common in 10, rare in 28 and absent from 9 (Brousseau 1987). Kerr (2002) lists the Ontario waters currently known to contain lake sturgeon; however, the relative abundance of sturgeon is not provided. The species is reported from 13 lakes/reservoirs and 25 streams/rivers in the south-central region, from 16 lakes/reservoirs and 46 streams/rivers in the northeast region and from 95 lakes/reservoirs and 30 streams/rivers in the northwest region (Kerr 2002: Figure 1). Landings of 6000 kg of lake sturgeon were reported from the commercial fishery during 2000 (Table 4) compared to the hundreds of thousands of kg harvested annually in the late 1800s and early 1900s (Brousseau 1987; Houston 1987). However, most of the fish harvested recently were not from populations that supported the historic commercial fisheries. Most of those populations have failed to recover from the exploitation of the late 19th and early 20th centuries. The commercial harvest is < 0.1% of that reported in 1895 and has declined by over 80% since 1964 (Brousseau 1987). At the same time, recreational angling has increased. Since 1984, recreational yieldsfrom this fishery have surpassed the commercial harvest (Brousseau 1987) and still are increasing where populations are abundant enough to support angling.

Table 4: Commercial Landings of Lake Sturgeon (tonnes) from Ontario1
LocationYear
1994199519961997199819992000
Lake Ontario-------
Lake Erie1-1-111
Lake St. Clair       
Lake Huron: Main Basin3344344
Lake Huron: Georgian Bay-------
Lake Huron: North Channel1111211
Eastern Lake Superior-------
Western Lake Superior-------
Northern inland lakes including Lake Nipigon411111-

1 Laurette Gagnon (Statistical Services, Fisheries and Oceans Canada. Data came from the Ontario Commercial Fishers Association, Blenheim, Ontario.

 

Table 5: Duration, Maximum and Minimum Commercial Harvest of Lake Sturgeon from Canadian Waterbodies
WaterbodyFirst
Year
Last
Year
MaximumMinimum
kg
kgYear
Lake Ontario11879192046 9091882454.5
Lake Erie118791964277 0901887454.5
Lake Erie11879194541 8181879909
Lake Huron118791943379 22718862 727
Lake Superior11885190455 0001902454.5
Lake Simcoe21881191513 1821902454.5
Lake of the Woods318921930111 4841895454.5
Lake Nipissing41900193075 00019031 400
Lake Winnipeg518851920446 13619003 409
Nelson River61907191011 68219094 091
Sipiwesk Lake61917191868 182191730 909
Sipiwesk Lake61921192966 36419241 818
Sipiwesk Lake61937194613 63619382 636
Sipiwesk Lake61953195914 04519544 318
Cumberland House, Saskatchewan River71900-15 50019282 700
Cumberland House, Saskatchewan River7    3 600
Ottawa River8-190728 84018986 500
Lake St. Francis8-1969 (native)8 70019641 600
Lake St. Louis8 1976 (native)39 750197215 360
St. Lawrence River9 (Quebec)19202005245 70019955 540

1 Baldwin et al.
2 McCrimmon and Skobe 1970;
3 Mosindy and Rusak 1991; 1978;
4 Brousseau 1987;
5 Harkness 1980;
6 Sopuck 1987;
7 Wallace 1991;
8 Houston 1987;
9 See Figure 15 in this document.

 

Table 6: Duration, Maximum and Minimum Combined Lake Sturgeon Harvest (kg) in Canadian and American Waters
LakeLake
Area
(km2)
YearMaximumMinimum
kg
KgYear
Ontario19 550189026 0901920909
Erie125 67018852 357 72719213 636
Huron159 8301885473 1811951@6 818
Michigan157 75018791 745 4541956@455
Superior182 0001885108 11819193 636
St. Clair11 2701879495 9091908*16 363
Lake of the Woods23 8461895>800 0001926<1 000

Data only include years when both countries were involved in the commercial fishery.

@State of Michigan waters closed to commercial sturgeon fishing from 1920-1950.
* After 1908, the American portion of Lake St. Clair did not yield more than 227 kg of fish/year so was not recorded as a contribution to the lake’s harvest data after this time (i.e. only the Canadian production after 1908 on Lake St. Clair).
1 Baldwin et al. 1978;
2 Mosindy and Rusak 1991.

Catches reported from Quebecover an extended time differ from the history of other commercially fished lake sturgeon populations in Canada. With the exception of the Second World War period, declared landings since 1920 were generally over 50 tons (and yields over 0.5 kg/ha) [Figure 15]. During this period, fisheries in parts of Quebec did not follow the scenario observed in other areas where initial, relatively high yields were followed by quick, rapid and prolonged declines. However, the catch records prior to 1983 for the Quebec part of the St. Lawrence River are considered inaccurate and must be seen as minimal (Dumont et al. 1987). Therefore, the true pattern of catches prior to the mid-1980s is uncertain; however, Fortin et al. (1993) and Dumont et al. (2000) estimated that between 15,000 to 30,000 sturgeon were harvested annually during the 1980s and 1990s.

Many lake sturgeon populations in DUs 6 and 8 are also shared with the United States. Therefore, these populations could be affected by management activities and harvesting activities that occur outside of Canadian waters. Trends in Canadian waters could be influenced by the population dynamics and movement of lake sturgeon in adjacent U.S. waters.


Population Trends by Designatable Unit

Western Hudson Bay (DU1)

Skaptason (1926) reported landings of up to 46,000 kg/yr for lake sturgeon from the Churchill River in the early 1920s, as well as the presence of lake sturgeon in the upper Churchill Riverat Granville and Indian lakes. Landings from 1939 to the late 1960s from the Churchill-Granville-Opachuanoa region ranged from 100 to 530 kg/yr (Lysack, pers. comm.). However, lake sturgeon catches thereafter dropped off and only sporadic catches were reported [1967 to 1968 (13 kg/yr), 1976 to 1977 (145 kg/yr), and 1986 to 1987 (70 kg/yr)], despite continued effort to fish for lake sturgeon (SERM 1996). Landings data suggest that the apparent availability of sturgeon had declined by over 98% between the 1920s and 1939. The few sturgeon reported are very large and probably very old (two in the last 5 years from the Upper Churchill River at Sandy Bay, Saskatchewan); the population is believed to be a remnant, but no other recent information on this population is available (Wallace, pers. comm.).


Figure 15: Declared Lake Sturgeon Commercial Landings in the Quebec Part of the St. Lawrence River (1920-2005)

Declared Lake Sturgeon Commercial Landings in the Quebec Part of the St. Lawrence River (1920-2005).

Landings data adapted from Robitaille et al. (1988) and from Marcel Bernard (unpublished data). NA = Not Available.

Mark-recapture and larval drift studies conducted in 2003 at the confluence of the Churchill and Little Churchill rivers provide a population estimate of 1812 ± 508 adult sturgeon (MacLean and Nelson 2005). This may however be an overestimate of the number of adult fish, as the assumptions applied in their (MacLean and Nelson 2005) Petersen single census estimate equation included that the captured fish were all adults, and that the population is stable and unexploited. The population is subject to subsistence harvest by Cree Nation peoples of the area (Tataskweyak, War Lake, Fox Lake, and York Factory Cree Nations), who are interested in a sustainable domestic fishery. The extent of subsistence usage is not known, nor is the degree of consistency of the subsistence fishery over the past decades. Of the 366 fish captured, ranging in size from 0.4-1.4 m (fork length) in length and 5.7 to 24.7 kg in weight, fish over 0.8 m fork length were taken as mature and included in the estimation. However, other reports (see Biology section) indicate many fish of this size would still be immature. Moreover, the low mean weight and age suggest that this is a population with no large, older mature fish, an indication of stress, and perhaps overexploitation.

In summary, historic landings information indicate a decline of over 90% between the 1920s and 1940s, with no evidence of subsequent substantial increase. Life history traits of harvested lake sturgeon in the 2000s are consistent with a population that has been subject to severe overexploitation and has not recovered. The only available population estimate, for part of one watershed in the DU, estimates between 1300 and 2300 adults, but may be biased upward.

Saskatchewan River (DU2)

There are early records of sturgeon, undoubtedly A. fulvescens, taken in the North Saskatchewan Riverand adjacent tributaries. For example, Moodie (1965, p. 59) noted a statement in a book by G.M. Grant that “sturgeon up to 10 kg (“25 pounds”) were caught in the Sturgeon River as late as 1872 by men of the Sandford Fleming expedition” (Nelson and Paetz 1992). Sturgeon no longer ascend this tributary as water levels now are too low to support a population (J. Nelson, Department of Zoology, University of Alberta, Edmonton, AB; personal communication).

In Alberta, a 1986 estimate for the population of the South Saskatchewan River was 510 potential spawners (those > 20 years of age). Assuming an equal sex ratio, and females spawning every 4 to 7 years, there would be 36 to 63 female sturgeon spawning per year in the Alberta portion of the South Saskatchewan River (Alberta Sustainable Resource Development 2002). The Alberta portion of the North Saskatchewan River has an estimated 190 mature fish (Alberta Sustainable Resource Development 2002). Again, assuming an equal sex ratio, and females spawning every 4 to 7 years, there are probably less than 100 mature females and only about 14 - 25 spawning females spawning annually in the North Saskatchewan River.

Although historic population levels are unknown and both populations have experienced significant declines in numbers, there are signs of positive recruitment (Nelson and Paetz 1992). However, the number of spawning fish remaining in each sub-population may be below the critical number required for genetic or demographic viability (Alberta Sustainable Resource Development 2002). The North Saskatchewan River has about 2.5 lake sturgeon per kilometre. In the summer of 2002, boat electrofishing (4.01 hours) near Medicine Hat on the South Saskatchewan River yielded 8 lake sturgeon, and an additional 13 lake sturgeon observed (age and size information not indicated), from a total sample of 821 fishes representing 17 species (D. Watkinson, Central and Arctic Region, Department of Fisheries and Oceans, Winnipeg, MB; personal communication).

Little is known about populations in the Saskatchewan River upstream of the E.B. Campbell Dam to the Alberta border. Historically, there were a total of 111 sites where lake sturgeon were known to occur in the Saskatchewan River watershed: 30 in the South Saskatchewan, 48 in the North Saskatchewan and 33 in the Saskatchewan (Smith 2003). Recent surveys indicate that lake sturgeon occurrences are now regularly reported from only 7 sites on the South Saskatchewan, 16 on the North Saskatchewan, and 12 on the Saskatchewan River, a loss of 76 sites (Smith 2003). Lake sturgeon are still present in Tobin Lake and at the Forks of the Saskatchewan (J. Durbin, Saskatchewan Environment, Regina, SK; personal communication), and anglers reported catches at several sites along both rivers, although the recreational fishery has been closed since 1996 (Smith 2003). In the summer of 2002, boat electrofishing (4.01 hours), downstream of Prince Albert on the North Saskatchewan River failed to find any lake sturgeon (Watkinson, pers. comm.). Boat electrofishing in large, deep, turbid rivers may not be effective in producing catches of lake sturgeon (Golder 1999). However, the method has been used successfully for lake sturgeon in shallow, less turbid waters such as those in the 2002 study area, and according to Kempinger (1996) is the most effective way to sample YOY.

In 1885 the typical weight of fish taken in Saskatchewan was in the order of 5 to 30 kg (Houston 1987). Angling records from the 1990s indicate that the typical weight of fish caught now ranges from 0.1 to 14 kg, although fish up to 30 kg and larger are still taken occasionally (Smith 2003). Similarly, test netting indicates that average size of sturgeon is much less than that earlier in the 20th century (Smith 2003).

Skaptason (1926) reported a quota corresponding to nearly 30,000 kg for lake sturgeon from the Saskatchewan River (from CumberlandLake downstream to Lake Winnipeg at Grand Rapids), although total population size was not estimated. In 1960, prior to construction of the E.B. Campbell Dam, it was estimated that there were about 10,000 to16,000 lake sturgeon, 8.2 kg and larger, in the Cumberland area (Wallace 1991). By 2001, there were estimated to be < 1300 sturgeon, 8.2 kg and larger, in the Saskatchewan River from the E.B. Campbell Dam in Saskatchewan to The Pas in Manitoba including the delta at Cedar Lake (Smith 2003; R. Wallace, Saskatchewan Environment, Saskatoon, SK, personal communication; Lysack unpubl. data). Therefore, this population has declined by more than 80% in 40 years, and even more than that in the last 80 years. Mark-recapture estimates indicate that the remaining lake sturgeon populations on the Saskatchewan River between Cumberland House, Saskatchewan and Cedar Lake, Manitoba may have declined by 50% since 1998 (Lysack, pers. comm.; Smith 2003).

Findlay et al. (1995) and Findlay (1995) report that the decline at Cumberland House became noticeable in the 1980s and Catch Per Unit Effort (CPUE) declined by 88% between 1982 and 1993. A similar decline at The Pas, Manitoba began in the
mid-1970s and within 10 years, CPUE and harvest statistics declined by 93% (Findlay 1995). A study conducted on the lower Saskatchewan River between The Pas and Cumberland House in 2000 (Bretecher and MacDonell 2001) yielded only 3 lake sturgeon with a CPUE of 0.002 (sturgeon/yard/day), which was similar to that reported by Wallace and Leroux (1999), but much lower than the CPUE for Alberta sections of the Saskatchewan River (0.029), or the CPUE for Ontario rivers (0.011) [Seylor 1997a,b]. Analysis of age-sex distributions for the Cumberland House harvest indicated a shift in the size-age composition of the catch between 1954 and 1965, during construction of the E.B. Campbell Dam, toward smaller, younger fish and that the trend was evident into the 1990s (Wallace 1991). Using a gonad index, Dick (unpubl. data) found that 97% of the lake sturgeon from the last commercial sturgeon fishery at Cumberland House were pre-spawners or juveniles.

Lake sturgeon harvest surveys conducted at Cumberland House and The Pas during the summers of 2000 and 2001 (North/South Consultants 2002) indicated a variety of fishing pressures that included 30 groups of anglers, 27 groups of domestic fishers, one group of commercial fishers and two fishing derby groups (all groups were accompanied by at least one First Nations member). The estimated harvest was 319 sturgeon annually for 2001 and 2002 with a CPUE of 0.0004/angler/hour. This harvest produces an estimate of an exploitation rate of 12.3%, which is unsustainable (North/South Consultants 2002). This may be an underestimate of exploitation rate because the survey methodology may not account for all domestic/subsistence use. Domestic/subsistence use is not reported systematically, but may be at a scale to present a concern for the sustainability of the Saskatchewan River lake sturgeon population (Brectecher and MacDonell 2001).

In summary, all estimates of population status in this DU suggest major reductions from the 1920s to the 1960s and 1970s, but these declines are difficult to quantify. Since the 1970s, there has been a further decline of as much as 80% in estimates of the populations comprising this DU. Current density estimates from a variety of sources are consistently very low, and estimates of the numbers of breeding females in various river branches do not exceed a few dozen spawners annually. The information on age and size composition of components of this DU are also consistently indicative of a population that has been affected by heavy exploitation and habitat disruptions for a long period, although recruitment to the populations is still occurring.

Nelson River (DU3)

Lake sturgeon are found throughout the Nelson River from Lake Winnipeg to Hudson Bay. Commercial fisheries established in 1907 initiated a pattern of exploitation followed by collapse and closure of the fishery. The fishery was opened and closed for periods of 4 - 9 years on at least 3 occasions between 1932 and 1970. In 1991 it was finally decided that commercial fishing on the Nelson was not sustainable and the remaining commercial fisheries were subsequently curtailed. Local domestic fisheries persist, however (MacDonnell 1995). Information on populations and harvests from the upper Nelson River are presented in Macdonald (1998).

A population estimate for lake sturgeon in the region of SipiweskLake in the Nelson River drainage in 2000 was 1200 adults with a high proportion of males (Macdonald, pers. comm.). In the past, this area provided most of the commercial harvest within the Nelson River (Sopuck 1987). Compared to initial population estimates of 12,000 lake sturgeon by Sunde (1961), or even the corrected estimates of 6,000 sturgeon for 1987 (Sopuck 1987), the population has declined by 80 - 90% between 1987 and 2000 (Macdonald, pers. comm.), and more than 90% since 1960. Current population estimates may be reduced even further as a substantial number of lake sturgeon were harvested from Sipiwesk Lake during the summer of 2000. Choudhury and Dick (1993), using a gonad index, found that 87% of the lake sturgeon from the last commercial sturgeon fishery in Sipiwesk Lake were pre-spawners or juveniles.

Impoundment of the Nelson River for hydroelectric generation has isolated sturgeon into a series of reservoirs and this fragmentation may be one of the major limiting factors for sturgeon in the Nelson system (Horne and Baker 1993).

About 500 fish were recorded in the area below the Limestone Dam on the Nelson Riverin 1998, but this concentration of lake sturgeon may be atypical. The large concentration may be related to changes in flows below Limestone Dam because flows drop at night and during the weekends as electrical demand drops, and/or it may be due to the large natural flow fluctuations observed on the Weir River (Macdonald, pers. comm.). However, the river below the dam is the longest stretch (100 km) of unimpounded water on the river and the lake sturgeon in this section may represent perhaps the last true riverine stock on the river (MacDonnell (1995). The Cross Lake and Playgreen Lake stocks are probably extirpated (Manitoba Department of Natural Resources 1994).

Studies conducted on the Lower Nelson at the Weir River, a tributary of the Nelson downstream of Limestone Dam and rapids, and the only known spawning location on the lower Nelson (MacDonnell 1997), during the 1990s indicate that lake sturgeon are still using the area, and that fish from the Hayes River may also utilize this as a spawning area (Barth and MacDonnell 1999). Catches varied from year to year from a low of 26 (15 adults) in 1994 (MacDonnell 1995), to a high of 355 (215 adults) in 1996 (MacDonnell 1997). Immature lake sturgeon made up about 30 – 40 % of the catch except in 1998 when immatures made up only 9.5% of the 232 fish (Macdonnell 1995, 1997, 1998; Barth and MacDonnell 1999). Small numbers of eggs (672 in 1997 and 82 in 1998) and larvae (< 100 annually) were collected in larval drift studies. Annual variation in catch is probably related to lake sturgeon spawning periodicity. The largest fish captured measured 1.6 m in length, weighed 30 kg, and was aged at 43 years. Throughout the study period, length and weight of the largest fish in the sample each year ranged from 1.4 to 1.6 m in length and 17.3 to 30 kg in weight (MacDonnell 1995, 1997, 1998, Barth and MacDonnell 1999). The lack of older, larger fish and low numbers of larvae are suggestive of a stressed population with low recruitment.

In 1990, several groups of 5 to 6 spawning sturgeon were observed at LandingRiver, a tributary of the Nelson River. This compares with reports of hundreds of lake sturgeon spawning at the site several decades ago (Macdonald, pers. comm.).

Sixty-nine lake sturgeon were captured and tagged in the GullLake area (between Split and Stevens lakes) in northern Manitoba during an intensive monitoring program in 1998. An additional 350 were tagged in a series of studies conducted in the Gull Lake area between 2001 and 2004 as part of an aquatic monitoring and environmental impact assessment program in relation to a proposed development of a hydroelectric generating station at Gull Rapids (upstream end of Stevens Lake) on the Nelson River. Spring gillnetting from 2001 to 2004 resulted in catches of 139 (2001) to 335 (2003) lake sturgeon annually (Barth and Mochnacz 2004; Barth 2005; Barth and Murray 2005; Barth and Ambrose 2006). The largest lake sturgeon caught was 1.54 m in length and weighed 54 kg. Although not aged, this fish was probably over 40 years of age. The 338 fish surveyed in 2004 ranged in length from 0.9 to 1.5 m and 5.4 – 31.3 kg in weight. The spring gillnetting in 2003 and 2005 indicated that about 18-19% of the fish captured were in spawning or post-spawning condition (Barth and Ambrose 2006) with about 80% immature or non-spawners. Few larvae were taken in larval surveys and fall netting resulted in the capture of only 3 lake sturgeon in 700 hours of fishing effort (Barth 2005). As in the case of the Weir River studies the population demographics suggest a stressed population with poor recruitment.

In summary, historic commercial landings data indicated large but unquantified declines in abundance of this DU since the first quarter of the 1900s. For the period since population estimates have become available for components of this DU, estimates of the largest component declined 80-90% from the early 1960s to the late 1990s, and have declined further thereafter. Dams in the river system have fragmented the population into largely isolated components, some of which appear to have been extirpated. All components that have been studied show few spawners annually, few eggs and larvae, a low proportion of mature sturgeon in samples, and a small maximum size of the mature sturgeon that are found.

Red-Assiniboine Rivers-Lake Winnipeg (DU4)

Large commercial fisheries for lake sturgeon once existed on Lake Winnipeg (Table 5). The largest single year catch of lake sturgeon was 446,136 kg, taken in 1900. Extensive sampling of fish populations in Lake Winnipeg has been conducted from the early 1970s to the present, using gill nets with a stretched mesh size ranging from 7.5 cm to 12.5 cm. Sampling consisted of 30 stations sampled twice a year from the early 1970s until the early 1980s. From the early 1980s to the present, sampling continued at four sites in each of the north and south basins of Lake Winnipeg where survey nets are set in conjunction with the commercial fisheries for other species. These surveys have not collected any lake sturgeon (K. Campbell, Manitoba Conservation, Gimli, MB; personal communication), and during the past 28 years, only three lake sturgeon have been observed in Lake Winnipeg, two under 2 kg and one about 15 kg (Campbell, pers. comm.).

Little is known about lake sturgeon populations in the AssiniboineRiverother than that they have virtually disappeared. There is no evidence of a naturally reproducing population in the Assiniboine River. Stocking of lake sturgeon in the Assiniboine River was undertaken between 1996 and 2002 by Manitoba Conservation and the Brandon Wildlife Association to determine the survival of fry (4000 stocked) and fingerlings (4156 stocked) produced in the lake sturgeon culture program of Manitoba Conservation. Extensive electrofishing (1995 to 2002) between the Red River and the City of Brandon has failed to detect any sturgeon, although the area close to Brandon was sampled only once. Most electrofishing was conducted further downstream near the crossing of Highway 34 and between Portage la Prairie and Winnipeg. Additional sampling was done upstream of Brandon near the confluence with the Qu’appelle River (Franzin, pers. comm.). Between 1998 and 2002, anglers reported catching over 280 lake sturgeon (lengths 20-100 cm) from the Assiniboine River in a 20-km stretch between Brandon and the Little Souris River (S. Matowski, Manitoba Conservation, Winnipeg, MB; personal communication), but it has not been established if these fish originated from natural reproduction or stocking. Lake sturgeon were also stocked in the Assiniboine River above the city of Winnipeg in 1997 (Dick, unpubl. data). Lake sturgeon remained in the vicinity of the stocking site due to cold water that fall, but no lake sturgeon were caught in the Assiniboine River, from the mouth to a few km upstream of the stocking site, by boat electrofishing in June 1998. Only one of these tagged lake sturgeon was recovered, at the mouth of the Red River (Dick, unpubl. data). There is no evidence of an extant, reproducing population in the Assiniboine River.

Local newspapers have reported angler catches of two lake sturgeon from the Red River below the St. Andrew’s Dam at Lockportin the last several years (Lysack, pers. comm.). One of the two fish was taken in the Red River in 1996 and measured 199 cm in length, setting a Manitoba angling record. The second fish, caught sometime later, measured within 5 cm of the other and may have been the same fish as this was a catch-and-release fishery. Given their size, they probably were born prior to the construction of the dam at Lockport. Between 1975 and 2002, only a few lake sturgeon captures were reported from 8.1 million angling hours on the Red River between Lockport and Netley Creek (Lysack 1986; Dick and Lysack, unpubl. data). No sturgeon were collected in the Red River from Lake Winnipeg to the American border during three boat electrofishing surveys in fall 2002, and spring and summer 2003 (Watkinson, pers. comm.).

Since 1997, Minnesota Department of Natural Resources and White Earth Indian Reservation have been releasing lake sturgeon into the U.S. waters of the Red River Drainage (Aadland et al. 2005). To date 375 adults, 3482 yearlings and 18 000 fingerlings (in 2002) have been released as part of a 20-year introduction plan that will see an annual release of 34 000 fingerlings and 600 000 fry (MN DNR 2002, as cited in Aadland et al. 2005). These fish are from Rainy River stock, which according to MacDonnell (pers. comm.) may have interbred with Winnipeg River stocks. This is unlikely at present given the limitation to upstream availability because of the natural and manmade barriers near the mouth of the river, preventing upstream movement (see Designatable Units and Figure 8). Only two lake sturgeon were captured during a study of fish movement at the City of Winnipeg floodway inlet control gates in 2005 (Graveline and MacDonnell 2005) and also is present in the Pigeon, Bloodvein, Poplar and Berens rivers that flow into the east side of Lake Winnipeg and is harvested by Aboriginal communities. However, virtually no population data are available for these rivers. Based on mark-recapture studies, the sturgeon population in Round Lake on the Pigeon River is estimated to have between 800 to 1000 individuals with very few (< 100) spawning females (Dick 2004). Recent telemetry studies have been conducted on the population in the Ontario portion of the Berens River. There is no information on the status of the population, but it is believed to be depressed from commercial fishing that took place in the 1950s (M. Schillemore, Ontario Ministry of Natural Resources, Red Lake, ON, personal communication). It is currently not exploited.

In summary, very large populations of lake sturgeon must have once existed in Lake Winnipeg, but these populations were depleted by commercial exploitation in the first decades of the 1900s. Historic sizes of population components in rivers in the range of this DU are poorly known, but all are known to be depleted. There is no firm evidence of naturally reproducing populations in the major watersheds of the Assiniboine or Red Rivers in recent years, although there have been small angling catches each year. A number of smaller rivers draining into Lake Winnipeg do continue to support lake sturgeon, but the little evidence available suggests that numbers of spawning females are unlikely to exceed a hundred annually in any component. Stocking has been undertaken in the major rivers of this DU since 1996, but survival rates appear to be low, and it would be at least one or two more decades before lake sturgeon stocked in these rivers could establish reproducing populations if they were to survive.

Winnipeg River-English River (DU5)

Large commercial fisheries for lake sturgeon once existed on the Winnipeg River, and in the 1930s catches were in excess of 35 000 kg. By 1947 the annual catch was less than 2000 kg at which time the fishery became economically non-viable. After a hiatus of 9 years the fishery was reopened and the 1956 take was in the order of 12 000 kg with a slight increase in 1957 and then a decline to about 5 000 kg in 1959 at which time the fishery was closed (Harris et al. 2000; Lysack, unpubl. data). Opened again in the 1960s, harvest continued to decline and the fishery was closed in 1970 due to mercury contamination (Harris et al. 2000). Since the 1970s only a few (~ 5 per year) have been taken as bycatch with other species (Harris et al. 2000). Today, they have essentially disappeared from a major historic lake sturgeon spawning and nursery area below the Seven Sisters Dam on the Winnipeg River (H. Letander, Sagkeeng First Nations Elder, personal communication), and from above Scott Rapids where few fish have been reported since the 1990s. Two fish were taken in 2 weeks of fishing above Scott Rapids (Lysack, unpubl. data), and 40 juvenile and sub-adults were taken in additional sampling in the late 1990s (Dick, unpubl data). In June 2000, two overnight gill net sets below Seven Sisters Dam on the Winnipeg River caught a single juvenile sturgeon.

Mark-recapture studies provided a mean estimate of 2352 sturgeon in the Winnipeg River from Seven Sisters to Point de Bois in the late 1990s (Guimon and Coucherne vs. The Queen 2001; Lysack unpubl. data), with catch per unit effort showing a downward trend (Figure 16). The mark-recapture study indicated that 72% of the fish caught and released were below the mean age of maturity, which was taken to be the same as that for fish in the Nelson River, i.e., 27 years (Lysack unpubl. data). This would produce an estimate of 660 mature fish, although according to lake sturgeon life history, only a portion of these would spawn annually. Moreover, the estimate of 2352 may be an overestimate since mark-recapture results were biased in some years (particularly the last 2 years of the study) by the removal of tagged and untagged fish by poachers. A third index demonstrating decline in this population is the CPUE determined for the Nutimik-Numano stock (Figure 16) that declined by 54.9% between 1983 and 1999.


Figure 16: Temporal Changes in Catch Per Unit Effort (CPUE) of the Nutimik-Numao Stock

Figure 16: Temporal changes in Catch Per Unit Effort (CPUE) of the Nutimik-Numao stock.

The upper line indicates CPUE from 1983-1988 and the lower line indicates CPUE from 1989-2004. CPUE changes provide a method of indexing temporal changes in stock size. The Nutimik-Numao stock is decreasing over time. From 1983 to 1988, 9 inch and 12 inch nets were set. A 5.5 inch mesh net was added to the 9 inch and 12 inch mesh nets in 1989. Since 1989, CPUE has decreased by 7.84 fish/net/night. This is a 54.9% decrease.

Lake sturgeon are not known to currently occur in the Black Sturgeon River (Harris et al. 2000).

Little is known about lake sturgeon in the English River, which like the Winnipeg River is fragmented by falls, rapids and dams. They have not been recorded from the upper reaches of the river by the Ontario Ministry of Natural Resources or by the local Native people (Lesley Barnes, Ontario Ministry of Natural Resources, Red Lake, Ontario; personal communication). Apparently they are taken occasionally in lower sections by Aboriginal people, and possibly anglers and poachers, but are quite rare there (Bruce Ranta, Ontario Ministry of Natural Resources, Kenora, Ontario; personal communication). Index netting in 1998 and 1999 in the portion of the river between Ball and Separation lakes failed to catch any lake sturgeon (McAughey, personal communication).

In summary, populations were large enough to produce commercial catches of over 35 000 kg in the 1930s, but these catches were unsustainable and the populations were unable to support catches of less than a third that size by the 1950s. Although commercial fisheries have been closed since the 1970s, lake sturgeon are uncommon to rare at sites that historically were major spawning beds, and lake sturgeon rarely are taken in angling and subsistence catches. A single mark-recapture estimate is available for a major portion of the population in the DU, and it suggests it is unlikely that the mature population exceeds 660 fish, of which less than half are female, and only a portion of these would spawn each year.

Lake of the Woods-Rainy River (DU6)

In the late 1800s, Lake of the Woods was known as the “greatest sturgeon pond in the world”; however, the annual commercial catch had fallen from an annual average of 225 000 kg to 1 000 kg by 1915 (Macins 1972:1). Some recovery of the Rainy River/Lake of the Woods population has been sustained since 1970, largely in response to water quality improvement. (Mosindy and Rusak 1991). The one remaining commercial licence held by a First Nation has remained closed by voluntary moratorium, following a government buyout of non-native licences in 1995. Current harvest by recreational and subsistence fisheries is estimated to be 5500 kg/year (T. Mosindy, Ontario Ministry of Natural Resources, Kenora ON; personal communication). Size of the population, comprised of adult and sub-adult fish >1 m in length, was estimated at 15 000 fish in 1990 (Mosindy and Rusak 1991) and over 50 000 fish in 2004 (Stewig 2005; Mosindy, pers comm.). At an average weight of 5 kg for fish in the 1-m length range, the harvest represents approximately 1100 fish or ~ 2% fishing mortality, which appears to be sustainable.

On Rainy Lake, research and stock assessment data are only available for the South Arm (a shared stock with Minnesota). The population is thought to be recovering (D. McLeod, OMNR, Fort Frances, ON, personal communication). Abundance may be increasing, but a full age structure has not been re-established (although estimates of age composition may be affected by the sampling gear), and recruitment is still variable. This population mixes with the lower Seine River population, and discreteness of the spawning components is unknown. The other two lake basins (North Arm and Redgut Bay) have not been assessed.

In the Seine River the population below Sturgeon Falls (Crilly) Dam is probably distinct from the population above the dam due to a migration barrier. Limited stock assessment on the lower Seine River in 1993-95 suggested the population was depressed, with low abundance, low mean age, and few old fish (McLeod, 1999). This population is subject to subsistence harvest, but the exploitation rate is unknown. The population in the upper Seine River has not been assessed, but is likely also low in abundance.

There has been limited assessment of the lake sturgeon population component in the Namakan Reservoir (including Namakan Lake, Sand Point Lake, Little Vermilion Lake, Crane and Kabetogama lakes in Minnesota, and Loon River), although an active commercial fishery existed until 2001 (Mcleod, personal communication). Adult fish are captured annually in fall walleye index netting. Commercial fishery observations indicate a healthier stock than Rainy Lake (good abundance and age structure, good recruitment of younger fish). The limited assessment effort has supported a conclusion that this population component is likely increasing, and still recovering. Lake sturgeon in the Namakan River (including Little Eva, Bill and Three Mile lakes) may be genetically similar to those in the Namakan Reservoir. Intermixing is likely occurring, but there may be discrete spawning stocks further upstream. Population status is not known, but limited sampling with nets conducted in the lower portion in fall 2005 found high catch rates of adults with no juveniles captured.

Lake sturgeon populations in Little Turtle Lake (including Big and Little Turtle rivers), Lac La Croix and Loon Lake have not been assessed and their status is unknown. In Quetico Provincial Park, lake sturgeon populations occur in Sturgeon Lake, Russell Lake, Tanner Lake and the Maligne River. The status of these populations is unknown, but lake sturgeon are regularly observed in these locations (B. Jackson, OMNR, Atikokan, ON personal communication).

In summary, lake sturgeon in this DU were reduced by commercial fishing in the 1800s, but there has been substantial rebuilding of many stock components, particularly in recent decades. A total population estimate is not available, but estimates of the Rainy River/Lake component of the Lake of the Woods component exceeds 50 000 lake sturgeon > 1 m, and exploitation rates appear sustainable. Some population components still have an age composition skewed towards younger fish, but in most components where estimates are available, the proportion of mature fish is either stable or increasing.

Southern Hudson Bay-James Bay (DU7)

This ecozone covers portions of northeastern Manitoba, northern Ontario and northern Quebec. Lake sturgeon occur at multiple locations in most of the major river systems in this area. Relatively unexploited sturgeon populations of unknown size occur in the Gods and Hayes rivers of northeastern Manitoba (Matowski, pers. comm.). The Hayes and Fox rivers had landings of 2800 kg in both 1939 and 1941 (Lysack, pers. comm.) Currently, the lake sturgeon population size and harvest, including the Aboriginal fishery, for the Gods/Hayes system is unknown, although 26 lake sturgeon were reported at the mouth of the Hayes River in 2005 (MacDonnell unpubl. data). A gillnetting study conducted on a 20-km reach of the Fox River (a tributary of the Hayes) between Great Falls and Rainbow Falls in 2004 provided an estimate of 625 (± 375) adult lake sturgeon (MacDonnell, unpubl. data).

Data from lake sturgeon populations for rivers in northern Ontario indicate that several have been impacted by hydroelectric dams (Seyler et al. 1996, 1997a, b, c; S. McGovern, Ontario Ministry of Natural Resources, Timmins, ON; personal communication). Estimates of lake sturgeon numbers in the Little Long Generation Station head pond on the Mattagami and the Goundhog and Kapuskasing rivers varied from 8 000 to 25 000 fish all ages in the late 1980s (Nowak and Jessop 1987; Sheenan and McKinley 1992). The Groundhog River sturgeon population, assessed in a study of adult sturgeon habitat by Seyler (1997a), appears to be quite healthy as there are good spawning runs and recruitment, and a broad age-class distribution. In the Moose River basin, the Kenogami River has populations consisting of predominantly younger-aged fish with a small proportion over 30 years of age (Ecologistics 1988). The Kwataboahegan, Cheepash, North French, Missinaibi and Onakawana rivers have relatively unimpacted, healthy populations (McGovern, pers. comm.). Lake sturgeon exist throughout the Abitibi River, but the population is substantially impacted in this system. Spawning habitat is marginal in areas impacted by dams, but some of the best lake sturgeon habitat in the system remains in the area downstream of Otter Rapids (McGovern, pers. comm), and the population in the Fredrickhouse River of the Abitibi drainage appears to be healthy (M. Gauthier, OMNR Cochrane, ON, personal communication).

Hydro-Québec (2001) reports that lake sturgeon are distributed widely in northern Quebec where they are present and widespread in the James Bay watershed and the basins of the Harricana, Nottaway, Broadback, Opincaca, Rupert, Eastmain and La Grande rivers; however, no population data are available. Ferguson and Duckworth (1997) also indicate that lake sturgeon are common in the La Grande and Rupert rivers, and in the area of Lake Abitibi, and absent in most of the other smaller rivers and lakes of the southern Hudson Bay/James Bay drainage. However, they report that lake sturgeon is rare in other rivers draining into James Bay and southern Hudson Bay, such as the Harricana, Nottaway and Eastmainrivers. As elsewhere, new fisheries often have been initially successful, but rarely sustainable. Commercial fisheries have been sporadically conducted in various locations on an annual quota basis based on a 0.08 kg/ha of total waterbody area (Fortin et al. 1992; Ferguson and Duckworth 1997). In a major part of the lake sturgeon distribution area in Northern Quebec, according to the James Bay and Northern Quebec Agreement (INAC 1993), lake sturgeon fishing since 1973 has been limited to Native people. South of this territory, recommendations have been made to close most of the fisheries in the James Bay watershed. Commercial fishing is now allowed in only three sections of the Upper Nottaway system (the Bell, Megiscane Est and Megiscane Ouest networks) where quotas of 0.1 kg/ha are thought to be sustainable based on available habitat, which is less than the total waterbody area (Fortin et al. 1992, Ministère des Resources naturelles et de la Faune, 2005).

Microsatellite DNA characterization studies of lake sturgeon from the Rupert and Eastmain-Opinaca watersheds (Bernatchez and Saint-Laurent 2004) indicate that lake sturgeon from the Rupert River are distinct from those of the Eastmain-Opinaca system, but no tests were made with lake strugeon from any other waterbodies. Lake sturgeon from the Rupert River are structured into three genetically distinct populations, but there is no evidence for structuring within the Eastmain-Opinaca population. Hydro-Québec is in the process of further development of hydroelectric power generation facilities, in an area where lake sturgeon populations have already been fragmented by the construction of dams and impoundments for large-scale production of hydroelectric power (see, for example,Hydro-Québec 2004a). The proposed development would result in a partial diversion of the Rupert River, essentially joining the Rupert and Eastmain watersheds (Hydro-Québec 2004 a, b). The Eastman River has already been connected to the La Grande River in the 1970s for hydroelectric production. All these modifications may have implications vis à vis future viability of either or both populations. Bernatchez and Saint-Laurent (2004) indicate that, in Quebec, lake sturgeon are likely a threatened or vulnerable species.

In summary, lake sturgeon population components in a number of watersheds through northeastern Manitoba, northern Ontario, and northwestern Quebec appear healthy, but generally are poorly quantified, and in a few watersheds, different reports of population status are inconsistent. Some components in Quebec and Ontario have been fragmented by hydroelectric dams, whereas hydroelectric developments have created the potential for connection between other watersheds supporting lake sturgeon. Many northern drainages have not been subjected to commercial fisheries, and where commercial fisheries have occurred, all but three small fisheries have been closed due to unsustainable harvesting in the past.

Great Lakes-Upper St. Lawrence (DU8)

Canadian waters of theGreat Lakeswere opened to commercial sturgeon fishing in 1879 (Prince 1905). Within a short time, populations of lake sturgeon throughout the Great Lakes, and in the surrounding waterbodies in Canada and the US, declined to less than 1% of their former strength (Hay-Chmielewski and Whelan 1997). Commercial catches for the Great Lakes are provided in Tables 4-6. Currently, lake sturgeon populations are extant at 63 sites in the Great Lakes and St. Lawrence basin; however, successful spawning is documented from only 20 sites (and is unknown for the rest). Spawning run size estimates are available for only 17 of these sites, and populations at only four of these sites are considered to have a spawning run greater than 500 adults (Holey et al. 2000). As many as 20 spawning populations in the Great Lakes may have been lost, but other populations are now showing signs of modest recovery (A. Dextrase, Ontario Ministry of Natural Resources, Peterborough, ON; personal communication).

Eight tributaries to Lake Superior in Ontarioare documented to have extant, self-sustaining populations of lake sturgeon, but the species is extirpated from six other tributaries (Holey et al. 2000). In spring 1998, it was estimated that the Kaministiquia River had 140 fish (15 adult fish/km of river). Population densities were estimated to be 1.2 fish/ha in the lower reaches and 5.0 fish/ha in the upper reaches (Stephenson, pers. comm.) A fall 2001 estimate for the Kaministiquia River was 188 adult fish for the 12 km of river netted during the survey (Friday and Chase 2006). In the spring of 2003 and 2004 it was estimated that the Black Sturgeon River had 96 (2003) and 103 (2004) adult spawning fish (Friday 2006). Spring assessment netting on the Goulais River was also conducted from 2000 to 2004. Low numbers were caught each year: 2000 (n=9), 2001 (n=11), 2003 (n=9) and 2004 (n=20) [Chase 2006]. During the spring of 2002 a survey of the Big Pic River and its tributaries was conducted. Very few fish were captured: Big Pic River (n=13), Kagiano River (n=3) and Black River (3) [H. Quinlan, USFWS, Ashland, WI; personal communication]. Kelso and Cullis (1996) reported that lake sturgeon are extirpated from Nipigon Bay of Lake Superior. Commercial harvest is prohibited in the waters of Lake Superior and its tributaries, but recreational and subsistence harvesting are still permitted.

Lake Nipigonwas first opened to commercial fishing in 1917 and lake sturgeon were quickly over-fished. The peak harvest of 42 273 kg was reported in 1924, but the fishery collapsed by 1930 and has not recovered. Although the commercial fishery no longer targets lake sturgeon, a 770-kg commercial quota was established in the early 1980s to cover incidental catches of lake sturgeon associated with the lake whitefish, lake trout and walleye gillnet fishery. The reported commercial harvest of lake sturgeon has averaged less than 45 kg annually since the quota was set. Recreational and subsistence harvest of lake sturgeon is assumed to be minimal; however, no harvest statistics are available for Lake Nipigon. There have been no programs initiated to assess lake sturgeon populations in Lake Nipigon since the work conducted by Harkness in the 1920s (Harkness and Dymond 1961). Ontario Ministry of Natural Resources staff with the Lake Nipigon Fisheries Assessment Unit have occasionally captured and released juvenile lake sturgeon while conducting periodic Fall Walleye Index Netting projects since 1996 (R. Salmon. OMNR, Nipigon, ON; personal communication).

In the Canadian portion of Lake Huron, lake sturgeon were known from 21 sites, and still are extant at eight of these sites (Holey et al. 2000). All of the extant populations are considered small, with the exception of the St. Clair River/Southern Lake Huron population. Successful reproduction is known at four sites, and unknown for the rest, although spawning is suspected to occur in several other tributaries of Lake Huron. The size of spawning runs is unknown except for the Mississagi River, where an estimate of the spawning run is in the order of 150 fish (Holey et al. 2000; Mohr, pers. comm). A commercial harvest is still permitted in the Ontario waters of Lake Huron. Currently there are 41 licensees with a quota of 12 000 kg, but the annual harvest has averaged 2 227 kg, approximately 325 fish. Recreational and subsistence fishing is also permitted, but no landing statistics are available.

Recently, Lake Huron populations have been assessed annually through a cooperative program with commercial fishers (Mohr 1996, 1997, 1998/99, 2000). A total of 648 fish were sampled in 2000, the largest number observed since the program started in 1995. Mohr (1998/99) reported that some populations of lake sturgeon in the Lake Huron drainage are localized, while others move in and out of an area. Thirty-nine tagged fish were recaptured in Lake Huron in 2000 and, of these, five fish had been tagged in Lake St. Clair, 10 in Saginaw Bay, Michigan and the remainder in the Ontario waters of Lake Huron. Fish tagged in the Ontario waters of Lake Huron also were caught in the Detroit River, St. Clair River and Saginaw Bay. A number of age-two fish were observed in 2000 in the North Channel and southern Lake Huron, and age-three fish were common in southern Georgian Bay, indicating recruitment in all basins of Lake Huron (Mohr 2000). The age-class distribution and average size is similar through the late 1990s with the largest fish recorded in the Mississagi River spawning run (Mohr 2000). Although 60% of the fish sampled were of legal harvestable size (1.09 m), only 28% were harvested (Mohr 2000). McMurtry et al. (1997) reported that lake sturgeon are extirpated in Lake Simcoe, in the Lake Huron basin.

Historically lake sturgeon in the St. Clair system were abundant; however, over-exploitation and habitat loss impacted the populations. The Michigan Department of Natural Resources (unpublished data 2005) estimated that lake surgeon in the St. Clair basin (including US waters) may be in the order of between 15 000 and 25 000 fish (all ages). However, Holey et al. (2000) report that the St. Clair and Detroit rivers have a small, extant population. Recent tag recapture data provide an estimate for the Lake St. Clair component of approximately 5000 individuals of all ages (B. Locke, Ontario Ministry of Natural Resources, Peterborough, unpublished data). Some of these may be migrants from the Detroit and/or St. Clair rivers or Lake Erie.

Commercial and recreational fisheries exist in the Ontario waters of Lake St. Clair. A substantial commercial sturgeon fishery existed on Lake St. Clair in the late 1800s but since 1920 only a small commercial fishery has continued. The annual quota of 1542 kg has not changed since 1986. Currently this fishery consists of two licences, one of which has been inactive recently, that annually take on average 772 kg of lake sturgeon using hook and line gear. Recent (1996) creel surveys for the Lake St. Clair basin indicated that the recreational harvest is between 30-50 fish/year in Ontario, primarily taken during spawning in the St. Clair River (Locke, unpubl. data). Ongoing aging studies from samples taken in the commercial fishery show that the age structure of the population has remained stable for the past 40 years (Locke, unpubl. data). This would indicate that there has been stable recruitment and mortality for at least the same period of time.

Historically, there were 4 population components known from Lake Erie, and three of the four are currently extant. The three lake sturgeon population components appear to be increasing in Lake Erie (L. Cargnelli, Ontario Ministry of Natural Resources, London, ON; personal communication), although no population contains more than 1000 individuals. The Lake Erie annual index-netting program catches some lake sturgeon, as does the commercial gill net fishery in the western basin; however, the origin of the individuals captured in the western basin may be Lake Huron or the St. Clair River system. Recent, increasingly frequent observations of juvenile sturgeon in Lake Erie (Locke, pers. comm.) indicate that there could be increased recruitment, or a strong year class present in the lake.

Historically, lake sturgeon were abundant in Lake Ontario, and provided a peak commercial harvest of over 225 000 kg in 1890. However, by 1900 the population component had declined such that the commercial catch was insignificant, and this fishery is often cited as a classic example of over-fishing (Christie 1973). Commercial harvest of lake sturgeon was banned in New York State in 1976, and in Ontario in 1978. Currently no commercial or recreational harvest is permitted in Lake Ontario waters. Although the role of habitat degradation in the decline is unclear, the damming and degradation of tributary spawning areas may have been a significant factor in the continued decline of these population components. In 1983, the species was listed as “threatened” in New York State (Carlson 1995).

Lake sturgeon populations in the Ontario waters of Lake Ontario appear to be showing some recovery (A. Mathers, Ontario Ministry of Natural Resources, Picton, ON; personal communication), although no population contains more than 1000 individuals. Prior to 1996, only two lake sturgeon were observed in the long-term surveys in eastern Lake Ontario and the Bay of Quinte. Since 1996, a total of 24 sub-adult lake sturgeon have been caught in survey gear. A few lake sturgeon were caught in 1997 in Lake Ontario index nettings, and commercial fishers report sturgeon in their nets as bycatch. Spawning fish have been noted in the Trent River below Dam 1, and in the Salmon River (Mathers, pers. comm.). Recently lake sturgeon have been documented to be reproducing at two of the five known historic spawning sites in Lake Ontario. Small populations of lake sturgeon are still found at two other historic sites but spawning has not been documented (Quinlan 2005), and status of the Amherst Island shoal population component is unknown (Holey et al. 2000).

Commercial fishermen have also reported the incidental capture of up to fifty juveniles annually in eastern Lake Ontario since 1996. This increase in lake sturgeon bycatch may be attributable to either increased natural reproduction within Lake Ontario and its tributaries, or movements of juvenile sturgeon stocked into waters adjacent to the eastern basin. Observations of adult lake sturgeon during the spawning period in the Black River (New York) and in the Trent River (Ontario) during the mid-1990s suggest that increased natural reproduction may be occurring. In studies of lake sturgeon on the lower Niagara River in 1998-1999, 33 sturgeon were collected, ranging from 0.3 to 1.3 m total length, indicating continuing recruitment to the population. In addition, increased incidental catches of lake sturgeon by recreational anglers and sightings by recreational SCUBA divers suggest that the lake sturgeon population in the lower Niagara River may be increasing. In the short-term it seems likely that lake sturgeon numbers will remain at low levels in Lake Ontario, but the increased catches of sub-adult sturgeon since 1996 are encouraging.

Lake sturgeon populations in the OttawaRiver basinonce were abundant (Small 1883; Dymond 1939) but have declined due to anthropogenic stresses (Toner 1943; Haxton 2002, 2006). In 1879, the commercial lake sturgeon fishery harvested fish from the Madawaska and Bonnechère rivers, and from lakes in Lanark and Renfrew counties, all of which are tributaries to the Ottawa River (Harkness and Dymond 1961). Today, lake sturgeon do not exist in these waterbodies; the chances of re-establishment are remote due to fragmentation and barriers (Seyler 1997a, b). Commercial harvest on the Ottawa River peaked at 28,800 kg in 1898 and never again attained similar levels (T. Haxton, OMNR, Kemptville, ON, unpubl. data). Currently, commercial harvesters are licensed for the Quebec portion of the Ottawa River only, from Quyon to Fort William. Quotas have been established at 0.1 kg/ha/yr of Quebec waters since the early 1990s (OMNR and Gouvernement du Québec Faune et Parcs 1999), and an average of 1250 kg of lake sturgeon was harvested from the Ottawa River from 2000 to 2005 (Haxton, unpubl. data). Despite strict harvest controls, there is little evidence of recovery throughout the Ottawa River watershed. Abundance varies substantially among river reaches (Haxton 2002). A study on the contribution of waterpower management, commercial harvest and contaminants to lake sturgeon population dynamics in the Ottawa River suggests that waterpower management has the greatest influence (Haxton, unpubl. data) although the results are not definitive.

Spawning has been documented in all Ottawa River reaches, but recruitment appears to be a limiting factor in several reaches (Haxton, unpublished data). In 1949, over 400 lake sturgeon were sampled on the spawning shallows below the Fitzroy Dam on the Ottawa River; only eight lake sturgeon were found in a repeat survey in 2001 (Haxton 2002), 58 were caught in 2003 (including 7 recaptures from 2001), and 17 in 2004 (including 3 recaptures from 2003). Based on the small number of recoveries Haxton (2006) estimated a spawning population of 202 (93-378; 95% C.I.). However because only males were observed in spawning condition and very few juveniles were found in this study (Haxton 2006), especially in comparison to the historical study and concurrent studies on other spawning shoals in the Ottawa River, the productivity of this component may be very low. Only limited recruitment appears to be occurring, and many of the fish observed may have been born before 1932, when the dam was completed. The influence of downstream migration or drift, and barriers impeding upstream migration are unknown for this population component.

There is a small commercial fishery on the Quebecportion of the Ottawa River between Quyon and Fort William (Haxton 2002). There are also commercial fisheries in Lake Timiskaming (286 fish/year) where the population is considered stable (Fortin et al. 1992; Nadeau, pers. comm.). There was a spring fishery of 25 to 30 fish for caviar in the Timiskaming-Abitibi district for 50 years, but this was discontinued in 2000 and fishing for lake sturgeon is restricted to June 15 through July 16, and to September 15 through October 31 (Nadeau, pers. comm.). There also is some subsistence fishing for lake sturgeon in the Timiskaming-Abitibi area.

In the St. Lawrence River, all 14 known Canadian population components are extant; 12 are considered small, two are large (Holey et al. 2000). Spawning is known to occur at six sites, probably not occurring at two sites, and unknown at six sites (Holey et al. 2000). Of the known spawning populations, four have spawning runs of <1000 individuals, the spawning runs in the St. Maurice population has ~1250 individuals, and the Des Prairies population has ~7000 individuals (Holey et al. 2000; Fortin et al. 2002). The growth characteristics of the upper St. Lawrence sturgeon populations (collected in 1993 and 1994) below the Robert Moses Dam near Massena, New York were examined by Johnson et al. (1998) and found to be similar to growth of lake sturgeon 25 years ago.

Catches in the St. Lawrence River remained fairly consistent from 1920 to 1984 (Figure 15; Dumont et al. 2000, 2006), a phenomenon not observed in fisheries elsewhere. This may be related to a number of factors. The St. Lawrence River sturgeon fishery started much later, and the major fishery is located on a large river/lake system (which makes it more difficult to remove all spawning fish as spawning occurs at several sites along the river). The annual harvest is also much lower than historically found elsewhere (e.g. Lake Erie, Lake of the Woods; Tables 5, 6). In addition, the fishery focused on sub-adult and smaller adults, rather than only on large individuals (Fortin et al. 1992). Dumont et al. (1987) also attributed the ability of this stock component to sustain harvest for 80 years without a collapse to the relatively high productivity of the system, the restriction of intensive commercial fishing to specific zones, leaving some sectors unexploited to act as reservoirs, and the high selectivity of the historically used commercial gill nets (19 to 20 cm stretched mesh).

However, by 1987 the St. Lawrence population component was considered overexploited. Dumont et al. (1987), and La Haye et al. (1992) reported trends in age structure, year-class strength (based on the age distribution of the 2- to 8-years-old segment), and abundance of mature females (over age 25), all indicating that overexploitation probably began in the mid-1970s, and was continuing. In addition the reported commercial catch remained very high after 1986 (mean 202 000 kg, range 152 000-259 000 kg), greatly surpassing the landings reported before 1983 (maximum ca 65 000 kg, but reporting was considered incomplete). In 1987, the St. Lawrence population component was considered overexploited due to high annual natural and fishing mortality rates of the exploited segment (ages 15 to 30), unbalanced age structure, low reproductive potential and excessive annual harvest (Dumont et al. 1987). The utilized spawning grounds were found to be reduced to primarily the upstream part of the system. Many previously used sites were no longer accessible or utilized because of various human interventions (Dumont et al. 1987; La Haye et al. 1992).

Between 1987 and 1991 a new management plan was gradually implemented to reduce the catch, provide more protection to spawners and strengthen enforcement (Dumont et al. 2000b, 2006). The fishing season and the number of fishing licences were reduced, long-lines and seines were banned, gillnet stretched mesh was restricted to 20 cm, and recreational fishing regulations were tightened. In the 1990s, research was undertaken to increase the knowledge of the characteristics of spawning grounds and juvenile habitats, and to develop an index of year-class strength in order to adjust harvests to expected changes to the age 15- to 25-year-old portion of the population component, which constitutes the majority of the harvest (Fortin et al. 1993; Nilo et al. 1997; Fortin et al. 2002).

The commercial harvest was sampled again in four fishing sectors in 1994 and further data were collected in 1998 (see Dumont et al. 2000b). The 1994 samples indicated that, from the early 1980s to the mid-1990s, length-at-age remained about the same, but the 1994 weights at given length were 7 to 15% lower. The observed decrease of somatic condition may have been related to changes in the trophic condition of the St. Lawrence River, possibly causing a decrease in the potential production of the lake sturgeon population component. Catch curves based on commercial catches showed that annual mortality rates were high throughout this period and the apparent age of recruitment to the fishery shifted substantially to older fish in Lac Saint-Louis (from age 16 to 23) and Lac Saint-Pierre (from age 14 to 20). Between 1994 and 1998, CPUE from the 20-cm gillnet fishery declined 7% in Lac Saint-Louis and 25% in Lac Saint-Pierre. The annual rate of decrease was similar (1.75%) but only the second decrease was statistically significant. Juvenile surveys indicated continuous production of cohorts between 1980 and 2001, but the index of year-class strengths decreased by 58% from 1984 to 1992 (rs = - 0.7; P = 0.02) (Dumont et al. 2000b). Two comparatively strong year classes were produced in 1993 and 1994, but subsequent ones were average and in one case weak. From 1995 to 1999, mark-recapture estimates of the number of mature females on the Des Prairies River spawning ground declined by 61%, from 1231 to 500 fish (Fortin et al. 2002). At the same time exports of lake sturgeon declined by 36%, and there also was a notable decrease (10%) in the weight of lake sturgeon of comparable size from the 1980s to 1994.

Taken together the evidence led to the conclusion that the 1987 management plan had failed to reverse the decline in stock status (Dumont et al. 2000b). A 200-tonne commercial catch quota was implemented in 1999, coupled with the obligation to tag and register each sturgeon carcass. The quota was decreased by 20% in 2000, 2001 and 2002 and has been maintained at 80 tonnes (which corresponds to about 11 000 sturgeon) thereafter. Since 2003, the fishing season has been shortened by two months (June 14 to July 31 and September 14 to October 15) [Ministère des Ressources naturelles et de la Faune du Québec 2005].

Concurrently, habitat conservation and improvement measures were implemented. The quality and area of two spawning grounds were successfully increased in the Des Prairies and Saint-Maurice rivers (Fortin et al. 2002; GDG Conseil 2001). In the Des Prairies River, the most important spawning ground of the system, egg-to-larvae survival has increased from an average of 0.8% before the improvement (1995-1996) to 5.4, 3.7 and 2.4% in 1997, 1998 and 1999, respectively (Fortin et al. 2002). Larval production was also increased from 3.9 million (1.2 to 8.6 million) between 1994 and 1996 to 7 million (2 to 12.8 million) between 1997 and 2003 (Fortin et al. 2002; Garceau and Bilodeau 2004). Three other spawning grounds were artificially created: in 1999 downstream of Beauharnois Generating Station, in 2001 in the Saint-François River, and in the fall of 2005 in the lower part of the Chaudière River (Trencia and Collin 2006). The first trial has not been successful (Environnement Illimité 2002), but the spawning bed created in the Saint-François River has been utilized for reproduction (Faucher and Abbott 2001; Alliance Environnement et al. 2002). Monitoring of the third trial has been planned for spring 2007; however, during the spring of 2006, following the completion of an industrial waste water treatment plan, many lake sturgeon were observed or caught in the Chaudière River for the first time since the 1950s (Guy Trencia, Regional Biologist (Chaudière-Appalaches Ministère des Ressources naturelles et de la Faune du Québec, Charny, QC; personal communication).

Dredging projects along the navigation channels and in harbour works were also analyzed and modified in order to protect juvenile habitats. Improvements in water flow management, erosion protection, and construction of sewage treatment facilities in the l’Assomption watershed have also improved two lake sturgeon spawning beds, resulting in a 10-fold increase in larval production. The project also raised enthusiasm among authorities and organizations where the species has become a regional symbol. The watershed management corporation has recognized the spawning areas as key sites for biodiversity and developed plans for their conservation and restoration (Dumas et al. 2003).

In the 1980s, new operating and discharge criteria were tested and applied to ensure access to a spawning ground used before the construction in 1960 of the Pointe-des-Cascades Dam at the confluence of the St. Lawrence River in Lac Saint-Louis. A new fishway was opened in 2001 (Paradis and Malo 2003), at the Saint-Ours dam, on the Richelieu River, but only a few large lake sturgeon have used it each year (Fleury and Desrochers 2004).

A multidisciplinary study was initiated in 2001 in order to quantify the impacts of water discharge variation on the St. Lawrence River fish community. A two-dimensional spatially explicit fish habitat model covered a large part of the lake sturgeon distribution area in the Quebec section of the St. Lawrence River. Surfaces of suitable lake sturgeon habitat increased with discharge. Recommendations concerning fish habitat protection in the fluvial St. Lawrence were presented to the International Joint Commission under the in-process evaluation of the regulation criteria of the Lake Ontario – St. Lawrence River system (Mingelbier et al. 2004, 2005ab).

In 2006, a hydroelectric project in the Courant Sainte-Marie, Montréal, was withdrawn in order to protect one of the rare remaining rapid sectors of the St. Lawrence River and to maintain free movements of migratory species (lake sturgeon, American shad, Alosa sapdissima, American eel, Anguilla rostrata, copper redhorse, Moxostoma hubbsi, etc.) [Dumont et al. 2005]. In the mid-1980s a similar project was rejected for similar environmental concerns at another site, the Lachine Rapids, located about 15 km upstream.

In summary, a variety of trends are seen in lake sturgeon population components in this large DU. Throughout the Great Lakes, lake sturgeon abundances are certainly much lower than they were historically, but self-sustaining population units are present in all the Great Lakes and many tributaries. The major declines were usually in late 1800 or early 1900s, following opening of commercial fisheries, which tended to follow patterns of “boom–and–bust”. Most commercial fisheries in the Great Lakes have been closed or greatly reduced for many decades, and usually recreational catches are tightly restricted as well. Many spawning components remain in the Great Lakes and their tributaries, and some monitoring programs suggest that abundances are increasing. However, age composition of essentially all population components tend to lack older fish, numbers of spawners are often low each year, and many traditional spawning sites are not used.

In the St. Lawrence River and its tributaries, commercial fisheries collapsed in only a few areas, such as the Ottawa River drainage basins. Commercial harvests showed no overall trend in much of the Quebec portion of the St. Lawrence basin, but assessments in the 1980s and again in 1990 both found the population components supporting the fisheries to be harvested unsustainably. Harvests have been reduced from over 200 000 kg to quotas of 80 000 kg. Enforcement has been improved, and there are preliminary indications that these catches may be sustainable. There have also been many initiatives to improve habitat quality, particularly in spawning and nursery areas, and to improve water flow regimes. There are indications that recruitment is improving in some portions of the range of this DU in Quebec, as a result of these habitat initiatives.


Population Trends in Native and Subsistence Fisheries

The above statistics do not include any subsistence fishing data that ranges from 50 to at least 200 fish/year in some communities in westernCanada. Historically, lake sturgeon were an important aspect of the diet where available, particularly in the Battleford and Cumberland House areas (Smith 2003). Sturgeon were at one time abundant in the lower Saskatchewan River (see Population sizes and trends - DU2 above), especially from the area of Cumberland House to Lake Winnipeg at Grand Rapids (Kew 1962). Since the time of earliest settlement in the area, fish for local use as food for humans and dogs has been a way of life for the Native peoples in the area of Cumberland, Suggi, Namew, and Windy lakes (Kew 1962).

The Native fishermen (both First Nations and Metis) of the area formed the Cumberland House Fishermen’s Cooperative in 1950 (incorporated 1951) and have undertaken management of the commercial fishery, which had existed since at least the 1920s (Skaptason 1926). A quota of 4 000 kg of sturgeon was established, but never met (Kew 1962) despite the fact that in the 1920s, that quota had been 30 000 kg. Commercial harvests in the Cumberland Lake area had peaked at 19 000 kg in 1935 with a subsequent decline to little more than 2 000 kg in 1939, followed by a recovery to about 8 000 kg in the early 1950s (Wallace 1991). Wallace (1991) estimated that a quota of 3200 kg was sustainable for the fishery prior to construction of the E.B. Campbell dam. Construction of the dam led to considerable loss and disruption of critical habitat and has resulted in a population decline in excess of 80% as well as decreased reproductive success in the remaining fish (see DU2 above). Although the Cooperative still holds an annual quota of 3200 kg, Native fishermen have voluntarily ceased commercial operations as a conservation measure (Wallace 1991; pers. comm.). A limited number of fish may still be caught for subsistence and cultural use.

According to Houston (1987), subsistence catches amount to approximately 2300 fish/year in northern Quebec [this value was considered to be about the same in 2002 (Nadeau, pers. comm.)]. Although most people contacted (see “Authorities Consulted” section) indicated that there was some subsistence fishing in their administrative regions, the numbers harvested generally are unknown. The subsistence fisheries are an important component of the current situation but there are few records on fish taken in traditional subsistence sturgeon fisheries.

Ojibway subsistence fisheries of the mid-19th century provide the only numerical estimate of the historical Aboriginal lake sturgeon harvest (Holzkamm and McCarthy 1988). During this time, fresh and dried sturgeon meat, as well as isinglass, a gelatinous material procured from the dried inner membrane of the sturgeon swim bladders, became important commercial by-products of the subsistence fisheries that were traded to the Hudson’s Bay Company (HBC) (Holtzkamm and McCarthy 1988). The HBC isinglass trade records have been used to estimate the traditional harvest of lake sturgeon by the Ojibway in the Lac la Pluie district (Holzkamm and McCarthy 1988). Between 1823 and 1885, records estimate an average yield of nearly 141,210 kg of fish/year. The harvest was estimated to be between 74 and 319 fish in 2002 (Bretecher and MacDonnell 2001). Holzkamm and McCarthy (1988) suggest that these records probably are low as the labour-intensive process of isinglass production may have resulted in the frequent discarding of the swim bladders of small catches, and of small fish, and any sturgeon retained by the Ojibway for personal use would not be accounted for in the HBC records (Holzkamm and Wilson 1988). Hopper and Power (1991) discussed a multi-species subsistence fishery, including lake sturgeon, in the Winisk River in northern Ontario. Forty lake sturgeon, averaging 8.7 kg, were harvested from October 1987 to September 1988 for a total of about 350 kg.

Traditional knowledge indicates that lake sturgeon have been a preferred species in the domestic fisheries of the Cree- and Ojibway-speaking people of the Nishnawbe-Aski Nation in northern Ontario for generations (Michalenko et al. 1991). Historically, families would gather in the spring at spawning sites in the Severn River and its tributaries, especially theWindigoRiverto capture spawning fish for food. The annual harvest was estimated to be 250 fish. A non-native commercial fishery established at Muskrat Dam in 1951 fished the population to the extent that the enterprise was no longer commercially viable within five or six years (Michalenko et al. 1991). A second non-native commercial operation managed by the Hudson Bay Company at Bearskin Lake from 1953 to the early 1970s suffered a similar fate. Since 1965, residents of Muskrat Dam have maintained a domestic food fishery for sturgeon, the annual harvest ranging from 200 to 400 fish, depending on need (Michalenko et al. 1991), and in some years very few fish are taken. Local (Native) fishermen report a decline in numbers in the Windigo and Severn rivers, which they attribute to the commercial fisheries of the 1950s, and the community has chosen not to continue commercial fishing, even though they have licences for a quota of 500 kg (Michalenko et al. 1991).

Lake sturgeon at one time were known in the Grand River as far upstream as Brantford, Ontario, and until at the least the 1920s, were taken by the local Mohawks for food (Jamieson 2005). Lake sturgeon are not known in the Grand River today, and there is a now a dam near the mouth of the river (Mandrak, personal observation).

Historical use of lake sturgeon by First Nations people has been well documented in Quebec. For example, in the archaeological site of Pointe-du-Buisson, at the confluence of the St. Lawrence River in Lac Saint-Louis, bones of lake sturgeon, channel catfish and catostomids dominate the remnants of meals identified and associated to the Late Woodland period circa 920-940 A.D. (Courtemanche 2003). Mohawk people of Kahnawake are still practising subsistence fishery in Lachine Rapids, at the outlet of Lac Saint-Louis.

The Cree of northern Quebec have a time honoured association with the land and its resources that is foundational to their culture, social-economic base, way-of-life, and lake sturgeon has always been a highly valued species. The rights of the Native inhabitants in respect of the lands and their resources were recognized in Canada’s first modern land claim settlement, The James Bay Northern Quebec Agreement of 1975 [(JBNQA) see, for example,INAC 1993]. The rights of Native peoples to the exclusive use of the wildlife resources of the area, including sturgeon, are set out in sub-sections of Section 24 of the Agreement.

Guaranteed harvest rights and limits were established and adopted in 1982 based on historic patterns of harvest as determined by a massive Native Harvesting Research (NHR) project carried out between 1972 to 1979 (James Bay and Northern Quebec Native Harvesting Research Committee 1982). The project did not undertake any background or baseline studies vis à vis estimates of population sizes and trends, etc., but did provide an indication of annual harvest levels by community over the period of the study and allowed for input of elders and the harvesters regarding insights on the state of the resource. Lake sturgeon harvests were reported regularly by fishermen in all communities (Table 7) and the results indicate that the average harvest was in the area of 2000 fish per year. Results varied from year to year, and community to community, reflecting local need and availability of fish; the total average for the region was 2311 fish. Age and size were not reported, but the fishing methods and pattern of harvest, as well as comments supplied by the fishermen indicated that usually only larger, mature fish are taken.

 

Table 7: Means of Estimated Annual Harvest of Sturgeon by Communities in Northern Quebec, 1974 to 1979
Northern Quebec CommunitiesMeans
Great Whale0
Fort George615
Paint Hills155
Eastmain94
Rupert House229
Nemaska63
Mistassini366
Wasapini789
Total2311

James Bay and Northern Quebec Native Harvesting Research Committee 1982.

The report (James Bay and Northern Quebec Native Harvesting Research Committee 1982) also indicates that the La Grande River system is the northern limit of the range as stated by Scott and Crossman (1998), and that sturgeon are not equally abundant in the river systems of the area as subsequently confirmed by Fortin et al. (1992) and Ferguson and Duckworth (1997). Despite the massive hydroelectric power development that has already been carried out in the region, and plans for further development, no other surveys of subsistence use have been conducted, nor have there been any studies to establish baseline data on status and abundance that would permit an evaluation of population sizes and trends (Hydro-Québec 2004 a, b). It is thought that subsistence use of sturgeon remains at the levels of the late 1970s and early 80s as reported in the Wealth of the Land report (INAC 1993), and these values have been used as a baseline for subsistence use in the mandated environmental impact assessments carried out by Hydro-Québec in support of hydroelectric power development projects (Environnement Illimité 2003; Hydro-Québec 2004 a, b). However, the historic values (INAC 1993) were reflected as guaranteed harvest levels in the JBNQA and not as quotas. Actual harvest levels in some areas (Rupert, Eastmain and Nottaway rivers) may in fact be in excess of those levels (Environnement Illimité 2003). Since the time of the NHR survey some 10 000 km of access roads have been developed in support of hydroelectric development (Hydro-Québec 2004a,b) and these roads have increased access to sturgeon habitat and facilitated harvests that were not possible at the time of the NHR survey.

Since the 1920s several attempts have been made to establish commercial fisheries in the James Bay territory; the largest of which was initiated with the backing of the Department of Indian Affairs and Northern Development in the late 1960s, and involved lakes Sakami, Evans, Némiscau and Mesgouez and a network of lakes near Matagami. The fishery was terminated in the 1970s when mercury contamination became an issue (A.F. Penn, Science Advisor, Grand Council of the Crees, Montréal, QC; personal communication). Another fishery on the Nottaway and Rupert rivers, initiated in 1989 was terminated in 1994, when, as elsewhere (Lake Timiskaming, Harricana River, and the above fishery), reductions in CPUE, as well as size and age structure, suggested that these fisheries were not sustainable. Although the results of an evaluation of the Nottaway/Rupert fishery were not available for use in the Fortin et al. (1992) report, they recommended that sustainable quotas where such fisheries existed should be based on the yield in kg/ha of suitable habitat, not total waterbody, and should be implemented for commercial and subsistence fisheries, and that no quota should exceed 0.1 kg/ha (Environnement Illimité 2003).